The Neuropathology of Dementia, Second edition

The Neuropathology of Dementia Completely rewritten and updated, this new edition is almost twice the size of its prede...

Author: Margaret M. Esiri | Virginia M. -Y. Lee | John Q. Trojanowski

9 downloads 1806 Views 56MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

The Neuropathology of Dementia

Completely rewritten and updated, this new edition is almost twice the size of its predecessor. Illustrated in colour throughout, and with contributions from the world’s leading authorities, it is the definitive reference on the neuropathology of dementia. It gives practical guidance to pathologists, describes the contribution of neuroimaging to diagnosis, and surveys the clinical features of dementia. New material includes: r Three entirely new chapters on neuroimaging, molecular diagnostics, and transgenic models. r Two chapters on tauopathies under new authorship. r A chapter under new authorship on synucleinopathies, which includes multiple system atrophy. From the reviews of the first edition: ‘This up-to-date and authoritative account will be invaluable for practising neuropathologists and a treasured work of reference for psychiatrists and neuroscientists with an interest in dementia.’ Nigel J. Cairns International Journal of Geriatric Psychiatry ‘It should undoubtedly be on the shelves of consultant histopathologists in a teaching hospital . . . I am delighted to have a copy and have already consulted it with more enthusiasm that I would have believed possible.’ Jennian F. Geddes Journal of the Royal Society of Medicine ‘. . . this is certainly a compulsory purchase for neuroscience centres. It deserves to live on into further editions whose evolution will map the rapid advances being made in many of the conditions covered within.’ P. G. Ince Neuroradiology ‘. . . this publication is a “must have” for every practising neuropathologist.’ Journal of Neuropathology and Experimental Neurology Margaret Esiri is Professor of Neurology in the University of Oxford Department of Clinical Neurology, and Honorary Consultant Neuropathologist at the Oxford Radcliffe NHS Trust. Virginia M.-Y. Lee is Professor of Pathology and Director of the Center for Neurodegenerative Disease Research at the University of Pennsylvania School of Medicine. John Trojanowski is Professor of Pathology, Director of the Alzheimer’s Disease Center and Director of the Institute on Aging at the University of Pennsylvania School of Medicine.

The Neuropathology of Dementia Second Edition

Edited by

Margaret M. Esiri Department of Neuropathology Radcliffe Infirmary, Oxford, UK

Virginia M.-Y. Lee Department of Pathology and Laboratory Medicine University of Pennsylvania School of Medicine Philadelphia, USA

and

John Q. Trojanowski Department of Pathology and Laboratory Medicine University of Pennsylvania School of Medicine Philadelphia, USA

published by the press syndicate of the university of cambridge The Pitt Building, Trumpington Street, Cambridge, United Kingdom cambridge university press The Edinburgh Building, Cambridge CB2 2RU, UK 40 West 20th Street, New York, NY 10011–4211, USA 477 Williamstown Road, Port Melbourne, VIC 3207, Australia Ruiz de Alarcón 13, 28014 Madrid, Spain Dock House, The Waterfront, Cape Town 8001, South Africa http://www.cambridge.org C Cambridge University Press 2004 This book is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2004 Printed in the United Kingdom at the University Press, Cambridge Typefaces Utopia 8.5/12 pt. and Dax

System LATEX 2ε

[TB]

A catalogue record for this book is available from the British Library Library of Congress Cataloguing in Publication data ISBN 0 521 81915 6 hardback The neuropathology of dementia/edited by Margaret M. Esiri, Virginia M.-Y. Lee and John Q. Trojanowski.–2nd ed. p.

cm.

Includes bibliographical references and index. ISBN 0 521 81915 6 (hardback) : alk. paper) 1. Dementia – Pathogenesis.

2. Postmortem changes.

3. Autopsy. I. Esiri, Margaret M.

II. Lee, V. M.-Y. (Virginia M.-Y.) III. Trojanowski, John Q. [DNLM:

1. Dementia – aetiology.

2. Dementia – pathology.

3. Dementia – complications.

WM 220 N4918 2004] RC521.N475 2004 616.8’3 – dc21

2003051547

Every effort has been made in preparing this book to provide accurate and up-to-date information which is in accord with accepted standards and practice at the time of publication. Nevertheless, the authors, editors and publisher can make no warranties that the information contained herein is totally free from error, not least because clinical standards are constantly changing through research and regulation. The authors, editors and publisher therefore disclaim all liability for direct or consequential damages resulting from the use of material contained in this book. Readers are strongly advised to pay careful attention to information provided by the manufacturer of any drugs or equipment that they plan to use. The publisher has used its best endeavours to ensure that the URLs for external websites referred to in this book are correct and active at the time of going to press. However, the publisher has no responsibility for the websites and can make no guarantee that a site will remain live or that the content is or will remain appropriate.

Contents

List of contributors List of abbreviations Preface to second edition

page vii xiii xv

1 Definition, clinical features and neuroanatomical basis of dementia

1

Thomas J. Grabowski and Antonio R. Damasio

2 Important anatomical landmarks in the brain in dementia

34

James H. Morris and Margaret M. Esiri

3 Practical approach to pathological diagnosis

48

Margaret M. Esiri and James H. Morris

4 Morphometric methods and dementia

75

Michael C. Irizarry

5 Safety precautions in laboratories involved with dementia diagnosis and research

82

Jeanne E. Bell

6 Molecular diagnosis of dementia

91

Vivianna M. D. Van Deerlin

7 Neuropathology of the ageing brain

113

` and John H. Morrison Patrick R. Hof, Thierry Bussiere

8 Neuroimaging Alzheimer’s disease

128

Arthur W. Toga, Michael S. Mega and Paul M. Thompson

9 Alzheimer’s disease

161

James H. Morris and Zsuzsanna Nagy

10 Down’s syndrome and Alzheimer’s disease

207

David M. A. Mann

11 Sporadic tauopathies: Pick’s disease, corticobasal degeneration, progressive supranuclear palsy and argyrophilic grain disease 227 Dennis W. Dickson

v

vi

Contents

12 Hereditary tauopathies and idiopathic frontotemporal dementias

18 Alcoholism and dementia 257

Mark S. Forman, John Q. Trojanowski and Virginia M.-Y. Lee

13 Vascular dementias

19 Hydrocephalus and dementia 289

330

21 Infectious (and inflammatory) diseases causing dementia

457

472

Harry V. Vinters

22 Schizophrenia and its dementia

497

Paul J. Harrison

23 Other diseases that cause dementia

509

Margaret M. Esiri

353

Benoit I. Giasson, Virginia M.-Y. Lee and John Q. Trojanowski

16 Huntington’s disease

20 Head injury and dementia Colin Smith, James A. R. Nicoll and David I. Graham

Gordon T. Plant, Jorge Ghiso, Janice L. Holton, Blas Frangione and Tamas Revesz

15 Parkinson’s disease, dementia with Lewy bodies, multiple system atrophy and the spectrum of diseases with α-synuclein inclusions

442

Margaret M. Esiri and Gary A. Rosenberg

James H. Morris, Hannu Kalimo and Matti Viitanen

14 Familial and sporadic cerebral amyloid angiopathies associated with dementia and the BRI dementias

427

Clive Harper and Richard A. Scolyer

376

24 Transgenic mouse models of neurodegenerative disease

533

David R. Borchelt, Joanna Jankowsky, Alena Savonenko, Gabriele Schilling, Jiou Wang and Guilian Xu

Jean Paul G. Vonsattel and Maxim Lianski

17 Human prion diseases James W. Ironside and Mark W. Head

402

Appendix: Dementia brain banks Index

558 564

Contributors

Jeanne E. Bell Neuropathology Unit, Pathology Department, Western General Hospital, Crewe Road, Edinburgh EH4 2XU UK David R. Borchelt Departments of Pathology and Neuroscience, Johns Hopkins University School of Medicine, 720 Rutland Avenue, Ross Building Room 558, Baltimore, MD 21205, USA Thierry Bussière Elan Pharmaceuticals, 800 Gateway Avenue, South San Francisco, CA 94080, USA Antonio R. Damasio Department of Neurology, The University of lowa, lowa City, IA 52240, USA Dennis W. Dickson Department of Pathology, Mayo Clinic, 4500 San Pablo Road Jacksonville, FL 32224, USA

vii

viii

List of contributors

Margaret M. Esiri Department of Neuropathology, The Radcliffe Infirmary, Woodstock Road, Oxford OX2 6HE UK

Clive Harper Neuropathology Unit, Department of Pathology, University of Sydney, Sydney NSW 2006, Australia

Mark S. Forman Center for Neurodegenerative Disease Research, Department of Pathology and Laboratory Medicine, University of Pennsylvania School of Medicine, 3400 Spruce Street, Philadelphia, PA 19104, USA

Paul J. Harrison Neurosciences Building, University Department of Psychiatry, Warneford Hospital, Oxford OX3 7JX UK

Blas Frangione Department of Psychiatry, New York University School of Medicine, New York, NY, USA Jorge Ghiso Department of Pathology, New York University School of Medicine, New York, NY, USA Benoit I. Giasson Center for Neurodegenerative Disease Research, Department of Pathology and Laboratory Medicine, University of Pennsylvania School of Medicine, Philadelphia, PA 19104, USA Thomas J. Grabowski Department of Neurology, 2RCP UIHC, 200 Hawkins Drive, Iowa City, IA 52230, USA David I. Graham Department of Neuropathology, University of Glasgow Institute of Neurological Sciences, Southern General Hospital, Glasgow G51 4TF, UK

Mark W. Head National Creutzfeldt–Jakob Disease Surveillance Unit, Division of Pathology, University of Edinburgh, Western General Hospital, Edinburgh EH4 2XU UK

Patrick R. Hof Elan Pharmaceuticals, South San Francisco, CA USA

Janice L. Holton Queen Square Brain Bank, Division of Neuropathology, Institute of Neurology, University College, London WC1N 3BG, UK

Michael C. Irizarry Alzheimer Disease Research Unit, Massachusetts General Hospital East, 114 16th Street, Charlestown, MA 02129, USA


James W. Ironside National Creutzfeldt–Jakob Disease Surveillance Unit, Division of Pathology, University of Edinburgh, Western General Hospital, Edinburgh EH4 2XU UK Joanna Jankowsky Departments of Pathology and Neuroscience, Johns Hopkins University School of Medicine, 720 Rutland Avenue, Ross Building Room 558, Baltimore, MD 21205, USA Hannu Kalimo Department of Pathology, Uppsala University Hospital, and Helsinki University Hospital Finland Virginia M.-Y. Lee Department of Pathology and Laboratory Medicine, HUP, Maloney 3rd floor, 36th and Spruce Streets, Philadelphia, PA 19104-4283, USA Maxim Lianski Department of Psychiatry, New England Medical Center, Tufts University, Boston, MA 02111, USA David M. A. Mann Clinical Neurosciences Research Group, Department of Medicine, University of Manchester, Oxford Road, Manchester, M13 9PT, UK

Michael S. Mega Laboratory of Neuro Imaging, Department of Neurology, UCLA School of Medicine, 710 Westwood Plaza, Los Angeles, CA 90095-1769, USA James H. Morris Department of Neuropathology, The Radcliffe Infirmary, Woodstock Road, Oxford OX2 6HE, UK John H. Morrison Fishberg Research Center for Neurobiology, Kastor Neurobiology of Aging Laboratories, Mount Sinai School of Medicine, New York, NY 10029, USA Zsuzsanna Nagy Department of Neuropathology, The Radcliffe Infirmary, Woodstock Road, Oxford OX2 6HE, UK James A. R. Nicoll Division of Clinical Neurosciences University of Southampton General Hospital Southampton UK Gordon T. Plant Department of Neurology, National Hospital for Neurology and Neurosurgery, Queen Square, London WC1N 3BG, UK Tamas Revesz Queen Square Brain Bank, Department of Molecular Pathogenesis, Institute of Neurology, University College, London WC1N 3BG, UK

ix

x


Gary A. Rosenberg Department of Neurology, University of New Mexico, 915 Camino de Salud NE, New Mexico USA Alena Savonenko Departments of Pathology and Neuroscience, Johns Hopkins University School of Medicine, 720 Rutland Avenue, Ross Building Room 558, Baltimore, MD 21205, USA Gabriele Schilling Department of Pathology, Johns Hopkins University School of Medicine, 720 Rutland Avenue, Ross Building Room 558, Baltimore, MD 21205, USA Richard A. Scolyer, Anatomical Pathology Department, Royal Prince Alfred Hospital, Missenden Road, Camperdown NSW 2050, Australia Colin Smith Neuropathology Laboratory Department of Pathology, University of Edinburgh, Alexander Donald Building, Western General Hospital, Crewe Road, Edinburgh EH4 2XU UK Paul M. Thompson Laboratory of Neuro Imaging, Department of Neurology, UCLA School of Medicine, 710 Westwood Plaza, Los Angeles, CA 90095-1769, USA

Arthur W. Toga Laboratory of Neuro Imaging, Department of Neurology, UCLA School of Medicine, 710 Westwood Plaza, Los Angeles, CA 90095-1769, USA John Q. Trojanowski Department of Pathology and Laboratory Medicine, HUP, Maloney 3rd floor, 36th and Spruce Streets, Philadelphia, PA 19104-4283, USA Vivianna M. D. Van Deerlin Department of Pathology and Laboratory Medicine, Hospital of the University of Pennsylvania, 7.103 Founders Pavilion, 3400 Spruce Street, Philadelphia, PA 19104, USA Matti Viitanen Division of Clinical Geriatrics, Karolinska Institutet, Huddinge University Hospital, Stockholm SE-14186, Sweden Harry V. Vinters Department of Pathology and Laboratory Medicine (Neuropathology), UCLA Medical Center, CHS 18-170, Los Angeles, CA 90095-1732, USA Jean Paul G. Vonsattel Brain & Tissue Resource Center, The Taub Institute – Columbia University, Babies Hospital, Tunnel T8, 3959 Broadway, New York, NY 10032, USA


Jiou Wang Department of Pathology, Johns Hopkins University School of Medicine, 720 Rutland Avenue, Ross Building Room 558, Baltimore, MD 21205, USA

Guilian Xu Department of Pathology, Johns Hopkins University School of Medicine, 720 Rutland Avenue, Ross Building Room 558, Baltimore, MD 21205, USA

xi

Abbreviations

Aβ AD ALS APP ApoE APOE ε CAA CJD CNS CSF CT DLB DRPLA FTD FTDP-17

beta amyloid peptide Alzheimer’s disease amyotrophic lateral sclerosis beta amyloid precursor protein apolipoprotein E protein apolipoprotein E alleles congophilic amyloid angiopathy Creutzfeldt–Jakob disease central nervous system cerebrospinal fluid computerized tomography dementia with Lewy bodies dentate-rubro-pallido-luysian atrophy frontotemporal dementia frontotemporal dementia with parkinsonism linked to chromosome 17 FTLD frontotemporal lobar degeneration GFAP glial fibrillary acidic protein GSS Gerstmann–Sträussler–Scheinker syndrome HD Huntington’s disease IHC/ICC immunohisto/cytochemistry MAP microtubule-associated protein MND motor neuron disease MRI magnetic resonance imaging MT microtubule(s) NFT neurofibrillary tangle PD Parkinson’s disease PDC Parkinson–dementia complex PHF paired helical filament(s) PiD Pick’s disease PrP prion protein PSP progressive supranuclear palsy SF straight filament(s) TG transgenic VaD vascular dementia WM white matter WT wild type

xiii

Preface to second edition

The first edition of this book was conceived with the intention of providing a practical guide to the neuropathological diagnosis of dementia. It was intended particularly for those who did not regard themselves as experts in this increasingly complex field. The relatively small number of authors was encouraged to share their diagnostic experience with others. This second edition still aims to provide practical assistance in this way, but it now goes further than that. The pace of scientific research advances in this field of dementia is such that it has become a formidable enough task for an expert to remain fully conversant with his or her own subfield, let alone with the field as a whole. Therefore we have aimed to widen the authorship greatly so that many chapters could be written by those specialists with research interests in the topic that is covered by each. We are delighted that those approached have responded so generously and enthusiastically since the outstanding contributions from the authors have made the second edition the most up-to-date and comprehensive monograph on the neuropathology of dementia. Because of the wider international authorship it has also been possible to extend the scope of the coverage of each topic to include more about research findings and the background science so that the book can now claim to provide comprehensive coverage of current understanding of the dementias rather than a more restricted practical guide. We hope, therefore, that a wider readership of neurobiologists and clinicians as well as pathologists will find much of interest here. Those familiar with the first edition will find that there are many new chapters and even more new authors contributing to the second edition. In addition, many more of the illustrations are now in colour in order to provide maximum assistance with diagnosis and to add extra clarity generally. The book has been rapidly produced and is therefore thoroughly up-to-date – an essential feature when covering a fast-developing field. We particularly want to thank our authors and publisher for their splendid efforts on which the success of the book depends. We hope that you, our readers, will find it comprehensive, readable and, above all, useful.

xv

1 Definition, clinical features and neuroanatomical basis of dementia Thomas J. Grabowski1 and Antonio R. Damasio1 1

Introduction Dementia is a frequent consequence of neurodegenerative diseases involving the cerebral cortex. Unlike stroke, encephalitis or head injury, which lead to relatively circumscribed and stable brain damage, degenerative disorders often affect many regions of the brain. The widespread changes in brain structure and the multiple signs of cognitive impairment that result from such changes have led to a conceptualization of the degenerative dementias, and especially of Alzheimer’s disease, as ‘diffuse’ pathological processes, but this is not strictly true. The degenerative dementias, including Alzheimer’s disease, do not affect the entire cerebral cortex equally. Instead, the degenerative dementias are associated with varied profiles of anatomic involvement, which can be tracked by quantitative histopathological and neuroimaging techniques. Association and limbic regions suffer the brunt of the damage. It is widely accepted that cognition is supported by distributed neural systems, and that it is susceptible to dissociation by focal brain damage. Despite continued uncertainties about the physiology underlying normal cognition, locally and globally, the pathological functional anatomy of many cognitive disorders is beginning to be elucidated. Classic examples of such disorders and their anatomic correlates include anterograde declarative amnesia, which is due to lesions of the hippocampal formation and adjacent mesial temporal lobe structures; aphasia, which is due to lesions in the left perisylvian cerebral cortex; ideomotor apraxia, which is due to lesions of the left parietal lobe; and simultanagnosia, which is due to bilateral lesions of the dorsal occipital and parietal lobes. The clinical manifestations

Department of Neurology, Division of Behavioral Neurology and Cognitive Neuroscience, Roy J. and Lucille A. Carver College of Medicine, The University of Iowa, USA

of degenerative processes clearly depend in part on which neural structures and systems are affected earliest and most extensively. It is now apparent that degenerative dementia can present with impairments resembling any of the classic ‘focal’ disorders listed above. In this introductory chapter, we review the clinical features of the cortical dementias, including their profiles of cognitive impairment. We then review the evidence that the profile of involvement of association and limbic regions determines the profile of cognitive impairment. These relationships recapitulate many of the relationships that have been observed in focal, stable brain damage. The syndrome of primary dementia (dementia without other neurological signs) reflects relative sparing of primary motor and sensory cortex and of subcortical structures. The prominent amnesic component of Alzheimer’s disease is accounted for by the early and salient involvement of mesial temporal structures. The data supporting this conclusion will be considered in some detail. Variant patterns of involvement of neocortical regions in Alzheimer’s disease have predictably altered clinical correlates. Lesion-deficit correlation is a promising approach in the study of other degenerative diseases, such as frontotemporal lobar atrophy and dementia with Lewy bodies. Since the first version of this chapter, remarkable progress has been made on several fronts in neuropathology and neurology. As the existence of this volume testifies, there now exists a greatly improved description of the neuropathology of dementia, due in part to increasing success in characterizing these disease processes at the molecular level, and in developing more specific staining techniques. This is particularly true for the frontotemporal

The Neuropathology of Dementia: Second Edition, ed. Margaret M. Esiri, Virginia M.-Y. Lee and John Q. Trojanowski. Published by Cambridge University Press. C Cambridge University Press 2004.

1

2

T. J. Grabowski and A. R. Damasio

lobar atrophies. At the same time, registries of research subjects with dementia and their associated autopsy series, have greatly extended the amount of data available for clinico-pathological correlation. These data have been leveraged in clinical research due to the universal availability of high resolution anatomic clinical imaging, which now routinely reveals a great deal of anatomic information in living patients with dementia (e.g. the degree of regional cerebral atrophy). Increasingly, sophisticated structural and functional imaging techniques are being used in clinical research, which are sensitive to losses of gray matter and metabolic activity early in the course of degenerative disease. Further, these quantitative techniques allow the investigation of the anatomic correlates of indices of impairment of particular aspects of cognition. These developments in the molecular biology and imaging of degenerative disease permit us to address the main questions of this chapter in greater detail than was previously possible.

Definitions and clarifications The ICD-10 definition of dementia illustrates the prevailing concept of dementia. A diagnosis of dementia requires: (a) impairment in short- and long-term memory; (b) impairment in abstract thinking, judgement, higher cortical function, or personality change; (c) memory impairment and intellectual impairment, which cause significant social and occupational impairments; and (d) the occurrence of these traits when patients are not in a state of delirium. As this set of criteria emphasizes, amnesia is a prominent feature of dementia, but the term implies broader cognitive impairment than amnesia per se. The paradigm on which this definition is based is Alzheimer’s disease (AD), in which anterograde amnesia is one of the inaugural and salient features. Not all degenerative dementias share this profile, however, and although it is unusual for anterograde memory to be completely normal in a patient with any of the degenerative dementias, the clinical picture of dementia is sometimes dominated by other impairments, for example, in executive function or language, rather than memory. Consequently, a somewhat broader definition of dementia is a better foundation for developing a neurological clinical approach. In this perspective, Dementia is an acquired and persistent impairment of intellectual faculties, affecting several cognitive domains, that is sufficiently severe to impair competence in daily living, occupation, or social interaction.

Dementia implies involvement of multiple neural systems, supported by multiple anatomic structures. Isolated impairment of memory, language, visuospatial abilities or higher visual processes do not qualify per se as dementia, and are best denoted by the terms amnesia, aphasia or agnosia. Also, dementia implies a decline from a previously attained level of intellectual function (in contradistinction to developmental encephalopathies), but progressive impairment is not required to diagnose dementia. Dementia may be static, as it commonly is when closed head injury or large cerebral infarcts are the causes. Nevertheless, it is true that most dementing illnesses are progressive. Definitions of dementia traditionally and explicitly exclude the sort of fluctuating encephalopathy which alters the level of consciousness and is known as a ‘confusional state.’ Although such fluctuating conditions are usually due to a metabolic disturbance, this exclusion criterion is problematic now that it has become clear that a fluctuating sensorium is a prominent feature in dementia with Lewy bodies (DLB). Finally, it must be stressed that dementia is a clinical term, not a pathological term. It is possible to have neuropathological changes of a particular degenerative disease without dementia. For example, pathological changes characteristic of AD may be associated with clinical mild cognitive impairment or with no demonstrable cognitive impairment. It is also possible to meet criteria for dementia without neuropathological changes in the brain, for example, in certain metabolic conditions or as a consequence of depression. Objective memory or other cognitive impairment, which is not sufficiently severe to impair activities of daily living is defined as mild cognitive impairment (MCI)(Peterson et al., 1999). Most cases of MCI are cases of mild and more or less circumscribed amnesia, but other forms of MCI exist, for example mild and more or less circumscribed anomia, and the mild impairments of executive dysfunction that presage frontotemporal dementia. However, these conditions are rarely recognized, and have hardly been studied. To our knowledge, no early autopsies of such cases have been reported. In the remainder of the chapter, we will use the terms amnesic MCI and MCI more or less interchangeably. Current data suggest that patients with formally defined MCI worsen to the point of being classifiable as demented at a rate of approximately 10–15% per year (Morris et al., 2001). Ultimately, more than half of these patients may develop dementia, though some do not, even after many years of follow-up. Mild cognitive impairment may be the earliest overt manifestation of AD in many patients. (But it is not

Definition of dementia

simultaneous with the appearance of the neuropathology of AD, which may precede symptoms by a number of years.) Whether all or most patients with AD pass through such a phase of ‘amnesic MCI’ remains to be determined; however, it is likely that amnesic MCI is the ordinary presentation of AD (Morris et al., 2001). Since the previous version of this chapter, MCI has become a focus of intense research interest. The observation that there are graded degrees of vulnerability to degenerative processes across brain regions and cortical laminae, and the fact that these diseases have presymptomatic phases, raise the question of what determines the transition from unaffected status to the status of MCI and to the status of dementia? What is the anatomic basis for the development of dementia? We review the evidence bearing on these questions for several degenerative conditions. The neuropathological basis of Alzheimer dementia will be discussed in the most detail, but we will also consider frontotemporal lobar degeneration (FTLD), Lewy body diseases; and Creutzfeldt–Jakob disease.

Clinical evaluation of dementia In order to provide context for a discussion of the anatomic basis of the dementias, we will review briefly the contemporary clinical approach to dementia. We then provide an overview of the clinical features of the degenerative disorders which cause dementia. The reader is referred to the appropriate chapters in this volume for a fuller consideration of these topics.

General clinical approach Neurologists and other clinicians are often called on to evaluate progressive or suspected progressive impairments of cognition. The first task of a clinician confronted with a patient suspected of dementia is to establish whether the patient’s intellectual ability has been diminished relative to its prior level, and whether the patient’s capacity to function in his/her accustomed occupational and social settings has been impaired. The clinician must decide whether the patient has apparently normal cognition, mild cognitive impairment, or dementia based on the formal evaluation of mental status (in the office and/or neuropsychology laboratory), coupled with estimation of the patient’s premorbid intellectual abilities. Interviewing a relative, friend or companion is a necessary component of the evaluation, since the value of the self-report of the patient, which is usually crucial information in a medical

encounter, is attenuated in this setting. Both overestimation and underestimation of the severity of one’s own cognitive impairment are common. The former is common in normal ageing, adjustment disorders and depression. The latter arises when patients have poor insight into, or even unawareness of, their illness (anosognosia), which is very common in AD and FTLD. Mild cognitive impairment is most often evident on tests of anterograde episodic memory, and on timed tests. Anterograde memory impairment in isolation is a common profile. The differential diagnosis in such cases includes other processes that preferentially affect mesial temporal lobe structures, such as epilepsy, anoxia, carbon monoxide intoxication, herpes simplex encephalitis, and paraneoplastic limbic encephalitis. However, a good history usually excludes most of these processes, and careful neuropsychological evaluation may demonstrate minor degrees of impairment in other domains of cognition. When dementia is present, the clinician should attempt to classify it as primary dementia, dementia ‘plus’ or secondary dementia. The syndrome of dementia without any abnormal signs on neurological examination is sometimes called primary dementia. Its presence implies that the brunt of the pathological process falls on association and limbic regions rather than on the primary cortices and subcortical regions. In other patients the neurological examination discloses dementia ‘plus’ elementary neurologic signs (e.g. dementia plus evidence of cerebrovascular disease, dementia plus parkinsonism, or dementia plus gait impairment). In each case, the additional neurological findings constitute an important diagnostic clue. In many of these diseases, the brunt of the pathological process is suffered by subcortical structures (white matter or grey matter) rather than association and limbic regions, and the cognitive profile will often be dominated by slowed cognition, apathy and poor problem solving rather than signs of cortical dysfunction, such as amnesia, agnosia or aphasia. Secondary dementia is dementia caused by potentially treatable medical or surgical conditions, such as medication effects, depression, metabolic derangements or surgically approachable structural problems, such as chronic subdural hematoma or orbitofrontal meningioma. Secondary dementia is generally identified by its association with past medical or psychiatric history, laboratory evaluation and structural imaging. The clinician should be prudent with respect to secondary dementia. The differential diagnosis of dementia is potentially vast, but the majority of possible conditions are neither reversible nor treatable given our current

3

4


knowledge. Although secondary dementia is not nearly as common as the leading degenerative conditions, some of these cases are treatable with a high degree of efficacy. Therefore, the clinician often concentrates on potentially treatable causes of dementia rather than adopting the wider point of view necessary to achieve a precise diagnosis. These include primary affective disorders, especially major depression (‘pseudodementia of depression’) and medication encephalopathy. Metabolic disorders (thyroid disease, hypovitaminosis B12, hypercalcemia, etc.), alcoholism, neurosyphilis and certain structural alterations (chronic subdural hematoma, normal pressure hydrocephalus, brain tumours) are also sought. Since the 1990s, it has become clear that the profile of cognitive impairment across cognitive domains is also helpful in diagnosis. In some cases, the impairment predominantly affects one domain (e.g. progressive non-fluent aphasia). In others, several aspects of cognition are affected, and on occasion, it is mainly the speed and efficiency of cognition that is affected. The latter pattern is commonly a consequence of subcortical disease. Secondary dementia usually has a subcortical profile, not a focal or broad cortical profile. As noted, AD usually presents as a primary dementia with salient anterograde amnesia, and other impairments. In practice, many clinicians begin to suspect an alternative diagnosis when impairments in some domains are found to be ‘disproportionate’ to anterograde memory impairment, i.e. when there is a deviation from the Alzheimer profile. Some of these non-Alzheimer profiles have been tentatively codified with diagnostic criteria, particularly those with salient executive or language dysfunction. Specific laboratory investigation of dementia still has limited diagnostic value, except for structural imaging, which is useful in most, if not all, patients to detect surgically approachable structural lesions that may cause or contribute to dementia (e.g. hydrocephalus), as well as evidence of cerebrovascular disease and asymmetric atrophy. Imaging is especially helpful in the evaluation of patients with suspected frontotemporal dementia. Executive function is difficult to evaluate in the clinic, but frontotemporal atrophy in these cases is usually obvious on MR images. As for functional imaging, its role is still being defined. Blood biomarkers have yet to be developed, and CSF biomarkers described to date have not been demonstrated to be sufficiently sensitive and specific to be useful. Genetic testing is becoming more important, but commercial tests are available only for the most common inherited form of AD (i.e. presenilin-1) and Huntington’s disease.

The main causes of dementia Alzheimer’s disease The most common aetiology of dementia is AD ease. The inaugural manifestation of AD is, almost always, memory impairment. More precisely, it frequently presents with an insidious anterograde amnesia for factual material. (Note that some diagnostic criteria, such as ICD-10 imprecisely equate anterograde memory impairment with ‘short-term’ memory impairment.) The constellation of primary dementia featuring anterograde memory impairment more or less corresponds to what has been termed the ‘dementia of Alzheimer type’. Some perceptive families also note, in retrospect, that patients with AD have had subtle alterations in personality and affective symptoms. Once manifest, the disease pursues a slowly progressive but relentless course over 8–12 years, without remission. Memory difficulty becomes more pronounced. Visuospatial disorientation, and impaired word-finding appear within the first 3 years or so. Other cognitive deficits (loss of semantic memory, aphasia, apraxia, executive impairment) ensue, combined with a striking lack of insight into the presence and/or degree of impairment (anosognosia). Gait impairment and other motor signs do not appear until the final stage of the disease, when cognitive dysfunction is very severe. The disease is equally notable for what it spares: patients regularly have no elementary motor or sensory deficits. It is as if all higher-ordered mental functions are singled out. The amnesic dementia of AD is the paradigm for the syndrome of primary dementia. Social graces are usually preserved for quite a while, giving way later to profound alteration in personality and sometimes psychosis, such as that manifested by Alzheimer’s original case (Adams & Victor, 1989; Cummings & Benson, 1992). The clinical research criteria for AD were formulated in 1984 by McKhann et al. These criteria, which are expected to be established by clinical examination and standardized brief mental status examination, and confirmed by neuropsychological tests, include: (a) deficits in two or more areas of cognition; (b) progressive worsening of memory and other cognitive function; (c) no disturbance of consciousness; (d) onset between ages 40 and 90; and (e) absence of other systemic or neurological disorders sufficient to account for the progressive cognitive defects. Note the similarity of these criteria to those of the ICD-10 for dementia, underscoring the fact that the prevailing definition of dementia is heavily influenced by the Alzheimer paradigm. Secondly, note the prominent exclusion criterion. These criteria do not permit a separation of AD from other degenerative dementias, notably FTLD


and DLB. Autopsy series have documented that the positive predictive value of the NINDS/ADRDA criteria is about 90%, although the negative predictive value is unclear. Although an amnesic presentation applies to the majority of patients with AD, a substantial number have variant presentations. Some patients, for example, have early and prominent extrapyramidal signs or myoclonus; these patients tend to decline more rapidly (Mayeux et al., 1985). Some patients have relatively greater language disturbance at the onset (Chui et al., 1985). The pattern of pathology in these patients has yet to be discovered. On the other hand, a subset of patients presenting with prominent higher-order visual impairment have a distinctive pattern of pathology. This presentation and other ‘focal’ cognitive presentations will be considered later in this discussion. Frontotemporal lobar degeneration FTLD is the term applied to a group of conditions that also present with the syndrome of primary dementia. Investigation of FTLD is one of the most active areas of research in degenerative dementia, and the concepts are still evolving. In general, this cluster of diseases is characterized by disproportionate atrophy of the anterior frontal and temporal lobes. The cortex and white matter are both affected heavily, and the affected gyri may narrow to a ‘knife-edged’ configuration. The term FTLD encompasses the disorder formerly referred to as Pick’s lobar atrophy, as well as frontal atrophy associated with motor neuron disease (Neary et al., 1990; Peavy et al., 1992); frontal atrophy without distinctive histopathology, also termed frontal lobe degeneration (FLD) (Brun & Passant, 1996); some cases of corticobasal degeneration and other rarer conditions. A striking gross pathological feature of these diseases is that the atrophy is often asymmetric, both with respect to the degree of involvement of the frontal and temporal lobes, and with respect to the degree of involvement of the right and left hemispheres. FTLD is usually manifested as a syndrome of frontotemporal dementia (FTD), in which the neuropsychological profile is dominated by defective executive function and social misconduct, rather than defects in memory and visuospatial function (Tissot et al., 1985; Knopman et al., 1989). The nomenclature is unfortunately confusing: FTLD is a general pathological term; FTD is a clinical term; and FLD is a term for a specific pathological subtype of FTLD. The clinical distinction of FTD from the dementia of AD can be difficult. As in AD, there is usually a paucity of non-cognitive neurological signs, and the disease progresses relentlessly over a 8–10-year period. Cases of FTD usually have prominent bilateral frontal involvement, though as mentioned above there is frequently left–right asymmetry. When the atrophy

is lateralized to the left, dysfluent speech is usually present, and the syndrome might be termed progressive aphasic dementia. When the atrophy is markedly lateralized to the left, there is a primary progressive aphasia, in which language disturbances dominate the clinical presentation, often for years, before more widespread deficits ensue. Two forms of progressive aphasia are now recognized: progressive non-fluent aphasia (PNFA), and semantic dementia. PNFA conforms more or less to a progressive aphasia of Broca type, with effortful, dysfluent, agrammatic speech, and is accompanied by left perisylvian (i.e. left inferior frontal and superior temporal) atrophy. Semantic dementia occurs in FTLD cases in which the atrophy is temporal-predominant, and left-lateralized. The earliest manifestation is frequently anomia, progressing to fluent, empty speech, impaired verbal comprehension, visual agnosia and general semantic loss. Corticobasal degeneration (CBD), though usually thought of as being associated with parietal lobar atrophy, serves well to illustrate the difficulties encountered in clinical diagnosis in FTLD. The problem cuts two ways: variability of phenotype and variability of regional pathology. Boeve et al. (1999) evaluated a series of 13 cases of clinically diagnosed CBD. The pathological diagnosis was CBD in only 7, the other cases being AD, PSP, Pick’s disease, non-specific histology, and CJD. Grimes et al. evaluated another series of 13 cases of pathologically confirmed cases of CBD. The clinical diagnosis was CBD in only four cases. A majority of the cases had dementia, which preceded the movement disorder. Speech difficulty, memory loss and dysexecutive presentations were common, and most of the patients did not have prominent apraxia at presentation. Surprisingly, the most consistently atrophied region was the frontal lobe (Grimes et al., 1999). Dementias with parkinsonism Dementia presenting in association with parkinsonism is often not due to AD and raises a differential diagnosis of idiopathic Parkinson’s disease (IPD), dementia with Lewy bodies (DLB), corticobasal degeneration (CBD), and progressive supranuclear palsy (PSP). The most common presentation is dementia with levodopa-responsive parkinsonism, which may be due either to idiopathic Parkinson’s disease (or combined AD and IPD) or dementia with Lewy bodies. Approximately 20 to 40% of patients with IDP eventually exhibit some degree of dementia, but in such cases dementia is not the presenting symptom, and the profile is usually one of ‘subcortical’ dementia, characterized by slowed cognition, impaired problem solving, and memory retrieval difficulty (with relative sparing of

5

6


recognition memory). In most such cases, concurrent AD is not present and cortical Lewy bodies, while they may be present in small numbers, are not prominent. In DLB, in contrast to IPD, cognitive symptoms are often the presenting complaint, and parkinsonism is often mild enough that treatment with levodopa is not recommended initially. In addition to the extrapyramidal signs, patients with DLB have a similar pattern of cognitive impairment to AD, but may have prominent and sustained fluctuations in performance (Ballard et al., 2001), and salient hallucinations. In addition, patients may have more prominent visuospatial impairment than is the norm in AD. The current consensus criteria (McKeith et al., 1996) for probable DLB are the presence of dementia and two of the three cardinal mentioned features (fluctuating sensorium/cognition, parkinsonism and visual hallucinations). Delusions, frequent falls and a REM sleep behaviour disorder are often present. Because of difficulty ascertaining the presence of fluctuation, the criteria appear insensitive in retrospective analyses, but perform better in prospective use (McKeith et al., 2000). Dementia with levodopa-unresponsive parkinsonism is usually due to either CBD or to PSP. In CBD, parkinsonism is usually markedly asymmetric and associated with signs of parietal lobe dysfunction, including apraxia and cortical sensory loss. Limb dystonia may be a presenting feature, and in some cases the alien limb phenomenon is prominent. Asymmetric, circumscribed parietal atrophy is often a feature, and tau-positive inclusions are found in neurons in basal ganglia. These features indicate that CBD is closely related to the frontotemporal lobar atrophies. Further underscoring the close relationship to FTLD, it is noteworthy that many cases which are classified as CBD at autopsy presented with dementia and frontal lobe atrophy (Grimes et al., 1999). PSP is usually characterized by parkinsonism, imbalance and vertical gaze abnormalities. The parkinsonism usually features axial rigidity. Supranuclear paresis of downgaze is particularly characteristic. Autopsy series have shown that a substantial number of cases of PSP present with dementia. The associated dementia was the paradigm for subcortical dementia (Albert et al., 1974). Dementia may precede the eye movement abnormalities by a year or more. Vascular dementia Vascular dementia is commonly considered as the second most common cause of dementia in the United States. The diagnosis of vascular dementia depends on both clinical assessment and neuroimaging. Hachinski developed a clinical scoring system for distinguishing between AD and

other degenerative and vascular dementia (Hachinski et al., 1975). The score embodies features such as a stroke-like course of illness (abrupt onset, stepwise deterioration, fluctuating course), the presence of atherosclerotic vascular disease risk factors (hypertension, history of strokes, evidence of atherosclerosis) and signs of focal brain damage. A meta-analysis reported that the Hachinski score correctly classified 76% of AD cases, 84% of vascular dementia and 12% of mixed dementia (Moroney et al., 1997). Present criteria go beyond the Hachinski ischaemic score, mainly by incorporating neuroimaging, which was not available routinely when the Hachinski scale was devised. Although this approach is useful in establishing an important role for vascular factors, it cannot distinguish between pure vascular dementia per se and mixed vascular and Alzheimer dementia. It is likely that many patients with vascular dementia have a component of AD (Hulette et al., 1997; Snowdon et al., 1997). Another set of diagnostic criteria for vascular dementia was drafted by an NINDS/AIREN task force. These research criteria diagnose probable vascular dementia on the basis of (a) dementia; (b) cerebrovascular disease evidenced by neurological signs and imaging; and (c) a relationship between (a) and (b) defined as the onset of dementia within 3 months of a recognized stroke or with abrupt onset and/or fluctuating stepwise progression. These criteria are relatively insensitive but very specific, and are meant to facilitate research rather than to be applied routinely clinically. The form of dementia associated with small vessel ischaemic disease does not conform well to the progressive amnesic paradigm (i.e. Alzheimer-type dementia). The more common clinical correlate of ischaemic vascular dementia is one of general slowing and inefficiency of thought, impaired memory retrieval, poor problem solving, and apathy (i.e. the profile which has been termed ‘subcortical dementia’). The cognitive defects are demonstrated most effectively with tests of executive function and of speeded information processing, neither of which is presented on the usual cognitive screening instruments such as the Mini-Mental State Examination. This is one reason why the prevalence of vascular dementia in the population has been difficult to ascertain. Normal pressure hydrocephalus Normal pressure hydrocephalus (NPH) is defined by the presence of ventricular enlargement, normal CSF pressure at the time of lumbar puncture, and a clinical picture of gait disorder, cognitive impairment and urinary incontinence. Many cases of NPH arise as a result of events leading to chronic impairment of CSF reabsorption, namely


subarachnoid hemorrhage, head injury, prior surgery or meningitis. The fully developed gait disorder is characterized by short slow steps, a failure to lift the feet from the floor, a wide base, and variable stepping force (‘vertical ataxia’). It is frequently the first component of the triad to appear and often the most prominent. Early, prominent gait impairment is a predictor of successful shunting. The cognitive impairment usually conforms to the subcortical profile, though it can be confused with Alzheimer’s disease or superimposed on it. Marked cognitive impairment is a predictor of unsuccessful shunting. Creutzfeldt–Jakob disease Prion diseases, though relatively uncommon, are another important cause of degenerative dementia. Creutzfeldt– Jacob disease (CJD) is the prototypical prion disease of humans. In the classic, myoclonic form of CJD dementia evolves rapidly after an insidious onset. The course usually includes a characteristic stage in which myoclonic jerks and startle responses are prominent. Highly characteristic periodic sharp waves are usually seen on the EEG at some phase of the illness. CJD is fundamentally a fulminant disorder in which the dissolution of cognition occurs in a few short weeks. Death follows in less than 6 months after onset. The profile of cognitive impairment varies from case to case. The Heidenhain variant of CJD presents as a rapid impairment of vision, even cortical blindness, due to spongiform changes beginning in the visual cortex. There is also a substantial variability in the pace of the disease. The variability relates in part to genetic factors, primarily to polymorphism at codon 129. Methionine homozygosity at codon 129 is characteristic of patients who develop the classic myoclonic form of CJD, whereas valine homozygosity or valine–methionine heterozygosity is associated with ataxic presentations, longer course, and less reliable diagnosis with EEG or 14-3-3 antigen assay. A diagnosis of CJD can be made when periodic sharp waves are found on EEG in a patient with subacute dementia, normal standard MRI and normal CSF. Brain biopsy may be avoided. In the setting of subacute dementia and normal imaging, CSF assay for the 14-3-3 protein is highly sensitive and specific, but CSF 14-3-3 may be elevated in other settings that lead to widespread neuronal loss, including HSV encephalitis, recent ischaemic stroke, hypoxic– ischaemic encephalopathy, and Hashimoto thyroiditis. In some cases, brain biopsy may still be the only way to establish the diagnosis and exclude potentially treatable conditions, such as primary angiitis of the nervous system and other inflammatory conditions (see Chapter 5 for safety considerations).

In 1996 a new form of prion disease was identified among young British and French patients with a novel clinical phenotype characterized by ataxia, sensory loss, and psychiatric symptoms. These patients are methionine homozygous at codon 129, have no periodic sharp waves on EEG, and no prion protein mutations. The pathological changes differ from classical CJD, and include extensive prion protein plaques surrounded by intense spongiform degeneration. These cases are thought to be due to bovine spongiform encephalopathy, which has crossed the species barrier.

Differential diagnosis The leading aetiology of dementia, accounting for at least half and possibly two-thirds of cases, is AD. Primary dementia with no other reasonable explanation is usually presumed to be AD. In AD, the anterograde amnesia may be moderate to severe before there is obvious impairment of other cognitive domains. The dementia of Alzheimer type, amnesic primary dementia, is the modal form of dementia, and departures from this profile (i.e. ‘disproportionate’ impairments in some other domain of cognition) raise the possibility of another diagnosis. In terms of prevalence, the second tier of dementing disorders is occupied by vascular dementia, FTLD, and DLB, all of which have been claimed to be the second most common cause of dementia. Vascular dementia is usually suggested by a history of stroke or significant cardiovascular disease, and compatible findings on imaging studies. FTLD may be suggested by a dysexecutive profile, but is probably most commonly recognized on the basis of the characteristic lobar atrophy on imaging studies. DLB is suggested by prominent psychiatric symptoms and parkinsonism, but may be difficult to distinguish from AD and IPD. All other causes of dementia are quite uncommon. Among the uncommon causes, hydrocephalic dementia (associated with gait difficulty and urinary incontinence), CBD (with asymmetric dopa-unresponsive parkinsonism, and apraxia), and CJD (associated with subacute course and often myoclonus) have the most distinctive presentations. Reversible secondary dementia is quite uncommon, probably accounting for about 1–2% of all cases of dementia.

Diagnostic challenges While diagnosis of dementia has certainly improved substantially, it is still not as accurate as one could wish. Clinical diagnoses of dementia rely on recognizing patterns of neurological and cognitive impairment, rather than

7

8


on specific biomarkers or other laboratory approaches. Diagnostic problems arise because the neurological findings are usually non-specific, and the cognitive profiles may overlap. For example, both AD and FTLD manifest defects in executive function and memory. By the time FTLD presents to a neurologist, the disease may be quite advanced, as suggested by the discovery of pronounced lobar atrophy, and at that stage the associated amnesia may be comparable in severity to that of AD. Problems also arise because the processes that cause dementia are not mutually exclusive, and multiple factors may coexist in any given patient. Given the high base rate of AD, the most common overlaps are of AD and small vessel cerebrovascular disease; AD and cortical Lewy bodies; AD and metabolic disorders; and AD and IPD. Superimposed diseases present challenges for both diagnosis and management. Degenerative, vascular, and metabolic factors must all be considered for their relative contribution(s) to the patient’s presentation as potential components of a dementing illness. In addition, depression may complicate AD, FTLD, parkinsonian syndromes or vascular dementia. Finally, in order to realize the potential benefit of therapies that slow down the natural history of dementia, the problem of early diagnosis will have to be tackled. Amnesic MCI which is an inaugural manifestation of AD will have to be distinguished from amnesic MCI with a benign prognosis. Additional challenges may lie ahead in detecting MCI at the inaugural stages in other degenerative diseases, such as FTLD and DLB.

The neuroanatomical basis of dementia We have asserted that the degenerative disorders manifesting as dementia are not diffuse processes, in pathological terms. We have also reviewed the major aetiologies of dementia, which may present different cognitive profiles. The question arises as to whether the selective distribution of pathological changes explains the pattern of cognitive impairments in dementia. Is there a correspondence between the profile of cognitive deficits and the degenerating neural systems for given conditions or cases? Such a correspondence is not obligatory. For example, the pathological hallmarks of a disease could conceivably cause little dysfunction in the tissue, and the dysfunction could arise instead from an associated process that does not alter macroscopic structure (e.g. a neurotransmitter deficiency.) We were drawn to attempt a structural explanation of the dementias because the early manifestations of AD resemble, in some instances, some of the focal cognitive syndromes

that have been observed in subjects with focal and stable brain lesions.

Normal functional neuroanatomy Cognition and behaviour emerge from the dynamic activity within and between the nodes of distributed neural systems, particularly among association, limbic, and paralimbic regions, and their supporting subcortical structures and motor cortices. Regions tend to be relatively specialized for one or more functional aspects. In other words, the distributed neural systems underlying different aspects of cognitive function are regionalized (Damasio & Damasio, 1989), and cognition is susceptible to dissociation by focal brain damage. A wealth of information has been obtained concerning the nature of the mapping of cognitive function onto brain structure, principally from the lesion method, and more recently also from functional neuroimaging. Though a systematic discussion of the basic findings is beyond the scope of this chapter, some general principles of functional anatomy are important. An overarching principle is that the posterior cortical regions (parietal, occipital, temporal lobes) are mainly concerned with perception and integration of sensory stimuli and with encoding knowledge, while anterior regions (frontal lobe) are mainly concerned with functions of an executive character, such as elementary and complex motor function, planning, manipulation of mental images, and strategic and complex allocation of processing resources. Another overarching principle of functional anatomy is hemispheric dominance. More than 90% of right-handers and at least 60% of left handers are left hemispheredominant for language, based on incidences of aphasia in stroke, as well as studies conducted with Wada tests. The right hemisphere plays a dominant role with respect to complex perceptual processing (for example, recognizing faces), emotion, the representation of body state and the spatial distribution of attention. The functional correlates of a given cortical region depend in part on the functional type of cortex (primary, unimodal association, heteromodal association (higher order), or paralimbic/limbic), and in part on its location relative to the sensory and motor portal regions. Unimodal sensory association regions have to do with perception in their proper sensory channel and are located near their primary regions. Thus the areas surrounding the primary visual cortex in the occipital lobe are concerned with higher aspects of vision, and when damaged lead to deficits that are confined to visually dependent tasks. Within these unimodal association regions there is substantial segregation of the sensory processing streams


representing the features and properties of individual entities, which are located in ventral occipital and temporal regions and those representing their location in space, which are located in lateral and dorsal occipital regions. Areas adjacent to the primary sensory cortex, in the parietal operculum or superior parietal lobule, are concerned with higher aspects of somatic sensation, and their damage leads to deficits confined to somatosensory tasks. The premotor and supplementary motor regions, just anterior to the primary motor cortex, are related to higher aspects of motor function, including motor preparation and planning and when damaged lead to motor disorders (e.g. akinesia). The heteromodal (higher order) areas are found primarily in the prefrontal regions, the inferior parietal lobules, the mesial parietal regions, and the inferotemporal regions. The proximity of heteromodal association regions to particular unimodal association regions also provides a useful heuristic for understanding their functions. The association cortex of the occipito-parietal junction, which is interposed between visual and sensory unimodal regions, is associated with visuospatial integration; damage to this region produces disorders such as optic ataxia and simultanagnosia. The association cortex of the occipitotemporal region is interposed between the visual and auditory unimodal regions and the mesial temporal limbic regions. Damage to these areas leads to impaired object recognition (agnosia), impaired naming (anomia), and semantic memory loss. The association cortex of the superior parietal lobule, near the primary somatosensory cortex, is associated with representing and orienting personal and extrapersonal space; damage to it causes spatial neglect and impairment in spatial orientation. The association cortex rostral to the premotor region is concerned with the manipulation of internally generated images, referred to as working memory, among other processes. Limbic regions, by virtue of their connections to homeostatic and autonomic systems, and the convergence/divergence pattern of cortical connectivity linking them to the association regions, are implicated in arousal, emotion, motivation, prioritization of attention and memory. Damage to these regions usually causes deficits of memory, emotion and decision making.

The convergence zone framework A theoretical neural systems account of the basis of certain aspects of cognition in the interaction of early association, limbic, and higher order association regions has been proposed by Damasio and colleagues (Damasio, 1989a,b, 1999; Damasio et al., 1990a; Damasio & Damasio, 1994).

Briefly, this framework distinguishes between images, i.e. explicit, on-line mental patterns of any sensory type (e.g. visual, auditory, somatosensory), some of which constitute the manifest mental contents of conscious experience; and dispositions, i.e. latent, nonconscious, knowledge that directs operations on images, e.g. the construction of images in the process of recall, the manipulation of images in thought, the generation of action, and the regulation of body processes. The principal neural substrates for images are the explicit, topographical, neural patterns (or maps) formed in areas of cerebral cortex located in and around the points of entry of sensory signals in the cerebral cortex (the early sensory cortices). These neural patterns continuously change under the influence of external and internal inputs involved, for example, in perception or recall. All knowledge, that which was accumulated in evolution and is innately available, and that which has been acquired through learning, exists in dispositional form (implicitly, covertly, non-consciously), with the potential to become an explicit image or action. The neural substrates for dispositions are distributed in higher-order (association) cortices, parts of the limbic cortices, and numerous subcortical nuclei (e.g. basal ganglia, amygdala). The framework posits that dispositions are held in neuron ensembles called convergence zones. Convergence zones are made of microcircuits and cannot be resolved individually by current neuroimaging techniques, although it is presumed that aggregates of many activated convergence zones constitute anatomically macroscopic sets that can be visualized by imaging techniques or selectively affected by neurologic disease. The process of anatomical selection of convergence zones, both during learning and during subsequent operation is probability driven, flexible, and individualized, but because of the constraints of the brain’s anatomical design, convergence zones involved in certain classes of tasks are likely to be found, in most individuals, in the same large-scale convergence region of the brain, e.g. temporal pole; anterior inferotemporal cortex; frontal operculum. One important set of constraints is the pattern of connectivity among cortical regions. For example, one consistent relationship in posterior cortices is that feedforward projections arising from the pyramids in layer III of a cortical region project to layer IV of the next higher-order region. Feedback projections arise from layer VI of the latter and project to layer II of the former. Feedforward/feedback architecture is a general property of the posterior sensory regions, tying together early and higher-order regions as large scale systems (Rockland & Pandya 1979, 1981). The entorhinal cortex is an important node of convergence/divergence among the pathways emanating from unimodal regions

9

10


proper to each sensory modality. Thus, in this framework, the entorhinal cortex is a convergence region, whose convergence zones ‘hold’ dispositions pertaining to patterns of activity across early sensory regions. The process of making these memories explicit in recall involves a role for these convergence zones in directing activity in the early regions.

The anatomic basis of dementia In the past, dementia was considered to be a ‘global’ encephalopathy. The idea that the pathology was vascular, or that ageing itself was in some way responsible for the disease, fostered the concept of truly diffuse degenerative pathology. But neither the cognitive impairment nor the pathology is usually diffuse. Diffuse neural degeneration would be incompatible with the phenomenon of primary dementia, in which patients may have advanced cognitive impairment, including virtually complete anterograde amnesia, yet no impairment of motor function, and indeed even have a preserved capacity for motor learning. Further, primary dementia is not a unitary entity – AD and FTLD both present this way, yet with different cognitive profiles. The typical patient with AD will present with anterograde amnesia and visuospatial disorientation, while the patient with FTLD will present with an insidious behaviour disorder and defective social decision-making, and demonstrate relatively good anterograde memory and visuospatial orientation on formal testing. Within the spectrum of FTLD, one patient presents with an apathetic, dependent, perseverative state; another with anomia, impaired speech comprehension and visual agnosia. In other subdisciplines of neurology, especially cerebrovascular disease, dissociations such as these are encountered commonly, and are explained by the anatomic properties of the nervous system damage. We now turn to discussing the anatomic basis of the dementias. The anatomic basis of cognitive impairment in Alzheimer’s disease The cholinergic hypothesis The disproportionate impairment of memory in AD was clearly perceived from the beginning, but it came to be explained by a subcortical abnormality under the ‘cholinergic hypothesis’ (Bartus et al., 1982). The backdrop that set the stage for the acceptance of this account included the following observations: (a) there was a cholinergic deficiency in the brain tissue of affected patients (Perry et al., 1977; White et al., 1977; Davies et al., 1988); (b) presynaptic cholinergic markers were severely depleted in the hippocampus and cerebral cortex; (c) the predominant

source of cortical cholinergic afferents, the basal nucleus of Meynert, undergoes marked changes in Alzheimer’s disease; and (d) neuritic plaques were known to be rich in cholinesterase activity, suggesting that the degenerating nerve terminals are the projections of cholinergic neurons (Perry et al., 1981). Because anticholinergic pharmaceuticals, such as scopolamine, cause significant memory impairment, it was reasoned that a deficiency of cholinergic neural transmission in the cortex, resulting from cholinergic deafferentation, might underlie the amnesia of AD. The cholinergic hypothesis appealed to the clinical sense of many investigators, who realized that cholinergic drugs were available for the treatment of Alzheimer’s disease. Twenty years after the proposal, there is now a consensus that cholinergic augmentation is modestly effective in improving attention, memory, concentration and psychiatric symptoms in AD. Although the relationship between regional differences in the severity of neural degeneration and regional differences in cholinergic fibre loss (Geula et al., 1998) has yet to be clearly worked out, it has become clear that memory impairment in AD is not primarily due to cholinergic deafferentation. In fact, recent autopsy studies of patients with mild cognitive impairment showed that there were elevated levels of choline acetyl transferase (a presynaptic marker) in the hippocampus and frontal cortex of patients with MCI, all of whom had histological AD. Up-regulation of cholinergic systems in MCI would seem to be incompatible with the cholinergic hypothesis of memory dysfunction (DeKosky et al., 2002). Not until a careful accounting of the distribution of the pathological hallmarks of Alzheimer’s disease was made did a well-founded alternative to the cholinergic hypothesis become available. This account focused on mesial temporal limbic structures, as attention had been directed there by a new paradigm for amnesia: the bilateral hippocampal formation lesion. Before considering this evidence in detail, we must review the major neuropathological features of AD. The pathological hallmarks of Alzheimer’s disease The pathology of AD is characterized by the accumulation of insoluble fibrous material in intracellular and extracellular locations. The intracellular pathology consists of taupositive neurofibrillary pathology (neurofibrillary tangles, neuropil threads, and dystrophic neuritis), granulovacuolar change, and Hirano bodies. The extracellular deposits are diffuse and senile amyloid plaques (SPs). Both SPs and neurofibrillary tangles (NFTs) are widespread in the brain in AD, and are felt to be directly related to the pathophysiology of the disease. How exactly they relate to this pathophysiology and to each other


remains unclear, and has been the focus of vigorous debate (Chapter 9). The focus here is whether some insight into their relationship to cognitive impairment is available from their spatial and temporal distribution. NFTs and SPs do not occur uniformly throughout the cerebrum (Mutrux, 1947), but rather exhibit characteristic regional, cellular and laminar specificities. Moreover, the distribution of SPs is somewhat different from that of NFTs. Because of their restricted distribution and because they appear to be playing a secondary pathophysiological role (i.e. the catabolism of the predominant neuropathological elements), GVC and Hirano bodies have not figured prominently in theoretical treatments of AD. We will have little more to say about them here. Neurofibrillary tangles and neuropil threads NFTs are insoluble intracellular structures composed primarily of accumulations of paired 10 nm helical filaments composed of the microtubule associated protein tau (Goedert, 1993). Tau in its normal state is thought to have the effect of stabilizing microtubule interactions, promoting their assembly. Tau in paired helical filaments is hyperphosphorylated, cross-linked, and ubiquitinated. The ubiquitination suggests that neurons recognize NFTs as abnormal but fail to catabolize them. Varying degrees of phosphorylation and post assembly processing result in several intermediate manifestations of tau (e.g. ‘stage zero tangles’ (Bancher et al., 1989), and dispersed filaments (Goedert, 1993). Some tangles occur in non-viable neurons (i.e. ‘tombstone neurons’) (Hyman et al., 1988a,b). A number of investigators have reported evidence for the accumulation of abnormally phosphorylated tau before the formation of NFTs (Bancher et al., 1989). Tau antibodies also recognize the dystrophic neurites which surround senile plaques and neuropil threads. These neuropil threads are frequently the distal dendrites of tangle-containing neurons (Braak & Braak, 1988). In some cases, the dystrophic neurites are found to localize to the predicted projection zone of tangle-containing neurons (Hyman et al., 1988a,b), but there is evidence that dystrophic neurites appear to be more widely distributed than either SPs or NFTs (Kowall & Kosik, 1987). NFTs are recognized in clinical histologic sections with sensitive silver stains, such as the Bielschowsky stain, or with Thioflavine S. Immunohistochemical techniques have become the mainstay of research studies, however, and are being used increasingly frequently in clinical pathology. Antibodies to tau protein are the current standard. NFTs are known to be features of other neurological diseases, for example Pick disease, progressive

supranuclear palsy, dementia pugilistica, Down syndrome, the Parkinson-dementia complex of Guam, and postencephalitic parkinsonism. Some investigators have accordingly proposed that tangle formation is a stereotyped cellular reaction that may be initiated by diverse pathophysiological mechanisms (Wisniewski et al., 1979). Senile plaques SPs in fully developed form are characterized by a focal spherical deposit of fibrillary amyloid surrounded by dystrophic neurites, reactive astrocytes and microglia. Plaque cores are composed of the beta amyloid protein, a hydrophobic fragment of a transmembrane glycoprotein known as the beta amyloid precursor protein. Monoclonal antibodies against beta amyloid have been used to demonstrate widely distributed non-neuritic plaques, known as A4- or diffuse plaques. These ‘diffuse plaques’ are not surrounded by dystrophic neurites. Beta amyloid is also found in leptomeningeal vessels (congophilic amyloid angiopathy, CAA). CAA correlates with numbers of senile plaques containing amyloid cores and is structurally indistinguishable from the amyloid in those structures (Vinters et al., 1988). SPs are detected clinically with a variety of staining techniques, including with Congo red, Thioflavin S, sensitive silver stains, but the current standard is antibodies to beta amyloid. Neuronal loss In addition to these major pathological features of AD, there is a striking loss of synapses and neurons in AD. Terry and colleagues found that the major component of neuronal loss is large neurons, especially those in layer III (Terry et al., 1964). No loss of small neurons could be identified. The neuronal dropout has been correlated with reductions in synaptic immunostaining (DeKosky & Scheff, 1990; Masliah et al., 1991). Neuronal loss appears to occur in the same cell populations that are vulnerable to NFT formation, though many or most of these neurons do not develop tangles (Gomez-Isla et al., 1997). Spatial and temporal distribution of Alzheimer changes Spatial distribution of Alzheimer pathology Blessed and colleagues, correlated the number of SPs in elderly subjects with a quantitative index of their degrees of dementia (Blessed et al., 1968), and concluded that the dementia of Alzheimer’s type bears a quantitative relationship to the peculiar pathology of the disease. An appreciation of the significance of the uneven distribution of this pathology was slower in coming, but it is now well accepted that there is a characteristic, hierarchical pattern of distribution and progression of pathology in Alzheimer’s disease. In cases of fullblown AD, investigators have documented that the most severe changes are in the hippocampus and in the association

11

12


Fig. 1.1. The anatomic distribution of Alzheimer pathology. Arnold et al. surveyed the distribution of NFTs in Alzheimer’s disease (upper tier), finding the greatest numbers in mesial temporal limbic structures. Tangles become progressively less numerous in paralimbic cortex, heteromodal association cortex, and unimodal association cortex. The smallest extent of neurofibrillary pathology that has been correlated with dementia encompasses areas 28, CA1 and the inferotemporal neocortex. Neuritic plaques were found in a somewhat different distribution (lower tier). For example, they tended to be more evenly distributed throughout the cortex than tangles are. Most investigators have reported that they have less specific regional and laminar distributions (reprinted from Arnold et al. (1999). Cerebral Cortex, 1, 103–16, Fig. 8, with permission from The Editor, Cerebral Cortex and Oxford University Press).

regions, while the least severe changes are in primary sensory and primary motor regions (Brun & Gustafson, 1976). In their comprehensive cross-sectional survey of the cerebral cortex in AD, Arnold and colleagues found NFTs in greatest numbers in limbic periallocortex (area 28) and allocortex (subiculum, area 51). Corticoid areas (accessory basal nucleus of amygdala, basal nucleus of Meynert) were somewhat less involved. In the neocortex, tangles became progressively less numerous in the following order: proisocortex (areas 11, 12, 23, 24, 35, 38 and anterior insula), heteromodal association cortex (areas 9, 32, 36, 37, 39,

46, superior temporal sulcus, posterior hippocampus), and unimodal association cortex (areas 7, 18, 19, posterior 20, posterior 21, posterior 22) (Fig. 1.1). Primary motor and sensory cortices were largely spared (Arnold et al., 1991). The only fields with consistently more than 50 NFTs/1.6 mm2 were the entorhinal cortex and the subiculum/CA1 zone of the hippocampus. In the cerebral cortex, the NFTs predominantly affected layers III and V (Pearson et al., 1985; Lewis et al., 1987; Arnold et al., 1991); in entorhinal cortex, the NFTs affected layers II and IV. The findings of Arnold et al. are consistent with the findings of several other investigators (Brun & Gustafson, 1976;


Pearson et al., 1985; Lewis et al., 1987; Braak & Braak, 1991; Delacourte et al., 1999). Chu and colleagues extended the study of Arnold to include fine-grained analysis of ventromedial frontal structures (areas 11, 12, 24, 25, 32, posterior orbito-frontal cortex, and anterior insula). These cortices comprise allocortex, agranular and dysgranular paralimbic cortex, and association isocortex. They found: (i) NFTs in (periallocortical) area 25 occur mostly in layer V, while layer III is largely spared; (ii) in area 11 (isocortex) and the anterior insula (periallocortex), both layers III and V are affected; (iii) the highest density of NFTs was observed in a narrow strip of agranular periallocortex along the posteriomedial margin of the orbitofrontal cortex, with only slightly lower counts in the dysgranular (proisocortical) sector anterior to it; (iv) The density of NFTs decreased both anteriorly and laterally in the anterolateral granular orbitofrontal (association) cortex (Chu et al., 1997; Van Hoesen et al., 2000). These findings indicate that the distribution of NFTs in frontal cortices conforms to the same hierarchical pattern seen in medial temporal and posterior sensory association cortices. Amyloid is found in a much different distribution. Plaques are relatively uncommon in limbic periallocortex and allocortex, and tend to be distributed more evenly throughout the cerebral cortex than NFTs. Although SPs are also found infrequently in primary sensory cortex, plaques without dystrophic neurites (diffuse, or primitive plaques) are found in these areas, as well as in the striatum (Braak & Braak, 1990a) and the cerebellum (Joachim et al., 1989), where neither neuritic plaques (NPs) nor NFTs are found. Most investigators have reported that NPs have less specific regional and laminar distribution patterns than NFTs, though they tend to be more numerous in cortical layers III and IV (Duyckaerts et al., 1986; Arnold et al., 1991; Braak & Braak, 1991; Hof et al., 1992). Quantitative MRI studies with voxel-based morphometry (VBM) are in good agreement with the above description of the pattern of anatomic involvement (especially in terms of neurofibrillary pathology). In mild AD, VBM demonstrated mesial temporal, posterior cingulate/precuneus, temporoparietal, and perisylvian neocortical atrophy (Fig. 1.2). Little frontal atrophy was found (Baron et al., 2001). Morphometric studies with handtraced regions of interest have also documented losses in septal grey matter, amygdala, hippocampus, and posterior cingulate (Callen et al., 2001). Temporal distribution of Alzheimer pathology All the lesions reported in AD have been observed (in much lower numbers) in the brains of non-demented elderly

individuals (Blessed et al., 1968; Dayan, 1970; Tomlinson et al., 1970; Miller et al., 1984; Ulrich, 1985; Crystal et al., 1988; Bouras et al., 1994). The pathological distinction of AD from ageing is based on quantitative criteria (Khachaturian, 1985; Mirra et al., 1991, Newell et al., 1999). The presence of Alzheimer pathology in non-demented elderly subjects was studied by Arriagada and colleagues, who found that NFTs occurred most frequently in the perirhinal cortex, entorhinal cortex, and CA1/subiculum. A few tangles were also be found in neocortical areas. The rank order of involvement was highly consistent across individuals: entorhinal cortex > perirhinal cortex > CA1 > amygdala > nucleus basalis of Meynert > area 20 > parasubiculum > area 21 > CA3/4, dentate gyrus, presubiculum, area 21 and area 41. The entorhinal cortex was involved in 24 of 25 cases, with lamina II and IV predominantly affected. In perirhinal cortex, lamina III was the affected layer. In hippocampus, CA1/subiculum was the only significantly affected area. The number of NFTs, but not SPs, correlated with increasing patient age. Alz-50 immunoreactive SPs were found only in those areas which are also invested with NFTs or Alz-50 immunoreactive neurons. Beta amyloid-immunoreactive SPs were found more extensively in the neocortex, and much less frequently in those limbic cortices where NFTs were found (Arriagada et al., 1992b). This spatial pattern of Alzheimer pathology in elderly persons is strikingly similar to the reported distribution of the lesion burden in AD. (Pearson et al., 1985; Hyman et al., 1990; Arnold et al., 1991; Braak & Braak, 1991; Arriagada et al., 1992a,b). Arriagada and colleagues concluded that elderly non-demented individuals and Alzheimer patients display the same hierarchical pattern of selective vulnerability, differences between these populations being quantitative rather than qualitative. Delacourte et al. (1999) generated converging results using autopsy material from 130 patients, including approximately 25 non-demented subjects, 10 with mild impairment, and 100 with dementia. Once again, there was a hierarchy of involvement, which they divided into ten pathological stages. The transentorhinal (perirhinal) region was the most frequently involved. In fact, no subject over 75 did not have involvement of this area. Although clinical status did not bear a one-to-one relationship with pathological stage, there was a monotonic relationship of the degree of impairment to the extent of pathological change. The hierarchical temporal and spatial pattern of involvement in Alzheimer’s disease underlies the staging system devised by Braak and Braak (1991), now widely used to report the extent of Alzheimer-related pathology (Fig. 1.3). These investigators found the pattern of development of

13

14


Fig. 1.2. The spatial pattern of atrophy in Alzheimer’s disease based on structural MR analysis. Results generated with voxel-based morphometry of T1 -weighted MR images of 19 Alzheimer’s disease brains compared to 16 age-matched controls. The colour scale, superimposed for reference on a normal scan, indicates the level of statistical confidence in local loss of grey matter in AD. Note the involvement of mesial temporal lobe, precuneus/posterior cingulate, and temporo-parietal association areas. These results are largely compatible with post-mortem studies of the distribution of pathology in AD. (Reprinted from Baron et al. (2001) Neuroimage, 14, 298–309, Fig. 2, with permission from Academic Press.)

NFTs to be much more consistent and useful for staging than the distribution of SPs. Transentorhinal (perirhinal) involvement precedes involvement of the entorhinal cortex proper. Braak and Braak distinguish two transentorhinal stages, followed by two limbic stages, and finally two stages with progressively severe isocortical involvement. Correlation of clinical and pathological stages Given the stereotypical anatomical stages of AD and the relationship of lesion burden to the degree of dementia, it is meaningful to speak of the anatomic correlates of mild cognitive impairment, and those of mild dementia. The transentorhinal stages appear to represent an asymptomatic preclinical phase of Alzheimer’s disease. The duration of the preclinical phase is a matter of great interest. Braak and colleagues have inferred that this phase may last for decades (Ohm et al., 1995). Bouras and colleagues studied a large number of unselected autopsies performed at a geriatric hospital (Bouras et al., 1994). Their cases, for which premorbid MMSE and clinical data were available, suggested that ‘mild cognitive impairment and disorientation’ was the maximal deficit seen in patients in whom NFTS did not yet invest the inferior temporal cortex. In the study of Delacourte et al. (1999), most subjects with MCI or CDR 0.5 had disease limited to mesial temporal structures,

though a few also had temporal polar and inferotemporal NFTs. All patients with neocortical tangles beyond the inferotemporal region were demented. Mitchell et al. (2002) reported that phosphorylated tau lesions in the parahippocampal region develop prior to the onset of clinical dementia and that their presence is correlated with the degree of episodic memory impairment, but not other cognitive impairments. Hyman and colleagues have reported a patient whose neuropsychological examination disclosed evidence of early Alzheimer’s disease and who expired of other causes only 7 weeks later. The autopsy disclosed NFTs in the entorhinal cortex, hippocampus, amygdala, dorsal raphe, and inferior temporal gyrus. Alz-50 positive SPs were limited to limbic areas; beta-amyloid immunoreactive SPs were widespread in the association cortex and amygdala (Hyman et al., 1991). Hof and colleagues reported pathological findings in an 82-year-old subject who was felt to have had incipient dementia at the time of death. No neuropsychological data had been obtained, but unusually reliable collateral historical data was available (Hof et al., 1992). In this patient, NFT counts similar to those of AD patients were found only in the perirhinal cortex, layers II and IV of the entorhinal cortex, and subiculum. NFTs were found in lesser numbers


Fig. 1.3. Pathological staging of Alzheimer disease. The Alzheimer’s disease pathological staging system of Braak and Braak, is based on the hierarchical pattern of the appearance of NFTs in various anatomic regions. The division points between stages are arbitrary, and do not necessarily correspond to specific clinical stages of AD, though the Braak stage does correlate with the severity of dementia. There are two transentorhinal stages, in which NFTs are confined to the perirhinal cortex, and which represent preclinical stages of AD. In stage III, NFTs begin to invest limbic regions, especially the cellular islands in layer II of the entorhinal cortex. These limbic changes become more pronounced in stage IV, involving entorhinal layer IV and hippocampal CA1 more heavily, but there is still minimal pathology in cortex (mainly in inferotemporal regions), and no cortical atrophy. Stages III and IV correspond roughly to the clinical stage of MCI and/or CDR 0.5. Stage V is characterized by severe isocortical involvement, especially in the inferotemporal and retrosplenial regions. There are minimal changes in primary motor and visual cortex. In Stage VI, all association regions of the cortex are affected severely, and NFTs appear in striatal and nigral regions, while the primary motor cortex remains more or less spared. Stages V and VI correspond to overt dementia in the great majority of patients. (Reprinted from Braak & Braak (1995) Neurobiol. Aging, 16, 271–84, Fig. 4, with permission from Elsevier.)

15

16


in hippocampal sectors and in the layers V and VI of the inferior temporal cortex. The nucleus basalis of Meynert had low NFT density but higher SP density. SPs were found throughout the neocortex. Gold et al. (2000) reviewed the relationship of Braak state to CDR score in nonagenarians, and found that Braak stage IV was invariably associated with dementia of some degree, and that some patients with Braak III also had dementia. Taken together, these data suggest that disease limited to the mesial temporal regions underlies amnesic MCI, and that the minimal lesion which can cause dementia in AD encompasses entorhinal cortex, CA1, and the neocortex of the inferior temporal lobe (Braak stage III-IV)(Grober et al., 1999). At this stage, neuritic plaques may be limited to limbic structures. Neuropsychological correlates of regional and laminar specific pathology We now turn to a more detailed consideration of the specific neural systems involved in AD, beginning with a finergrained analysis of it mesial temporal pathology. The medial temporal lobe Amnesic mild cognitive impairment is associated with mesial temporal Alzheimer pathology. The association of bilateral mesial temporal lesions with amnesia was first reported by Glees and Griffith (1952). In 1953, patient H.M. underwent bilateral mesial temporal lobe resection for the treatment of intractable epilepsy and developed a profound and irrevocable anterograde amnesic syndrome (Scoville & Milner, 1957; Corkin, 1984). The case of H. M. became the paradigm of the amnesic syndrome. Numerous subsequent lesion studies have confirmed that the inevitable consequence of bilateral hippocampal formation lesions is anterograde amnesia. One of the most focal and instructive cases is that of ZolaMorgan and Squire’s patient, RB, who became amnesic as a result of an hypoxic–ischaemic intraoperative event. When he died 5 years later, an autopsy disclosed that the loss of neurons was limited to field CA1 of the hippocampus (ZolaMorgan et al., 1986). Though the nature of the operations it accomplishes is debated, these studies solidly establish the hippocampal formation as an essential neural substrate for the acquisition of new factual memory. The predilection in AD for early NFT formation in mesial temporal structures offers an alternative explanation to the cholinergic hypothesis for the early amnesia of Alzheimer’s disease. Anatomy of medial temporal limbic structures (see also Chapter 2) The entorhinal cortex (EC, Brodmann’s area 28) occupies the anterior part of the parahippocampal gyrus. The transition from entorhinal cortex to temporal isocortex

occurs as interdigitations of allo- and isocortical laminae rather than as a smooth transition (Braak & Braak, 1985). Layer II of the entorhinal cortex contains clusters of large stellate neurons which form intracortical cellular islands. The main cortical afferent tract to the hippocampus, the perforant pathway, has its cell bodies in layer II and the superficial portion of layer III. The perforant pathway ‘perforates’ the subiculum and crosses the hippocampal fissure to terminate in the outer two-thirds of the molecular layer of the dentate gyrus, synapsing there with the outer dendritic branches of the dentate gyrus granule cells. At the entorhinal border of this perirhinal or ‘transentorhinal’ cortex (Brodmann’s area 35), the cellular islands of layer II amalgamate to become Braak’s pre-alpha layer. The pre-alpha layer gradually moves obliquely through the outer cellular layers to lie in a deeper position at the isocortical border of this zone. The entorhinal cortex sits at the apex of an orderly system of feedforward and feedback connections from the occipital, parietal, temporal, and limbic lobes (Van Hoesen et al., 1986; Jones & Powell, 1970; Seltzer & Pandya, 1978; Pandya & Yeterian, 1985). Primary sensory cortices project (primarily via pyramids in layer III) to the surrounding early association cortices. These connections are reciprocated, primarily by pyramids in layer V (Rockland & Pandya, 1979; Barbas, 1986). Early association cortices (e.g. Brodmann areas 18, 19) project to higher-order cortices; these connections are also reciprocated. Eventually the polymodal sensory association cortices converge on the entorhinal cortex (Kosel et al., 1982; Van Hoesen, 1982). There are also afferents to the EC from limbic (cingulate gyrus) and unimodal sensory association cortices. The EC is essentially the only portal from the posterior cerebral cortex into the hippocampus. Thus, the entorhinal cortex is the paradigm of a convergence region, in the framework of Damasio & Damasio (1994). The hippocampus proper is an allocortical structure with distinct subfields (see Chapter 2). Area CA4 (CA=cornu ammonis, Ammon’s horn) is found within the hilus of the dentate gyrus. Proceeding around Ammon’s horn to the subiculum, one sequentially encounters the fields CA3, CA2, and then CA1. The subiculum also has recognized subdivisions, which are interposed between the hippocampus proper and the entorhinal cortex: presubiculum, prosubiculum, parasubiculum. Although the hippocampus projects to several structures, including the amygdala and the hypothalamus, it has essentially only one cortical projection: via the subiculum, to layer IV of the entorhinal cortex. Thus, anatomical evidence strongly suggests that the entorhinal cortex is in a strategic location with respect to hippocampal-posterior


cortical interactions, as all direct communication between these structures bottlenecks there. The amygdala is a subcortical limbic structure, which lies immediately anterior to the hippocampus. It has connections with several structures related to memory: entorhinal cortex, hippocampus, basal nucleus of Meynert, dorsomedial nucleus of the thalamus. Like the entorhinal cortex, it is also an area of convergence of multimodal information from all sensory modalities. There are major connections from the amygdala to the hypothalamus and to the autonomic centres of the brainstem. Although the role of the amygdala in memory is not fully understood, it is fair to say this structure is indispensable in the processing of stimuli with affective significance. Early pathological changes in AD isolate the hippocampus The earliest Alzheimer changes may occur in the pre-alpha layer of area 35. At this stage, all indications are that the condition is asymptomatic. It is not known whether there are measurable memory impairments resulting from an isolated, complete lesion of perirhinal cortex (let alone this layer in particular), although lesions of the perirhinal and entorhinal cortex in animals lead to amnesia (Zola-Morgan et al., 1989). As discussed earlier, patients presenting with amnesic MCI have pathology that has moved beyond the transentorhinal stage to also involve the entorhinal cortex. Involvement of layers II and IV of the entorhinal cortex therefore correlates with the appearance of amnesia in AD. Layer II contains most of the cells of origin of the strategic perforant pathway, and layer IV receives the subicular afferents that constitute the only cortical projection of the hippocampus. Hyman and colleagues have convincingly demonstrated the invariable involvement of the perforant pathway’s cells of origin (Hyman et al., 1986b) and projections (Hyman et al., 1988a,b) in AD. In the hippocampus proper, the CA1/subicular zone and especially the prosubiculum is invariably heavily involved with SPs and NFTs in AD. It is this hippocampal zone that sends the projection to the entorhinal cortex. By precluding hippocampalcortical interaction, these early changes of AD are likely to functionally isolate the hippocampus and contribute to the early and prominent amnesia so characteristic of AD. These relationships illustrate the principle that NFTs mark the dysfunctional neural systems in AD. Changes in amygdala are also likely to take a toll. Certain nuclei of this complex structure are severely and consistently affected (Kromer-Vogt et al., 1990). The most severely affected nuclei, in terms of numbers of NFTs, are the accessory basal and cortical nuclei. The mediobasal nucleus is moderately affected and the medial, lateral, laterobasal and central nuclei are relatively spared. SPs are found in a

somewhat different distribution: accessory basal and medial basal > cortical, lateral, laterobasal, and medial nuclei (Hyman et al., 1990). Of the nuclei which project heavily to the hippocampus, accessory basal, and lateral, the accessory basal is severely affected with NFTs. The accessory basal nucleus is also one of the major sources of afferents to the entorhinal cortex, which terminate in layer III. A4 amyloid protein deposition in this layer has been demonstrated. Hippocampal and entorhinal projections to the amygdala arise from EC layer IV and from the CA1/subiculum zone, both of which are severely affected by AD and contain numerous NFTs. In summary, the cells of origin of most of the interconnecting projections of the EC, the hippocampal formation, and the amygdala are vulnerable in AD and are marked by NFT formation. Specifically affected are the perforant pathway, the projection from EC layer IV to amygdala, and the projections of the prosubiculum to the entorhinal cortex and amygdala. The association regions In general, the syndrome of primary dementia (dementia without elementary neurologic signs) arises from involvement of limbic and association cortex with sparing of primary cortex and subcortical structures. The dementia of Alzheimer type, which presents a profile of primary dementia with salient amnesia, arises because of the anatomic profile of involvement, in which mesial temporal limbic and temporal association regions are affected most severely. Indeed, in these regions, Alzheimer’s disease appears to affect selectively the very neurons upon which large scale cortico-cortical interconnectivity depends. NFTS are found primarily in the large pyramids of layer III, which feed forward to higher order association cortices, and in the large pyramids of layer V, which feed back to earlier association cortices (Rockland & Pandya, 1979). Data supporting the presence of a vulnerable subpopulation of neurons that respects the hierarchical arrangement of corticocortical connectivity has also been reviewed by DeLacoste and White (1993; see also the commentary which follows). The pattern of cognitive impairment of moderately advanced Alzheimer’s disease (e.g. Braaks’ first isocortical stage) might have been predicted from the hierarchical vulnerability of the neocortical association cortices. The pattern is similar to that which results from focal and stable damage to the same structures. The similarity is well illustrated by a patient registered with the Division of Cognitive Neuroscience at the University of Iowa, who suffered selective, bilateral destruction of the following brain structures due to herpes simplex encephalitis: entorhinal cortex, hippocampus, amygdala, temporal pole (BA 38), anterolateral

17

18


Fig. 1.4. The pattern of anatomic involvement in a patient with herpes encephalitis. The figure shows axial T1 -weighted MR images of the brain of Boswell, in the chronic epoch following herpes simplex encephalitis. Note that the disease destroyed most of the limbic and paralimbic areas, including the hippocampal formation, amygdala, temporal pole, insula, and orbitofrontal region bilaterally, together with adjacent inferotemporal and right prefrontal regions. The distribution of these lesions resembles the distribution of neurofibrillary tangles in AD.

and anteroinferior temporal cortices (BA 20, 21, anterior 22, portions of 37), the entire basal forebrain (septum, nucleus basalis of Meynert, nucleus accumbens) and the posterior orbitofrontal cortex (Fig. 1.4). The pattern of involvement in this patient, known as Boswell, is strikingly analogous to the regional pattern of involvement in Alzheimer’s disease, in which the mesial temporal structures and temporal pole are severely affected, the area known as IT (areas 20, 21, and 37) and the posterior orbitofrontal cortex are also affected, though less severely, and the early sensory and motor cortices are relatively spared. Boswell’s cognitive impairments include complete anterograde amnesia for factual knowledge, retrograde amnesia for all unique entities and events, and a deficit of knowledge for certain categories of nonunique factual knowledge. Boswell’s ability to learn motor skills, his basic perceptual and motor abilities, his ability to sustain attention, and semantic knowledge are relatively unimpaired. Cognition aside, neurological examination discloses no abnormalities. There is a striking overall correspondence between Boswell’s deficits and those of Alzheimer patients. AD patients are, as a rule, free of elementary motor and sensory impairments, as might be expected from the virtual sparing of primary sensory and motor cortices. Like Boswell, these patients have a relatively intact capacity for the acquisition of perceptuomotor skills whose retrieval does not require the generation of a conscious internal representation (i.e. procedural or motor learning). Finally, AD patients display defective knowledge of unique objects and events, while manifesting relative sparing of categorical knowledge. They fail to recognize faces of family members and forget their names, but have no difficulty recognizing the gender of

faces, for example. AD patients also manifest higher order sensory and motor impairments (agnosia and apraxia), consistent with the involvement of the highest-order association cortices. Cortical and subcortical regions related to autonomic function Chu and colleagues have related the laminar pattern of NFT formation in ventromedial frontal cortices to abnormalities of autonomic function in Alzheimer patients (Chu et al., 1997). For example, area 25, the posterior orbito-frontal cortices, and the anterior insula are important in central autonomic regulation, via afferents from layer V. These connections are disrupted by neurofibrillary pathology in layer V. In behavioural studies, Chu found that AD patients fail to develop sympathetic skin conduction responses to emotionally charged visual stimuli, as do patients with focal lesions in these same cortices. These same patients exhibit defective personal and social decision making as their prominent behavioural abnormality. The findings are in keeping with the somatic marker hypothesis advanced by Damasio and colleagues, which postulates the reactivation of visceral and musculoskeletal states as a component in the process of normal personal and social decision-making (Damasio et al., 1991). AD also involves brainstem sites. Parvizi and colleagues (1998, 2000, 2001) have recently systematically surveyed the brainstem in 32 patients with AD and 26 normal controls. (see also Rub et al., 2001). As is the case in the cerebrum, brainstem structures were involved selectively by neurofibrillary tangles and neuritic plaques. Most of the changes were in the upper brainstem (Fig. 1.5). Most of the involved nuclei displayed neurofibrillary


Fig. 1.5. Alzheimer pathology in the brainstem. The Figure displays the distribution of Alzheimer changes in the periaqueductal grey matter (PAG) and nearby structures charted from thioflavin S-stained sections (yellow circles are NFTs, green diamonds are dense core senile plaques, and orange squares are diffuse plaques). The PAG is affected mainly by plaques, not tangles, which localize in these sections to the dorsal raphe nuclei (DRN) and the Edinger–Westphal (EW) nucleus. (Reprinted from Parvizi et al. (2000) Ann. Neurol., 48, 344–53, Fig. 2, with permission from John Wiley.)

tangles, or plaques, but rarely both. Many of the involved nuclei are involved in regulation of autonomic function, including the parabrachial nucleus, the periaqueductal grey, the Edinger–Westphal nucleus, the intermediate reticular zone, the dorsal motor nucleus of the vagus, and the nu-

cleus tractus solitarius. Other involved nuclei are sources of ascending projection systems, for example, the raphe nuclei. The rostral raphe nuclei were severely involved, while the caudal raphe nuclei were completely uninvolved. The brainstem seems to offer clues about the factors underlying

19

20


Progressive dysexecutive state

Progressive apraxia

Progressive higher visual disorder

Progressive aphasia

Fig. 1.6. Four patterns of cerebral lobar atrophy. The Figure displays T1 -weighted MR images from four patients with focal cerebral atrophy, after 3D reconstruction and digital deletion of extracerebral structures. Clockwise from top left, the images depict cerebral atrophy in frontal-predominant frontotemporal lobar atrophy, paracentral atrophy in corticobasal degeneration, left perisylvian lobar atrophy, and posterior cortical atrophy, in which the salient cognitive impairments were executive dysfunction, apraxia, aphasia, and higher visual impairment, respectively. Note the sparing of the pre- and postcentral gyri in frontotemporal dementia and progressive aphasia, indicating that these processes respect the distinction between primary and association cortex.

selective vulnerability to AD. Its involvement may also explain behavioural, affective, and autonomic manifestations of AD. Patterns of neocortical involvement in variant presentations The modal presentation of AD is amnesic, but variant presentations also occur, and the distribution of pathology in these cases is pertinent to this discussion. The best characterized variant is the visual variant (aka posterior variant) of AD, which is by far the most common cause of the syndrome of posterior cortical atrophy (Fig. 1.6). These cases of AD present with

higher visual disturbances, rather than with amnesia (Cogan, 1985; Hof et al., 1990a; Hof & Bouras, 1991; Levine et al., 1993). The small number of such cases in which systematic neuropathological observations have been made has revealed an atypical distribution of neocortical neurofibrillary tangles, and contributes substantially to the confidence with which neuropsychological deficits can be correlated with regional Alzheimer lesion burden. Levine and colleagues (Levine et al., 1993) reported the case of a 59-year-old executive whose first difficulty was reading. At the time of his first neurological examination, 2 years later, problems in locating and identifying objects by


sight dominated the neuropsychological profile. Although the history indicated subtle memory difficulty, no objective memory impairment was detected by the examiners. His illness pursued a 12-year, relentlessly progressive course which was remarkable for progressive visual dysfunction and the relatively late development of marked amnesia and aphasia. Careful neuropsychological and psychophysical observations were made during his illness. At autopsy, striking occipitoparietal atrophy was found. NFTs were abundant in the primary (area 17) and association (areas 18 and 20) visual cortices, the posterior cingulate gyrus, hippocampus, amygdala, and temporal isocortex. There were far fewer tangles in the frontal lobes than in the parietal and occipital lobes, which is the reverse of the usual pattern. Beta-amyloid immunoreactive plaques were most numerous in visual association cortex and frontal cortex, but neuritic plaques made up high percentages of the plaques in the primary visual cortex and visual association cortex. The unusual occurrence of large numbers of NFTs and NPs in the primary visual cortex, and the disproportionate numbers of NFTs and NPs in the visual association cortices, in this case correlate well with the patient’s well-documented clinical course. Hof and colleagues reported eight cases pathologically similar to the case of Levine et al. Although far less clinical information was reported, they were all said to have Balint’s syndrome. The pathological findings were homogeneous (Hof et al., 1990a): (i) Brodmann’s areas 17, 18, 19 had much higher numbers of NFTs and NP than did typical AD control cases. In the deep layers of area 17, the number of NFTs was 63-fold higher than AD controls. Moreover, there was a striking dropout of Meynert cells in layers V and VI of the primary visual cortex, cells which are normally not affected in AD. (ii) There were greatly increased numbers of NFTs in the superior colliculus, relative to AD controls. (iii) There were significantly fewer NFTs in prefrontal cortices (areas 9, 45, 46) relative to AD controls. (iv) Parietal area 7b and the posterior cingulate (area 23) had higher incidences of NPs than AD controls. The hippocampus was severely affected in all subjects. This group had previously shown that the affected pyramidal cells in the occipital cortex in AD are those giving rise to long corticocortical connections (Hof & Morrison, 1990). Hof and Bouras (1991) have also reported a case of an 89-year-old Alzheimer patient who was found to have a prominent defect in visual object recognition at the time of diagnosis, in the context of ‘slight memory disturbances and disorientation’. Tactile agnosia and prosopagnosia occurred later in the course. At autopsy, disproportionate occipito-temporal atrophy was found. The distribution of NFTs was similar to Hof’s cases of AD-Balints, with

the following exceptions: In the middle temporal gyrus (BA 21), there were more NFTs than AD controls and in the posterior parietal cortex (BA 7b), strikingly fewer NFTs than controls (Hof & Bouras, 1991). The authors interpreted their findings to reflect relative interruption of the ventral visual stream in the case of AD-visual agnosia, and relative interruption of the dorsal stream in AD-Balint’s. Although more cases with careful antemortem neuropsychological testing will be required to put this interpretation on a solid footing, their cases do help to establish the visual variant of AD. Galton et al. (2000) also described six pathologically confirmed cases with aphasic presentations, both fluent and non-fluent. All had severe temporal lobe involvement, almost always left-lateralized. All had some degree of memory and visuospatial impairment, and all had mesial temporal lobe tangles, though mesial temporal pathology in the two non-fluent cases was less impressive than neurofibrillary pathology in the temporal neocortex. They also described two ‘biparietal cases’, presenting with alexia, agraphia and apraxia. These patients also had severe mesial temporal involvement, but the parietal regions were affected more than other neocortical areas. Johnson et al. (1999) described three Alzheimer patients with early and salient impairment on tests of executive function. These patients were not distinguished from standard AD patients on other tests. At autopsy they had more tangles in frontal cortex than in entorhinal cortex, the reverse of the usual Alzheimer pattern. Yet another unusual Alzheimer variant was reported by Jagust and colleagues (Jagust et al., 1990). This patient manifested a clinical course that was generally typical for AD, but was exceptional because of a progressive left hemiparesis with associated signs of corticospinal tract dysfunction. At autopsy, in addition to advanced and classically distributed AD pathology, there was focally severe atrophy of the right postcentral gyrus (BAs 3,1,2). This area was invested with numerous NFTs. As with the visual variant cases, the exception proves the rule: atypical clinical findings were explained by an atypical distribution of pathology. Crystal and colleagues reported a case of a woman presenting with left-sided asterognosis and pseudoathetosis and preserved memory, in whom a right frontal biopsy revealed AD. A CT scan showed cortical atrophy which was worse on the right side. Further pathological data was not obtained (Crystal et al., 1982). Thus the increasingly well-documented cases of variant Alzheimer presentations show that a broad range of cognitive presentations is possible, and at the same time

21

22


reinforce the idea that the cognitive impairments in AD are related to the anatomic distribution of cortical pathology. Although dramatically focal cases are rare (Hof culled the eight cases of AD-Balints from 2500 autopsies), minor differences in cognitive profiles between AD patients are common, and some investigators have also tried to relate specific aspects of these profiles to regional pathology. For example, a measure of apathy has been correlated with tangle density in the anterior cingulate, and a measure of agitation has been correlated with tangle density in the orbitofrontal cortex (Tekin et al., 2001; Migneco et al., 2001). The degree of dyscalculia has been correlated with metabolic reductions in the left parietal and inferior temporal regions (Hirono, et al., 1998), and the degree of spatial and temporal disoriention with reductions in the right mesial parietal (areas 7 and 23) and the hippocampus (Giannakopoulos et al., 2000). Many similar studies are under way.

Are plaques or tangles better correlated with clinical symptoms of Alzheimer’s disease? Which manifestation of AD, NFTs or SPs, is more directly related to AD’s clinical symptoms? If the amnesia of AD is any indication, it appears that the distribution of the neurofibrillary changes provides a good correlate to the neuropsychological impairments of AD patients, at least in the early stages of AD. The degree to which amyloid deposition contributes is unclear. For one thing, regions containing amyloid plaques without dystrophic neurites (striatum, cerebellum, primary cortex), do not appear to be dysfunctional. In addition, there are well-documented cases of non-demented elderly individuals with extensive SPs in neocortex. SPs are distributed less extensively and specifically than NFTs. A quantitative study failed to establish a significant correlation between SP burden and severity of dementia, whereas the correlation between NFT burden and severity of dementia was highly significant (Arriagada et al., 1992a). Wilcock and Esiri (1982) and Bouras et al. (1994) have reported similar findings. This last finding is discrepant in relation to the seminal work of Tomlinson, Blessed, and colleagues, the first to show a relationship between SPs and dementia by regression analysis. One explanation for this discrepancy might be the inclusion of non-demented individuals in the sample in Blessed’s series. When only affected individuals are considered, a trend is present but correlation does not reach statistical significance (Terry et al., 1991; Arriagada et al., 1992b). Moreover, these investigators did not attempt to correlate tangles with dementia.

Note that the idea that NFTs are markers of the neural systems which are dysfunctional in AD does not address the question of which lesion type is most closely related to the aetiology of AD. It is possible that NFT formation is a cellular reaction to a more proximate insult (i.e. amyloid deposition) that occurs in only some vulnerable cell populations, and leads to the cognitive dysfunction as a secondary phenomenon (Selkoe, 2000 ). Specifically, it is possible that only a certain class of neurons, large corticocortical projecting pyramidal cells, react in this characteristic way. Some investigators have called attention to a poorer correlation between plaques and tangles and cognitive impairment in the later stages of AD (Terry et al., 1991). These investigations demonstrated that neuronal and synaptic loss made a better correlate (Neary et al., 1986) of global neuropsychological indices in these cases. This possibly reflects the failure of NFTs and plaques to accumulate with time in association cortex (Mann et al., 1988). Another possibility is that not all neurons destined to degenerate develop NFTs. Nevertheless, it appears to be the population of cells which is marked by NFT formation which drops out.

Reconciling findings from functional neuroimaging Neuroimaging studies are also informative about the anatomic basis of dementia (Chapter 8). As mentioned above structural imaging studies using voxel-based morphometry have generated findings in early AD that are compatible with the post-mortem studies discussed earlier. Complementary information is generated by functional imaging studies. For example, functional imaging with PET has identified metabolic asymmetries in individuals with AD, which correlate with neuropsychological findings (Foster et al., 1983). Haxby and colleagues reported a longitudinal study of 11 demented subjects. Right–left asymmetries of parietal glucose metabolism, identified with positron emission tomography, correlated significantly with neuropsychological discrepancies between visuospatial and language abilities in these patients. The detected asymmetries were stable over time, or tended to become more pronounced (Haxby et al., 1990). It should be noted, however, that the brain regions implicated in early AD by functional imaging studies differ in some important ways from those implicated by structural imaging studies. It has been known for some time that the salient functional imaging correlate of AD is hypometabolism in the parietal and temporal association regions (Duara et al., 1986). More recent findings confirm this finding, but suggest that metabolic reduction may occur even earlier in the posterior cingulate region (Minoshima


et al., 1997). There is a discrepancy between these findings and those of pathological and voxel-based MR morphometry studies, which implicate the mesial temporal lobes as the site of earliest change. A similar point was made by Mega et al. (1997), who reported a poor correlation between metabolic reductions assessed by PET, and NFT density assessed by histological examination of cryosectioned postmortem tissue slices from the same patient. (The patient, who had severe dementia (CDR 3), had previously arranged to donate his body to the UCLA medical school; the PET scan was performed when death appeared imminent.) More sensitive image analysis techniques seem to show that the discrepancy is partly one of degree. Voxel-based morphometry and related structural MRI approaches do show posterior cingulate and temporoparietal atrophy in addition to mesial temporal atrophy in presymptomatic individuals (Fox et al., 2001), and in patients with mild AD (Baron et al., 2001). Nevertheless, in PET images posterior cingulate and association area reductions are more prominent than mesial temporal reductions, particularly in the presymptomatic and MCI phases. Although partial volume effects may make parahippocampal reductions difficult to detect by visual inspection of PET images, they are marginal even in quantitative analyses (Johannsen et al., 2000). The discrepancy is probably explainable by the fact that functional images mainly reflect metabolic activity at those locations where metabolic demands are greatest, that is, at the nerve terminals, rather than at the neuronal cell bodies (Schwartz et al., 1979; Logothetis et al., 2001). In other words, loss of a neuron cell body might reduce metabolic activity most at its distant projection sites. Thus, association area reductions may reflect, at least in part, the effects of loss of neurons in locations which project to them, notably the mesial temporal regions. Direct evidence for this idea was provided by Meguro et al. (1999), who showed in monkeys that entorhinal and perirhinal ablation causes temporoparietal, posterior cingulate, and parahippocampal hypometabolism. Others (Mielke et al., 1996; Shukla & Bridges, 1999; Matsuda et al., 2002) have suggested the same interpretation. One of the important contributions of functional imaging studies has been to focus attention on the posterior cingulate. Because focal lesions of the area are very rare, the consequences of loss of function of this region are not completely understood (but see Damasio, 1999 for an account of the role of this region and the surrounding medial parietal cortex in consciousness). Another important contribution has been to demonstrate directly that damage to one anatomic region can have dramatic functional effects on remote regions (Meguro et al., 1999). Finally, functional

imaging studies regularly demonstrate the systematic involvement of association areas and systematic sparing of primary cortical areas in AD. The anatomic basis of cognitive impairment in frontotemporal lobar atrophy Unlike AD, in which the pathology is similar in character and distribution across cases, the pathology underlying FTLD varies appreciably. A minority of these cases have distinctive intracellular inclusions (e.g. Pick bodies), but most do not. It is difficult to predict the precise pathology from the clinical presentation, and sometimes the diagnosis turns out to be another degenerative condition not usually classified as a frontotemporal lobar atrophy, such as PSP or CBD. Classification and diagnosis of this cluster of diseases, most of which involve deposition or defective expression of tau (Feany & Dickson, 1996; Zhukareva et al., 2001), is an area of active research. FTLD, like AD, is characterized by disproportionate involvement of association and limbic areas, with relative sparing of subcortical and primary cortical areas. It is usually not accompanied by elementary neurologic signs. In FTLD, the brunt of the degenerative process falls on prefrontal and/or anterior temporal cortical regions, though other association areas are not altogether spared. Unlike AD, in which it is unusual to have more than a 20% interhemispheric difference in lesion burden, FTLD is regularly asymmetric, both in terms of the relative degree of involvement of the right and left hemispheres and the relative degree of frontal and temporal involvement. No pathological staging system corresponding to that of Braak and Braak is currently available for FTLD. Although there are few studies relating the cognitive deficits to the quantitative histopathological index (but see Arnold et al., 2000), there are studies correlating the deficits to the degree of atrophy that is present. Recently, computational imaging techniques (e.g. voxel-based morphometry) have been used to demonstrate the consistent correlates of clinical syndromes (e.g. semantic dementia) across groups of patients (Mummery et al., 2000; Rosen et al., 2002). The profile of cognitive impairment in FTLD is also much more variable than it is in AD. The phenotype may even vary within kindreds sharing the same tau gene mutations (Bird et al., 1999). Currently, FTLD is held to have three major clinical manifestations, namely frontal variant frontotemporal dementia, progressive non-fluent aphasia, and semantic dementia, but this nosology significantly understates the degree of variability in the presentations of FTLD. Although the clinical phenotype does not reliably predict the

23

24


underlying pathology, it does reliably predict the location of the brunt of that pathology. The pronounced clinical variance of FTLD makes it possible to inquire whether the profile of gross anatomic involvement relates systematically to the profile of cognitive impairment. There is an emerging consensus that this is indeed the case. We will consider each entity briefly. Frontal variant FTD is a syndrome of insidious alteration in personality, social conduct, emotionality, and aspects of executive function and attention. The inaugural manifestations contrast with those of AD, in which the typical patient is densely amnesic (Perry & Hodges, 2000). In more advanced disease, the degree of memory impairment is comparable in the two disorders. Frontal variant FTD is accompanied by atrophy of the dorsolateral and orbitofrontal prefrontal association regions. Left-lateralized prefrontal atrophy is accompanied by more language impairment than AD, and right-lateralized cases by more perseverative behaviour than AD (Razani et al., 2001). Patients with FTD have significantly less impairment of visuoconstruction than there is in AD, consistent with the relative sparing of parietal cortex. Progressive non-fluent aphasia is a disorder characterized by the insidious onset of non-fluent speech with agrammatism, phonemic paraphasias, and anomia, in the context of preserved executive and nonverbal function. Progressive non-fluent aphasia is accompanied by left frontal and temporal atrophy that affects the perisylvian region disproportionately. The left temporal pole and the left hippocampus and amygdala are only affected mildly (Mesulam, 1982, 2001). Semantic dementia is a syndrome of insidious and progressive language impairment in which there is a loss of the mapping between word forms and their meaning (Hodges et al., 1992; Hodges & Miller, 2001). Most commonly, the inaugural symptom is anomia, followed later by impaired comprehension and semantic paraphasias. Semantic memory loss also becomes prominent. Speech remains fluent. When the process is right-lateralized, anomia and semantic memory loss occur together, but the process is usually left-lateralized, and in this case, anomia is present out of proportion to any semantic impairment, at least in the early stages (Lambon Ralph et al., 2001). Semantic dementia is accompanied by temporal lobar atrophy that affects all anterior temporal structures, including the hippocampus, amygdala, entorhinal cortex, temporal pole, middle temporal gyrus, inferior temporal gyrus, and occipitotemporal gyrus. Among all these structures, the atrophy observes an anteroposterior gradient, worse anteriorly. In all cases, the anterior inferior and middle temporal gyri

are prominently involved (Chan et al., 2001). Although the process is usually worse on the left side, bilateral temporal atrophy occurs, as does a degree of ventromedial frontal atrophy. Frontotemporal dementia and progressive non-fluent aphasia strongly resemble the familiar cognitive syndromes that accompany focal, stable damage to the prefrontal regions and to the left anterior perisylvian region, respectively, and are associated with atrophy which is worst in these areas. The syndrome of semantic dementia does not have a well-characterized counterpart in the lesion literature, but patients with sequelae of herpes simplex encephalitis do commonly have semantic memory deficits when the process has damaged temporal polar and inferotemporal regions, the same portions of the temporal lobe affected in semantic dementia. Patients with damage confined to the left temporal pole (e.g. the many patients who have undergone left anterior temporal lobectomies) may have anomia, but usually do not display lexical semantic deficits (Saykin et al., 1995; Langfitt & Rausch, 1996; Tranel et al. 1997; Damasio et al., in press), but the atrophy in semantic dementia is not confined to the temporal pole or the left hemisphere. The anatomic basis of cognitive impairment in corticobasal degeneration Although not considered a syndrome of FTLD, CBD is another asymmetric lobar degenerative disorder that is accompanied by the accumulation of tau pathology. It is frequently accompanied by progressive ideomotor limb and buccal apraxia. It also commonly has elementary neurologic signs that include asymmetric cortical sensory loss, parkinsonism and myoclonus. Functional imaging studies of patients with progressive apraxia due to CBD showed metabolic reductions relative to AD in the paracentral region and thalamus; and relatively better metabolism in the posterior cingulate, mesial temporal, and orbitofrontal areas. Metabolic asymmetries were more pronounced in the CBD group than in the AD group (Hirono et al., 2000; Okuda et al., 2001). Structural imaging studies show atrophy of the paracentral area (Kitagaki et al., 2000), and histopathological studies have demonstrated tau-positive pathology in subcortical locations, as well as ballooned neurons, gliosis, and neuronal loss in parietal and precentral gyri. The involvement of the basal ganglia is inconsistent (Cordato et al., 2001). Thus CBD is exceptional among the lobar atrophies in that it involves sensorimotor cortex early but is no exception to the rule that dementias without elementary neurological signs spare primary cortex. As noted in a previous


section, series of clinically or pathologically defined cases of CBD have documented considerable regional overlap with FTLD and AD, and some of the imaging studies cited here may include patients with other diagnoses. The consistency of the relationship of paracentral atrophy to the syndrome of progressive apraxia underscores the notion that the profile of cognitive impairment and neurologic signs bears a closer relationship to the regional distribution of pathology than to the precise character of that pathology.

The anatomic basis of cognitive impairment in dementia with lewy bodies The hallmarks of DLB are parkinsonism, fluctuation of cognition and sensorium, and visual hallucinations. The profile of cognitive impairment is not well distinguished from AD, except that there is a greater impairment of attention and of visuoperception and visuoconstruction, and there may be less memory impairment than in AD (Connor et al., 1998). Interpretation of this cognitive profile is difficult because Alzheimer changes are commonly superimposed on DLB (i.e. the Lewy body variant of AD). In these patients, the contribution of the Lewy body pathology to the clinical picture has been difficult to determine (Stern et al., 2001). Pure DLB cases are less common, and, unfortunately, many studies do not distinguish adequately between these two forms of DLB. The distribution of cortical LBs has been studied in autopsied cases and it has been demonstrated that there is a consistent distribution of LBs, involving the following regions: substantia nigra > entorhinal cortex > insula > frontal cortex > hippocampus > occipital cortex (Gómez-Tortosa et al., 2000). However, there is no clear correlation between the regional distribution of LBs and the neuropsychological profile, and the number of LBs relates only weakly to disease duration. These authors did not find a consistent anatomical profile corresponding to cognitive fluctuation, or delusions. Moreover, they are present in far fewer numbers than NFTs and SPs are in AD. In cases of pure DLB, some studies have suggested that neuronal loss could not be detected in the superior temporal sulcus or in the entorhinal cortex (Gomez-Isla et al., 1997). Similar results had also been obtained by Samuel et al. (1997), who found no reduction in anti-synaptophysin reactivity in mid-frontal cortex in pure DLD cases.

Hallucinations In an interesting PET study of autopsy-defined cases of AD and DLB, Minoshima et al. found that, DLB was associated

with widespread association area reductions, including the posterior cingulate and parietal regions characteristic of AD, but that occipital cortical metabolism, including the calcarine cortex, is lower in DLB than in AD. The relationship was seen for both pure DLB and Lewy body variant AD (Minoshima et al., 2001). This may reflect a correlate of the deficient visual processing in these patients, and perhaps of hallucinations. Harding et al. (2002) found that, among patients with DLB or IPD, the cases with well-formed visual hallucinations were associated with high densities of LBs in the amygdala, parahippocampal, and inferior temporal cortex. They concluded that the distribution of temporal lobe LBs was more related to visual hallucinations than to the degree of dementia per se.

Cholinergic hypothesis revisited There is a cholinergic deficiency in DLB, as in AD. This deficiency is not due to superimposed AD. In fact, the deficiency is more prominent than it is in AD, and equally or more pronounced in patients with pure DLB as in those with Lewy body variant AD (Samuel et al., 1997; Tiraboschi et al., 2000). Lippa et al. (1994) and other investigators have reported that LBs occur in the basal nucleus of Meynert. Moreover, there have been a number of reports of amelioration of the fluctuation and psychiatric symptoms in DLB with cholinergic augmentation (McKeith et al., 2000). These data suggest a role for cholinergic deafferentation in DLB. Given that DLB is distinguished by neuropsychiatric symptoms, and that these symptoms respond significantly to cholinergic agents, it is of interest that cholinergic augmentation may be particularly helpful for neuropsychiatric symptoms in AD, too. Cummings and Kaufer (1996) proposed that the cholinergic deficit in AD contributes substantially to the neuropsychiatric symptoms in that disorder. Minger et al. (2000) found significant correlations between loss of cholinergic markers and psychiatric symptoms in patients with AD. In summary, in DLB, cortical Lewy bodies occur in a hierarchical distribution, and are distributed somewhat differently than NFTs are in AD. It has been difficult to correlate specific aspects of DLB with specific locations of LBs. Visual cortex hypometabolism and inferotemporal cortex Lewy bodies may somehow underlie formed visual hallucinations, but more investigations are needed. Evidence of cholinergic deficiency, and resolution of some of the hallmark symptoms in some patients after treatment with cholinomimetic agents, suggest that cholinergic system pathology plays an important role. Progress has been hampered by the relative rarity of pure DLB cases,

25

26


difficulty quantitating fluctuation, and the inconsistent attention paid to the basal forebrain and cholinergic markers. The anatomic basis of cognitive impairment in dementia with Creutzfeldt–Jakob disease In contrast to AD, FTLD, CBD, and DLB, Creutzfeldt– Jakob disease (CJD) does not present a hierarchical profile of anatomic involvement, or a stereotyped profile of cognitive dysfunction, and clinicopathologic correlation cannot be carried out at the group level. Case-by-case correlation has been severely hampered by the absence of reliable changes on classic structural neuroimaging studies. The disease progresses rapidly, and by the time the patient is examined at autopsy, end-stage changes are usually present. However, the cognitive profile is often focal early in the course. One well described presentation is the Heidenhain variant, characterized by severe higher visual dysfunction. It has been shown recently that diffusion-weighted MRI is very sensitive to regional cortical changes in CJD, correlating well with pathological findings (Mittal et al., 2002) and with metabolic reductions on functional imaging studies (Na et al., 1999). Preliminary evidence therefore suggests that, in this disorder too, the pattern of anatomic involvement determines the profile of cognitive dysfunction.

Conclusions The various disorders causing cortical dementia are associated with different clinical profiles and different profiles of involvement of association and limbic regions. The associations of specific impairments with specific profiles of cortical involvement recapitulate many of the associations that have been observed in focal and stable brain damage. This is true even for AD, which was traditionally considered a diffuse degenerative disease, but in which NFTs mark specific dysfunctional neuralsystems and regions. The anatomic correlates of variant cognitive presentations of AD, FTLD and CBD show that the anatomic pathological profile is altered as might be predicted by the neuropsychological profile. Thus the cognitive deficits which characterize these diseases are explained by selective disruption of neural systems. The clinical principles according to which symptoms are correlated with cortical pathology in focal brain damage are also applicable to degenerative disease.

Acknowledgements Supported by a grant from the Mathers Foundation to ARD.

REFERENCES Adams, R. D. & Victor, M. (1989). Principles of Neurology, 4th edn. New York: McGraw-Hill. Albert, M. L., Feldman, R. G. & Willis, A. L. (1974). The ‘subcortical dementia’ of progressive supranuclear palsy. Journal of Neurology, Neurosurgery and Psychiatry, 37, 121–30. Arnold, S. E., Hyman, B. T., Flory, J., Damasio, A. R. & Van Hoesen G. W. (1991). The topographical and neuroanatomical distribution of neurofibrillary tangles and neuritic plaques in the cerebral cortex of patients with Alzheimer’s disease. Cerebal Cortex, 1, 103–16. Arnold, S. E., Han, L.-Y., Clark, C. M., Grossman, M. & Trojanowski, J. Q. (2000). Quantitative neurohistological features of frontotemporal degeneration. Neurobiology of Aging, 21, 913–19. Arriagada, P. V. & Hyman, B. T. (1990). Predominance of A-4 over Alz-50 positive plaques in demented elderly (abstract). Neurobiology of Aging, 11. Arriagada, P. V., Growdon, J. H., Hedley-Whyte, E. T. & Hyman, B. T. (1992a). Neurofibriallary tangles but not senile plaques parallel duration and severity of Alzheimer’s disease. Neurology, 42, 631–9. Arriagada, P. V., Marzloff, K. & Hyman B. T. (1992b). Distribution of Alzheimer-type pathologic changes in nondemented elderly individuals matches the pattern in Alzheimer’s disease. Neurology, 42, 1681–8. Ballard, C., O’Brien, J., Gray, A. et al. (2001). Attention and fluctuating attention in patients with dementia with Lewy bodies and Alzheimer disease. Archives of Neurology, 58, 977–82. Bancher, C., Brunner, C., Lassmann, H. et al. (1989). Accumulation of abnormally phosphorylated tau precedes the formation of neurofibrillary tangles in Alzheimer’s disease. Brain Research, 477, 90–9. Barbas, H. (1986). Pattern in the laminar origin of corticocortical connections. Journal of Comparative Neurology, 252, 415–22. Baron, J. C., Chetelat, G., Desgranges, B. et al. (2001). In vivo mapping of gray matter loss with voxel-based morphometry in mild Alzheimer’s disease. NeuroImage, 14, 298–309. Bartus, R. T., Dean, R. L. III, Beer, B. & Lippa, A. S. (1982). The cholinergic hypothesis of geriatric memory dysfunction. Science, 217, 408–14. Bird, T. D., Nochlin, D., Poorkaj, P. et al. (1999). A clinical pathological comparison of three families with frontotemporal dementia and identical mutations in the tau gene (P301L). Brain, 122, 741–56. Blessed, G., Tomlinson, B. E. & Roth, M. (1968). The association between quantitative measures of dementia and of senile change in the cerebral grey matter of elderly subjects. British Journal of Psychiatry, 114, 797–811. Boeve, B. F., Maraganore, D. M., Parisi, J. E. et al. (1999). Pathologic heterogeneity in clinically diagnosed corticobasal degeneration. Neurology, 53, 795–800. Bouras, C., Hof, P. R., Giannakopoulos, P., Michel, J.-P. & Morrison, J. H. (1994). Regional distribution of neurofibrillary tangles and senile plaques in the cerebral cortex of elderly patients: a


quantitative evaluation of a one-year autopsy population from a geriatric hospital. Cerebral Cortex, 4, 138–50. Braak, H. & Braak, E. (1985). On areas of transition between entorhinal allocortex and temporal isocortex in the human brain. Normal morphology and lamina-specific pathology in Alzheimer’s disease. Acta Neuropathologica, 68, 325–32. (1988). Neuropil threads occur in dendrites of tangle-bearing nerve cells. Neuropathology and Applied Neurobiology, 14, 39–44. (1990a). Alzheimer’s disease: Striatal amyloid depositis and neurofibrillary changes. Journal of Neuropathology and Experimental Neurology, 49, 215–24. (1991). Neuropathological staging of Alzheimer-related changes. Acta Neuropathologica, 82, 239–59. Brun, A. & Gustafson, L. (1976). Distribution of cerebral degeneration in Alzheimer’s disease. Acta Psychiatrica Nervenkr, 223, 15–33. Brun, A. & Passant, U. (1996). Frontal lobe degeneration of nonAlzheimer type. Structural characteristics, diagnostic criteria and relation to other frontotemporal dementias. Acta Neurologica Scand Suppl, 168, 28–30. Callen, D. J., Black, S. E., Gao, F., Caldwell, C. B. & Szalai, J. P. (2001). Beyond the hippocampus: MRI volumetry confirms widespread limbic atrophy in AD. Neurology, 57, 1669–74. Chan, D., Fox, N. C., Scahill, R. I. et al. (2001). Patterns of temporal lobe atrophy in semantic dementia and Alzheimer’s disease. Annals of Neurology 49, 421–3. Chu, C. C., Tranel, D., Damasio, A. R. & Van Hoesen, G. W. (1997). The autonomic-related cortex: pathology in Alzheimer disease. Cerebral Cortex, 7, 86–95. Chui, H. C., Teng, E. L., Henderson, V. W. & Moy, A. C. (1985). Clinical subtypes of dementia of the Alzheimer type. Neurology, 35, 1544–50. Cogan, D. G. (1985). Visual disturbances with focal progressive dementing disease. American Journal of Ophthalmology, 100, 68–72. Connor, D. J., Salmon, D. P., Sandy, T. J., Galasko, D., Hansen, L. A. & Thal, L. J. (1998). Cognitive profiles of autopsy-confirmed Lewy body variant vs pure Alzheimer disease. Archives of Neurology, 55, 1352. Cordato, N. J., Halliday, G. M., McCann, H. et al. (2001). Corticobasal syndrome with tau pathology. Movement Disorders, 16, 656–67. Corkin, S. (1984). Lasting consequences of bilateral medial temporal lobectomy: clinical course and experimental findings. Seminars in Neurology, 4, 249–59. Crystal, H. A., Horoupian, D. S., Katzman, R. & Jotkowicz, S. (1982). Biopsy-proved Alzheimer disease presenting as a right parietal syndrome. Annals of Neurology, 12, 186–8. Crystal, H., Dickson, D., Fuld, P. et al. (1988). Clinico-pathologic studies in dementia: nondemented subjects with pathologically confirmed Alzheimer’s disease. Neurology, 38, 1682–7. Cummings, J. L. & Benson, D. F. (1992). Dementia: A Clinical Approach. Boston: Butterworth-Heinemann.

Cummings, J. L. & Kaufer, D. (1996). Neuropsychiatric aspects of Alzheimer’s disease: the cholinergic hypothesis revisited. Neurology, 47, 876–83. Damasio, A. R. (1989a). Time-locked multiregional retroactivation: a systems level proposal for the neural substrates of recall and recognition. Cognition, 33, 25–62. (1989b). The brain binds entities and events by multiregional activation from convergence zones. Neural Computation, 1, 123–32. (1999). The Feeling of What Happens. New York: Harcourt Brace & Company. Damasio, A. R. & Damasio, H. (1994). Cortical systems for retrieval of concrete knowledge: the convergence zone framework. In Large-Scale Neuronal Theories of the Brain, ed. C. Koch, Cambridge, pp. 61–74. MA: MIT Press. Damasio, A. R., Damasio, H., Tranel, D. & Brandt, J. P. (1990a). Neural regionalization of knowledge access: preliminary evidence. In Symposia on Quantitative Biology. Vol. 55, pp. 1039–47. New York: Cold Spring Harbor. Damasio, A. R., Van Hoesen, G. W. & Hyman, B. T. (1990b). Reflections on the selectivity of neuropathological changes in Alzheimer’s disease. In Modular Deficits in Alzheimer-Type Dementia, ed. M. F. Schwartz, pp. 83–99. Cambridge, MA: MIT Press. Damasio, A. R., Tranel, D. & Damasio, H. (1991). Somatic markers and the guidance of behavior: theory and preliminary testing. In Frontal Lobe Function and Dysfunction, ed. H. S. Levin, H. M. Eisenberg & A. L. Benton, pp. 217–99. New York: Oxford University Press. Damasio, H. & Damasio, A. R. (1989). Lesion Analysis in Neuropsychology. New York: Oxford University Press. Damasio, H., Tranel, D., Grabowski, T. J. et al. (2003). Neural systems behind word and concept retrieval. Cognition, in press. Davies, L., Wolska, B., Hilbich, C. et al. (1988). A4 amyloid protein deposition and the diagnosis of Alzheimer’s disease. Neurology, 38, 1688–93. Dayan, A. D. (1970). Quantitative histological studies on the aged human brain. Acta Neuropathologica, 16, 95–102. De Lacoste, M.-C. & White, C. L. (1993). The role of cortical connectivity in Alzheimer’s disease pathogenesis: a review and model system. Neurobiology of Aging, 14, 1–16. DeKosky, S. T., Ikonomovic, M. D., Styren, S. D. et al. (2002). Upregulation of choline acetyltransferase activity in hippocampus and frontal cortex of elderly subjects with mild cognitive impairment. Annals of Neurology, 51, 145–55. Delacourte, A., David, J. P., Sergeant, N. et al. (1999). The biochemical pathway of neurofibrillary degeneration in aging and Alzheimer’s disease. Neurology, 52, 1158–65. Duara, R., Grady, C., Haxby, J. et al. (1986). Positron emission tomography in Alzheimer’s disease. Neurology, 36(7), 879– 87. Duyckaerts, C., Hauw, J. J., Bastenaire, F. et al. (1986). Laminar distribution of neocortical senile plaques in senile dementia of the Alzheimer type. Acta Neuropathologica, 70, 249– 56.

27

28


Feany, M. B. & Dickson, D. W. (1996). Neurodegenerative disorders with extensive tau pathology: a comparative study and review. Annals Neurology, 40, 139–48. Foster, N. L., Chase, T. N., Fedio, P., Patronas, N. J., Brooks, R. A. & Di Chiro, G. (1983). Alzheimer’s disease: Focal cortical changes shown by positron emission tomography. Neurology, 33, 961–65. Fox, N. C., Crum, W. R., Scahill, R. I., Stevens, J. M., Janssen, J. C. & Rossor, M. N. (2001). Imaging of onset and progression of Alzheimer’s disease with voxel-compression mapping of serial magnetic resonance images. Lancet, 358, 201–5. Galton, C. J., Patterson, K., Xuereb, J. H. & Hodges, J. R. (2000). Atypical and typical presentations of Alzheimer’s disease: a clinical, neuropsychological, neuroimaging and pathological study of 13 cases. Brain, 123, 484–98. Geula, C., Mesulam, M. M., Saroff, D. M. & Wu, C. K. (1998). Relationship between plaques, tangles, and loss of cortical cholinergic fibers in Alzheimer disease. Journal of Neuropathology and Experimental Neurology, 57, 63–75. Giannakopoulos, P., Gold, G., Duc, M., Michel, J. P., Hof, P. R. & Bouras, C. (2000). Neural substrates of spatial and temporal disorientation in Alzheimer’s disease. Acta Neuropathologica (Berlin), 100, 189–95. Glees, P. & Griffith, H. B. (1952). Bilateral destruction of the hippocampus (cornu ammonis) in a case of dementia. Monatsschrift Fur ¨ Psychiatrie und Neurologie, 123, 193–204. Goedert, M. (1993). Tau protein and the neurofibrillary pathology of Alzheimer’s disease. Trends in Neurosciences, 16, 460–5. Gold, G., Bouras, C., Kovari, E. et al. (2000). Clinical validity of Braak neuropathological staging in the oldest-old. Acta Neuropathologica, (Berlin), 99, 579–82. Gomez-Isla, T., Hollister, R., West, H. et al. (1997). Neuronal loss correlates with but exceeds neurofibrillary tangles in Alzheimer’s disease. Annals of Neurology, 41, 17–24. Gómez-Tortosa, E., Irizarry, M. C., Gómez-Isla, T. & Hyman, B. T. (2000). Clinical and neuropathological correlates of dementia with Lewy bodies. Annals of NY Academy Sciences, 920, 9–15. Grimes, D. A., Lang, A. E. & Bergeron, C. B. (1999). Dementia as the most common presentation of cortical-basal ganglionic degeneration. Neurology, 53, 1969–74. Grober, E., Dickson, D., Sliwinski, M. J. et al. (1999). Memory and mental status correlates of modified Braak staging. Neurobiological Aging, 20, 573–9. Hachinski, V. C., Iliff, L. D., Zilhka, E. et al. (1975). Cerebral blood flow in dementia. Archives of Neurology, 32, 632–7. Harding, A. J., Broe, G. A. & Halliday, G. M. (2002). Visual hallucinations in Lewy body disease relate to Lewy bodies in the temporal lobe. Brain, 125, 391–403. Haxby, J. V., Grady, C. L., Koss, E. et al. (1990). Longitudinal study of cerebral metabolic asymmetries and associated neuropsychological patterns in early dementia of the Alzheimer type. Archives of Neurology, 47, 753–60. Hirono, N., Mori, E., Ishii, K. et al. (1998). Regional metabolism: associations with dyscalculia in Alzheimer’s disease. Journal of Neurology, Neurosurgey and Psychiatry, 65, 913–16.

Hirono, N., Ishii, K., Sasaki, M. et al. (2000). Features of regional cerebral glucose metabolism abnormality in corticobasal degeneration. Dementia and Geriatric Cognition Disorders, 11, 139–46. Hodges, J. R. & Miller, B. (2001). Frontotemporal dementia (Picks’s disease). In Early Onset Dementia: A Multidisciplinary Approach. pp. 284–303. Oxford: Oxford University Press. Hodges, J. R., Patterson, K., Oxbury, S. & Funnell, E. (1992). Semantic dementia. Progressive fluent aphasia with temporal lobe atrophy. Brain, 115, 1783–806. Hof, P. R. & Bouras, C. (1991). Object recognition deficit in Alzheimer’s disease: possible disconnection of the occipitotemporal component of the visual system. Neuroscience Letters, 122, 53–6. Hof, P. R. & Morrison, J. H. (1990). Quantitative analysis of a vulnerable subset of pyramidal neurons in Alzheimer’s disease: II. Primary and secondary visual cortex. Journal of Comparative Neurology, 301, 55–64. (1990a). Selective disconnection of specific visual association pathways in cases of Alzheimer’s disease presenting with Balint’s syndrome. Journal of Neuropathology and Experimental Neurology, 49, 168–84. Hof, P. R., Bierer, L. M., Perl, D. P., Delacourte, A. B. L., Bouras, C. & Morrison, J. H. (1992). Evidence for early vulnerability of the medial and inferior aspects of the temporal lobe in an 82 year old patient with preclinical signs of dementia. Archives of Neurology, 49, 946–53. Hulette, C., Nochlin, D., McKeel, D. et al. (1997). Clinicalneuropathologic findings in multi-infarct dementia: a report of six autopsied cases. Neurology, 48, 668–72. Hyman, B. T., Van Hoesen, G. W., Kromer, L. J. & Damasio, A. R. (1986b). Perforant pathway changes and the memory impairment of Alzheimer’s disease. Annals of Neurology, 20, 472–81. Hyman, B. T., Kromer, L. J. & Van Hoesen, G. W. (1988a). A direct demonstration of the performant pathway terminal zone in Alzheimer’s disease using the monoclonal antibody Alz-50. Brain Research, 450, 392–7. Hyman, B. T., Van Hoesen, G. W., Wolozin, B. L., Davies, P., Kromer, L. J. & Damasio, A. R. (1988b). Alz-50 antibody recognizes Alzheimer-related neuronal changes. Annals of Neurology, 23, 371–9. Hyman, B. T., Van Hoesen, G. W. & Damasio, A. R. (1990). Memoryrelated neural systems in Alzheimer’s disease: an anatomic study. Neurology, 40, 1721–30. Hyman, B. T., Arriagada, P. V., McKee, A. C., Ghika, J., Corkin, S. & Growdon, J. H. (1991a). The earliest symptoms of Alzheimer disease: anatomic correlates. Society for Neuroscience Abstracts, 352. Hyman, B. T., Flory, J. E., Arnold, S. E. et al. (1991b). Quantitative assessment of ALZ-50 immunoreactivity in Alzheimer’s disease. Journal of Geriatric Psychiatry and Neurology, 4, 231–35. Jagust, W. J., Davies, P., Tiller-Borcich, J. K. & Reed, B. R. (1990). Focal Alzheimer’s disease. Neurology, 40, 14–19. Joachim, C. L., Mori, H. & Selkoe, D. J. (1989). Amyloid -protein deposition in tissues other than brain in Alzheimer disease. Nature, 341, 226–30.


Johannsen, P., Jakobsen, J. & Gjedde, A. (2000). Statistical maps of cerebral blood flow deficits in Alzheimer’s disease. European Journal of Neurology, 7, 385–92. Johnson, J. K., Head, E., Kim, R., Starr, A. & Cotman, C. W. (1999). Clinical and pathological evidence for a frontal variant of Alzheimer disease. Archives of Neurology, 56, 1233–9. Jones, E. G. & Powell, T. P. S. (1970). An anatomical study of converging sensory pathways within the cerebral cortex of the monkey. Brain, 93, 793–820. Khachaturian, Z. S. (1985). Diagnosis of Alzheimer’s disease. Archives of Neurology, 42, 1097–105. Kitagaki, H., Hirono, N., Ishii, K. & Mori, E. (2000). Corticobasal degeneration: evaluation of cortical atrophy by means of hemispheric surface display generated with MR images. Radiology, 216, 31–8. Knopman, D. S., Christensen, K. J., Schut, L. J. et al. (1989). The spectrum of imaging and neuropsychological findings in Pick’s disease. Neurology, 39, 362–8. Kosel, K. C., Van Hoesen, G. W. & Rosene, D. L. (1982). Nonhippocampal cortical projections from the entorhinal cortex in the rat and rhesus monkey. Brain Research, 244, 202–14. Kowall, N. W. & Kosik, K. S. (1987). Axonal disruption and aberrant localization of Tau protein characterize the neuropil pathology of Alzheimer’s disease. Annals of Neurology, 22, 639–43. Kromer-Vogt, L. J., Hyman, B. T., Van Hoesen, G. W. & Damasio, A. R. (1990). Pathological alterations in the amygdala in Alzheimer’s disease. Neuroscience, 37, 377–85. Lambon Ralph, M. A., McClelland, J. L., Patterson, K., Galton, C. J. & Hodges, J. R. (2001). No right to speak? The relationship between object naming and semantic impairment: neuropsychological evidence and a computational model. Journal of Cognitive Neuroscience, 13, 341–56. Langfitt, J. T. & Rausch, R. (1996). Word-finding deficits persist after left anterotemporal lobectomy. Archives of Neurology, 53, 72–6. Levine, D. N., Lee, J. M. & Fisher, C. M. (1993). The visual variant of Alzheimer’s disease: a clinicopathologic case study. Neurology, 43, 305–13. Lewis, D. A., Campbell, M. J., Terry, R. D. & Morrison, J. H. (1987). Laminar and regional distributions of neurofibrillary tangles and neuritic plaques in Alzheimer’s disease: a quantitative study of visual and auditory cortices. Journal of Neuroscience, 7, 1799–808. Lippa, C. F., Smith, T. W. & Swearer, J. M. (1994). Alzheimer’s disease and Lewy body disease: a comparative clinicopathological study. Annals of Neurology, 35, 81–8. McKee, A. C., Kosik, K. S. & Kowall, N. W. (1991). Neuritic pathology and dementia in Alzheimer’s disease. Annals of Neurology, 30, 156–65. McKeith, I., Del Ser, T., Spano, P. et al. (2000). Efficacy of rivastigmine in dementia with Lewy bodies: a randomised, doubleblind, placebo-controlled international study. Lancet, 356, 2031–6. McKeith, I. G., Galasko, D., Kosaka, K. et al. (1996). Consensus guidelines for the clinical and pathologic diagnosis of dementia with Lewy bodies (DLB): report of the consortium on DLB international workshop. Neurology, 47, 1113–24.

McKhann, G., Drachmann, D., Folstein, M., Katzman, R., Price, D. & Stadlan, E. M. (1984). Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ARRDA Work Group under the auspices of Dept. of Health & Human Services Task Force on Alzheimer’s Disease. Neurology, 34, 940. Mann, D. M. A., Marcyniuk, B., Yates, P. O., Neary, D. & Snowden, J. S. (1988). The progression of the pathological changes of Alzheimer’s disease in frontal and temporal neocortex examined both at biopsy and at autopsy. Neuropathology and Applied Neurobiology, 14, 177–95. Masliah, E., Terry, R. D., Alford, M., DeTeresa, R. & Hansen, L. A. (1991). Cortical and subcortical patterns of synaptophysin-like immunoreactivity in Alzheimer’s disease. American Journal of Pathology, 138, 235–46. Matsuda, H., Kitayama, N., Ohnishi, T. et al. (2002). Longitudinal evaluation of both morphologic and functional changes in the same individuals with Alzheimer’s disease. Journal of Nuclear Medicine, 43, 304–11. Mayeux, R., Stern, Y. & Spanton, S. (1985). Heterogeneity in dementia of the Alzheimer type: evidence of subgroups. Neurology, 35, 453–61. Mega, M. S., Chen, S. S., Thompson, P. M. et al. (1997). Mapping histology to metabolism: coregistration of stained whole-brain sections to premortem PET in Alzheimer’s disease. NeuroImage, 5, 147–53. Meguro, K., Blaizot, X., Kondoh, Y., Le Mestric, C., Baron, J. C. & Chavoix, C. (1999). Neocortical and hippocampal glucose hypometabolism following neurotoxic lesions of the entorhinal and perirhinal cortices in the non-human primate as shown by PET. Implications for Alzheimer’s disease. Brain, 122, 1519–31. Mesulam, M. M. (1982). Slowly progressive aphasia without generalized dementia. Annals of Neurology, 11, 592–8. (2001). Primary progressive aphasia. Annals of Neurology, 49, 425–32. Mielke, R., Schroder, R., Fink, G. R., Kessler, J., Herholz, K. & Heiss, W. D. (1996). Regional cerebral glucose metabolism and postmortem pathology in Alzheimer’s disease. Acta Neuropathologica, (Berlin), 91, 174–9. Migneco, O., Benoit, M., Koulibaly, P. M. et al. (2001). Perfusion brain SPECT and statistical parametric mapping analysis indicate that apathy is a cingulate syndrome: a study in Alzheimer’s disease and nondemented patients. NeuroImage, 13, 896–902. Miller, F. D., Hicks, S. P., D’Amato, C. J. & Landis, J. R. (1984). A descriptive study of neuritic plaques and neurofibrillary tangles in an autopsy population. American Journal of Epidemiology, 120, 331–41. Minger, S. L., Esiri, M. M., McDonald, B. et al. (2000). Cholinergic deficits contribute to behavioral disturbance in patients with dementia. Neurology, 55, 1460–7. Minoshima, S., Giordani, B., Berent, S., Frey, K. A., Foster, N. L. & Kuhl, D. E. (1997). Metabolic reduction in the posterior cingulate cortex in very early Alzheimer’s disease. Annals of Neurology, 42, 85–94. Minoshima, S., Foster, N. L., Sima, A. A., Frey, K. A., Albin, R. L. & Kuhl, D. E. (2001). Alzheimer’s disease versus dementia with

29

30


Lewy bodies: cerebral metabolic distinction with autopsy confirmation. Annals of Neurology, 50, 358–65. Mirra, S. S., Heyman, A., McKeel, D. et al. (1991). The consortium to establish a registry for Alzheimer’s disease (CERAD). Part II. Standardization of the neuropathologic assessment of Alzheimer’s disease. Neurology, 41, 479–86. Mitchell, T. W., Mufson, E. J., Schneider, J. A. et al. (2002). Parahippocampal tau pathology in healthy aging, mild cognitive impairment, and early Alzheimer’s disease. Annals of Neurology, 51, 182–9. Mittal, S., Farmer, P., Kalina, P., Kingsley, P. B. & Halperin, J. (2002). Correlation of diffusion-weighted magnetic resonance imaging with neuropathology in Creutzfeldt–Jakob disease. Archives of Neurology, 59, 128–34. Moroney, J. T., Bagiella, E., Desmond, D. W. et al. (1997). Metaanalysis of the Hachinski Ischemic Score in pathologically verified dementias. Neurology, 49, 1096–105. Morris, J. C., Storandt, M., Miller, P. et al. (2001). Mild cognitive impairment represents early-stage Alzheimer disease. Archives of Neurology, 58, 397–405. Mummery, C. J., Patterson, K., Price, C. J., Ashburner, J., Frackowiak, R. S. & Hodges, J. R. (2000). A voxel-based morphometry study of semantic dementia: relationship between temporal lobe atrophy and semantic memory. Annals of Neurology, 47, 36–45. Mutrux, S. (1947). Diagnostic diff,rentiel histologique de la maladie d’Alzheimer et de la démence sénile. Monatsschrift Fur ¨ Psychiatrie und Neurologie, 113, 114–17. Na, D. L., Suh, C. K., Choi, S. H. et al. (1999). Diffusion-weighted magnetic resonance imaging in probable Creutzfeldt–Jakob disease. Archives of Neurology, 56, 951–7. Neary, D., Snowden, J. S., Mann, D. M. A. et al. (1986). Alzheimer’s disease: a correlative study. Journal of Neurology, Neurosurgery and Psychiatry, 49, 229–37. Neary, D., Snowden, J. S., Mann, D. M. A., Northen, B., Goulding, P. J. & MacDermott, N. (1990). Frontal lobe dementia and motor neuron disease. Journal of Neurology, Neurosurgery and Psychiatry, 53, 23–32. Newell, K. L., Hyman, B. T., Growdon, J. H. & Hedley-Whyte, E. T. (1999). Application of the National Institute on Aging (NIA)Reagan Institute criteria for the neuropathological diagnosis of Alzheimer disease. Journal of Neuropathology and Experimental Neurology, 58, 1147–55. Ohm, T. G., Muller, H., Braak, H. & Bohl, J. (1995). Close-meshed prevalence rates of different stages as a tool to uncover the rate of Alzheimer’s disease-related neurofibrillary changes. Neuroscience, 64, 209–17. Okuda, B., Tachibana, H., Kawabata, K., Takeda, M. & Sugita, M. (2001). Comparison of brain perfusion in corticobasal degeneration and Alzheimer’s disease. Dementia and Geriatric Cognition Disorders, 12, 226–31. Pandya, D. N. & Yeterian, E. H. (1985). Architecture and connections of cortical association areas. In Cerebral Cortex, vol. 4, ed. A. Peters & E. G. Jones. New York: Plenum Press. Parvizi, J., Van Hoesen, G. W. & Damasio, A. (1998). Severe pathological changes of parabrachial nucleus in Alzheimer’s disease. NeuroReport, 9, 4151–4.

Parvizi, J., Van Hoesen, G. W. & Damasio, A. (2000). Selective pathological changes of the periaqueductal gray matter in Alzheimer’s disease. Annals of Neurology, 48, 344–53. Parvizi, J., van Hoesen G. W. & Damasio, A. R. (2001). The selective vulnerability of brainstem nuclei to Alzheimer’s disease. Annals of Neurology, 49, 53–66. Pearson, R. C. A., Esiri, M. M., Hiorns, R. W., Wilcock, G. K. & Powell, T. P. S. (1985). Anatomical correlates of the distribution of the pathological changes in the neocortex in Alzheimer disease. Proceedings of the National Academy of Sciences USA, 82, 4531–4. Peavy, G. M., Herzog, A. G., Rubin, N. P. & Mesulam, M.-M. (1992). Neuropsychological aspects of dementia of motor neuron disease: a report of two cases. Neurology, 1004–8. Perry, E. K., Perry, R. H., Blessed, G. & Tomlinson, B. E. (1977). Necropsy evidence of central cholinergic deficits in senile dementia. Lancet, 1, 189. Perry, R. H., Blessed, G., Perry, E. K. & Tomlinson, B. E. (1981). Histochemical observations on cholinesterase activities in the brains of elderly normal and demented (Alzheimer type) patients. Journal of the Neurological Sciences, 51, 279–87. Perry, R. J. & Hodges, J. R. (2000). Differentiating frontal and temporal variant frontotemporal dementia from Alzheimer’s disease. Neurology, 54, 2277–84. Peterson, R. C., Smith, G. E., Waring, S. E. et al. (1999). Mild cognitive impairment: clinical characterization and outcome. Archives of Neurology, 56, 303–8. Razani, J., Boone, K. B., Miller, B. L., Lee, A. & Sherman, D. (2001). Neuropsychological performance of right- and leftfrontotemporal dementia compared to Alzheimer’s disease. Journal of the Internatinal Neuropsychology Society, 7, 468–80. Rockland, K. S. & Pandya, D. N. (1979). Laminar origins and terminations of cortical connections of the occipital lobe in the rhesus monkey. Brain Research, 179, 3–20. (1981). Cortical connections of the occipital lobe in the Rhesus monkey: interconnections between areas 17, 18, 19 and the superior temporal sulcus. Brain Research, 212, 249–70. Rosen, H. J., Gorno-Tempini, M. L., Goldman, W. P. et al. (2002). Patterns of brain atrophy in frontotemporal dementia and semantic dementia. Neurology, 58, 198–208. Rub, U., Del Tredici, K., Schultz, C., Thal, D. R., Braak, E. & Braak, H. (2001). The autonomic higher order processing nuclei of the lower brain stem are among the early targets of the Alzheimer’s disease-related cytoskeletal pathology. Acta Neuropathologica (Berlin), 101, 555–64. Samuel, W., Alford, M., Hofstetter, R. & Hansen, L. (1997). Dementia with Lewy bodies versus pure Alzheimer disease: differences in cognition, neuropathology, cholinergic dysfunction, and synapse density. Journal of Neuropathology and Experimental Neurology, 56, 499–508. Saykin, A. J., Stafiniak, P., Robinson, L. J. et al. (1995). Language before and after temporal lobectomy: specificity of acute changes and relation to early risk factors. Epilepsia, 36, 1071–7. Schwartz, W. J., Smith, C. B. & Davidsen, L. (1979). Metabolic mapping of functional activity in the hypothalamoneurophypophysial system of the rat. Science, 205, 723–5.


Scoville, W. B. & Milner, B. (1957). Loss of recent memory after bilateral hippocampal lesions. Journal of Neurology, Neurosurgery and Psychiatry, 20, 11. (2000). Toward a comprehensive theory for Alzheimer’s disease. Hypothesis: Alzheimer’s disease is caused by the cerebral accumulation and cytotoxicity of amyloid beta-protein. Annals of N Y Academy Sciences, 924, 17–25. Seltzer, B. & Pandya, D. N. (1978). Some cortical projections to the parahippocampal area in the rhesus monkey. Experimental Neurology, 50, 146–60. Shukla, C. & Bridges, L. R. (1999). Regional distribution of tau, betaamyloid and beta-amyloid precursor protein in the Alzheimer’s brain: a quantitative immunolabelling study. NeuroReport, 10, 3785–9. Snowdon, D. A., Greiner, L. H., Mortimer, J. A., Riley, K. P., Greiner, P. A. & Markesbery, W. R. (1997). Brain infarction and the clinical expression of Alzheimer disease. The nun study. Journal of the Academic Medical Society, 277, 813–17. Stern, Y., Jacobs, D., Goldman, J. et al. (2001). An investigation of clinical correlates of Lewy bodies in autopsy-proven Alzheimer disease. Archives of Neurology, 58, 460–5. Tekin, S., Mega, M. S., Masterman, D. M. et al. (2001). Orbitofrontal and anterior cingulate cortex neurofibrillary tangle burden is associated with agitation in Alzheimer disease. Annals of Neurology, 49, 355–61. Terry, R. D., Gionatas, N. K. & Weiss, M. (1964). Ultrastructural studies in Alzheimer’s presenile dementia. American Journal of Pathology, 44, 269–97. Terry, R. D., Masliah, E., Salmon, D. P. et al. (1991). Physical basis of cognitive alterations in Alzheimer’s disease: synapse loss is the major correlate of cognitive impairment. Annals of Neurology, 30, 572–80. Tiraboschi, P., Hansen, L. A., Alford, M. et al. (2000). Cholinergic dysfunction in diseases with Lewy bodies. Neurology, 54, 407–11. Tissot, R., Constantinidis, J. & Richard J. (1985). Pick’s disease. In Handbook of Clinical Neurology, vol. 2, ed. J. A. M. Fredericks, pp. 233–46. Amsterdam: Elsevier Science Publishers. Tomlinson, B. E., Blessed, G. & Roth, M. (1970). Observations on the brains of demented old people. Journal of the Neurological Sciences, 11, 205–42. Tranel, D., Damasio, H. & Damasio, A. (1997). A neural basis for the retrieval of conceptual knowledge. Neuropsychologia, 35, 1319–27. Ulrich, J. (1985). Alzheimer changes in nondemented patients younger than sixty five: a possible early stage of Alzheimer’s disease and senile dementia of Alzheimer’s type. Annals of Neurology, 17, 273–7. Van Hoesen, G. W. (1982). The primate parahippocampal gyrus: New insights regarding its cortical connections. Trends in Neurosciences, 5, 345–50. Van Hoesen, G. W., Hyman, B. T. & Damasio, A. R. (1986). Cellspecific pathology in neural systems of the temporal lobe in Alzheimer’s disease. In Progress in Brain Research, ed. D. F. Swaab, M. Fliers, W. A. Mirmiran, W. A. Van Gool, F. Van Haaren. Amsterdam:Elsevier.

Van Hoesen, G. W., Parvizi. J. & Chu, C. C. (2000). Orbitofrontal cortex pathology in Alzheimer disease. Cerebral Cortex, 10, 243–51. Vinters, H. V., Miller, B. L. & Pardridge, W. M. (1988). Brain amyloid and Alzheimer disease. Annals of Internal Medicine, 109, 41–54. White, P., Hiley, C. R., Goodhardt, M. J. et al. (1977). Neocortical cholinergic neurons in elderly people. Lancet, 1, 668–70. Wilcock, G. K. & Esiri, M. M. (1982). Plaques, tangle and dementia. A quantitative study. Journal of the Neurological Sciences, 56, 343–56. Wisniewski, K., Jervis, G. A., Moretz, R. C. & Wisniewski, H. M. (1979). Alzheimer neurofibrillary tangles in diseases other than senile and presenile dementia. Annals of Neurology, 5, 288–94. Zhukareva, V., Vogelsberg-Ragaglia, V., Van Deerlin, V. M. D. et al. (2001). Loss of brain tau defines novel sporadic and familial tauopathies with frontotemporal dementia. Annals of Neurology, 49, 165–75. Zola-Morgan, S., Squire, L. R. & Amaral, D. (1986). Human amnesia and the medial temporal region: enduring memory impairment following a bilateral lesion limited to the CAI field of the hippocampus. Journal of Neuroscience, 6, 2950–67. Zola-Morgan, S., Squire, L. R., Amaral, D. & Suzuki, W. A. (1989). Lesions of perirhinal and parahippocampal cortex that spare the amygdala and hippocampal formation produce severe memory impairment. Journal of Neuroscience, 9, 4355–70.

FURTHER READING Alzheimer, A. (1907). On a peculiar disease of the cerebral cortex. Allgemeine Zeitschrift fur ¨ Psychiatrie und Psychischgerichtliche Medizin, 64, 146. American Psychiatric Association Task Force on Nomenclature and Statistics. (1994). Diagnostic and Statistical Manual of Mental Disorders, 4th edn, revised. Benson, D. F., Davis, R. J. & Snyder, B. D. (1988). Posterior cortical atrophy. Archives of Neurology, 45, 789–93. Blocq, P. & Marinesco, G. (1892). Sur les lesions et la pathogenie de l’epilepsie dite essentielle. Seminars Medicin (Paris), 12, 445. Braak, H. & Braak, E. (1990b). Neurofibrillary changes confined to the entorhinal region and an abundance of cortical amyloid in cases of senile and presenile dementia. Acta Neuropathologica, 80, 479–86. Damasio, A. R. & Tranel, D. (1991). Disorders of higher brain function. In Comprehensive Neurology, ed. R. N. Rosenberg, pp. 639–57. New York: Raven Press. Damasio, H., Grabowski, T. J., Tranel, D. et al. (1996). A neural basis for lexical retrieval. Nature, 380, 499–505. Davies, P. & Maloney, A. J. F. (1976). Selective loss of central cholinergic neurons in Alzheimer’s disease. Lancet, 2, 1403. DeKosky, S. T. & Scheff, S. W. (1990). Synapse loss in frontal cortex biopsies in Alzheimer disease: correlation with cognitive security. Annals of Neurology, 27, 457–64.

31

32


Esiri, M. M., Pearson, R. C. A., Steele, J. E., Bowen, D. M. & Powell, T. P. S. (1990). A quantitative study of the neurofibrillary tangles and the choline acetyltransferase activity in the cerebral cortex and the amygdala in Alzheimer’s disease. Journal of Neurology, Neurosurgery and Psychiatry, 53, 161–5. Fisher, C. M. (1989). Binswanger’s encephalopathy: a review. Journal of Neurology, 236, 65–79. Francis, P. T., Palmer, A. M., Snape, M. & Wilcock, G. K. (1999). The cholinergic hypothesis of Alzheimer’s disease: a review of progress. Journal of Neurology, Neurosurgery and Psychiatry, 66, 137–47. Gibb, W. R. G., Luthert, P. J. & Marsden, C. D. (1989). Corticobasal degeneration. Brain, 112, 1171–92. Grant, I. & Adams, K. M. (1994). Neuropsychological Assessment of Neuropsychiatric Disorders. New York: Oxford University Press (in press). Hirono, N., Mori, E., Ishii, K. et al. (2001). Neuronal substrates for semantic memory: a positron emission tomography study in Alzheimer’s disease. Dementia and Geriatric Cognition Disorders, 12, 15–21. Hof, P. R., Bouras, C., Constantinidis, J. & Morrison, J. H. (1989). Balint’s syndrome in Alzheimer’s disease: specific disruption of the occipito-parietal visual pathway. Brain Research, 493, 368–75. Hof, P. R., Cox, K. & Morrison, J. H. (1990b). Quantitative analysis of a vulnerable subset of pyramidal neurons in Alzheimer’s disease: I. Superior frontal and inferior temporal cortex. Journal of Comparative Neurology, 301, 44–54. Hyman, B. T. (1999). Clinicopathologic correlates in temporal cortex in dementia with Lewy bodies. Neurology, 53, 2003–9. Hyman, B. T., Damasio, A. R., Van Hoesen, G. W. & Barnes, C. L. (1984). Alzheimer’s disease: cell specific pathology isolates the hippocampal formation. Science, 225, 1168–70. Hyman, B. T., Van Hoesen, G. W. & Damasio, A. R. (1986a). Glutamate depletion of the perforant pathway terminal zone in Alzheimer’s disease. Society for Neuroscience Abstracts, 11, 458. Hyman, B. T., Wenniger, J. J. & Tanzi, R. E. (1993). Nonisotopic in situ hybridization of amyloid beta protein precursor in Alzheimer’s disease: expression in neurofibrillary tangle bearing neurons and in the microenvironment surrounding senile plaques. Molecular Brain Research, 18, 253–8. Ishii, K., Yamaji, S., Kitagaki, H., Imamura, T., Hirono, N. & Mori, E. (1999). Regional cerebral blood flow difference between dementia with Lewy bodies and AD. Neurology, 53, 413–16. Katzman, R. (1977). Normal pressure hydrocephalus. In Dementia, 2nd edn., ed. C. E. Wells, pp. 69–92. Philadelphia: FA Davis. Katzman, R. & Terry, R. D. (1983). The Neurology of Aging. American Academy of Neurology, Washington DC. Katzman, R., Terry, R., DeTeresa, R. et al. (1988). Clinical, pathological, and neurochemical changes in dementia: a subgroup with preserved mental status and numerous neocortical plaques. Annals of Neurology, 23, 138–44. Knopman, D. S., Mastri, A. R., Frey, W. H., Sung, J. H. & Rustan, T. (1990). Dementia lacking distinctive histologic features: a

common non-Alzheimer degenerative dementia. Neurology, 40, 251–6. Knowles, R. B., Wyart, C., Buldyrev, S. V. et al. (1999). Plaque-induced neurite abnormalities: implications for disruption of neural networks in Alzheimer’s disease. Proceedings of the National Academy Sciences, USA, 96, 5274–9. Logothetis, N. K., Pauls, J., Augath, M., Trinath, T. & Oeltermann, A. (2001). Neurophysiological investigation of the basis of the fMRI signal. Nature, 412, 150–7. McKeith, I. G., Fairbairn, A. F., Bothwell, R. A. et al. (1994). An evaluation of the predictive validity and inter-rater reliability of clinical diagnostic criteria for senile dementia of Lewy body type. Neurology, 44, 872–7. Mann, D. M. A. & Esiri, M. M. (1988). The site of the earliest lesions of Alzheimer’s disease. New England Journal of Medicine, 318, 789–90. Marsden, C. D. (1978). The diagnosis of dementia. In Studies of Geriatric Psychiatry, ed. A. D. Isaacs & F. Post, pp. 95–118. Chichester: John Wiley. Marsden, C. D. (1985). Assessment of dementia. In Handbook of Clinical Neurology, ed. J. A. M. Frederiks, pp. 221–32. Amsterdam: Elsevier. Morris, J. C., Heyman, A., Mohs, R. C. et al. (1989). The consortium to establish a registry for Alzheimer’s disease (CERAD). Part I. Clinical and neuropsychological assessment of Alzheimer’s disease. Neurology, 39, 1159–65. Neary, D., Snowden, J. S., Northen, B. & Goulding, P. (1988). Dementia of frontal lobe type. Journal of Neurology, Neurosurgery and Psychiatry, 51, 353–61. Perry, E. K., Tomlinson, B. E., Blessed, G., Bergmann, K., Gibson, P. H. & Perry, R. H. (1978). Correlation of cholinergic abnormalities with senile plaques and mental test scores in senile dementia. British Medical Journal, 2, 1457–9. Price, J. L., Davis, P. B., Morris, J. C. & While, D. L. (1991). The distribution of tangles, plaques and related immunohistochemical markers in healthy aging and Alzheimer’s disease. Neurobiology of Aging, 12, 295–312. Rebeck, G. W. & Hyman, B. T. (1993). Neuroanatomical connections and specific regional vulnerability in Alzheimer’s disease. Neurobiology of Aging, 14, 45–47, discussion 55–6. Rogers, J. & Morrison, J. H. (1985). Quantitative morphology and regional and laminar distributions of senile plaques in Alzheimer’s disease. Journal of Neuroscience, 5, 2801–8. Rosenberg, R. N. (1993). A causal role for amyloid in Alzheimer’s disease: the end of the beginning. Neurology, 43, 851–6. Roth, M., Tomlinson, B. E. & Blessed, G. (1966). Correlations between scores for dementia and counts of ‘senile plaques’ in cerebral grey matter of elderly subjects. Nature, 209, 109–10. Scheff, S. W., Sparks, D. L. & Price, D. A. (1993). Quantitative assessment of synaptic density in the entorhinal cortex in Alzheimer’s disease. Annals of Neurology, 34, 356–61. Selkoe, D. J. (1991). The molecular pathology of Alzheimer’s disease. Neuron, 6, 487–98.


Sevush, S., Leve, N. & Brickman, A. (1993). Age at disease onset and pattern of cognitive impairment in probable Alzheimer’s disease. Journal of Neuropsychiatry, 5, 66–72. Terry, R. D., Hansen, L. A., DeTeresa, R., Davies, P., Tobias, H. & Katzman, R. (1987). Senile dementia of the Alzheimer type without neocortical neurofibrillary tangles. Journal of Neuropathology and Experimental Neurology, 46, 262–8. Van Hoesen, G. W., Hyman, B. T. & Damasio, A. R. (1991). Entorhinal cortex pathology in Alzheimer’s disease. Hippocampus, 1, 1–8.

Whitehouse, P. J., Price, D. L., Struble, R. G., Clark, A. W., Coyle, J. T. & DeLong, M. R. (1982). Alzheimer’s disease and senile dementia: loss of neurons in the basal forebrain. Science, 215, 1237–9. Wright, C., Geula, C. & Mesulam, M.-M. (1993). Neuroglial cholinesterases in the normal brain and in Alzheimer’s disease: relationship to plaques, tangles, and patterns of selective vulnerability. Annals of Neurology, 34, 373– 84.

33

2 Important anatomical landmarks in the brain in dementia James H. Morris1 and Margaret M. Esiri2 1 2

Radcliffe Infirmary John Radcliffe Hospital, Oxford, UK

In this chapter we summarize the more important structures in the brain with which it is essential to be familiar when studying the pathological basis of dementia. As described in Chapter 3 many dementing conditions are impossible to distinguish on naked eye examination of the brain since they do not display gross regional pathology. In order to reach the correct diagnosis, it is necessary to select the appropriate areas for more detailed examination. To do that requires knowledge of the parts of the brain that are significant in the particular context of dementia. For more detailed information textbooks of neuroanatomy such as Paxinos (1990), Heimer (1995), Parent (1996), or Nolte (2001) should be consulted.

Cerebral cortex Chapter 1 has already emphasized the crucial importance of the cerebral cortex for the cognitive functions which deteriorate in dementia. The cerebral cortex can be divided anatomically into a phylogenetically older and simpler allocortex consisting of the hippocampus and closely related entorhinal cortex, subiculum and olfactory regions, and the remaining, much more voluminous and phylogenetically more recent, neocortex.

Hippocampus, subiculum and entorhinal cortex (archicortex allocortex) These structures have already been mentioned in Chapter 1 but because of their importance it is worth providing a brief supplementary account here. Excellent reviews of the structure of the human hippocampus and

related cortex can be found in Amaral and Insausti (1990) and Duvernoy (1988). The entorhinal cortex lies in the uncus and anterior parahippocampal gyrus and forms an intermediate type of cortex between the complex six-layered neocortex of the temporal lobe and the simpler, basically three-layered, cortex of the hippocampus. Entorhinal cortex, parasubiculum, presubiculum and subiculum lie in continuity with each other in the medial temporal lobe ribbon of cortex seen in coronal slices of the brain, curving dorsally and then laterally to join the curled-up structure of the hippocampus in the floor of the inferior horn of the lateral ventricle (Fig. 2.1). The hippocampus has a head, a body and a tail and extends 4–4.5 cm in an antero-posterior direction from just behind the amygdala to the splenium of the corpus callosum. Anteriorly, the hippocampus is relatively expanded (the head) (Fig. 2.2), whereas more posteriorly it is compact, as seen at the level of the lateral geniculate body (the body) (Fig. 2.3). The hippocampus contains two interlocked cortical ribbons. One, the cornu ammonis, which is a continuation from the subiculum, is enclosed at its termination in a second, the dentate fascia (Fig. 2.1). The hippocampus contains pyramidal and non-pyramidal cells like the neocortex, but here they form one main layer sandwiched between two cell-sparse layers. In the cornu ammonis the three layers are made up of a deep molecular layer, largely devoid of cell bodies, but containing dendrites and axons, in which three strata can be distinguished – the strata radiatum, lacunosum and moleculare. Adjacent to this is the stratum pyramidale, layer of large pyramidal neurons. The pyramidal cells show regional variations in their arrangement, which provides the main basis for dividing the cornu ammonis


34

Important anatomical landmarks

Fig. 2.1. Drawing of a transverse section of the hippocampus and adjacent cortex (modified from Duvernoy, 1988). Arrow indicates the hippocampal sulcus. 1: hippocampus; 2: fimbria; 3: lateral geniculate body; 4: choroid fissure and tela choroidea; 5: stria terminalis; 6: tail of caudate nucleus; 7: temporal horn and choroid plexuses; 8: collateral eminence; 9: collateral sulcus; 10: parahippocampal gyrus; 11: entorhinal area; 12: ambient cistern; 13: mesencephalon; 14: parasubiculum; 15: presubiculum; 16: subiculum proper; 17: prosubiculum; 18: transverse fissure.

Fig. 2.3. Drawing of a transverse section of the body of the hippocampus. CA1: CA1 field of hippocampus; CA2: CA2 field of hippocampus; CA3: CA3 field of hippocampus; cf: choroidal fissure; COS: collateral sulcus; f: fimbria; PaS: parasubiculum; PrS: presubiculum; PHG: parahippocampal gyrus; V: inferior horn of ventricle.

Fig. 2.4. Close-up drawing of a transverse section of the body of the hippocampus, redrawn from Duvernoy (1988). CA1–CA4: fields of the hippocampus (cornu ammonis). 1: alveus; 2: stratum oriens; 3: stratum pyramidale; 4: stratum incidum; 5: stratum lacunosum; 6: stratum moleculare; 7: hippocampal sulcus; 8: stratum moleculare of dentate gyrus; 9: stratum granulosum of dentate gyrus; 10: polymorphic layer of dentate gyrus; 11: fimbria; 12: margo denticulatus; 13: fimbrio dentate sulcus; 14: hippocampal sulcus; 15: subiculum.

Fig. 2.2. Drawing of a transverse section of the head of the hippocampus. AHA: amygdala hippocampal area; CA1: CA1 field of hippocampus; EC: entorhinal cortex; hf: hippocampal fissure; ot: optic tract; PRC: perirhinal cortex; PrS: presubiculum; S: subiculum; SSa: sulcus semiannularis; V: interior horn of ventricle.

(CA) into four subdivisions – CA1-4 (Lorente de No 1934) (Figs. 2.4 and 2.6). CA1 is the largest zone, extending from the subiculum as a broad band of triangular-shaped, wellspaced neurons to the point where the band of cells narrows abruptly and the cell bodies become more oval in shape in CA2. CA3 neurons are slightly more widely dispersed

35

36

J. H. Morris and M. M. Esiri

column of the fornix (Fig. 2.5). The efferent hippocampal fibres in the fornix, derived principally from the large hippocampal and subicular pyramidal cells, are distributed to the lateral septal region in the basal forebrain, mamillary bodies of the hypothalamus and anterior nucleus of the thalamus. The dentate fascia (Fig. 2.4) also contains three main layers, one richly and two sparsely populated with neurons: the innermost polymorphic layer containing mainly the axons of the granule cells, next to this the granule cell layer containing tightly packed small neurons and the outermost molecular layer containing axons and dendrites.

Fig. 2.5. Diagram to show the connections of the hippocampus with the limbic system with the fornix and mamillo-thalamic tract shown in black (redrawn from Duvernoy, 1988). The limbic sulcus is divided into: 1: cingulate sulcus; 2: subparietal sulcus; 3: anterior calcarine sulcus; 4: collateral sulcus; 5: rhinal sulcus. The limbic gyrus (fine dots) is divided into: 6: subcallosal gyrus; 7: cingulate gyrus; 7’: isthmus; 8: parahippocampal gyrus; 9: anterior segment of the uncus. The intralimbic gyrus (with large dots) is divided into: 10: prehippocampal rudiment (precommissural hippocampus); 11: indusium griseum (supracommissural hippocampus); 12: hippocampus (retrocommissural hippocampus); 13: mamillary body; 14: anterior perforated substance; 15: anterior commissure; 16: septal nuclei covered by the paraterminal gyrus; 17: precommissural fornix; 18: postcommissural fornix (anterior column); 19: mamillothalamic tract; 20: anterior thalamic nucleus; 21: medial thalamic nucleus; 22: body of fornix; 23: corpus callosum; 24: crus of fornix.

than those of CA2 and they curve in a band to enter the region bounded by the dentate fascia. CA4 refers to the area containing widely scattered, large neurons enclosed by the dentate fascia granule cells. This region is sometimes also referred to as the end folium. (Alternative terminologies have been put forward by Rose (1927) and Vogt and Vogt (1937), their H1 zone being equivalent to CA1, H2 to CA2 and CA3 and H3 to CA4.) The third, outer, layer of the cornu ammonis, called the stratum oriens contains a few scattered interneurons, but consists mainly of the basal dendrites and axons of the pyramidal cells. The axons collect up into a fibre tract called the alveus (Fig. 2.4). This is the chief efferent outflow tract from the hippocampus and it sweeps over its ventricular surface to form a delicate, compact tract, the fimbria. The fimbria arches from the posterior hippocampus superiorly and medially to form the fornix, which lies beneath the corpus callosum, directed rostrally. At the genu of the corpus callosum it bends downwards and back as the

Connections of the hippocampus (Fig. 2.6) One source of afferent fibres to the hippocampus is the medial septal nucleus via the fimbria to the pyramidal cells of the cornu ammonis and the granule cells of the dentate fascia. This is a cholinergic supply. There are also a few commissural fibres derived from the contralateral

Fig. 2.6. Diagram of the main connections of the hippocampus (redrawn and modified from Williams and Warwick, 1980).


body formation in normal aging (see below) and even more in Alzheimer’s disease (Chapter 9). They are also liable to die as a consequence of hypoxia/ischaemia (Chapter 13) or severe, prolonged epilepsy (Chapter 23). The small pyramidal cells of the dentate fascia granule cell layer as well as the large pyramidal cells, are susceptible to Pick body formation in Pick’s disease (Chapter 11). The whole hippocampus and adjacent temporal lobe, amygdala and interconnected cingulate gyrus is liable to be destroyed in herpes simplex encephalitis (Chapter 21).

Neocortex Fig. 2.7. Diagram of the internal connections of the subiculum and hippocampus (redrawn from Duvernoy, 1988). ABCDE are parts of the sequential chains forming the principal pathways. A’: these perforant fibres join the apical dendrites of the pyramidal neurons directly. Cornu ammonis: 1: alveus; 2: stratum pyramidale; 3: axon of pyramidal neurons; 4: Schaffer collateral; 5: stratum radiatum and lacunosum; 6: stratum moleculare; 7: hippocampal sulcus. Gyrus dentatus: 8: stratum moleculare; 9: stratum granulosum; 10: polymorphic layer. GD: gyrus dentatus; CA3, CA1: fields of the cornu ammonis; SUB: subiculum.

hippocampus also via the fornix and fimbria. However, the chief afferent connections of the hippocampus are derived from the nearby entorhinal area via the perforant path which crosses the subiculum and ends predominantly on the dendrites of the granule cells in the molecular layer of the dentate fascia. There is a lesser perforant path projection from entorhinal cortex to subiculum and the strata moleculare and lacunosum of CA1 and CA3 regions of the hippocampus. This is a glutamatergic projection. Connections then run via the axons of the granule cells of the dentate fascia, which traverse the polymorphic layer to end on dendrites of pyramidal cells in the molecular layer of CA3 and CA4 (Fig. 2.7). The granule cell axons, known as mossy fibres, are also glutamatergic and have an exceptionally high level of zinc in their endings. Axons of CA3/4 neurons enter the alveus, but before doing so give off prominent collateral branches known as Schaffer collaterals, which run in the strata radiatum and lacunosum of the cornu ammonis and synapse there with CA1 pyramidal cell apical dendrites. Axons of CA1 neurons enter the alveus from whence they are distributed via collaterals to the subiculum as well as via the fimbria to its terminations in the lateral septal region, mamillary bodies and thalamus. The large pyramidal cells, particularly those of CA1, of the hippocampus are highly susceptible to neurofibrillary tangle formation, granulovacuolar degeneration and Hirano

The neocortex is built on a common structural plan, being composed of about 3–500 m diameter columns of neurons of many different types (Fig. 2.8). Each column lies perpendicular to the surface and functions as a unit, but has intimate connections with many other units (Fig. 2.8(b)). Allocortex shows more variable arrangements of neurons. Between allo- and neo-cortex there are transitional zones. At current levels of understanding of dementing processes it is helpful to distinguish between two main categories of cortical neurons: pyramidal and non-pyramidal (Jones, 1986). Many of the cellular pathologic changes characteristic of degenerative disease are found in pyramidal cells. Pyramidal cells have a triangular-shaped cell body, a long axon arising from the base of the pyramid, and a long apical dendrite extending towards the pial surface of the cortex. They outnumber non-pyramidal cells but the proportions are not constant either regionally or throughout the depth of the six-layered columns. Pyramidal cells are particularly prominent in motor cortex, where the largest of them form the Betz cells. In all regions pyramidal cells are concentrated in cortical layers 2, 3, 5 and 6. Most vary in approximate diameter from 10–50 m, but the Betz cells measure up to 120 m. The efferent and afferent connections of the cortex have been deduced principally in laboratory animals including primates, but the essential features are probably applicable to man. These studies have shown that pyramidal cells are the main, though not the only, source of efferent axons from neocortex. Depending on their origin, the efferent axons are directed towards one or more of four target tissues: (i) other areas of cortex via ipsilateral and contralateral (callosal) connections; (ii) the amygdala and basal ganglia (chiefly claustrum, caudate nucleus and putamen); (iii) the specific and non-specific nuclei of the thalamus and (iv) the brainstem and spinal cord. Pyramidal cells release excitatory neurotransmitters at their terminals, particularly glutamate and aspartate. In addition to having a wide influence on other regions of the brain, pyramidal cells

37

38


are also subject to a wide variety of inputs from their rich afferent supply. The afferents are derived from: (i) other distant and more local cortical pyramidal cells; (ii) local non-pyramidal cells and (iii) subcortical neurons. The latter include cholinergic afferents from the nucleus basalis of Meynert (excitatory), noradrenergic afferents from the locus ceruleus (excitatory) and serotonergic afferents from the raphe nuclei (excitatory) (see below). Given the key role of these cells in linking parts of the cortex together, it is not difficult to understand that their degeneration results in dementia. Non-pyramidal cells are distributed in all layers of the cortex, but are particularly numerous in layer 4. They take a variety of forms but they are generally of small size, 5– 15 m. Most of them are interneurons with connections chiefly confined to their near vicinity particularly, though not exclusively, within their own column of neurons and the columns immediately adjacent. They employ a variety of neurotransmitters including acetylcholine (Ach), -amino butyric acid (GABA) and the neuropeptides somatostatin, substance P, enkephalins, calcitonin generelated peptide, corticotrophin releasing factor, VIP, bombasin and neurotensin and their influence is thought to be predominantly inhibitory. Most studies of dementing diseases have so far found less pathology in these neuronal populations than in pyramidal cells, though loss of nonpyramidal cells occurs even in normal ageing and can be severe (Braak & Braak, 1986). An important exception are the large stellate cells found in clusters in layer 2 of the entorhinal cortex (see below). These are highly susceptible to neurofibrillary tangle formation in normal ageing as well as in Alzheimer’s disease.

Fig. 2.8. Drawings to illustrate the principles of cortical neuronal connectivity. (a) The internal connectivity in a cortical column outlined by the cylinder. The discs in layer IV (SPEC. AFF CYLINDER) indicate the sites of arborisation of specific afferents which this column of neurons shares with adjacent columns. Corticocortical afferents (arrow bottom, centre) terminate at all levels of the column and, in layer I, beyond it. Pyramidal neurons (punctate shading) provide the efferent outflow from the column

and are illustrated on the right. Interneurons of excitatory type (spiny stellate = SS, dashed shading) and inhibitory type (basket cells = BC in lower laminae, small basket cells = SBC in lamina II, axonal tuft cells = ATC and axoaxonic cells = AAC) are illustrated on the left side. CDB (= cellule a` double bouquet) cells, inhibitory interneurons that act specifically on inhibitory interneurons and therefore have a disinhibitory effect, are shown on the right-hand side. (b) ‘A’ shows the manner in which cortical columns such as that shown in (a) interact with other columns. Ipsilateral connections are derived mainly from pyramidal neurons in lamina III (shown left in outline) and contralateral connections from pyramidal neurons in layers II–VI (shown left in black) TH = thalamus. ‘B’ illustrates a single, Golgi-stained corticocortical afferent orientated as in the left-hand column of ‘A’, but at higher magnification to show the profuse branching in all laminae. ((a) and (b) modified and redrawn from Szentágothai, 1978, 1979 and Eccles, 1984.)


Fig. 2.9. Brodmann (1909) map of the lateral surface of the human cerebral hemisphere.

normal ageing. These are therefore suitable areas to sample when considering a diagnosis of Alzheimer’s disease. In contrast, frontal (e.g. areas 9 and 46), parietal (e.g. area 7) and occipital (e.g. areas 18 and 19) association cortex, though regularly affected by tangle formation in younger cases of Alzheimer’s disease, may have few or no tangles in the very elderly cases and so are less sensitive areas to sample than the temporal lobe areas. On the other hand, some areas of the temporal lobe, such as periamygdaloid cortex of the uncus (area 34) and parahippocampal gyrus (area 28), may contain at least a few tangles even in normal undemented elderly people, so these areas lack selectivity for Alzheimer’s disease. Some areas (e.g. primary motorcortex (area 4), primary somatosensory (area 1–3), primary visual (area 17) and primary auditory (area 41) cortices lack tangles even in severe, young cases of the disease and are therefore unsuitable for sampling (Pearson et al., 1985; Esiri et al., 1990). To take other examples: Pick’s disease most commonly affects frontal and temporal poles (areas 8–12 and 38) and diffuse Lewy body disease the anterior cingulate cortex (area 24), insula (area 16) and parahippocampal gyrus (area 28), so these are good areas to sample to confirm the presence of these respective diseases.

Glial cells and vascular supply of cortex

Fig. 2.10. Brodmann (1909) map of the medial surface of the human cerebral hemisphere.

Regional sub-division of the cerebral cortex Brodmann in 1909 described regional sub-divisions of the cerebral cortex which remain widely used and are based on slight differences in the laminar architecture (Figs. 2.9 and 2.10). Although other maps were subsequently produced and some controversies persist, subsequent anatomical and physiological studies have generally underlined the functional significance and anatomical value of these Brodmann maps of the cortex. In sampling the cerebral cortex for the purpose of identifying microscopical changes characteristic of different dementing conditions, some Brodmann areas are much more useful than others. The aim is to sample regions with high selectivity and sensitivity for the pathological changes under consideration. In Alzheimer’s disease, neurofibrillary tangles are readily found in the temporal lobe neocortex of Brodmann areas 22 and 38, areas that do not normally contain tangles in

Scattered in the neocortex and archicortex are many astrocytes, satellite oligodendroglial and microglial cells, the latter being members of the macrophage lineage of cells. The vascular supply of the neocortex comes from inwardly directed arterioles given off from leptomeningeal branches of the three main cerebral arteries – anterior, middle and posterior. The hippocampus derives its blood supply chiefly from the anterior and posterior choroidal arteries. Venous drainage from the neocortex is chiefly via superficial veins to the nearby venous sinuses. From the hippocampus small veins enter the vein of Galen and straight sinus via the basilar vein.

Cerebral white matter The main cerebral white matter occupies much of each cerebral hemisphere. It contains myelinated fibres passing to the cerebral cortex from other parts of the cortex, thalamus and brainstem, and reciprocal fibres passing in the reverse direction. Some of these fibres are collected into more or less compact tracts such as the pyramidal tract and the corpus callosum. Others run more diffusely, crisscrossing each other in the centrum semiovale (Fig. 2.11). Cortico-cortical fibres that are distributed locally run in

39

40


Fig. 2.11. Coronal slice across the cerebral hemispheres at the level of the mamillary bodies (arrows). CS = centrum semiovale.

immediately sub-cortical fibre bundles. Scattered between the myelinated fibres are glial cells of all three main types: astrocytes, oligodendrocytes and microglia, with oligodendrocytes predominating. Much of the blood supply to the white matter is carried in long arterioles which course, unbranching, through the cortex before breaking up in the sub-cortical and deep white matter to supply a coarser network of capillaries than is found in the cortex. The venous drainage of the white matter is chiefly towards the ventricles where there are subependymal tributaries of the deep cerebral veins, which join the vein of Galen and drain into the straight sinus. The cerebral white matter is affected in a variety of different dementing diseases most notably some forms of ischaemic dementia, (Chapter 13), AIDS (Chapter 21) and progressive multifocal leucoencephalopathy (Chapter 21). If the main cerebral white matter is extensively demyelinated in other white matter diseases such as multiple sclerosis or leukodystrophy dementia also results. Cerebral white matter is also much reduced in volume in many degenerative conditions including Pick’s disease, Alzheimer’s disease, Huntington’s disease, in alcoholic brain damage or after severe head injury. This loss of white matter in neurodegenerative diseases probably mainly reflects Wallerian degeneration due to loss of neurons. Periventricular white matter is at risk of damage in normal pressure hydrocephalus (Chapter 19).

Amygdala This subcortical collection of grey matter lies just in front of the hippocampus and the inferior horn of the lateral

ventricle, in the dorsomedial portion of the deep temporal lobe (Fig. 2.12). It has a number of constituent nuclei which can be considered here to form two main groups: the basolateral and cortico-medial nuclei. Seen in cross-section in a coronal slice through the mamillary bodies, the amygdala appears as an almond-shaped structure separated on its medial and inferior aspect by a thin layer of white matter from the neighbouring peri-amygdaloid cortex of the uncus. The cortico-medial nuclei have close afferent connections with the olfactory system via the lateral olfactory stria, and with the hypothalamus, brainstem, nucleus of the horizontal limb of the diagonal band and bed nucleus of the stria terminalis. Their neurons project to other nuclei in the amygdala, medial frontal cortex, dorsomedial thalamus and hypothalamus. The basolateral nuclei receive fibres from the substantia innominata, the thalamus, hypothalamus, brainstem and cerebral cortex, including allocortex. They project back to many of these regions and to the bed nucleus of the stria terminalis and striatum. There are large, medium and smallsized neurons in the amygdala, some of them pyramidallike. The latter, particularly in the cortico-medial nuclei, are susceptible to neurofibrillary tangle formation in Alzheimer’s disease (Chapter 9) and to Lewy body formation in diffuse Lewy body disease (Chapter 15).

Nucleus basalis of Meynert (basal nucleus) (Fig. 2.13) A great deal has been written in the last two decades or so about the basal nucleus of Meynert (for review see Saper, 1990). The reason for this interest is that animal studies in the 1970s established it as the source of most of the acetylcholine in the cerebral cortex (the remainder coming from local interneurons) (Mesulam & van Hoesen, 1976), a transmitter that was found at about the same time to be selectively depleted in the cortex in Alzheimer’s disease (Bowen et al., 1977; Davies, 1979). The basal nucleus forms a collection of large multipolar neurons in the substantia innominata strung out from the level of the optic chiasm rostrally to the mamillary bodies caudally. The cell group merges anteriorly with similar cholinergic cells in the nuclei of the vertical and horizontal limbs of the diagonal band. Its anterior part lies ventral to the anterior commissure, a discrete band of transversely running myelinated fibres, while its posterior part lies ventral to the globus pallidus (Figs. 2.13 and 3.10). Its cells are predominantly cholinergic and project widely to all parts of the hemisphere. The nucleus basalis is involved in the pathology of both Alzheimer’s disease (Chapter 9) and Parkinson’s disease (Chapter 15).


Fig. 2.12. Drawing of a coronal slice through the anterior temporal lobe, amygdala and basal ganglia (redrawn from Duvernoy, 1988). 1: hippocampal head (digitationes hippocampi); 2: lateral nucleus of the amygdala; 2 : basal nucleus of the amygdala; 2 : accessory basal nucleaus of the amygdala; 2 : cortical nuclei of the amygdala; 3: anterior commissure; 4: anterior perforated substance; 5: ansa lenticularis; 6: optic tract; 7: hypothalamus; 8: anterior column of fornix; 9: interventricular foramen; 10: ventral anterior thalamic nucleus; 11: anterior thalamic nucleus; 12: caudate nucleus; 13: genu of internal capsule; 14: globus pallidus, pars medialis; 15: globus pallidus, pars lateralis; 16: putamen; 17: claustrum; 18: insula; 19: precentral gyrus; 20: lateral sulcus; 21: superior temporal gyrus; 21 : superior temporal sulcus; 22: middle temporal gyrus; 23: inferior temporal gyrus; 24: gyrus fusiformis; 25: parahippocampal gyrus; 26: tentorium cerebelli; 27: posterior cerebral artery; 28: internal carotid artery within the cavernous sinus; 29: hypophysis (posterior lobe).

Basal ganglia The number of different nuclei that are included among the basal ganglia is somewhat variable, but here we shall consider the following: caudate nucleus, putamen, globus

pallidus and sub-thalamic nucleus (Figs. 2.12 and 3.11). These heterogenous masses of grey matter contain nerve cells of varying size, shape and connectivity, together with glial cells, bundles of interspersed myelinated fibres and blood vessels. The caudate nucleus and putamen, referred

41

42


(b)

(a)

GP

Fig. 2.13. Nucleus basalis of Meynert. (a) Diagrams showing the position of the nucleus (in black) at different antero-posterior levels of the human ventral forebrain. (b) Photograph of a luxol fast-blue/cresyl violet stained section of the basal ganglia. An arrow points to the nucleus basalis. (c) Photomicrograph of a Nissl-stained section of the basal nucleus to show the large, multipolar neurons that characterize it. ac: anterior commissure; al: ansa lenticularis; Amy: amygdala; C: central nucleus of the amygdala; Cl; claustrum; Coa : anterior cortical nucleus of the amygdala; DB: diagonal band; EC: entorhinal cortex; ec: external capsule; fx: fornix; GP: globus pallidus; ic: internal capsule; lot: lateral olfactory tract; NBM: nucleus basalis of Meynert; ot: optic tract; P: putamen; PAC: periamygdaloid cortex; Sl: substantia innominata; 35: cortical area 35 of Brodmann. (See also Fig. 3.10.)


Fig. 2.13. (c)

to collectively as the striatum, are functionally and structurally similar and perform important tasks related to the control of movement. They receive a massive afferent input from the cerebral cortex (gluatamtergic) and additional afferents from the intra laminar thalami nuclei, the substantia nigra (dopaminergic) and brainstem raphe nuclei (serotonergic). Their afferent connections are with the globus pallidus and substantia nigra and, via connections of these structures with the thalamus and motor parts of the cerebral cortex. The most abundant cell type, accounting for more than 90% of cells in the caudate and putamen, is a small neuron with spiny dendrites that contains the neurotransmitter gamma amino butyric acid (GABA). There are additional separate populations of large (cholinergic) interneurons and somatostatin and other peptidergic-immunoreactive neurons. The globus pallidus consists of two parts, the external and internal segments, both of which receive their main afferents from the striatum. Most efferents from the internal segment project to the ventral thalamus and corticomedian nucleus of the thalamus, while those from the external segment are directed towards the subthalamic nucleus, where connections are made with the substantia nigra and back to the globus pallidus. The striatum is particularly severely affected in Huntington’s disease (Chapter 16) and some forms of multiple system atrophy (Chapter 15). The globus pallidus and subthalamic nucleus are foci of damage in

progressive supranuclear palsy and corticobasal degeneration (Chapter 11).

Thalamus and hypothalamus The thalamus and hypothalamus are further heterogeneous collections of subcortical grey matter nuclei that need to be considered in a few dementing diseases. They form the walls of the third ventricle, the hypothalamus lying ventral to the thalamus which forms a very large mass of grey matter medial to the internal capsule and ventral and partly posterior to the basal ganglia. The major groups of thalamic nuclei can be recognized, separated from each other by narrow bands of white matter – the internal medullary lamina – the median, lateral and anterior groups. The nuclei within the thalamus that are of particular interest in dementia are the dorsomedial nucleus the anterior nucleus and some of the intralaminar nuclei (Fig. 2.14). The dorsomedial nucleus forms part of a circuit involving the pre-frontal cortex (to which it projects) and the amygdala (which projects to it). The anterior nuclear complex, like the dorsomedial nucleus, forms part of the limbic system, and projects to cingulate cortex, entorhinal cortex and subiculum as well as to non-limbic cortex, receiving reciprocal connections from most of these areas and from the mamillary bodies. The intralaminar nuclei divide the medial nuclei from the lateral and ventral nuclei and include rostral and caudal groups. The dorsomedial

43

44


Fig. 2.14. Drawing of a transverse section through the thalamus at the level of the mamillary bodies (MB). ANT: anterior nucleus of thalamus; CC: corpus callosum; CN: caudate nucleus; dmN: dorso-medial nucleus of thalamus; F: fornix; GP: globus pallidus; IC: internal capsule; ic: interthalamic connection; IML: intramedullary lamina and nuclei of thalamus; LDT: lateral dorsal nucleus of thalamus; P: putamen; rMN: midline nuclei of thalamus; TV: third ventricle; VAN: ventral anterior nucleus of thalamus.

Fig. 2.15. Drawing of a transverse section of the cerebral hemispheres showing the position of some hypothalamic nuclei (redrawn and modified from Snell, 1992). CC: corpus callosum; LV: lateral ventricle; CN: caudate nucleus; Th: thalamus; ic: interthalamic connection; P: putamen; GP: globus pallidus; ot: optic tract; IN: infundibular nucleus; TN: tuberomammillary nuclei; VN: ventromedial nucleus; LHA: lateral hypothalamic; F: fornix; DN: dorsomedial nucleus; PV: paraventricular nucleus; TV: third ventricle.

and anterior nuclei are closely involved in memory function and in the pathology of Alzheimer’s disease (Chapter 9) and fatal familial insomnia (Chapter 17). The dorsomedial nucleus is also often affected in Wernicke’s encephalopathy (Chapter 18) and in dementia associated with thalamic degeneration (Chapter 23). The hypothalamus contains many nuclei that exert central control over the autonomic nervous system. However, some of its nuclei are prominently involved in dementing processes – the mamillary bodies (Figs. 2.11 and 2.14), lateral hypothalamic area (Fig. 2.15) and the less readily identified suprachiasmatic nucleus, for example (Braak & Braak 1987).

Brainstem Structures that are of chief interest in the mid-brain are the substantia nigra, the raphe nuclei, the superior corpora quadrigemina, the red nuclei and the periaquaductal grey matter (Fig. 2.16). The darkly pigmented dopaminergic neurons of the substantia nigra are readily seen in a transverse cut across the mid-brain at the level of the corpora quadrigemina, lying ventral and lateral to the red nuclei and dorsal to the cerebral peduncles (Fig. 2.16). There are two parts to the substantia nigra which lie side-by-side: the ventrolateral pars reticularis and the dorsomedial pars

Fig. 2.16. Drawing of a transverse section of the upper midbrain (redrawn and modified from Snell, 1992). sCQ: superior corpora quadrigemia; CA: cerebral aqueduct; pGm: periaqueductal grey matter; mnTN: mesencephalic nucleus of trigeminal nerve; nON: nucleus of oculomotor nerve; MLF: medial longitudinal fasciculus; rf: reticular formation; RN: red nucleus; SN: substantia nigra; dRT: decussation of rubrospinal tracts; CC: crus cerebri; mL: medial lemniscus; sL: spinal lemniscus; tL: trigeminal lemniscus.


Fig. 2.17. Drawing of the upper pons to show the position of the median raphe (MnR) superior cerebellar peduncles (scp) and locus ceruleus (LC). PnO pontine reticular nucleus, Pyr pyramidal tract, mlf median longitudinal fasciculus, CA cerebral aqueduct.

compacta. The cells of the pars compacta are selectively damaged in Parkinson’s disease (Chapter 15), and form one of the sites of pathology in progressive supranuclear palsy (Chapter 11) and some forms of multiple system atrophy (Chapter 15). The large serotonergic cells of the dorsal and median raphe form two large midline nuclei lying one above the other in the mid-brain and pons (Fig. 2.17). They are involved in the pathology of Alzheimer’s disease (Chapter 9), Parkinson’s disease (Chapter 15) and progressive supranuclear palsy (Chapter 11). The superior corpora quadrigemina, the more rostral pair of dome-like protruberances on the dorsal surface of the mid-brain (Fig. 2.16), are damaged in progressive supranuclear palsy. The periaqueductal grey matter (Fig. 2.16) is at risk of damage in Wernicke’s encephalopathy (Chapter 18) and progressive supranuclear palsy. The superior cerebellar peduncles (Fig. 2.17) lying to either side of the upper end of the fourth ventricle in the lower mid-brain, are often atrophied in progressive supranuclear palsy. In the pons the noradrenergic pigmented nucleus, the locus ceruleus, lying just ventral to the lateral angle of the floor of the lower aqueduct and rostral part of the fourth ventricle, is at risk in Alzheimer’s disease and Parkinson’s disease, while the tegmental nuclei and medial longitudinal fasciculus (Figs. 2.16–2.18) are damaged in progressive supranuclear palsy. The pontine nuclei are damaged in the pontocerebellar variety of multiple system atrophy (Fig. 2.17) (Chapter 15), and the median raphe in Parkinson’s disease (Chapter 15).

Fig. 2.18. Drawing of a transverse section of the medulla (redrawn and modified from Snell, 1992). iMV: inferior medullary vellum; FV: cavity of fourth ventricle; mvN: medial vestibular nucleus; icP: inferior cerebellar peduncle; VaN: vagus nerve; io: inferior olive; HyN: hypoglossal nerve; Pyr: pyramid; ArN: arcuate nuclei; amFi: anterior median fissure; mL: medial lemniscus (also see Fig. 2.16); TT: tectospinal tract; Na: nucleus ambiguus; stnTN: spinal tract and nucleus of trigeminal nerve; rf: reticular formation; nTS: nucleus of tractus solitarius; DVn: dorsal vagal nucleus; Hn: hypoglossal nucleus; mIF: medial longitudinal fasciculus.

The medulla (Fig. 2.18) is less obviously affected than the more rostral parts of the brainstem, the pons and midbrain, in dementing diseases. However, it is worth noting here the positions of the pigmented dorsal vagal nuclei, damaged in Parkinson’s disease, the inferior olives which suffer secondary retrograde degeneration if the Purkinje cells of the cerebellum are destroyed by disease, and the hypoglossal nucleus and medullary pyramids which degenerate in some cases of motor neuron disease with dementia (Chapter 12).

The cerebellum (Fig. 2.19) The division of the cerebellum into midline vermis and lateral hemispheres is worth noting. Certain diseases, particularly alcoholic cerebellar degeneration, affect the vermis more than the hemispheres. Widespread cerebellar cortical damage occurs in Creutzfeldt–Jacob disease, Gerstmann Sträussler syndrome (Chapter 17), ponto-cerebellar atrophy and paraneoplastic disease (Chapter 23). The deep cerebellar white matter may share in the pathology of white matter diseases such as progressive multi-focal leukoencephalopathy and AIDS (Chapter 21). The dentate nucleus of the cerebellum, forming a wavy band of large, deeply

45

46


Fig. 2.19. Drawing of the cerebellum from above, with the vermis bisected and the hemispheres displaced laterally (redrawn and modified from Snell, 1992). eCA: entrance into cerebral aqueduct; FV: cavity of the fourth ventricle; MS: median sulcus; rCH: right cerebellar hemisphere; eCC: entrance into central canal; fFV: floor of fourth ventricle formed by medulla oblongata; VC: vermis of cerebellum; fFVP: floor of fourth ventricle formed by pons.

situated neurons above the roof of the fourth ventricle, is selectively damaged in progressive supranuclear palsy. If cells from the dentate nuclei are lost their outflow tracts in the superior cerebellar peduncles, whose position has already been noted lateral to the fourth ventricle, in the upper pons and mid-brain, become atrophic.

REFERENCES Amaral, D. C. & Insausti, R. (1990). Hippocampal formation. In Paxinos, G. (ed.) The Human Nervous System. Academic Press NY pp. 711–56. Bowen, D. M., Smith, C. B., White, P. et al. (1977). Chemical pathology of the organic dementias. II Quantitative estimation of cellular changes in post mortem brains. Brain, 100, 427–53. Braak, H. & Braak, E. (1986). Ratio of pyramidal cells versus nonpyramidal cells in the human frontal isocortex and changes in ratio with ageing and Alzheimer’s disease. Progr. Brain Res., 70, 185–212. (1987). The hypothalamus of the human adult: chiasmatic region. Anat. Embryol., 175, 315–30.

Brodmann, K. (1909). Vergleichende Lokalisation lehre der Gross hirnrinde in ihren Prinzipiem dargestellt auf Gruand des Zellenbaues. J. A. Barth, Leipzig. Davies, P. (1979). Neurotransmitter-related enzymes in senile dementia of the Alzheimer type. Brain Res., 171, 319– 27. Duvernoy, H. M. (1988). The Human Hippocampus. Bergmann Verlag, Munich. Eccles, J. C. (1984). The cerebral cortex: a theory of its operation. In Jones, E. G. & Peters, A. (eds.) Cerebral Cortex Vol 2 Functional Properties of Cortical Cells. Plenum Press, MS. pp. 1–36. Esiri, M. M., Pearson, R. C. A., Steele, J. E., Bowen, D. M. & Powell, T. P. S. (1990). A quantitative study of the neurofibrillary tangles and the choline acetyltransferase activity in the cerebral cortex and the amygdala in Alzheimer’s disease. J. Neurol. Neurosurg. Psychiatry, 53, 161–5. Heimer, L. (1995). The human brain and spinal cord. Functional Neuroanatomy and Dissection Guide. 2nd edn. Springer, Berlin. Jones, E. G. (1986). Neurotransmitters in the cerebral cortex. J. Neurosurg., 65, 135–53. Lorente de No (1934). Studies on the structure of the cerebral cortex II. Continuation of the study of the Ammonic system. J. Psychol. Neurol., 46, 113–77.


Mesulam, M.-M. & van Hoesen, G. W. (1976). Acetylcholinesteraserich projections from basal forebrain of the rhesus monkey to neocortex. Brain Res., 109, 152–7. Nolte, J. (2001). The Human Brain. Mosby. Parent, A. (1996). Carpenter’s Human Neuroanatomy 9th edn. Williams and Wilkins. Paxinos, G. (ed.) (1990). The Human Nervous System. Academic Press, New York. Pearson, R. C. A., Esiri, M. M., Hiorns, R. W., Wilcock, G. K. & Powell, T. P. S. (1985). Anatomical correlates of the distribution of the pathological changes in the neocortex in Alzheimer’s disease. Proc. Natl Acad. Sci. (USA), 82, 4531–4. Rose, M. (1927). Allocortex bei Tier und Mensch die sogennante Richrinde beim Menschen und beim Affen. J. Psychol. Neurol., 34, 261–401. Saper, C. B. (1990). Cholinergic system. In Paxinos, G. (ed.) The

Human Nervous System. Academic Press, New York. pp. 1095– 114. Snell, R. S. (1992). Clinical Neuroanatomy for Medical Students. 3rd edn. Little, Brown and Co. Szentágothai, J. (1978). The neuron network of the cerebral cortex: a functional interpretation. Proc. R. Soc. Lond. Ser. B, 201, 219–48. (1979). Local neuron circuits of the neocortex. In Schmitt, F. O. & Worden, F. G. (eds.) The Neuroscience Fourth Study Program. MIT Press, Cambridge, MS. pp. 399– 415. Vogt, C. & Vogt, O. (1937). Sitz und Wesen der Krankheiten in Lichte der topistischen Hirnforschung und des Variereus der Tiere, erster Teil. J. A. Barth, Leipzig, p. 457. Williams, P. L. & Warwick, R. (1980). Gray’s Anatomy. 36th edn. Churchill Livingstone, London.

47

3 Practical approach to pathological diagnosis Margaret M. Esiri1 and James H. Morris2 1 2

Radcliffe Infirmary, Oxford, UK John Radcliffe Hospital, Oxford, UK

The post-mortem examination in cases of dementia It has been said, with a good deal of truth, that the answer to every question in medicine is in three parts: first, take a history, second, make a physical examination, and third, perform the relevant special tests and investigations. For the pathologist, the first two parts of this rubric are fulfilled by reading the patient’s chart. The third is the performance of what, in at least one sense, is the ultimate diagnostic test, the post-mortem examination.

History and examination The clinical information available to the pathologist called upon to perform an autopsy examination on a case of dementia is extraordinarily variable. At one extreme is the patient who has been studied over an extended period where the quality and extent of cognitive failure has been documented and, often, a presumptive pathological diagnosis is made. This type of history is often supplemented by more or less objective tests of intellectual function and the results of numerous investigations. Patients submitted to this degree of investigative rigour are often in centres that have a particular interest or active research programme into dementia. In these circumstances it can (we hope) be assumed that there is good liaison between the clinical service and the pathology department and the cases will be dealt with according to protocol. The opposite end of this particular spectrum is the patient coming to autopsy examination who is reported to have an unspecified degree of cognitive decline, variously described in imprecise terms. In such cases recourse to

considerable ingenuity is required to form an idea of the nature and severity of the decline. Clues can sometimes be gleaned from the nursing notes or even from the patient’s address. Residents of nursing homes are often more disabled than those living in the community, and the nursing home itself offers a source of an experienced assessment of the ability to perform the activities of daily living in a community. Whatever information is available, it should be carefully studied before the autopsy is commenced. It may well contain useful pointers to the type of disease present and indicate aspects of the general autopsy that require particular attention. It is unwise to place too great a reliance on the diagnosis made during life since clinical accuracy in the diagnosis of the cause of dementia is not always very high (Gilleard et al., 1992; Nagy et al., 1998; Knopman et al., 2001) (Table 3.1). However, the pathologist has, at his disposal, the additional important fact that the patient has died. The length of the history may have significant safety implications for the performance of the post-mortem, most particularly in relation to the diagnosis of Creutzfeldt–Jakob disease and other prion diseases (see Chapter 5). Prion diseases The possible presence of a prion disease (Chapter 17) in cases of dementia should be a sufficient stimulus to ensure careful scrutiny of the patient’s notes before performing an autopsy on a patient with dementia. Special safety requirements need to be fulfilled before performing an autopsy on a case of suspected Creutzfeldt–Jakob disease, Gerstmann– Sträussler syndrome, fatal familial insomnia (Medori et al. 1992) or any other suspected prion disease (Advisory Committee on Dangerous Pathogens, HMSO, 1994; Committee


48

Practical approach to pathological diagnosis

Table 3.1. Accuracy of clinical diagnosis of Alzheimer’s disease Study

Type

Criteria

Non-systemic diagnosis Todorov et al. (1975) Sulkava et al. (1983) Perl et al. (1984) Mölsä et al. (1985) Neary et al. (1986) Kokmen et al. (1987) Wade et al. (1987) Homer et al. (1988) Joachim et al. (1988) Ettlin et al. (1989) Mendez et al. (1992) Blacker et al. (1994) Klatka et al. (1996)

R P R P P R R P R P R R P&R

Systematic diagnosis ¨ Muller & Schwartz (1978) Alafuzoff et al. (1987) Martin et al. (1987) Morris et al. (1988) Crystal et al. (1988) Tierney et al. (1988) Boller et al. (1989) Jellinger et al. (1990) Risse et al. (1990) Burns et al. (1990) Galasko et al. (1994) Gearing et al. (1995) Victoroff et al. (1995) Lim et al. (1999)

R R P P P P R R P P P P P P

n

AD (%)

AD only (%)

Less established Less established Limited information Less established Less established Less established Less established Consensus Limited information Less established Variable NINCDS-ADRDA Variable

273 27 26 28 24 32 39 13 150 13 650 36 170

220(81) 22(81) 21(81) 23(82) 18(75) 23(72) 35(9) 6(46) 127(90) 12(92) 505(78) 33(92) 149(88)

149(56) 22(81) (9(35) 20(71) – 20(63) 28(72) 6(46) 105(74) 7(54) 390(60) – 96(65)

ICDA-8 DSM III NINCDS-ADRDA NINCDS-ADRDA NINCDS and DSM III NINCDS-ADRDA NINCDS-ADRDA NINCDS, DSMIII, ICD9 NINCDS and DSM III NINCDS and ADRDA NINCDS-ADRDA NINCDS-ADRDA

41 32 11 26 13 22 44 273 25 50 137 106 163 114

37(90) 26(81) 11(100) 26(100) 11(85) 18(82) 35(80) 246(90) 17(68) 42(84) 123(90) 92(87) 134(82) 95(83)

– 17(53) – 16(62) – 17(77) – 236(86) 16(64) 34(68) 80(58) 47(44) – 34(30)

NINCDS-ADRDA

R (retrospective), P (prospective). Source: Adapted from Mendez et al. (1992).

on Health Care Issues, ANA, 1986; Bell & Ironside, 1993; Budka et al., 1995). (Methods for handling this material are discussed in Chapter 5.) The most likely prion disease to be encountered is Creutzfeldt–Jakob disease (CJD) and the notes should be looked through to see if this diagnosis has been considered. It should be specifically considered in cases with a rapidly progressive course and death from natural causes within a year or less of onset of dementia. It is our practice to regard anyone who becomes severely demented and dies in a year or less to have CJD unless there is good reason to think otherwise. In this context it is worth remembering that 50% of cases of pathologically confirmed CJD die within 4 months and 90% within a year of onset. Two years after onset 95% of all cases of CJD will have died (Brown et al., 1986; Brown, 1988a). By contrast, less than 10% patients with Alzheimer’s

disease have died within a year of presentation to a psychiatrist or geriatrician, and approximately 40% are still alive 5 years later. The insidiousness of the onset of symptoms in Alzheimer’s disease, in fact, means that survival from the time of onset of symptoms is considerably longer than these figures suggest, with an estimated 50% survival of 8.1 years (Barclay et al., 1985). Although the ‘1-year rule’ will certainly catch most cases of Creutzfeldt–Jakob disease, it is not foolproof, as approximately 5% of patients survive for more than 2 years following diagnosis (Brown et al., 1984, 1986, 1987) and there is one report of survival for as long as 16 years (Cutler et al., 1984). As a group, these long survivors tend to have a higher frequency of familial incidence, younger age at onset and a lower frequency of accompanying myoclonus (see below). The longer survival cases may have a clinical course not

49

50

M. M. Esiri and J. H. Morris

dissimilar from Alzheimer’s disease from which they cannot be reliably distinguished on clinical grounds. It is not necessarily a great comfort to know that these cases are usually very easy to recognize on microscopic examination as they have prominent spongiform change. As well as rate of progression, another pointer in the history to the possible presence of prion disease is the occurrence of adventitious movements. In Creutzfeldt–Jakob disease there is typically a marked startle myoclonus, that comes on during the course of the disease. Even if the diagnosis of CJD has not apparently been considered by the patient’s medical advisors, the prudent pathologist should consider it in cases of a progressive dementia with abnormal movements. However, ‘en medecin, comme l’amour, pas de jamais, pas de toujour ’, for myoclonus is a rather common occurrence and occurs in a large number of neurological conditions, including, as might be surmised, some cases of Alzheimer’s disease (Haltia et al., 1994). Parenthetically, although usually thought of as a dementing disease, CJD has other guises. It has been our experience that ‘paraneoplastic syndrome’ is one of the most frequent suggested clinical diagnoses in atypical cases that prove to have CJD on pathologic examination. Like CJD, paraneoplastic disease is subacutely progressive and may affect grey matter in both the cerebellum and cerebral hemispheres. It is particularly likely to cause confusion with those cases of CJD where the cerebellar symptoms predominate in the early clinical stages. Prion disease, in addition to causing difficulty by masquerading as other neurological conditions in its clinical expression, may also be cryptic in its pathological expression. As the report by Collinge and colleagues (Collinge et al., 1990) demonstrates, prion disease is not always, or necessarily, accompanied by spongiform change when the brain is examined microscopically. Currently therefore, we are in the unfortunate situation of not knowing the full range of either the clinical or the pathological expression of the prion diseases. The counsel of perfection in this circumstance would be to perform all autopsies as if they were possible cases of prion disease but this is hardly a practicable course of action. In our view, while it is very necessary to take appropriate precautions, it is also important not to over-react to the possible risks of exposure to prions. The infectivity of prions, although demonstrable, is not high and the only established cases of nosocomial transmission to humans have been from corneal transplants (Duffy et al., 1974), contaminated stereotactic electrodes, cadaver-derived dural implants (Centre for Disease Control, 1987), cadaver-derived human growth and gonadotrophin hormone administration (Brown, 1988b; Centre for Disease Control, 1985). In this context it is appropriate to remember that numerous

autopsies were performed on patients with prion diseases before the ‘infective’ nature of prions was demonstrated. Given the comparatively cavalier way in which these autopsies were conducted and the brain tissue handled, mortuary technicians and neuropathologists should have been at a substantial risk of contracting prion disease if performance of an autopsy was a significant exposure. The risk, if there is one, is probably in handling the tissue in the histology laboratory where at least one histology technician who was known to have handled cases of CJD has died of CJD, although there is no evidence that demonstrates that the disease was contracted by exposure to the tissue (Miller, 1988). This potential risk can however be removed by treatment of tissue to be processed by formic acid (Brown et al., 1990) which effectively eliminates infectivity from tissue specimens (see Chapter 5). Returning to the history, in patients who have not been well investigated and who have uncharacterized cognitive decline, perhaps the first question to be asked is whether the patient has a true dementia or a confusional state brought about by metabolic derangements. Pointers to a confusional state are a marked variability in the degree of impairment, and usually, relatively short duration. It is always helpful to establish if possible the length of time for which the patient has been cognitively impaired. A sense of the rate and character of the progression of the decline is also very informative. For example, a patient with a cognitive decline that does not seem to have undergone significant progression for several years is not, in general, likely to be suffering from a neurodegenerative disease since, once symptomatic, these tend to be progressive. The manner of the presentation can also be relevant. It is commonplace that the symptoms that first bring a patient to medical attention markedly influence the manner of the patient’s progression through the medical establishment, and also the way in which they are investigated. Patients with Alzheimer’s disease who present with alteration in mood or affect are much more likely to find their way to a psychiatrist, while those with, for example, language disorders, are usually referred to a neurologist. It is revealing of the state of medical practice that the outcome, in terms of history taken, character of the physical examination and investigations performed, is often very different in these two groups of patients. A formal neurological examination is always useful, as it will almost always include some assessment of the mental state. Within the neurological examination, it is worthwhile looking for any evidence of focal neurological signs and symptoms, and their temporal relationship, if any, to the onset and/or progression of the decline in cognitive function. Myoclonus and cerebellar symptoms have already


been mentioned in relation to prion diseases, but Parkinsonian symptoms, choreiform movements, and signs of upper and lower motor neurone loss are all encountered in different forms of dementia. Stroke syndromes are also important, whether they be fixed impairments or transient ischaemic attacks. Bilaterally upgoing plantar responses as an isolated neurological sign are in general not a useful observation, as they have no localising value and do not differentiate between structural and metabolic disease. Another aspect of the history, which is worth specific attention, is the family history. There are a number of dementing diseases that are familial, Huntington’s disease for example, and many of the degenerative and other forms of dementia have a familial component. Both Alzheimer’s and Pick’s disease have a familial representation, as does Creutzfeldt–Jakob disease. However, it has to be admitted that this is usually one of the least well-documented aspects of a patient’s history. Investigations Neuroimaging Of all the investigations, neuroradiological imaging studies probably have the most immediate and direct value and should be looked for specifically. The quality of neuroimaging now available makes it perfectly feasible to make a reasonable assessment of the degree and distribution of, for example brain atrophy, in life. Indeed, it is probably quite reasonable to suppose that neuroimaging, particularly MRI, is able to provide at least as good an estimate of the degree of atrophy as neuropathological examination. The potential of MRI for functional studies of the nervous system is only just beginning to be exploited and it is likely that more advanced forms of electronic manipulation of MRI derived information will be a fruitful field of investigation of many forms of neurological disease in the relatively near future. Our personal experience of correlations of hippocampal size between CT and pathology in Alzheimer’s disease (Jobst et al., 1992, 1994) and MRI and pathology in temporal lobe epilepsy indicates that imaging and pathology findings give very good correlation in both ventricular size and severity of pathology. Correlation between imaging and pathology, is not, however, always this good or this simple. The situation in vascular disease for example is very different. Neuroradiological phenomena such as the so-called ‘unidentified bright objects’ (UBOs) and leukoariosis, have a less consistent association with either pathology or clinical phenomenology, so that neuroimaging findings should not be accepted uncritically (Smith et al., 2000; Schmidt et al., 1999). In some circumstances more arcane forms of neuroimaging may also be available. The most notable are single

photon emission computed tomography (SPECT), which gives an indication of the distribution of blood flow in the brain, and positron emission tomography (PET), which gives information about the patterns of metabolic activity. Although these imaging methods are not commonly available (particularly the very expensive PET scanning), patterns of activity in different diseases are beginning to be recognized which are suggestive of specific pathological diagnoses. The post-mortem examination Consent to post-mortem examination Most post-mortem examinations of the brain in cases of dementia are made for ‘medical interest’ and therefore require the consent of the patient’s next of kin. Sometimes, indeed, the patient’s relatives initiate the request for post-mortem examination to satisfy themselves as to the cause of the devastating illness which they have witnessed in a loved one. There is also an increasing awareness on the part of relatives that many of the dementing diseases involve a genetic risk which may have been passed to other family members as well. In agreeing to a post-mortem examination the next of kin need also to take account of any wishes about this matter that may have been expressed by the patient before he or she became demented. In some prospective studies it may have been possible for clinicians to ascertain the patient’s wishes, if they became acquainted with him/her at an early stage of the disease, when they judged that informed consent could still be given. It is, in any case, an important responsibility of the pathologist to ensure that informed consent to neuropathological examination of the brain has been obtained. This will normally involve retention of the brain for at least a period of months, and this should be fully explained to relatives. It is a further responsibility of the pathologist to ensure that information about the neuropathological diagnosis is fed back to the referring clinician and, if they have requested it, the patient’s family. In addition to providing a neuropathological diagnosis, retention of the brain after death offers an opportunity for additional investigations to be performed, for example, on fresh frozen brain samples. These include genetic studies which may be of relevance to other family members as well. It is therefore appropriate that relatives should be made aware of such investigations, if they are contemplated, and their consent obtained. Genetic implications of the results of such investigations for the rest of the family should be offered if the relatives indicate that they wish to know these. In this way the recommendations of recent reports on informed consent to medical investigation (Royal College of Pathologists, 2000; General Medical Council,

51

52


1998; Department of Health, 2001; Medical Research Council, 1999) will have been fully complied with. Neuropathological examination in other cases of dementia will take place as part of a post mortem examination ordered by the coroner to ascertain a cause of death. The immediate cause of death is likely to be a complication such as pneumonia or pulmonary embolus that has arisen as a secondary consequence of dementia. Coroners will take differing views on whether, in these circumstances, determination of the cause of dementia is a necessary part of determining the cause of death but it is our opinion that, whenever possible, the relatives’ consent should be sought to examination of the brain in these circumstances. It is our experience that most families will readily agree to this examination, and indeed, for tissue from the brain to be used for research and education, if the reasons why this is valuable are fully explained to them. We believe pathologists should always be willing to participate in such explanations whenever invited to do so.

Examining the extracranial organs As indicated above, pathological processes that cause dementia are rarely the direct cause of death. Most patients with progressive dementia die as a result of complications of their inactive, often terminally mute, and unresponsive state with bronchopneumonia, pulmonary embolism, urinary tract infections or terminal septicaemia. A few die as a result of the direct systemic effects of the primary disease, such as pneumocystis pneumonia in AIDS or complications of liver cirrhosis in alcoholics or subjects with Wilson’s disease. Cardiac deaths are also common in demented subjects, particularly in those with ischaemic dementia. In addition to providing a cause of death the systemic autopsy examination may also show other findings significantly related to the dementing process. Where particular types of pathological process are suspected, the following aspects of the general autopsy examination require particular attention:

Vascular dementia State of the heart (particularly valves’ sources of emboli in left atrium and ventricle, size of the left ventricle, patency of foramen ovale) and great vessels (particularly the aorta, internal carotid arteries in the neck and the vertebral arteries; sources of fat emboli; sources of emboli from leg veins if foramen ovale patent; evidence of infarction in other organs, especially the kidneys and spleen; evidence of widespread vasculitis and/or renal lesions of systemic lupus erythematosus or hypertension; evidence of neoplasia (carcinoma, lymphoma); evidence of sickle cell anaemia.

Infective causes Fungal or tuberculous infections in lungs/paranasal sinuses; Whipple’s disease; syphilis; systemic features of AIDS or another cause of immunosuppression such as sarcoidosis or lymphoma; parasitic infestations such as toxoplasmosis or cysticercosis. Intracerebral tumour or hydrocephalus Extra-cerebral neoplasia; Paget’s disease. Alcoholic intoxication Evidence of repeated trauma, liver cirrhosis, peripheral neuropathy. Dialysis dementia Renal disease and aluminium deposits in bones. Other metabolic disease Liver disease (hepatic encephalopathy, Wilson’s disease); pancreatic or extra pancreatic insulinoma (hypoglycaemia); adrenal atrophy (adreno-leukodystrophy); mitochondrial myopathy, cardiomyopathy (mitochondrial cytopathies); peripheral neuropathy (vitamin deficiencies, some storage disorders, porphyria). Paraneoplastic disease Small cell lung carcinoma, ovarian carcinoma in women, lymphoma, germ cell tumour. The spinal cord should be removed for later examination if the following are suspected: Gerstmann–Sträussler syndrome, familial amyloid angiopathy, motor neuron disease with dementia, metabolic storage disorders, adreno-leukodystrophy, multiple system atrophy, Hallervorden–Spatz disease, infective conditions, multiple sclerosis, vitamin deficiencies, cases with clinical evidence of peripheral neuropathy, paraneoplastic syndrome. In the latter two conditions samples of sensory and autonomic ganglia should also be removed for microscopic examination. Samples of peripheral nerves and muscles should also be removed from cases of suspected motor neuron disease, metabolic storage diseases, adreno-leukodystrophy, vitamin deficiency, peripheral neuropathy and paraneoplastic syndrome. Examination of fresh brain For the pathologist with little experience in examining demented brains, a systematic approach to the examination of the brain is the most effective way of building the familiarity that will permit reliable differentiation between the, in some cases very wide, range of normal appearances and genuine pathology. If a number of cases are to be examined and compared, it is almost essential that they be handled in a similar and systematic fashion. Where rare conditions are to be studied, it is usually necessary to collect material


from a number of different sources and this sort of study is made either difficult or impossible if the brain has been handled in very different ways. As will be seen in relation to a number of causes of dementia the lack of systematic examination and recording is a considerable handicap to progress in understanding. The foregoing can be read as a plea for neuropathologists to abandon subsidiarity and relinquish a little of their sovereignty to adopt a common minimum standard or protocol for the examination of the brain in cases of dementia. It is our opinion that, for the examination of cases of dementia, there is a very good product already available. Under the leadership and guidance of Drs Al Heyman and Sue Mirra, the Consortium to Establish a Registry for Alzheimer’s Disease (CERAD) has performed the invaluable service of producing a neuropathology assessment instrument (Mirra et al., 1991) that, although concentrating on Alzheimer’s disease, is specifically applicable to all causes of dementia and which is subject to continual review and periodic updating (Gearing et al., 1995; Newell et al., 1997; Hyman & Trojanowski, 1997; NIA 1997). The CERAD protocol is essentially a research tool and, properly applied, is too detailed for what might be called routine diagnostic use, but its outline and principles are entirely applicable to a standard diagnostic approach. On pages 62–63 and 181 we reproduce a scaled-down version of CERAD that we recommend to readers for this purpose. After opening the skull the following features should be sought before removing the brain. Chronic or acute subdural haematomas (small subdural haematomas are not uncommon as a consequence of brain atrophy or in the presence of an intra-ventricular shunt; occasionally large subdural haematomas are a cause of dementia). Meningeal thickening (though in chronic meningitis this may be more evidenced at the base of the brain than over the vertex). Cerebrospinal fluid may be readily removed if required from the pre-pontine cisterns during removal of the brain from the cranial cavity. The brain should then be removed carefully and attention paid particularly to avoiding tears of the mid-brain after the cerebral hemispheres have been freed but the contents of the posterior fossa are still anchored. After removal, the brain should be weighed and placed in a bowl to allow its natural shape to be retained. The main cerebral vessels at the base should be carefully examined and the extent of atheroma and any sites of arterial stenosis or occlusion noted. Aneurysms should be sought at branch points on the circle of Willis and middle cerebral arteries and any fresh subarachnoid haemorrhage or rusty

discoloration of leptomeninges, either generalized or localized, suggestive of old subarachnoid haemorrhage, described. A search should also be made for old contusions suggestive of head injury, particularly on the inferior surfaces of the frontal and inferolateral surfaces of the temporal lobes. Any surface softenings or tumours should be delineated and described. Frequently, the cerebral gyri appear atrophied and the extent of such atrophy and its distribution and extent should be documented. Atrophy may be more readily appreciated if the leptomeninges are stripped from the surface. The cerebellar folia and brainstem should also be examined for evidence of atrophy. At this stage a strong lead as to the cause of the patient’s dementia may have been obtained. For example, Creutzfeldt–Jacob disease is very likely if the history of dementia was very short, perhaps two to three months, from normal mental state to mute and unresponsive, no abnormalities were found in the general autopsy examination other than a cause of death such as bronchopneumonia or pulmonary embolism, and the brain appears normal or shows only slight generalized cerebral atrophy. With the same post mortem findings but a long history, measured in years, Alzheimer’s or Parkinson’s disease; if there was a small cell lung carcinoma and a normal appearing brain, paraneoplastic syndrome is very likely; if the lymph nodes appeared lymphomatous and the brain unremarkable, progressive multifocal leukoencephalopathy is possible. If there is gross atrophy, particularly confined to temporal and frontal poles, the diagnosis is almost certainly Pick’s disease. However, it is equally possible that no clear lead may emerge at this stage or indeed until the microscopy is available. Retention of fresh frozen brain samples It may frequently be advisable to retain fresh samples of brain or cerebrospinal fluid for microbiological, molecular biological, biochemical or frozen section analysis before proceeding to fix the brain. Many centres undertaking research on dementia have defined requirements for fresh frozen tissue. The addresses of brain banks receiving fresh tissue in the UK, USA and elsewhere are listed in the Appendix. In some centres one half of the brain or one cerebral hemisphere may be removed and frozen intact, or after slicing to allow gross pathology to be identified first. In most degenerative diseases the pathological changes are more or less symmetrically distributed and it is a likely safe assumption, at least to a first approximation, that the frozen tissue contains the same pathology as the equivalent area of the fixed hemisphere. However, if a precise match of pathology and frozen tissue analysis is essential, it is necessary to take immediately adjacent small samples for this purpose. Some

53

54


pathological processes are not symmetrical and freezing half a brain or hemisphere in such cases is more problematic. This is particularly true of ischaemic dementia, some infective causes of dementia and neoplasias. Since these causes may be impossible to exclude at this stage there is a case for limiting removal of fresh samples to small blocks (1 to 5 grams) of cerebral cortex from most cases since this will suffice for many research requirements. When any necessary fresh samples have been removed from the brain and any external features of interest noted and photographed, the remainder should be fixed by suspension from the basilar artery in a large closed container of fixative (commonly neutral 10% formalin). After 3 to 4 weeks the fixed brain will be ready for slicing.

(a)

Examination of fixed brain Before slicing the brain the formalin should be poured out of its container and replaced with water for one to two days to avoid excessive exposure to formalin vapour. Weight and volume changes in normal and demented There is a concept that might be called ‘spurious precision’ where a parameter is measured with great precision but can be interpreted only very approximately (the measurement of lactic dehydrogenase (LDH) springs to mind as a good example). So it is with brain weight; although easily measured to the nearest tenth of a gram, the range of normal values is so wide that it is impossible to interpret with equivalent precision. Any value between 1000 and 1500 g could be within the normal range or indicate a significant degree of atrophy. The very great variation in brain size that this difference in weights implies is illustrated in Fig. 3.1.

Fig. 3.1. Variation in size of non-atrophic brains. Both these patients were considered in life to be neurologically normal. The brain on the left weighed 860 g, and that on the right 1760 g.

(b) Fig. 3.2. Comparison of (a) a young and (b) an aged normal brain. There is slight widening of the sulci in the aged normal, but the changes are quite inconspicuous.

Several autopsy studies indicate that there are modest reductions in brain weight and volume, particularly in the cerebrum, in old age (see Chapter 7). Computerized tomography scan on healthy old people confirm these changes. They start about the age of 50 years and amount to a loss of weight and volume of the cerebral hemispheres of about 2–3% per decade (Davis & Wright, 1977; Miller et al., 1980). Compared to the brain of a young person, the brain from an elderly person thus shows some slight narrowing of the cortical gyri and widening of the sulci (Fig. 3.2), and there is also slight collagenous thickening of the leptomeninges. Brain weights in those 70–80 years average 1344 g for men and 1213 g for women (Dekaban & Sadowsky, 1978). Comparable average brain weights in young adults are 1450 g for men and 1290 g for women (Dekaban & Sadowsky, 1978). Loss of volume of brain substance in old age is generally


Fig. 3.3. Ventral view of a moderately atrophic brain. The conspicuous feature is the widening of the sulci towards the temporal poles. There is also some expansion of the sulci at the frontal poles.

greater for white matter (11% loss between 70 and 85 years) than grey matter (2–3% loss between 70 and 85 years), but the greatest loss in one study was in the subiculum of the hippocampus (28% between 70 and 85 years) (Anderson et al., 1983). Other external features The external configuration of the brain can reveal generalized or focal atrophy or focal lesions. The assessment of the degree of atrophy cannot be complete before the brain is sliced, but some brains show clear evidence of widening of the sulci (Figs. 3.3 and 3.4) that is usually easiest to see in the frontal and temporal lobes. If it is important to define or document the distribution and degree of sulcal widening,

Fig. 3.4. A lateral view of the same brain as Fig. 3.3. Widening of the sulci is apparent in both the frontal and temporal lobes.

Fig. 3.5. Removal of the meninges permits a more accurate estimation of the degree of cortical atrophy. In this case, where the atrophy is more severe than in Fig. 3.4, there is marked widening of the sulci in the frontal, temporal and parietal lobes. In this case, in which a diagnosis of Alzheimer’s disease was made on microscopic examination, there is no obvious sparing of the primary cortex bordering the Rolandic sulcus.

it is often helpful to strip off the leptomeninges (Fig. 3.5). Although this has the advantage of revealing the configuration of the gyri more clearly, it also removes most of the arterial supply to the cortex and it may be appropriate to strip the meninges from one hemisphere only. The absence of significant atrophy can also be important information; in a rapidly progressive dementia such as CJD, the external appearance of the brain is usually normal. External inspection of the fixed brain will also confirm the features identified when it was fresh. Sometimes areas of softening that were overlooked at an earlier stage become evident since fixation accentuates the difference in consistency between normal and softened brain. The vascular anatomy should also be noted and any softening should be related to the arterial territories of supply. Occasionally softening may be due to necrotizing encephalitis, for example due to toxoplasmosis or herpes simplex encephalitis. In these cases the softened areas will not conform to arterial territories or boundary zones. Failure to conform to arterial boundaries is also found in diffuse ischaemia where the parasagittal and lateral occipital arterial boundary zones are the most vulnerable to grossly visible damage. Examination of brain slices This is best done systematically, and as an aid to this there is considerable merit in laying the slices out on a tray which enables an overall impression to be gained and encourages the sequential examination of the slices. At this stage the hind brain can be separated from the cerebrum by slicing through the mid-brain. The

55

56


conventional manner of slicing the fixed cerebrum from cases of dementia is at one centimetre intervals in the coronal plane. Unless there is a need to compare horizontal slices with CT scan appearances, coronal slices are recommended because the hippocampal formation is better seen in this plane. The first slice is taken through the mamillary bodies which are readily identified at the base of the brain. When the hemispheres have been sliced in this manner the deep aspects – cerebral white matter and deep grey matter as well as the hippocampus, insula and ventricles – can be examined for the first time. Assessment of ventricular size provides useful diagnostic information, focal areas of deep softening or focal atrophy can be identified, and the features of interest photographed and reported. The hind brain is similarly examined in slices transverse to the brainstem with the cerebellum attached. Alternatively, the cerebellum can be removed by slicing through the cerebellar peduncles which affords an opportunity for making a mid-line cut through the vermis of the cerebellum and an oblique lateral cut through the hemispheres and dentate nuclei. Cerebral atrophy Ventricular system Assessment of ventricular size and configuration is an important part of the estimation of the degree of atrophy (Fig. 3.6). The range of normal variation in the size of the lateral ventricles within the older population is considerable, and only major degrees of ventricular enlargement can be unequivocally interpreted. The conventional places to examine are the angle and curvature between the head of the caudate and the internal capsule of the frontal poles, and, in particular, the volume of the temporal horn. The normal variation in the width of the third ventricle is less than that seen in the lateral ventricles, and a significantly barrel shaped third ventricle is almost always an indication of atrophy. Ventricular dilatation can also be seen in brains sectioned in the CT plane (Fig. 3.7), but it is difficult to compare the degree of ventricular dilatation in brains sectioned in a CT plane with that seen in brains sectioned in the coronal plane. One caveat in the examination of ventricular size and configuration is that the brain should have been properly suspended during fixation. All too often, asymmetric flattening of the gyri over the convexities indicates not a focal mass lesion but the distortion that comes from sitting on the bottom of a bucket. When this has occurred, there will usually be considerable distortion of the ventricular configuration. One aspect of atrophy as manifested by ventricular dilatation that can be assessed on gross examination of brain slices is its distribution. Atrophy is rarely entirely uniform,

and differences in the relative dilatation of the frontal, temporal and occipital poles of the lateral ventricles is sometimes conspicuous. Pick’s disease is a good example, there being cases where the frontal and temporal lobes can be affected to quite different degrees producing quite different degrees of dilatation of the frontal and temporal horns. Asymmetric atrophy, and indeed asymmetry in general, should also be looked for. The fourth ventricle is not often significantly enlarged in dementia, except in those diseases that affect the cerebellum and where there is usually a significant history of ataxia. One condition where gross examination of the fourth ventricle can be very helpful is in progressive supranuclear palsy (PSP), a condition that can be mistaken for Parkinson’s disease. In both conditions there is degeneration and depigmentation of the substantia nigra, but in PSP, there is also degeneration of the deep grey nuclei of the cerebellum which results in degeneration of the superior cerebellar peduncle. This results in a marked relative expansion of the superior end of the fourth ventricle where it merges with the aqueduct of Sylvius that can be seen on gross inspection of brain slices. Examination of the brain parenchyma Within each slice we have found it helpful to examine the cortical ribbon, white matter and deep grey nuclei separately. It has to be admitted that, with most cases of degenerative dementia, the gross examination of brain slices is seldom decisive. In cerebrovascular disease by contrast, the gross examination can be very important and in this circumstance it can be very helpful to have some sort of pro-forma of brain slices on which to record the observations (see Chapter 13). Cerebral cortex Many, if not most, of the causes of dementia affect the cerebral cortex and its assessment is therefore a very important part of the examination of the brain in cases of dementia. Cortical atrophy In most cases of dementia secondary to degenerative disease the cortical ribbon is not visibly attenuated or narrowed and the cortical degeneration is reflected only in the widening of the cortical sulci and expansion of the ventricles. However, when there is severe atrophy, the cortical ribbon can be seen to be reduced in thickness and altered in character even when examined by the naked eye. Grossly visible cortical atrophy is particularly easy to see in Pick’s disease because of the focal nature of the atrophy and is particularly noticeable in the contrast between the preservation of the cortex of the posterior superior temporal gyrus


(a)

(c)

(b)

XBWH-902 AB86-21

(d)

A76-330

(e) Fig. 3.6. Variations in ventricular size (a)–(e). Progressive expansion of the cerebral ventricles. (a) Normal brain. (b) Mild expansion of the third ventricle and the Sylvian fissure, widening of the angle of the lateral ventricles and the beginnings of expansion of the temporal horns of the lateral ventricles. (c) Further widening of the angles of the lateral ventricles with more pronounced barrel shaped expansion of the third ventricle and significant expansion of the temporal horns of the lateral ventricles. (d ), (e) Marked expansion of the temporal horns of the lateral ventricles with obvious atrophy of the hippocampi. There is also gross expansion of the remainder of the lateral ventricles.

57

58


Fig. 3.7. Ventricular enlargement is also apparent in a CT plane brain slice. In this example, the expansion of the temporal horns of the lateral ventricles is particularly conspicuous.

and the atrophic cortex of the adjacent middle temporal gyrus and insula. Other cortical changes Those that can be assessed are predominantly vascular. Focal infarcts can of course generally be seen, and their pattern is sometimes very informative (and occasionally misleading). One pattern that can stand out is that associated with the so-called ‘watershed’ or ‘boundary-zone’ lesions that are prominent parasagittally and tend to expand over the lateral occipital gyri. This pattern is usually quite recognizable and is associated with generalized reductions in cerebral perfusion, and in long-term survivors, widespread cortical damage and cognitive impairment. However, too much reliance should not be put on apparent distribution. We have had the experience of diagnosing probable boundary zone infarction by gross inspection only to discover on microscopic examination that the

Fig. 3.8. Hippocampal and amygdala atrophy in AD: a brain slice at the level of the posterior thalamus showing marked hippocampal atrophy and accompanying expansion of the temporal horns of the lateral ventricles. In the more anterior slice, above, there is marked expansion of the temporal horn of the lateral ventricle anterior to the hippocampus in the normal location of the amygdala.

appearance was a result of myriad microatheroemboli, which, because of their very small size had impacted in the most distal arterial branches. In another case, the distribution of the cortical and white matter damage superficially appeared to be consistent with boundary zone ischaemia, but microscopy showed the presence of chronic progressive multifocal leukoencephalopathy (PML). Hippocampus and amygdala The hippocampus should always be examined specifically. In Alzheimer’s disease, for example, both hippocampi are often atrophic (Fig. 3.8). However, asymmetries are always worth looking for, and, if visible on macroscopic examination are very likely to be pathologically significant. Imaging and tissue studies have


shown that there is very little difference in size between the left and right hippocampi in normals. Where there is a detectable difference in size, microscopic examination of both hippocampi is recommended. Although strictly a subcortical grey nucleus, it is convenient to examine the amygdala at the same time as the hippocampus since the anterior hippocampus is coextensive with the amygdala which, at least grossly, merges with the entorhinal cortex of the uncus of the temporal lobe. Atrophy of the amygdala is most easily detected by the expansion of the anterior extent of the temporal pole of the lateral ventricle that is its result (Fig. 3.8). In the cortical dementias the degree of amygdala atrophy is variable, but is usually very noticeable in Pick’s disease when the temporal lobe is affected and it is also often affected in Alzheimer’s disease. White matter Specific features of the white matter that are relevant in the examination of the brain from a patient with dementia are: (i) volume, (ii) appearance, and (iii) texture. Volume The white matter volume is really an aspect of atrophy and is most easily assessed as an inverse reflection of ventricular size. In patients with severe cortical disease there is almost always a correspondingly severe loss of white matter most marked in the affected lobes. As with cortical atrophy this is most easily seen in one of the focal atrophies, Pick’s disease again being the most dramatic. However, there are a few situations, such as for example Binswanger’s subcortical leukoencephalopathy, where the white matter may be disproportionately reduced in volume and the grey matter relatively preserved. This usually produces a picture of atrophy where the ventricular enlargement is out of proportion to the sulcal widening. In this circumstance making an assessment of the thickness of the anterior corpus callosum is sometimes helpful. In patients with not much apparent cortical atrophy, significant attenuation of the anterior corpus callosum may be a good indicator of severe white matter damage. Appearance Linear streaks of tissue infarction along the course of a long penetrating artery, the presence of lacunar infarction and a peppering with tiny holes (cribriform state) may all be indicators of the presence of diffuse vascular disease in the white matter. Other white matter lesions are the characteristic sharp-edged irregular, grey, translucent plaques of multiple sclerosis or the tiny punctate lesions coalescing almost imperceptibly into larger translucent areas of demyelination of PML (a rare cause of cognitive impairment). Occasionally, much more dramatic findings crop up; we have seen subacute cognitive decline as a

result of bihemispheric radiation/chemotherapy leukoencephalopathy, and almost total hemispheric demyelination in a patient carrying a clinical diagnosis of schizophrenia with a ‘degenerative neurological disease’ which proved to be adrenoleukodystrophy. The vital clue here was the observation in the general autopsy of very small adrenal glands!

Texture Normal white matter of the fixed brain is firm but not rubbery. Focal necrotizing lesions produce marked local softening that is easily detected by running a gloved finger lightly over the surface of the cut surface of the brain. Sometimes the edge of these lesions is particularly easy to detect. With diffuse white matter loss, such as that seen in subcortical leukoencephalopathy (Binswanger’s disease) the white matter has a much more rubbery texture, a finding that is usually accompanied by the ‘pockmarked’ appearance of the cribriform state.

Basal ganglia The basal ganglia consist chiefly of the caudate, putamen and globus pallidus. Gross examination will reveal atrophy, and, often some indication of any degree of vascular disease. Atrophy of the caudate/putamen is most conspicuous in conditions such as Huntington’s disease and other diseases principally affecting the extrapyramidal motor system. However, it is important to remember that it can also occur in Alzheimer’s and, particularly, Pick’s disease, in which latter condition it can be as dramatic as in Huntington’s disease. In Pick’s disease, the presence of the associated lobar atrophy serves unequivocally to distinguish it from Huntington’s disease. In hypertensive vascular disease, the expansion of the Virchow-Robin spaces around the lenticulo-striate arteries that supply the caudate/putamen can some times be very conspicuous (and may be misidentified as lacunes by inexperienced pathologists and by neuroradiologists interpreting scans) (Fig. 3.9). Small vessel vascular disease in the basal ganglia often produces a slightly pockmarked appearance of these nuclei (and also of the thalamus as all the deep grey nuclei tend to be affected by vascular disease to a similar extent (something that is not true of degenerative disease where the atrophy tends to be focal and affect specific nuclei leaving others relatively intact). The basal nucleus of Meynert cannot be effectively evaluated by gross examination, and it must therefore be sampled microscopically. The easiest way to locate it is to find the anterior commissure and sample the rectangular block of tissue that extends underneath the putamen/globus pallidus (lentiform nucleus) immediately inferior and posterior to the commissure (Figs. 3.10 and 2.13).

59

60


Fig. 3.9. Dilatation of the perivascular Virchow–Robin spaces in the basal ganglia. This appearance can sometimes be misinterpreted as lacunar disease on imaging studies, but the presence of an artery within each ‘lacune’ demonstrates their true nature.

merits a specific evaluation on gross examination of the brain. Within the hypothalamus, one of the most readily evaluable structures relevant to patients with dementia are the mamillary bodies. These may be atrophic and discoloured in the amnestic dementia (Korsakoff’s psychosis) most especially associated with alcoholism. The dorsomedial thalamus also usually shows marked neuronal loss and consecutive gliosis, but this cannot be appreciated on gross examination of the brain. In the subthalamus, the subthalamic nucleus (body of Luys) can be seen as a lozenge shaped region just medial to the upper extremity of the substantia nigra (see Fig. 13.11) It is not clearly visible in all sections that include the substantia nigra and a little experience is required to decide whether the nucleus is truly atrophic or the section does not contain a good section of the nucleus. The most important condition in which it is clearly atrophic is PSP (Fig. 3.11). Atrophy of this region can produce a characteristic pattern of expansion of the third ventricle where the expansion is of the inferior portion of the ventricle (cf. Fig. 3.11 with Fig. 13.6). Brainstem – mesencephalon, pons and medulla The most readily evaluable regions on naked eye examination of the brainstem are the pigmented nuclei, the substantia nigra and the locus ceruleus. Loss of pigmentation in these nuclei occurs in a quite large number of different conditions.

Fig. 3.10. A section of brain showing the location of the nucleus basalis of Meynert inferior to the anterior commissure. This photograph is almost at its anterior margin and the preferred block would be from the block face posterior to this cut.

Thalamus, hypothalamus and subthalamus One of the rarer associations of dementia is with damage to or degeneration of the thalamus and this may be reflected in bilateral thalamic changes. The syndrome of thalamic dementia has been described in association with a number of different causes of bilateral thalamic damage including prion disease, ischaemia and infarction secondary to occlusion of thalamic perforating arteries, tumour infiltration, trauma, and necrotising infection. These processes may all have manifestations elsewhere in the brain, but from the perspective of the dementia, the thalamic damage may be the most significant and therefore the thalamus

Substantia nigra The substantia nigra is a major component of the extrapyramidal motor system and the principal source of dopaminergic input to the caudate/putamen. Many of the extrapyramidal syndromes, notably idiopathic Parkinson‘s disease, multiple system atrophy and PSP (Fig. 3.12), with and without dementia exhibit depigmentation of the substantia nigra. However, Alzheimer’s disease and Pick’s disease, the quintessential cortical dementias, as well as frontal lobe dementia, have a significant incidence of associated degeneration of the substantia nigra. To reinforce the importance of examining the substantia nigra in cases of dementia, some examples of what appear to be clinically typical cases of Alzheimer’s disease, prove on neuropathological examination to have only the changes of idiopathic Parkinson’s disease with cortical Lewy bodies (the clue here is often a normal weight brain in a patient with a long history of progressive dementia). Hence, the substantia nigra should always be specifically examined in cases of dementia. Locus ceruleus Depigmentation of the locus ceruleus can be seen grossly and is most obvious in cases of idiopathic Parkinson’s disease, PSP and Alzheimer’s disease. Indeed,


Fig. 3.11. A photograph of the thalamus in a case of progressive supranuclear palsy (PSP). The location of the atrophic subthalamic nucleus is indicated by the arrow. The loss of tissue from the subthalamic region has resulted in dilation of the inferior part of the third ventricle while its superior portion is undilated. Comparison with Fig. 3.6(d ) shows the more conventional ‘barrel’-shaped atrophy of the third ventricle with the less selective tissue loss that is seen in Alzheimer’s disease.

none of these diseases is likely to be present in a brain with a normally pigmented locus ceruleus. Cerebellum Cerebellar changes in patients with decline in cognitive function are seen most often in the amnestic dementia of alcoholism (Korsakoff’s syndrome), multiple system atrophy, and progressive supranuclear palsy. Significant folial atrophy is reflected both in the external appearance of the cerebellar hemispheres and by expansion of the fourth ventricle (Fig. 3.13). In alcoholic cerebellar disease, the ataxia is truncal rather than affecting the limbs and the atrophy is most prominent in the vermis. To obtain an adequate section through the tissue, the cerebellum should be detached from the brain stem and sectioned sagittally, just to one side of the midline. It should be noted that to the inexperienced eye the folia of the superior vermis almost always seem to be a little atrophic. Selection of tissue blocks In some cases a strong suspicion of the pathological diagnosis will have been formed by the time the fixed brain has been examined with the naked eye. The evidence obtained from the brain slices may have yielded further diagnostic clues, although it is commonly still difficult at this stage

to positively identify some diseases, including Alzheimer’s disease. Brain blocks for histological examination should be selected at this stage and should be chosen to display as readily as possible the pathological features of the disease process suspected. Most laboratories now use paraffin embedding exclusively or occasionally supplement this with resin embedding. Because it is essential to be certain that a required structure will be included in the sections taken it is advisable to mark the block in such a way that the technician knows which surface to cut. In many cases it is necessary, in addition to taking blocks to confirm a suspected non-Alzheimer’s disease, to take blocks to see if Alzheimer’s disease is also present. This is because Alzheimer’s disease is very common, cannot be excluded without histological examination and not infrequently occurs together with other pathology such as ischaemic dementia or Parkinson’s disease. For the pathologist who is not used to examining the brain, the complexity of the organ seems to produce almost insurmountable problems of obtaining an adequate sample. Compounding this difficulty is the additional problem that in the neuropathological examination of dementia there are often no very obvious focal lesions to be seen on macroscopic examination. Fortunately, there

61

62


(a)

(b)

Fig. 3.12. (a) The pale substantia nigra in a case of idiopathic Parkinson’s disease. (b) Normally pigmented mesencephalon.

are ways in which this problem can be both simplified and systematised. At the risk of uttering a truism, it is perhaps worthwhile to state that the general principle to follow is to sample the areas that provide diagnostic information. It might also usefully be said that the giant sections so beloved of neuropathologists, although pleasing to the eye, and very helpful in the assessment of diffuse white matter loss, are not essential for adequate sampling of the brain in dementia. Schema of selection In this situation adherence to a protocol, such as that produced by CERAD, for the examination of brains with dementia is very useful because it provides a scheme of tissue block selection. The sections listed below are based on those recommended by CERAD, augmented to permit the diagnosis of dementing diseases other than Alzheimer’s disease. In selecting areas to sample where there are no focal lesions that need to be separately sampled the brain has been divided into (i) cortex and white matter; (ii) deep grey nuclei; (iii)

brainstem; and (iv) cerebellum. We have found the following brain samples useful in the diagnosis of dementia: Cortex and white matter (i) Middle frontal gyrus. (ii) Hippocampus and parahippocampal gyrus: many people sample the posterior hippocampus at the level of the lateral geniculate body. (iii) Superior and middle temporal gyri. (iv) *Anterior cingulate cortex: a good section to examine for cortical Lewy bodies. (v) *Amygdala and entorhinal cortex: sensitive to early plaque formation. (vi) Inferior parietal lobule: the white matter in this plane is particularly sensitive to Binswanger changes. (vii) *Occipital cortex including primary optic (striate) cortex. Deep grey nuclei (viii) Caudate/putamen: just posterior to the anterior commissure. When sampled at this level, a good general


(a)

(b) Fig. 3.13. (a) Normal fourth ventricle. (b) A case of PSP where the fourth ventricle is dilated as a consequence of degeneration and neuronal loss in the deep cerebellar nuclei and atrophy of the outflow fibres in the cerebellar white matter.

impression of the basal nucleus of Meynert can be obtained. (ix) Thalamus/Subthalamus: at the level of the subthalamic nucleus. Also permits an additional examination of the substantia nigra and the red nucleus, if the latter is not present in the section of mesencephalon. Brainstem (x) Mesencephalon: Substantia nigra, periaqueductal grey, raphe nuclei. (xi) Pons: locus ceruleus, raphe nuclei, basis pontis. (xii) Medulla: Olive, dorsal motor nucleus of vagus. Cerebellum (xiii) Cerebellar hemisphere: a section that includes dentate nucleus. * Optional

These 13 sections will, in our experience, permit a diagnosis to be made in most cases of dementia. They have the additional merit that if, after looking at them, the diagnosis is still not evident they can be referred to other, and perhaps more experienced, authorities without serious embarrassment as to the adequacy of the sample. They do not, however, quite cover all possible eventualities and the following additions would be recommended in specific situations. (i) If subcortical leukoencephalopathy is suspected: deep white matter in the centrum semiovale, parietal lobe and occipital lobe. (ii) In amnestic dementia when Korsakoff’s psychosis is suspected: hypothalamus through the mamillary bodies. (iii) Cerebellar vermis: as above. (iv) If there is a suspicion of motor neuron loss the XII nerve nucleus needs to be examined. Choice of stains Stains to be used on sections from the blocks selected are to some extent the subject of personal preference. Most diagnoses can be made using haematoxylin and eosin alone but the abnormalities can be much more readily discerned in many instances if additional stains are also used. Silver stains have in the past been particularly valuable for the argyrophilic plaques and neurofibrillary tangles of Alzheimer’s disease, Pick bodies in Pick’s disease and axonal pathology. In many laboratories they have, however, been replaced by immunostains (see below). The selection of a silver stain is markedly influenced by individual laboratory skills and preferences but there are very significant differences in what is revealed by different silver stains (Fig. 3.14). There is, however, wide acceptance of the value of the modified Bielschowsky stain for the demonstration of plaques and tangles that occur in Alzheimer’s Disease. One feature of silver stains that needs to be recognized (and is well known to all who use them regularly) is the considerable variability in sensitivity of the different stains, and the influence of variations in staining technique when the same stain is used in different laboratories (and in the same laboratory at different times and by different personnel). This has been demonstrated on a number of occasions (Duyckaerts et al., 1990) and most recently by Dr Mirra and her collaborators in CERAD, who submitted sections from eight cases of Alzheimer’s Disease of varying severity together with 2 controls to 18 different laboratories in the USA and Canada (Mirra et al., 1994). The different laboratories (and pathologists) were asked to produce quantitative and semi-quantitative measures to assess the cases. The results showed that while there was good agreement among the centres where

63

(a)

(b)

(c)

(d)

Fig. 3.14. A comparison of some different staining methods in a case of Alzheimer’s disease. (a) A haematoxylin and eosin stained section of cerebral cortex showing prominent pyramidal neurons and a cortical blood vessel that contains amyloid in the wall. ×50 (b) Bielschowsky silver stained adjacent section: in the parenchyma both neuritic plaques and neurofibrillary tangles are stained and the amyloid in the blood vessel shows conspicuous silver staining. ×50 (c) A methenamine silver (MS) stain shows at least as many senile plaques, but does not show clearly the neuritic component of the plaques or the neurofibrillary tangles. ×50 (d ) A cross-stain by comparison shows only the neuritic elements of the plaques, the neuropil threads and the neurofibrillary tangles ×50.


Fig. 3.15. Immunocytochemistry of congophilic angiopathy using a polyclonal antibody raised against vascular amyloid. Many vessels and some plaque core are conspicuously stained. There is also some subpial amyloid deposition. ×50.

semiquantitative measures were used, quantitative measures produced significant differences among the raters that reflected major differences in stain sensitivity and staining technique. These findings must engender a certain caution towards the current tendency to the adoption of purely numerical criteria in the diagnosis of, particularly, Alzheimer’s disease. Myelin stains are very helpful in delineating white matter pathology, Nissl stain for neuron cell bodies, congo red for vascular amyloid deposits, periodic acid Schiff (PAS) in lysosomal storage disorders, a reticulin stain in lesions of Wernicke’s encephalopathy and phosphotungstic acid haematoxylin, Holzer or Cajal’s gold chloride method for displaying gliosis. Almost all laboratories supplement these basic tinctures with an increasing variety and range of immunocytochemical stains of significant epitopes in the wide variety of different diseases encountered in patients with dementia. Immunocytochemistry has a very useful place in showing A4 peptide in congophilic angiopathy (Fig. 3.15) and argyrophilic plaques: prion protein in the non-neuritic plaques of Gerstmann Sträussler syndrome: the hyperphosphorylated form of the microtubule-associated tau protein to assess neurofibrillary tangles, other tau inclusions and the neuritic components in Alzheimer’s disease plaques: -synuclein or ubiquitin to show cortical Lewy bodies:

glial fibrillary acidic protein (GFAP) to show gliosis: and Ricin communis agglutinin-lectin or certain macrophage antibodies such as HAM 56 to show macrophages and microglial cell reactions. Ulex europeaus lectin or an antibody to CD34 is valuable for demonstration of endothelium. Many of the required antibodies are available commercially. Immunocytochemistry or in situ hybridization are also useful for confirming the presence of viral or parasitic organisms such as measles in subacute sclerosing pan encephalitis (SSPE), JC virus in progressive multifocal leukoencephalopathy (PML), cytomegalovirus (CMV) in glial nodule encephalitis, human immunodeficiency virus (HIV) in AIDS encephalopathy and toxoplasma in toxoplasmosis. Microscopic examination One of the major benefits of using a protocol such as that produced by CERAD for microscopic examination is that it engenders a systematic approach to histological evaluation. Users quite rapidly develop a familiarity with the range of microscopic appearances that are encountered in the different regions of the brain when they are systematically and repeatedly sampled and examined in a standard way. This familiarity noticeably sharpens the ability of detect more subtle microscopic differences which can be of great value

65

66


in difficult cases. The use of protocols is also very useful in training and helps to inculcate the habit of systematic examination. The microscopic features characteristic of the individual dementing diseases are described in the chapters that follow. We believe, however, that it may be worth making a few general points about the microscopic examination of the brain in cases of dementia. First, it is important to know which types of cell and their locations are susceptible to developing inclusions characteristic of particular diseases. By searching in the wrong type of neuron, or in the wrong layer of cortex, the relevant findings may be missed. Two examples will suffice: Pick bodies may be difficult to find in the neocortex in some cases of Pick’s disease, but are more readily found in the cells of the dentate fascia of the hippocampus; and typical oligodendroglial inclusions of PML are more readily found at or just beyond the margins of the demyelinated foci rather than in their centres. Secondly, it is necessary to be aware of a common tendency to under-estimate nerve cell loss in a nucleus in which there is selective loss of only some of the cells; the atrophy that the nucleus may have undergone causes the remaining cells to be more closely crowded together and the overall neuron density may be little altered. This can occur in Huntington’s disease in which only small neurons of the caudate nucleus are lost and the large ones are crowded together because of the global atrophy of the whole nucleus. Sometimes the degree of cell loss is better evaluated by examining the size, axonal density or depth of myelin staining in the outflow tract of the relevant nucleus – for example, the superior cerebellar peduncles in the case of the cerebellar dentate nuclei, or the myelinated fibre bundles in the caudate-putamen. Either way, when examining grey matter nuclei or outflow tracts, it is essential to have available for comparison normal brain sections of comparable areas using the same stains. Mild or even moderate nerve cell loss is notoriously difficult to evaluate by simple inspection and resort to morphometric methods may be required (Chapter 4). Useful information may be gleaned from the glial stains which are likely to show definite gliosis in the presence of significant neuron loss (note that the converse is not true: gliosis need not imply local neuron loss since the degeneration of the afferent input to a nucleus may also result in gliosis without the neuron cell bodies themselves being affected). Thirdly, some of the abnormal cytopathology, particularly that of Alzheimer’s disease, commonly needs to be at least crudely quantified, and in some contexts the counting of neurofibrillary tangles or argyrophilic plaques may be necessary. In other contexts picture matching is a relatively

objective method of assessing the number of discrete structures in a microscopic field and one to which the human eye is well-attuned. Arriving at a final pathological diagnosis Some cases of dementia present no problems of pathological diagnosis: such is the case, for example, if there are severe changes of Alzheimer’s disease with no other pathology and the clinical history is consonant with this diagnosis; or if there is a strong family history of dementia and the pathology conforms to one of the inherited diseases such as Huntington’s disease or Pick’s disease; or a small cell lung carcinoma is discovered along with features of paraneoplastic encephalitis, despite this condition not having been clinically considered. However, in a significant minority of cases there is difficulty at arriving at a final pathological diagnosis. Three dilemmas account for most such cases: first, there is mixed pathology present and it is uncertain how much weight should be given to each; secondly, there are mild pathological changes but perhaps they are within normal limits for the age of the patient; or thirdly, there are no pathological features of any dementing disease, yet the patient was clearly clinically demented. A fourth related problem is whether to make a final pathological diagnosis of a dementing disease in the absence of a history of dementia. These situations are briefly discussed in turn below. The problem of mixed pathology The most common forms of mixed pathology are the co-existence of Alzheimer’s disease with cerebrovascular disease and of Alzheimer’s disease with Parkinson’s disease. These are discussed below. Alzheimer’s disease occurring with Down syndrome is a special case and is considered in Chapter 10. Other combinations occur occasionally; Alzheimer’s disease with a cerebral glioma or meningioma; Pick’s disease and Alzheimer’s disease; multiple sclerosis and Alzheimer’s disease; Alzheimer’s disease and Creutzfeldt–Jacob disease and cerebrovascular disease with Parkinson’s disease. These have all occurred together in our personally studied cases or have been reported together in cases in the literature. Most combined pathology occurs in the elderly and the commonest diseases are those found together most frequently. This raises the suspicion that their co-existence is often a matter of chance but it may also represent increased susceptibility to one disease in the face of the other. In reporting such cases it is necessary to try to determine the relative contributions of the two diseases specifically to the clinical dementia, as distinct from other clinical neurological features that may also have been present.


Cerebrovascular disease and Alzheimer’s disease It is increasingly recognized that AD and cerebrovascular disease occur together quite commonly (see Table 13.1, for example). Nevertheless, it has been difficult, in cases in which both diseases are present, to make an assessment of the relative importance of the two diseases in producing dementia in any given case. Some progress has been made recently, however, in reaching towards at least the means to make such an assessment. To be able to do this, we need accurate information about which features, both of AD and cerebrovascular disease, are particularly associated with dementia. For AD there has been evidence for some years that neurofibrillary pathology is more important than plaque pathology in promoting dementia (Esiri & Wilcock 1982; McKee et al., 1991; Arriagada et al., 1992; Nagy et al., 1995; Berg et al., 1998). Synaptic pathology has also been considered to be of major significance to cognitive decline (Masliah et al., 1994; Heinonen et al., 1995; DeKosky et al., 1996; Liu et al., 1996, 1999; Heffernan et al., 1998). For cerebrovascular disease some early studies suggested that major cerebral infarcts were of particular importance (Tomlinson et al., 1970). However, the study of Esiri et al. (1997) demonstrated that subcortical vascular disease in white matter and deep grey matter and congophilic angiopathy were particularly correlated with dementia. Recent studies of cases in which both Alzheimer and vascular types of pathology were present have also been informative about how the two types of pathology interact to produce dementia. First, in a study of prospectively assessed nuns, Snowdon et al. (1997) showed that a cortical plaque density sufficient to meet Khachaturian (1985) criteria for the pathological diagnosis of AD was more likely to be associated with dementia if subcortical lacunes were also present than if they were absent. Secondly, Nagy et al. (1997) showed that, in pathologically confirmed AD, a lower density of neurofibrillary tangles in neocortex was needed to produce a given level of cognitive decline if cerebrovascular disease was also present than if it was absent. Thirdly, cases of AD with vascular disease had comparable AD pathology but lower cognitive scores than cases with AD pathology only (Heyman et al., 1998). Fourthly, Goulding et al. (1999) found a significant inverse relationship between cerebrovascular disease severity and Braak stage in demented subjects. Finally, in a study based on the Braak staging of AD pathology (Braak & Braak, 1991) of cases of dementia and controls (in which stages 1 and 2 are usually asymptomatic, stages 3 and 4 may show relatively mild symptoms, and stages 5 and 6 invariably show severe symptoms) (see Chapters 1 and 9) the presence of cerebrovascular disease significantly lowered the cognitive score at Braak stages 1 and 2, and

showed a trend to do so at stages 3 and 4 but no such trend was seen at stages 5 and 6 (Esiri et al., 1999). In yet another study of an unselected community sample of 209 elderly subjects the main determinants of dementia were neuritic plaques and neurofibrillary tangles but a subsidiary determinant was multiple forms of cerebrovascular disease, usually subcortical small vessel disease combined with either lacunes or infarcts (Esiri et al., 2001). From these studies we can suggest the following. (i) When abundant neurofibrillary AD pathology is present in neocortex, this is sufficient, on its own, to produce severe dementia and any vascular disease that is present is likely to have contributed relatively little to the dementia, at least in its later stages. (ii) When plaques are moderate or abundant in neocortex but neurofibrillary pathology is sparse or absent any cerebrovascular disease, particularly subcortical vascular disease, with or without infarcts, is likely to have contributed significantly to dementia. (iii) Congophilic angiopathy is likely to have contributed to dementia, particularly if it is severe, and whether or not there is sufficient associated plaque pathology to meet CERAD criteria for the diagnosis of AD. The reasons for the common association of AD with cerebrovascular disease are complex but probably include the following: (i) Both types of pathology are very common in old age in an unselected population (Esiri et al., 2001). (ii) There are many common risk factors shared by the two conditions. These include hypertension (Skoog et al., 1996; Launer et al., 1995), the ε4 allele of apolipoprotein E (Kalman et al., 1998; Hofman et al., 1997; Marin et al., 1998; Ji et al., 1998; Chapman et al., 1998), atrial fibrillation (Ott et al., 1997), diabetes mellitus (Leibson et al., 1997; Ott et al., 1997), hypercholesterolaemia (Evans et al., 2000) and hyperhomocysteinaemia (Clarke et al., 1998; Evers et al., 1997). (iii) Some vascular pathology (amyloid angiopathy) is a recognized component of AD pathology. (iv) There is some evidence that cerebrovascular disease may predispose to AD pathology (Jendroska et al., 1995). Alzheimer’s disease and Parkinson’s disease Alzheimer’s disease and Parkinson’s disease are both common neurodegenerative diseases affecting the cytoskeleton of neurons in overlapping parts of the brain: nucleus basalis, locus ceruleus, hypothalamus, raphe nuclei, amygdala and, if diffuse Lewy body disease is included as part of the spectrum of Parkinson’s disease, the cerebral cortex. In patients who present with Parkinsonism, approximately

67

68


one third go on to become frankly demented and most of these will be found to have Alzheimer’s changes on neuropathological examination. As noted by Katzman (1993a), it is lucky that Martin et al. (1973) reviewed this topic before the advent of l-dopa treatment of Parkinson’s disease and found a similar fraction becoming demented so that the dementia cannot be attributed to treatment with l-dopa. Conversely, about 15% of patients with Alzheimer’s disease are found to have Parkinsonian changes at post-mortem (Ditter & Mirra, 1987; Gearing et al., 1995) although the extrapyramidal motor symptoms in these patients are often not typical of Parkinson’s disease. The prevalence of Lewy bodies in normal aged patients is not easy to establish with certainty and estimated have ranged between 4.5% and 10.5%. However, Perry and colleagues (1990), in a study that was careful to exclude patients with neuropsychiatric disease concluded that in patients without neuropsychiatric disease the prevalence was only 2.3% (3/131). Despite suggestions that the association is a statistical artefact of chance co-existence of two common diseases (Quinn et al., 1986) it seems possible that their coexistence is more than a matter of chance. This is supported by the unpublished observation that in an unselected community-based autopsy population of elderly subjects the presence of Lewy bodies in the substantia nigra greatly increased the chance (11-fold) of argyrophilic plaques being present in the cerebral cortex. It has been suggested by Hansen et al. (1990) that the combination of Alzheimer’s disease and frequent cortical Lewy bodies constitutes a separate clinically and pathologically recognisable entity that they have called the Lewy body variant of Alzheimer’s disease. Although the cellular mechanisms for this association are unknown, similar pathogenetic mechanisms, perhaps involving free radical damage, may operate in the two diseases. At a practical level, pathological changes of the two diseases coexist sufficiently frequently that when features of one disease are found, features of the other should be looked for. As with Alzheimer’s disease and cerebrovascular disease, it is reasonable to expect that in the presence of both diseases symptoms of dementia may supervene at a lesser degree of severity of pathology than if each disease was present on its own. The problem of mild pathological changes, perhaps within normal limits for age There are no widely accepted criteria for distinguishing categorically between cellular changes of the type found in Alzheimer’s disease in normal ageing and Alzheimer’s disease itself. Attempts to produce such criteria have been made (Khachaturian, 1985) but have not been universally

accepted. Improved criteria are in the process of being refined (Tierney et al., 1988, Mirra et al., 1991; Gearing et al., 1995; Newell et al., 1997; Hyman & Trojanowski, 1997; NIA, 1997). The present situation is galling enough for neuropathologists who regularly examine brains from patients with dementia, but it is perhaps even more confusing for general pathologists who see far fewer cases. We believe that the most reliable pathological feature of Alzheimer’s disease, at present, is the presence of neurofibrillary tangles and neuritic plaques in neocortical association cortex. There is relatively little overlap here between ageing and Alzheimer’s disease. Picture matching of argyrophilic plaque densities is also considered helpful in distinguishing between normal ageing changes and Alzheimer’s disease, but in our experience the range of plaque densities found in elderly subjects with no prospectively detected dementia can be remarkably wide and substantially overlap the range seen in Alzheimer’s disease. This may reflect variable thresholds before clinical dementia supervenes, these thresholds perhaps being related to previous intellectual performance or cognitive ‘reserve’ function, although this has not been systematically studied. There is also much variation of the literature in the thoroughness with which intellectual deterioration has been looked for and this will clearly influence the frequency with which such deterioration is found. As discussed above, the normal thresholds that may be considered to divide Alzheimer change in normal ageing from Alzheimer’s disease almost certainly need modifying downwards in the face of additional pathology, most commonly cerebrovascular disease or Parkinson’s disease. Whether, after applying such modified criteria, one is justified in making a final pathological diagnosis of Alzheimer’s disease is, however, open to question. The two other diagnostic problems that regularly present themselves, are a history without a diagnosis, and its converse, a diagnosis without a history! The problem of no pathological features to account for clinical dementia There can be few, if any, pathologists who claim always to be able to produce a diagnosis in cases of clinical dementia. Almost all the large clinicopathological series of cases of dementia contain a small residuum of cases that have defied the best efforts of the pathologist. In recent series some of which were highly selected, the percentage is very low, varying from 0 to 3.0% (Gearing et al., 1995), but in unselected cases the percentages are almost certainly higher. To account for these undiagnosed cases, it is tempting to envisage a whole world of novel pathological mechanisms


of which we are currently entirely ignorant. There may indeed be some such examples within the cases that currently defy diagnosis but there are a number of rather more prosaic explanations. First, although it is impossible to be certain, it seems likely that the most frequent explanation for the inability to find morphological changes in the brain in a patient with what is clinically described as dementia, is that the patient was suffering not from dementia but from depression. It has been well demonstrated that there are a few cases of clinical dementia which are virtually indistinguishable from depression. In one clinical series the proportion of such cases was put at 8% (Marsden & Harrison, 1972), and the percentage is likely to vary depending on the energy and expertise with which the clinical diagnosis is pursued. Although not currently feasible routinely, it is likely that advances in neurochemical knowledge will allow us to identify this group in the future. Secondly, it is well recognized that some chronic drug intoxications without described neuropathology can induce a clinical state indistinguishable from organic dementia (Oxbury, 1991). These drugs include tranquillizers, barbiturates, l-dopa, isoniazid, and lithium. Thirdly, other primarily psychiatric pseudodementias are also rarely encountered. Fourthly, there are a few families with inherited spongiform encephalopathy in which some demented members have shown no pathological abnormalities despite the presence of clear-cut pathology in others (Medori et al., 1992). It may turn out to be the case that further investigation of such cases by screening for a prion disease mutation is worthwhile. Fifthly, the patient may have been suffering from a rare or undescribed dementing disease that is missed by the pathologist. Finally, just as the thresholds at which Alzheimer-type changes are considered to be Alzheimer’ disease are applicable to an average population, and a higher burden of such change can probably be carried by someone of above average intellect and education without developing dementia (Katzman, 1993b), so it may be speculated that someone of below average intellect or functional reserve may develop dementia while his or her burden of Alzheimer-type change is still within the range of normal. The problem of whether to make a final pathological diagnosis of a dementing disease in the absence of a history of dementia The converse of the patient with a history but no diagnosis is that of morphological changes of a specific diagnosis in the

absence of any history of disease. This situation is probably most frequently encountered in brains containing numerous plaques and tangles that would, with the appropriate history be confidently diagnosed as Alzheimer’s disease. This problem really occurs in two forms. The more common is that of the patient who has no history because he has not been medically examined, so that the problem is really that of an unknown rather than a definitively negative history. There is no ‘right’ way to classify such cases, but CERAD adopts the position that all such cases should be classified as possible Alzheimer’s disease, no matter how severe the morphological changes. The other, less common form of this problem is that of the patient with significant morphological findings who has been medically examined and found to be within the normal range of mental function. There are two general reasons that might explain this type of finding. First, there is almost always a time lag between the last mental state examination and the death of the patient, during which significant deterioration in mental function might have taken place. The second possible explanation is that the criteria used for the diagnosis are themselves faulty in that they are not closely related to mental function. This is particularly the case in Alzheimer’s disease where total plaque count and amyloid burden, are not closely related to cognitive scores. Hence, making the diagnosis of AD on the basis of plaque counts alone (particularly if ‘diffuse’ plaques are included) may lead to an inappropriate diagnosis of AD. The pathologist must be a faithful chronicler of what she or he finds. If well marked pathological changes are present that warrant the name of a pathological dementing condition then that name should be applied, albeit with the qualification that it was an incidental finding. We do not favour the application of altered pathological criteria for Alzheimer’s disease based on the clinical presence or absence of dementia (Khachaturian, 1985). We are too well aware of how easily dementia can be clinically overlooked or dismissed in elderly subjects for this to be advocated.

Brain biopsies in patients with dementia Brain biopsy is rarely resorted to for the diagnosis of the cause of dementia in the very elderly, but is employed to a varying extent in different centres in younger subjects and in patients with rapidly progressive undiagnosed disease. There are several reasons for the sparing use of brain biopsy for the diagnosis of dementia. First, most cases are likely to be due to Alzheimer’s disease, a disease for which there is so far no well-founded treatment and most clinicians

69

70


Table 3.2. Dementing conditions diagnosable on brain biopsy Condition

Major pathological finding

Alzheimer’s disease Pick’s disease Creutzfeldt-Jacob disease Gerstmann-Sträussler syndrome Cerebral arteritis Familial amyloid angiopathies Progressive multifocal leukoencephalopathy Herpes simplex encephalitis Subacute sclerosing panencephalitis eukodystrophy Whipple’s disease Syphilis – general paresis Intravascular lymphoma

Plaques and tangles Pick bodies/cells Spongiform change Prion amyloid plaques

Diffuse glioma/multiple metastases HIV encephalitis

Vascular changes Specific vascular amyloid Viral inclusion bodies Viral inclusion bodies Viral inclusion bodies White matter changes PAS positive macrophages Inflammation/organisms Perivascular/parenchymal tumour Tumour Multinucleated macrophages and HIV-specific antigens

consider it unethical to undertake an invasive procedure such as brain biopsy, to diagnose a disease for which there is no treatment. Secondly, there are non-invasive diagnostic methods such as single-photon emission tomography (SPET), position emission tomography (PET), magnetic resonance imaging (MRI) and computerised axial tomography (CT) scanning, which have helped to push recent diagnostic accuracy in cases of progressive dementia to 80– 90%, at least in Alzheimer’s disease (Tierney et al., 1988; Joachim et al., 1988). However, with all the sophistication of modern imaging and investigative technique it is important not to forget old-fashioned diseases and older simpler methods of diagnosis. One of us had had the experience of diagnosing neurosyphilis on a brain biopsy. The diagnosis had not been considered prior to biopsy, and the serological tests not performed. It is also necessary to be aware that brain biopsy itself may not yield a diagnosis because the small sample of cortex and white matter that is removed may lack specific diagnostic features. Finally, there is a remote but real risk of transmitting Creutzfeldt–Jakob disease by contamination of neurosurgical equipment as a result of performing biopsy, if this is the disease responsible (see Chapter 17). The range of possible pathology to be found

in a brain biopsy from a case of progressive dementia is very wide (Table 3.2). Details can be found in appropriate chapters of this book. If a decision to perform a brain biopsy is made, the biopsy is usually taken from the frontal or temporal lobe cortex of the non-dominant hemisphere. It should include subcortical white matter so that there is an opportunity to diagnose white matter diseases such as progressive multifocal leukoencephalopathy, multiple sclerosis, a leukodystrophy or HIV encephalopathy as well as predominantly cortical diseases such as Alzheimer’s disease, Pick’s disease or Creutzfeldt–Jakob disease. The sample should be divided, with a small portion being taken for resin embedding and electron microscopy, another small sample snap frozen for cytostat sectioning and the remainder fixed in neutral, buffered formalin and then embedded in paraffin wax. It is wise to treat the specimens as potentially infective as occasional cases may be due to human immunodeficiency virus infection, progressive multifocal leukoencephalopathy or Creutzfeldt–Jakob disease. Paraffin sections should be stained with haematoxylin and eosin, a myelin stain, a stain for reactive astrocytes, an axonal stain and methods for showing neurofibrillary tangles, argyrophilic plaques and vascular amyloid deposits. Electron microscopy may be helpful if a viral infection or leukodystrophy is suspected. Cryostat sections are useful for some histochemical, immunocytochemical or nucleic acid hybridization procedures.

REFERENCES Advisory Committee on Dangerous Pathogens (1994). Precautions for work with human and animal transmissible spongiform encephalopathies. 1st. edn. HMSO, London. Alafuzoff, I., Iqbal, K., Friden, H., Adolfson, R. & Winblad, B. (1987). Histopathological criteria for progressive dementia disorder. Clinico-pathological correlation and classification by multivariate data analysis. Acta Neuropathol., 74, 209–25. Anderson, J. M., Hubbard, B. M., Coghill, G. R. & Slidders, W. (1983). The effect of advanced old age on the neurone content of the cerebral cortex. J. Neurol. Sci., 58, 235–46. Arriagada, P. V., Growdon, J. H., Hedley-Whyte, E. T. & Hyman, B. T. (1992). Neurofibrillary tangles but not senile plaques parallel duration and severity of Alzheimer’s disease. Neurology, 42, 631–9. Barclay, L. L., Zemcov, A., Blass, J. P. & Sansone, J. (1985). Survival in Alzheimer’s disease and vascular dementias. Neurology, 35, 834–40. Bell, J. E. & Ironside, J. W. (1993). How to tackle a possible Creutzfeldt–Jakob disease necropsy. J. Clin. Path., 46, 193–7. Berg, L., McKeel, D. W., Jr., Miller, J. P. et al. (1998). Clinicopathologic studies in cognitively healthy aging and Alzheimer’s


disease: relation of histologic markers to dementia severity, age, sex, and apolipoprotein E genotype. Arch. Neurol., 55, 326–35. Blacker, D., Albert, M. S., Bassett, S. S., Go, R. C., Harrell, L. E. & Folstein, M. F. (1994). Reliability and validity of NINCDSADRDA criteria for Alzheimer’s disease. The National Institute of Mental Health Genetics Initiative. Arch. Neurol., 51, 1198– 204. Boller, F., Lopez, O. L. & Moossy, J. (1989). Diagnosis of dementia: clinico-pathologic correlations. Neurology, 39, 76–9. Braak, H. & Braak, E. (1991). Neuropathological staging of Alzheimer related changes. Acta Neuropatholo., 82, 239–59. Brown, P. (1988a). The clinical neurology and epidemiology of Creutzfeldt–Jakob disease, with special reference to iatrogenic cases. In Novel infectious Agents and the Central Nervous System. Bock, G., Marsh, J. (eds.) Ciba Foundation Symposium 135, John Wiley, Chichester. (1988b). Human growth hormone therapy and Creutzfeldt–Jakob disease: a drama in three acts. Pediatrics, 81, 85–92. Brown, P., Rodgers-Johnson, P., Cathala, F., Gibbs, C. J. & Gajdusek, D. C. (1984). Creutzfeldt–Jakob disease of long duration: clinicopathological characteristics, transmissibility, and differential diagnosis. Ann. Neurol., 16, 295–304. Brown, P., Cathala, F., Castaigne, P. & Gajdusek, D. C. (1986). Creutzfeldt–Jakob disease: clinical analysis of a consecutive series of 230 neuropathologically verified cases. Ann. Neurol., 20, 597–602. Brown, P., Cathala, F., Raubertas, R. F., Gajdusek, D. C. & Castaigne, P. (1987). The epidemiology of Creutzfeldt–Jakob disease: conclusion of a 15 year investigation in France and review of the world literature. Neurology, 37, 895–904. Brown, P., Wolff, A. & Gajdusek, D. C. (1990). A simple and effective method for inactivating virus infectivity in formalin-fixed tissue samples from patients with Creutzfeldt–Jakob disease. Neurology, 40, 887–90. Budka, H., Aguzzi, A., Brown, P. et al. (1995). Tissue handling in suspected Creutzfeldt–Jakob disease and other human spongiform encephalopathies. Brain Pathol, 5, 319–22. Burns, A., Luthert, P., Levy, R. et al. (1990). Accuracy of clinical diagnosis of Alzheimer’s disease. Br. Med. J. 301, 1026. Center for Disease Control (1985). Fatal degenerative neurologic disease in patients who received pituitary-derived human growth hormone. MMWR, 34, 359–60, 365–6. (1987). Rapidly progressive dementia in a patient who received a cadaveric dura mater graft. MMWR, 36, 49–50, 55. Chapman, J., Wang, N., Treves, T. A., Korczyn, A. D. & Bornstein, N. M. (1998). ACE, MTHFR, factor V Leiden, and APOE polymorphisms in patients with vascular and Alzheimer’s dementia. Stroke, 29, 1401–4. Clarke, R., Smith, A. D., Jobst, K. A., Refsum, H., Sutton, L. & Ueland, P. M. (1998). Folate, vitamin B12, and serum total homocysteine levels in confirmed Alzheimer disease. Arch. Neurol., 55, 1449– 55. Collinge, J., Owen, F., Poulter, M. et al. (1990). Prion dementia without characteristic pathology. Lancet, 336, 7–9.

Committee on Health Care Issues, American Neurological Association (1986). Precautions in handling tissues, fluids and other contaminated materials from patients with documented or suspected Creutzfeldt–Jakob disease. Ann. Neurol., 19, 75–7. Crystal, H., Dickson, D., Fuld, P. et al. (1988). Clinicopathologic studies in dementia: Non-demented subjects with pathologically confirmed Alzheimer’s disease. Neurology, 38, 1682–7. Cutler, N. R., Brown, P. B., Narayan, T., Parisi, J. E., Janota, F. & Baron, H. (1984). Creutzfeldt–Jakob disease: a case of 16 years duration. Ann. Neurol., 15, 107–10. Davis, P. J. M. & Wright, E. A. (1977). A new method for measuring cranial cavity volume and its application to the assessment of cerebral atrophy at autopsy. Neuropathol. Appl. Neurobiol., 3, 341–58. Dekaban, A. S. & Sadowsky, D. (1978). Changes in brain weights during the span of human life: relation of brain weights to body heights and body weights. Ann. Neurol., 4, 345–56. DeKosky, S. T., Scheff, S. W. & Styren, S. D. (1996). Structural correlates of cognition in dementia: quantification and assessment of synapse change. Neurodegeneration, 5, 417–21. Department of Health (2001). Reference guide to consent for examination or treatment. www.doh.gov.uk/consent Ditter, S. M. & Mirra, S. S. (1987). Neuropathologic and clinical features of Parkinson’s disease in Alzheimer disease patients. Neurology, 37, 754–60. Duffy, P. C., Wolf, J., Collins, G., DeVoe, A. G., Streeten, B. & Cowan, D. (1974). Possible person-to-person transmission of Creutzfeldt–Jakob disease. N. Engl. J. Med., 290, 692–3. Duyckaerts, C., Delaere, P., Hauw, J. J. et al. (1990). Rating of the lesions in senile dementia of the Alzheimer type: concordance between laboratories. A European multi-centre study under the auspices of EURAGE. J. Neurol. Sci., 97, 295–323. Esiri, M. M. & Wilcock, G. K. (1982). Plaques, tangles and dementia. A quantitative study. J. Neurol. Sci., 56, 343–56. Esiri, M. M., Wilcock, G. K. & Morris, J. H. (1997). Neuropathological assessment of the lesions of significance in vascular dementia. J. Neurol. Neurosurg. Psychiatry, 63, 749–53. Esiri, M. M., Nagy, Z., Smith, M. Z., Barnetson, L. & Smith, A. D. (1999). Cerebrovascular disease and threshold for dementia in the early stages of Alzheimer’s disease [letter]. Lancet, 354, 919–20. Esiri, M. M., Matthews, F., Brayne, C. & Ince, P. G. (2001). Pathological correlates of late onset dementia in a multicentre communitybased population in England and Wales. Lancet, 357, 169–75. Ettlin, T. M., Staehelin, H. B., Kischka, U. et al. (1989). Computed tomography, electroencephalography, and clinical features in the differential diagnosis of senile dementia. A prospective clinico-pathologic study. Arch. Neurol., 46, 1217–20. Evans, R. M., Emsley, C. L., Gao, S. et al. (2000). Serum cholesterol, APOE genotype, and the risk of Alzheimer’s disease: a population-based study of African Americans. Neurology, 54, 240–2. Evers, S., Koch, H. G., Grotemeyer, K. H., Lange, B., Deufel, T. & Ringelstein, E. B. (1997). Features, symptoms, and

71

72


neurophysiological findings in stroke associated with hyperhomocysteinemia. Arch. Neurol., 54, 1276–82. Galasko, D., Hansen, L. A., Katzman, R. et al. (1994). Clinical– neuropathological correlations in Alzheimer’s disease and related dementias. Arch. Neurol., 51, 888–95. Gearing, M., Mirra, S. S., Hedreen, J. C., Sumi, S. M., Hansen, L. A. & Heyman, A. (1995). The consortium to establish a registry for Alzheimer’s disease (CERAD). Part X. Neuropathology confirmation of the clinical diagnosis of Alzheimer’s disease. Neurology, 45, 461–6. General Medical Council (1998). Seeking patients’ consent: the ethical considerations. www.gmc-uk.org Gilleard, C. J., Kellett, J. M., Coles, J. A. et al. (1992). The St George’s dementia bed investigation study: a comparison of clinical and pathological diagnosis. Acta Psychiatr. Scand., 85, 264–9. Goulding, J. M., Signorini, D. F., Chatterjee, S. et al. (1999). Inverse relation between Braak stage and cerebrovascular pathology in Alzheimer predominant dementia. J. Neurol. Neurosurg. Psychiatry, 67, 654–7. Haltia, M., Viitanen, M., Sulkava, R. et al. (1994). Chromosome 14encoded Alzheimer’s disease: genetic and clinicopathological description. Ann. Neurol., 36, 362–7. Hansen, L., Salmon, D., Galasko, D. et al. (1990). The Lewy body variant of Alzheimer’s disease: a clinical and pathological entity. Neurology, 40, 1–8. Heffernan, J. M., Eastwood, S. L., Nagy, Z., Sanders, M. W., McDonald, B. & Harrison, P. J. (1998). Temporal cortex synaptophysin mRNA is reduced in Alzheimer’s disease and is negatively correlated with the severity of dementia. Exp. Neurol., 150, 235–9. Heinonen, O., Soininen, H., Sorvari, H. et al. (1995). Loss of synaptophysin-like immunoreactivity in the hippocampal formation is an early phenomenon in Alzheimer’s disease. Neuroscience, 64, 375–84. Heyman, A., Fillenbaum, G. G., Welsh-Bohmer, K. A. et al. (1998). Cerebral infarcts in patients with autopsy-proven Alzheimer’s disease: CERAD, part XVIII. Consortium to Establish a Registry for Alzheimer’s Disease. Neurology, 51, 159–62. Hofman, A., Ott, A., Breteler, M. M. et al. (1997). Atherosclerosis, apolipoprotein E, and prevalence of dementia and Alzheimer’s disease in the Rotterdam Study. Lancet, 349, 151–4. Homer, A., Honavar, M., Lantos, P., Hastie, I., Kellet, J. & Millard, P. (1988). Diagnosing dementia: do we get it right. Br. Med. J., 297, 894–6. Hyman, B. T. & Trojanowski, J. Q. (1997). Consensus recommendations for the postmortem diagnosis of Alzheimer disease from the National Institute on Aging and the Reagan Institute Working Group on diagnostic criteria for the neuropathological assessment of Alzheimer disease. J. Neuropathol. Exp. Neurol., 56, 1095–7. Jellinger, K., Danielczyk, W., Fischer, P. & Gabriel, E. (1990). Clinicopathological analysis of dementia disorders in the elderly. J. Neurol. Sci., 95, 239–58. Jendroska, K., Poewe, W., Daniel, S. E. et al. (1995). Ischemic stress induces deposition of amyloid beta immunoreactivity in human brain. Acta Neuropathol, 90, 461–6.

Ji, Y., Urakami, K., Adachi, Y., Maeda, M., Isoe, K. & Nakashima, K. (1998). Apolipoprotein E polymorphism in patients with Alzheimer’s disease, vascular dementia and ischemic cerebrovascular disease. Dement. Geriatr. Cogn. Disord., 9, 243–5. Joachim, C. L., Morris, J. H. & Selkoe, D. J. (1988). Clinically diagnosed Alzheimer’s disease: autopsy results in 150 cases. Ann. Neurol., 24, 50–6. Jobst, K. A., Smith, A. D., Szatmari, M. et al. (1992). Detection in life of confirmed Alzheimer’s disease using a simple measurement of medial temporal lobe atrophy by computed tomography. Lancet, 340, 1179–83. Jobst, K. A., Smith, A. D., Szatmari, M. et al. (1994). Rapidly progressing atrophy of medial temporal lobe in Alzheimer’s disease. Lancet, 343, 829–30. Kalman, J., Juhasz, A., Csaszar, A. et al. (1998). Increased apolipoprotein E4 allele frequency is associated with vascular dementia in the Hungarian population. Acta Neurol. Scand., 98, 166–8. Katzman, R. (1993a). Clinical and epidemiological aspects of Alzheimer’s disease. Clin. Neurosci., 1, 165–70. (1993b). Education and the prevalence of dementia and Alzheimer’s disease. Neurology, 43, 13–20. Khachaturian, Z. S. (1985). Diagnosis of Alzheimer disease. Arch. Neurol., 42, 1097–105. Klatka, L. A., Schiffer, R. B., Powers, J. M. & Kazee, A. M. (1996). Incorrect diagnosis of Alzheimer’s disease. A clinicopathologic study. Arch. Neurol., 53, 35–42. Knopman, D. S., DeKosky, S. T., Cummings, J. L. et al. (2001). Practice parameter: diagnosis of dementia (an evidence-based review). Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology, 56, 1143–53. Kokmen, E., Offord, K. P. & Okazaki, H. (1987). A clinical and autopsy study of dementia in Olmsted county, Minnesota, 1980–1981. Neurology, 37, 426–30. Launer, L. J., Masaki, K., Petrovitch, H., Foley, D. & Havlik, R. J. (1995). The association between midlife blood pressure levels and late-life cognitive function. The Honolulu–Asia Aging Study. JAMA, 274, 1846–51. Leibson, C. L., Rocca, W. A., Hanson, V. A. et al. (1997). The risk of dementia among persons with diabetes mellitus: a population-based cohort study. Ann. NY Acad. Sci., 826, 422–7. Lim, A., Tsuang, D., Kukull, W. et al. (1999). Cliniconeuropathological correlation of Alzheimer’s disease in a community-based case series. J. Am. Geriatr. Soc., 47, 564–9. Liu, X., Erikson, C. & Brun, A. (1996). Cortical synaptic changes and gliosis in normal aging, Alzheimer’s disease and frontal lobe degeneration. Dementia, 7, 128–34. Liu, X., Passant, U., Risberg, J., Warkentin, S. & Brun, A. (1999). Synapse density related to cerebral blood flow and symptomatology in frontal lobe degeneration and Alzheimer’s disease. Dement. Geriatr. Cogn. Disord., 10, 64–70. McKee, A. C., Kosik, K. S. & Kowall, N. W. (1991). Neuritic pathology and dementia in Alzheimer’s disease. Ann. Neurol. 30, 156–65.


Marin, D. B., Breuer, B., Marin, M. L. et al. (1998). The relationship between apolipoprotein E, dementia, and vascular illness. Atherosclerosis, 140, 173–80. Marsden, C. D. & Harrison, M. F. G. (1972). Outcome of investigations of patients with presenile dementia. BMJ, 2, 249–52. Martin, E. M., Wilson, R. S., Penn, R. D. et al. (1987). Cortical biopsy results in Alzheimer’s disease: correlation with cognitive deficits. Neurology, 37, 1201–4. Martin, W. E., Loewenson, R. B., Resch, J. A. & Baker, A. B. (1973). Parkinson’s disease. Clinical analysis of 100 patients. Neurology, 23, 783–90. Masliah, E., Mallory, M., Hansen, L., DeTeresa, R., Alford, M. & Terry, R. (1994). Synaptic and neuritic alterations during the progression of Alzheimer’s disease. Neurosci. Lett., 174, 67–72. Medical Research Council (1999). Human tissue and biological samples for use in research. Report of the MRC Working Group to develop operational and ethical guidelines. Medori, R., Tritschler, H.-J., LeBlanc, A. et al. (1992). Fatal familial insomnia, a prion disease with a mutation at codon 178 of the prion protein gene. N. Engl. J. Med., 326, 444–9. Mendez, M. F., Mastri, A. R., Sung, J. H. & Frey, W. H. (1992). Clinically diagnosed Alzheimer disease: neuropathologic findings in 650 cases. Alzheimer Dis. Assoc. Disord., 6, 35–43. Miller, A. K., Alston, R. L. & Corsellis, J. A. (1980). Variation with age in the volumes of grey and white matter in the cerebral hemispheres of man: measurements with an image analyser. Neuropathol. Appl. Neurobiol., 6, 119–32. Miller, D. C. (1988). Creutzfeldt–Jakob disease in histopathology technicians. N. Engl. J. Med., 318, 853–4. Mirra, S. S., Heyman, A., McKeel, D. et al. (1991). The Consortium to Establish a Registry for Alzheimer’s Disease (CERAD). II Standardization of the neuropathologic assessment of Alzheimer’s disease. Neurology, 41, 479–86. Mirra, S. S., Gearing, M., McKeel, D. W. et al. (1994). Interlaboratory comparison of neuropathology assessments in Alzheimer’s disease, A Study of the Consortium to Establish a Registry of Alzheimer’s Disease (CERAD) J. Neuropath. Exp. Neurol., 53, 303–15. Mölsä, P. K., Paljävi, L., Rinne, U. K., Rinne, J. O. & Säkö, E. (1985). Validity of clinical diagnosis in dementia, a prospective clinicopathological study. J. Neurol. Neurosurg. Psychiatry, 48, 1085– 90. Morris, J. C., McKeel, D. W., Fulling, K., Torack, R. M. & Berg, L. (1988). Validation of clinical diagnostic criteria for Alzheimer’s disease. Ann. Neurol., 24, 17–22. ¨ Muller, H. F. & Schwartz, G. (1978). Electroencephalograms and autopsy findings in geropsychiatry. J. Gerontol., 4, 504–13. Nagy, Z., Esiri, M. M., Jobst, K. A. et al. (1995). Relative roles of plaques and tangles in the dementia of Alzheimer’s disease: correlations using three sets of neuropathological criteria. Dementia, 6, 21–31. (1997). The effects of additional pathology on the cognitive deficit in Alzheimer disease. J Neuropathol. Exp. Neurol., 56, 165–70. Nagy, Z., Esiri, M. M., Hindley, N. J. et al. (1998). Accuracy of clinical operational diagnostic criteria for Alzheimer’s disease in

relation to different pathological diagnostic protocols. Dement. Geriatr. Cogn. Disord., 9, 219–26. National Institute on Aging and Reagan Institute Working Group on diagnostic criteria for the neuropathological assessment of Alzheimer’s disease (1997). Consensus recommendations for the post mortem diagnosis of Alzheimer’s disease. Neurobiol Aging 18(4S)S1–S2. Neary, D., Snowden, J. S., Bowen, D. M. et al. (1986). Neuropsychological syndromes in presenile dementia due to cerebral atrophy. J. Neurol. Neurosurg. Psychiatry, 49, 163–74. Newell, K., Hyman, B. T., Growdon, J. H. & Hedley-Whyte, E. T. (1997). Evaluation of NIA-Reagan Institute criteria for the neuropathological diagnosis of Alzheimer’s disease. J. Neuropathol. Exp. Neurol., 56, 593. Ott, A., Stolk, R. P., van Harskamp, F. et al. (1997). Diabetes mellitus and the risk of dementia in an elderly population. In Ott, A. Risk of Dementia. Delft, Juidels en Brinkman BV, pp. 49–62. Oxbury, J. M. (1991). Dementia. In Swash, M. & Oxbury, J. M. (eds.). Clinical Neurology. Churchill Livingstone, Edinburgh, pp. 112– 19. Perl, D. P., Pendlebury, W. W. & Bird, E. D. (1984). Detailed neuropathologic evaluation of banked brain submitted with clinical diagnosis of Alzheimer’s disease. In Wurtman, R., Corkin, S. & Growden, J. (eds.) Alzheimer’s disease: advances in basic research and therapies. CBSM Cambridge, MA p. 463. Perry, R. H., Irving, D. & Tomlinson, B. E. (1990). Lewy body prevalence in the aging brain: relationship to neuropsychiatric disorders, Alzheimer type pathology and catecholaminergic nuclei. J. Neurol. Sci., 100, 223–33. Quinn, N. P., Rossor, M. N. & Marsden, C. D. (1986). Dementia and Parkinson’s disease – pathological and neurochemical considerations. Br. Med. Bull., 42, 86–90. Risse, S. C., Raskind, M. A., Nochlin, D. et al. (1990). Neuropathological findings in patients with clinical diagnoses of probable Alzheimer’s disease. Am. J. Psychiat., 147, 168–72. Royal College of Pathologists (2000). Guidelines for the retention of tissues and organs at post-mortem examination. Schmidt, R., Fazekas, F., Kapeller, P., Schmidt, H. & Hartung, H. P. (1999). MRI white matter hyperintensities: three-year followup of the Austrian Stroke Prevention Study. Neurology, 53, 132– 9. Skoog, I., Lernfelt, B., Landahl, S. et al. (1996). 15-year longitudinal study of blood pressure and dementia. Lancet, 347, 1141–5. Smith, C. D., Snowdon, D. A., Wang, H. & Markesbery, W. R. (2000). White matter volumes and periventricular white matter hyperintensities in aging and dementia. Neurology, 54, 838–42. Snowdon, D. A., Greiner, L. H., Mortimer, J. A., Riley, K. P., Greiner, P. A. & Markesbery, W. R. (1997). Brain infarction and the clinical expression of Alzheimer disease. The Nun Study. JAMA, 277, 813–17. Sulkava, R., Haltia, M., Paetau, A. et al. (1983). Accuracy of clinical diagnosis in primary degenerative dementia: correlation with neuropathological findings. J. Neurol. Neurosurg. Psychiatry, 46, 9–13.

73

74


Tierney, M. C., Fisher, R. H., Lewis, A. J. et al. (1988). The NINCDSADRDA work group criteria for the clinical diagnosis of probable Alzheimer’s disease: a clinicopathologic study of 57 cases. Neurology, 38, 359–64. Todorov, A. B., Go, R. C. P., Constantinidis, J. & Elston, R. C. (1975). Specificity of the clinical diagnosis of dementia. J. Neurol. Sci., 26, 81–98.

Tomlinson, B. E., Blessed, G. & Roth, M. (1970). Observations on the brains of demented old people. J. Neurol. Sci., 11, 205–42. Victoroff, J., Mack, W. J., Lyness, S. A. & Chui, H. C. (1995). Multicenter clinicopathological correlation in dementia. Am. J. Psychiatry, 152, 1476–84. Wade, J. P. H., Mirsen, T. R., Hachinski, V. C. et al. (1987). The clinical diagnosis of Alzheimer’s disease. Arch. Neurol., 44, 24–9.

4 Morphometric methods and dementia Michael C. Irizarry Alzheimer Disease Research Unit Massachusetts General Hospital – East, Charlestown, MA, USA

Introduction

Profile-counting methods

Classical neuropathology implements gross and microscopic methods to determine diagnosis and neuroanatomical localization of pathology. Diagnostic standards for dementing illnesses have been refined as semi-quantitative grading systems for dementia pathology, allowing rough clinical-pathological correlations. These staging systems include the Consortium to Establish a Registry for Alzheimer’s Disease (CERAD) and the Braak and Braak criteria for Alzheimer’s disease (AD), and consensus neuropathological criteria for dementia with Lewy bodies (DLB) (Braak & Braak, 1991; Mirra et al., 1991; McKeith et al., 1996). For more detailed quantitative analysis of the neuropathology of dementia, sensitive morphometric methods have been developed. The dominant method in recent years is stereology, which applies a series of rules to overcome biases in counting objects (e.g. neurons) on a slide (Sterio, 1984; Gundersen et al., 1988a,b). Stereological approaches to the assessment of anatomical volumes, neurite length, synapse counts, neuron counts and neuron size have utilized such varied resources as gross brains, MRI scans, light microscopy, confocal microscopy, and electron microscopy (Geinisman et al., 1996; Everall et al., 1999; Peterson, 1999; Roberts et al., 2000). This chapter reviews principles of stereology as applied to anatomical volume measurements and neuron counting at the light microscope level, with a particular emphasis on AD. Other aspects of stereology, such as using the rotator or nucleator for particle volume and cycloid intercepts for surface area, can be found in recent reviews and texts (Gundersen et al., 1988a; Howard & Reed, 1998).

Since it is inefficient to count all the neurons in a particular brain region, methods have been developed to estimate neuron number from a sampling of microscopic fields. Sampling strategy is a critical consideration in morphometric analysis. From an appropriate sampling of the entire population assessed, estimates and statistical inferences can be made regarding the population. Statistical sampling theory has yielded techniques that reduce bias (the systematic deviation of the estimated value from the true value) and increase precision (the statistical variance between the estimate and the true value). The most common counting methods utilized are (i) profile-counting and (ii) stereologybased (Geuna, 2000). In typical profile-counting methods, cross-sectional neuron profiles are counted in mounted sections to determine a neuron density measurement per unit area. The assumption of this method – that cross-sectional profiles are proportional to total number of neurons – is not always justified, since the probability of a neuron being counted in cross section is affected by the size, shape, and orientation of the neuron; section thickness and orientation; and changes in tissue due to atrophy and processing (Mayhew & Gundersen, 1996). Artefacts due to sectioning and the optical depth of field of microscope objectives include overprojection (‘Holmes effect’ – the observed image includes cells outside the focal plane), and truncation (‘lost caps’ – tangentially sectioned cells are too small to be counted in the focal plane) (Peterson, 1999). Corrections have been devised for large spherical particles and small spherical particles (Abercrombie, 1946; Weibel & Gomez, 1962; Ebbesson & Tang, 1965).


75

76

M. C. Irizarry

Stereology-based counting methods, including those described in this chapter, reduce the biases and assumptions of profile-counting methods. Profile-counting methods may still be adequate if there is not significant variability in the size, shape, and orientation of objects counted. Nonetheless, for profile-counting and other methods requiring models to correct for differential probabilities of objects being counted, the sampling strategy and its biases should be carefully considered and validated by calibration studies with stereological methods (Geuna, 2000).

Stereology-based counting methods Stereology-based methods reduce the methodological biases and assumptions associated with profile-counting techniques. Stereology assesses specific features of twodimensional measurements obtained from sections spanning an anatomical region to calculate three-dimensional characteristics of the structure. In neuropathological studies involving dementia, these three-dimensional characteristics include volume; total neuron, synapse, or inclusion count; or fibre length. The principal tenets of stereology are that (i) specific sampling strategies should be employed to minimize biases, inaccuracies, and variabilities introduced by the counting and sampling technique, and that (ii) a count of objects in a measured volume (the reference volume Vref ) is more meaningful than density measurements. The methods that have been devised to satisfy these conditions for neuron counting are: systematic random sampling; reference volume Vref estimation using the Cavalieri principle; numerical volume density NV determination using the optical disector; and the determination of total neuron number from Vref and NV or from the fractionator method.

Systematic random sampling Stereological methods require unbiased sampling techniques. Unbiased sampling implies that all parts of the brain region of interest (the reference volume, Vref ) have an equal probability of being sampled, i.e. that the sampling is free from methodological bias. (Poor experimental design and experimental error can still lead to other forms of bias despite ‘unbiased’ sterological technique.) The sampling strategy most commonly utilized for stereological methods is systematic random sampling (Gundersen & Jensen, 1987). Practically, systematic random sampling begins by obtaining serial sections spanning the anatomical region of interest. Starting from a random start point (between 1 and n), every n th section is evaluated (Fig. 4.1(a)). Thus, if there

are 100 serial sections through the hippocampus, and every 10th section is to be evaluated, a random section from 1–10 is chosen (e.g. the 3rd section), and every 10th section thereafter is chosen (sections 3, 13, 23, . . . , 93). Within each of these chosen sections, systematic random sampling is again utilized to determine in which microscopic fields to count neurons. A rectangular grid is overlaid over the region of interest, and from a random start point (between 1 and m), every mth intersection point within the region is evaluated at high power for counting (Fig. 4.1(b)). At a minimum, a microscope with x-, y- stage encoders is usually required to count under high magnification at the coordinates selected from the grid. Thus, systematic random sampling is applied at all sampling levels; beginning with the selection of serial sections spanning an anatomical region, and then within these sections in the selection of microscopic fields of view.

Volume determination Classical methods of determining brain volume include calculation from brain weight and measuring displacement (Davis & Wright, 1977). Stereological methods apply the Cavalieri principle, which estimates volume from the cross-sectional areas of equally spaced parallel sections (Cavalieri, 1966; Gundersen et al., 1988a,b). Brain and brain region volume can be determined from point counting or outlining the region in serial sections. In the point counting method, a point counting grid is overlaid over each section (Fig. 4.2). The number of points overlying the region of interest is proportional to the area of that region within the section. Given the points over the region in each slice, the slice thickness, and the distribution of the points in the grid, the volume of the region of interest (the reference volume Vref ) can be calculated. As an alternative to grid counting over every slice, grid counting can be performed in a systematic random sampling of slices spaced equally apart spanning the reference volume. In a rectangular grid with points d distance apart horizontally and vertically, the area associated with each point (designated ap ) is d 2 . If Pi points intersect the region of interest in each slice i, then the area of the region in slice i (Ai ) can be estimated as: Ai = Pi ×

a p

= Pi × d 2

If in a systematic random sampling, m parallel sections equally separated by distance t span the region of interest, then the reference volume of the region is estimated as: Vref = t ×

m i=1

Ai = t × d 2 ×

m i=1

Pi

Morphometric methods

Fig. 4.1. Diagram of systematic random sampling and the optical disector to count neurons in the human mesial temporal lobe (drawings not to scale). (a) Medial view of human temporal lobe sectioned coronally with section thickness 50 m. Every 10th section beginning with a random start point between 1 and 10 (in this case the 3rd section) is selected (gray stripes). Since every 10th section is 1 . (b) Coronal view of hippocampus from one of the sections in (a). A grid of squares, each used, the section sampling fraction is ss f = 10 with dimensions 500 m x 500 m (area of each square = 250 000 m2 ), is overlaid over the section. Every 5th intersection point overlying the hippocampus, beginning with a random start point between 1 and 5 (in this case the 2nd intersection), is selected to be sampled with an optical disector (small unfilled squares). (c) At these intersection coordinates, an optical disector of dimensions 50 m × 50 m is applied to count neurons (grey triangles) through the depth of the section. Neurons are counted as they come into focus (after the first focal plane) if they are within the box (1) or they touch the upper margin or right margin (2) of the box (thin lines), but not if they touch the lower margin (4), left margin (3), or margin extensions of the box (thick lines). The area of the disector is 1 of the area of a single grid square. Since every fifth grid intersection is counted with a a(dis) = 50 m × 50 m = 2500 m2 , which is 100 1 1 = 500 . (d ) If only a fraction of disector, the fraction of the total area of each section sampled, the area sampling fraction, is as f = 15 × 100 20 m the entire 50 m section thickness is counted by the optical disector (e.g. 20 m), then the depth sampling fraction is ds f = 50 m = 25 . These values for ssf, asf, and dsf are used in the fractionator technique.

77

78

M. C. Irizarry

Fig. 4.2. Point counting method for area determination applied to a mouse hippocampal section. A low power Nissl stain of a 40 m coronal mouse brain section overlaid by points (‘+’) 500 m apart. The area associated with each point ( ap ) is 500 m × 500 m = 250 000 m2 . 13 points (P ‘*’) fall within the hippocampal region which is outlined with a grey line. By point counting, the hippocampal area in this section is approximately A = P × ap = 13 points × 250 000 m2 /point = 3 250 000 m2 . This approximates the automated area of the outlined region of 3 554 000 m2 . The areas from a systematic random sampling of sections spanning the hippocampus can be used to calculate the hippocampal volume by the Cavalieri principle.

Several image analysis systems (e.g. NIH image) can also be used to manually or in an automated fashion outline the region of interest Ai in each slice/section, as long as the dimensions of the captured image are calibrated.

Numerical volume density NV and total neuron Number N determination Stereological methods are based on the principle that a count of objects in a measured volume is more meaningful than density measurements. Measurements of density may be biased by atrophy, edema, or shrinkage during tissue processing. In addition, bias is introduced by the size, shape, and orientation of the object being counted. For instance, larger objects cover more sections than smaller objects, and are liable to be overcounted. These problems are minimized by estimating the absolute number of objects N within a defined volume, which is the product of a three-dimensional density of objects (the numerical

volume density NV , as determined using the optical disector) and the reference volume Vref of the region assessed. An alternative method – the optical fractionator – enables determination of the total number of neurons without explicit reference to volume density or volume.

Optical disector Counting strategies that minimize biases involve the optical disector and optical fractionator (Gundersen et al., 1988a). The optical disector requires a microscopic section thick enough to have more than one plane of focus at a high numerical aperture objective, such as an 100× oil objective. The optical disector is a counting box of a defined x and y dimension, with thickness being the section thickness z. Similar to a hemocytometer, objects touching the left margin or lower margin are not counted, while objects touching the right or upper margins are counted, in addition to counting objects completely within the box (Fig. 4.1(c)). Extending similar limitations to avoid overcounting in the z-axis,


objects counted divided by the total volume of all the disectors. Thus, given a disector with dimensions x and y, and tissue thickness z, the volume of each disector counting box is v(dis) = x × y × z. If Qi− objects are counted in each disector i, and a total of P disectors sample the entire reference volume over all sections, then the numerical volume density NV of neurons in the reference volume is approximated by the total number of neurons counted Qi− per total disector volume, v(dis) = P × v(dis) : P

NV =

i=1

Qi−

P × v(dis)

The total number of neurons N is the reference volume as determined by the Cavlieri method multiplied by the volume density: N = NV × Vref Rather than applying the counting box to the entire depth of the section, a microscope equipped with a motorized z-axis stage encoder or a microcator can be use to count within a specified depth distance h in the centre of the section, which represents a fraction of the total section thickness z (height sampling fraction, hs f = hz , Fig. 4.1(d )). This may reduce edge of section artifacts or ‘missed-tops’, but is susceptible to variations in tissue thickness with processing.

Fig. 4.3. Optical disector applied to human CA1. A 50 m × 50 m optical disector is applied to a 50 m thick section. Objects in the first plane of focus (single arrow), (a) are not counted, but subsequent neurons (double arrows) are counted as their nucleoli come into focus (b) and (c ). In this case 3 neurons are counted (Q− = 3).

objects in the superior plane of section are not counted, but objects that come into focus as the focal plane of the objective is moved through the section are counted (Fig. 4.3). Application of the optical disector is most commonly accomplished by viewing a high resolution video image of the microscope field and overlaying the counting frame on the image (Howard & Reed, 1998). The optical disector counting frame is applied to a fraction of the entire reference volume defined by systematic random sampling. Within each section to be counted, a grid is overlaid on the reference volume. The optical disector is applied to a systematic random sampling of the grid intersections within the reference volume. The numerical density of objects can be estimated by the total number of

Fractionator The fractionator is a method of determining total number of objects (in this case neurons) without specifically calculating reference volume or volume density (Gundersen et al., 1988a,b). The technique requires knowing what fraction of the entire reference volume has been counted using a disector. This can be derived from the fraction of sections counted (section sampling fraction, ssf, Fig. 4.1(a)), the fraction of area of the individual sections measured with disectors (area sampling fraction, asf, Fig. 4.1(b)), and the fraction of the depth of the tissue counted in disectors (height sampling fraction hsf Fig. 4.1(d)). If using systematic random sampling, every nth section of thickness z is evaluated, the section sampling fraction is ss f = n1 . A square grid is overlayed over each section, with the dimensions of each grid box being d × d; at each intersection point of the grid, an optical disector of length and width dimensions x × y is used to count neurons; then the area sampling fraction is the area of the disector (a (dis) = x × y) relative to the area of a grid square (d 2 ): as f =

a (dis) d2

The area sampling fraction can also incorporate a systematic sampling of the grid intersection points (for example,

79

80

M. C. Irizarry

see Fig. 4.1). If the height of the disector h is not the entire depth of the section z, then the height sampling fraction is hs f = hz . If Qi− neurons are counted in P disectors, representing a known fraction of the total region, then total neuron number is (West, 1993): P 1 1 1 × × Qi × N= hs f as f ss f i=1 Coefficient of error The disector counting chamber is used to sample the volume density in regions evenly spaced throughout the reference volume. An essential aspect of these methods is determining the sampling strategy and the number of sections and disector probes to be used. Stereological methods allow an estimate of the coefficient of error (CE) of the measurement technique – the amount of variability introduced by the sampling technique rather than biological variability. Approximations of the CE, calculated from the number of counts and disectors in each section, have been published (West, 1993; Gundersen et al., 1999). From pilot studies, the amount of sampling can be adjusted to reduce the CE below 0.10. For most studies, this requires counting about 100 neurons in an unbiased, systematic fashion (West, 1993).

Stereology and the neuropathology of Alzheimer’s disease Quantitative methods have been applied to characterize the vulnerability of specific brain regions to AD neuropathology. Amyloid plaques and neurofibrillary tangles occur in stereotypical anatomic distributions in the brain. Amyloid plaques occur in cortical and limbic brain regions and the hippocampal dentate gyrus, while sparing other hippocampal subregions. Neurofibrillary tangles occur in hippocampus and association cortex, as well as in subcortical nuclei such as the nucleus basalis of Meynert, but are infrequent in primary motor or sensory cortices (Arnold et al., 1991). Mapping of neuronal loss in AD, however, has been more complicated, requiring normative age-matched data, and considerations of specific brain region differences; atrophy; sampling; tissue preparation; and heterogeneity of cortical lamina. Careful experimental design is required in choosing the appropriate brain regions implicated in dementia (e.g. hippocampus, limbic cortex, association cortex, nucleus basalis of Meynert), and in assessing specific lamina or subfields within the anatomical region (e.g. CA1 in hippocampus or layer II in entorhinal cortex), and ensuring the appropriate age- and sex- matched controls. Stereological assessment using systematic random sampling and

the optical disector counting box can minimize the biases and variability that may be introduced by atrophy, sampling, and specimen handling (Hyman et al., 1998). Several stereology-based studies document a hierarchy of susceptibility to neuronal loss in AD, initially in the CA1 region of the hippocampus, followed by entorhinal cortex, especially layer 2, and then association cortex. West demonstrated that compared to non-demented controls, AD was associated with 68% loss of neurons in CA1 (West, 1994). Gomez-Isla, et al., extended these studies to additional regions vulnerable to AD neuropathology – the entorhinal cortex and a high order association area, the superior temporal sulcus region (STS). In a stereological analysis of the entorhinal cortex, there was progressive loss of neurons with Clinical Dementia Rating Scale (CDR) stage, ranging from 32% loss at CDR 0.5 to 69% loss at CDR 3. Layer 2 was particularly affected (Gomez-Isla et al., 1996). In the STS, there was minimal neuronal loss in early AD, but progressive loss with increasing disease duration, up to 75% loss in late AD (Gomez-Isla et al., 1997). In the STS analysis, neurons were counted in a reference volume consisting of a full thickness 700 m strip of cortex on a single 50 m section at a defined coronal level. Since this was a deviation from the standard stereological orthodoxy, the method was calibrated by a pilot study evaluating the entire STS. Volume densities were found to be homogeneous throughout the entire length of the STS, so the volume density calculated from the single section yielded a precise estimate of that of the entire STS. The method, however, underestimated atrophy in the antero-posterior direction, biasing toward the null hypothesis. In mixed AD/DLB, the STS neuronal loss was less than that of pure AD; there was no significant STS neuronal loss in pure DLB, or in normal aging (Gomez-Isla et al., 1999).

Conclusions The goal of morphometric methods applied to dementia is to efficiently quantitate neuropathological features of relevance with precision and a minimum of methodological and experimental bias. The stereological methods reviewed in this chapter for counting neurons (Cavalieri volume determination, systematic random sampling, optical disector, and optical fractionator) when incorporated with appropriate experimental design (e.g. AD versus age and sex-matched controls; relevant brain regions such as hippocampus, entorhinal cortex, and association cortex) have clarified the regional and temporal neuroanatomic vulnerability to neuronal loss in AD, and the relative preservation of neurons with normal ageing.


REFERENCES Abercrombie, M. (1946). Estimation of nuclear population from microtome sections. Anat. Rec., 94, 239–47. Arnold, S. E., Hyman, B. T., Flory, J., Damasio, A. R. & Van Hoesen, G. W. (1991). The topographical and neuroanatomical distribution of neurofibrillary tangles and neuritic plaques in the cerebral cortex of patients with Alzheimer’s disease. Cereb. Cortex, 1, 103–16. Braak, H. & Braak, E. (1991). Neuropathological staging of Alzheimer related changes. Acta Neuropathol, 82, 239–59. Cavalieri, B. (1966). Geometria degli indivisibili. Torino, Unione Tipografico, Editrice. Davis, P. J. M. & Wright, E. A. (1977). A new method for measuring cranial cavity volume and its application to the assessment of cerebral atrophy at autopsy. Neuropath. Appl. Neurobiol., 3, 341–58. Ebbesson, S. O. E. & Tang, D. (1965). A method for estimating the number of cells in histological sections. J. Roy. Microscop., 84, 449–64. Everall, I. P., Heaton, R. K., Marcotte, T. D. et al. (1999). Cortical synaptic density is reduced in mild to moderate human immunodeficiency virus neurocognitive disorder. HNRC Group. HIV Neurobehavioral Research Center. Brain Pathol., 9, 209– 17. Geinisman, Y., Gundersen, H. J., Van der Zee, E. & West, M. J. (1996). Unbiased stereological estimation of the total number of synapses in a brain region. J. Neurocytol., 25, 805–19. Geuna, S. (2000). Appreciating the difference between designbased and model-based sampling strategies in quantitative morphology of the nervous system. J. Comp. Neurol., 427, 333–9. Gomez-Isla, T., Price, J. L., McKeel, D. W. Jr et al. (1996). Profound loss of layer II entorhinal cortex neurons occurs in very mild Alzheimer’s disease. J. Neurosci., 16, 4491–500. Gomez-Isla, T., Hollister, R., West, H. et al. (1997). Neuronal loss correlates with but exceeds Neurofibrillary tangles in Alzheimer’s disease. Ann. Neurol., 41, 17–24. Gomez-Isla, T., Growdon, W. B., McNamara, M. et al. (1999). Clinical–neuropathological correlates in temporal cortex in dementia with Lewy bodies. Neurology, 53, 2003–9. Gundersen, H. J. & Jensen, E. B. (1987). The efficiency of systematic sampling in stereology and its prediction. J. Microscopy, 147, 229–63.

Gundersen, H. J., Bagger, P., Bendtsen, T. F. et al. (1988a). The new stereological tools: disector, fractionator, nucleator and point sampled intercepts and their use in pathological research and diagnosis. APMIS, 96, 857–81. Gundersen, H. J., Bendtsen, T. F., Korbo, L. et al. (1988b). Some new, simple and efficient stereological methods and their use in pathological research and diagnosis. APMIS, 96, 379–94. Gundersen, H. J., Jensen, E. B., Kieu, K. & Nielsen, J. (1999). The efficiency of systematic sampling in stereology – reconsidered. J. Microscopy, 193, 199–211. Howard, C. V. & Reed, M. G. (1998). Unbiased Stereology. New York, Springer-Verlag. Hyman, B. T., Gomez-Isla, T. & Irizarry, M. C. (1998). Stereology: a practical primer for neuropathology. J. Neuropath. Exp. Neurol., 57, 305–10. McKeith, I., Galasko, D., Kosaka, K. et al. (1996). Consensus guidelines for the clinical and pathological diagnosis of dementia with Lewy bodies (DLB): report of the consortium on DLB international workshop. Neurology, 47, 1113–24. Mayhew, T. M. & Gundersen, H. J. (1996). If you assume, you can make an ass out of you and me: a decade of the disector for stereological counting of particles in 3D space. J. Anat., 188, 1–15. Mirra, S. S., Heyman, A., McKeel, D. et al. (1991). The consortium to establish a registry for Alzheimer’s disease (CERAD). Part II. Standardization of the neuropathologic assessment of Alzheimer’s disease. Neurology, 41, 479–86. Peterson, D. A. (1999). Quantitative histology using confocal microscopy: implementation of unbiased stereology procedures. Methods (Duluth), 18, 493–507. Roberts, N., Puddephat, M. J. & McNulty, V. (2000). The benefit of stereology for quantitative radiology. Br. J. Radiol, 73, 679–97. Sterio, D. C. (1984). The unbiased estimation of number and sizes of arbitrary particles using the disector. J. Microscopy, 134, 127–36. Weibel, E. R. & Gomez, D. M. (1962). A principle for counting tissue structures on random samples. J. Appl. Physiol., 17, 343–8. West, M. (1993). New stereological methods for counting neurons. Neurobiol. Aging, 14, 275–85. West, M. J. (1994). Differences in the pattern of hippocampal neuron loss in normal ageing and Alzheimer’s disease. Lancet, 344, 769–72.

81

5 Safety precautions in laboratories involved with dementia diagnosis and research Jeanne E. Bell Pathology Department, Western General Hospital, Edinburgh, UK

Introduction Laboratories concerned with the diagnosis of dementia, and research into dementing conditions, will come into contact sooner or later with brain tissue from patients in whom the dementia is due to the presence of an infective agent. Concern about the possible presence of such microorganisms is one important reason for ensuring that good laboratory practice and carefully devised safety precautions are implemented wherever dementia diagnosis and research is undertaken. This chapter is concerned with discussion of the likely hazards and risks involved in this kind of work and the strategies for safe practice which should be in place. The concept of hazard focuses on the level of danger associated with the particular agent under consideration, while risk depends not only on the nature of the hazard but also on the likelihood of exposure (ACDP, 1995a). Consideration of the hazards and their likely risk in relation to dementing illnesses forms the basis of the containment protocols which have been developed to limit the danger posed by these hazards. The general principles of containment need to be considered in relation both to the process which is under way, for example the examination of a biopsy or the conduct of an autopsy, and the setting in which that process takes place, for example the laboratory or the post-mortem room. Laboratories which specialize in examining dementia cases may well be part of a formally constituted brain bank that stores tissues for research purposes and also dispatches tissue samples to the wider research community (Cruz-Sánchez & Tolosa, 1995). Particular safety precautions may well apply to individual brain banks that focus on infective dementias (Bell & Ironside, 1997). With respect

to containment of infective agents, some have advocated that universal precautions should be adopted for all autopsies as part of a similar approach in the wider health care setting (CDC, 1987; Kibbler, 1997) – this debate is briefly reviewed in this chapter. Finally, procedures are described for coping with accidents involving exposure to infected or potentially infected brain tissue within the post mortem room or in laboratories. The guiding principles for safe working in these conditions include comprehensive risk assessment, the development of clear standard operating procedures, and proper training and supervision of the staff involved.

The hazards Although the emphasis in this chapter is on infective agents as the major hazard group, physical agents and potentially dangerous items of equipment should not be ignored. Electrical equipment and chemical reagents are in use in every mortuary and laboratory. Some laboratories use radioisotopes and ionizing radiation for investigation of human tissue and the staff involved require training in radiation protection. The dangers of sharp instruments, such as scalpels and microtome blades, are minimized by adherence to standard operating procedures. Safety precautions associated with these hazards are embodied in the Control of Substances Hazardous to Health Regulations (COSHH, 1999; HSAC, 2002a,b). The major source of concern with respect to postmortem tissues relates to the certain or likely presence of microbiological agents which are hazardous to human health. These agents are categorized by the Advisory


82

Safety precautions in laboratories

Committee on Dangerous Pathogens (1995a) in four groups according to their nature and the level of containment required for their safe handling. Categorization depends in part on whether the diseases caused by these pathogens are treatable or preventable, on the mode of transmission including inhalation and/or inoculation and also the severity of the disease that they may produce. The hazardous infective agents most likely to be present in central nervous system (CNS) tissue include prions and the human immunodeficiency virus (HIV). Other bloodborne viruses, including hepatitis B and hepatitis C, may be present within the vascular compartment of CNS tissue. All of these agents are Category 3 pathogens (ACDP, 1995a, 2000), which pose a relatively grave hazard to humans because they cause serious or untreatable diseases. Other Category 3 pathogens, including most mycobacteria, and even Category 2 bacteria such as meningococci and some streptococci are likely to be dangerous if significant exposure occurs (Healing et al., 1995; Young & Healing, 1995). The hazard status of other newly discovered agents such as hepatitis G remains uncertain and classification in Category 3 is appropriate (Jarvis et al., 1996). Additional pathogens may be encountered in AIDS cases, which are not normally hazardous but which may pose risks to health care workers who have impaired immune status for whatever reason. Some viruses, such as papovavirus, which causes progressive multifocal leukoencephalopathy, and most fungi are included within this opportunistic class of organisms. Syphilis is rarely encountered nowadays in autopsy cases in developed countries except perhaps in the context of HIV infection. It may be surprising to note that the pathogens causing rabies, tickborne encephalitis and yellow fever are also classified only as Category 3 pathogens (Table 5.1), but examination of such cases is restricted to a small number of designated UK laboratories. They should not be encountered in most laboratories handling CNS tissue in the UK. Category 4 organisms, including Lassa fever and Ebola viruses, are not generally found in the UK, although the danger of importing these infections by jet travel is a real possibility. Routine good laboratory and mortuary practice will prevent transmission of some organisms, such as methicillinresistant Staphylococcus aureus (MRSA), which are of more concern in the clinical setting (HSAC, 2002b). The exact assignment of some bacteria and viruses to Group 2 or to Group 3 may be somewhat unexpected in certain instances and this is illustrated in Table 5.1, although this table does not represent an exhaustive list of all bacteria and viral agents listed in the ACDP categorization. The full categorization of pathogens can be viewed at the ACDP website (www.doh.gov.uk/acdp).

Table 5.1. ACDP Categorization of bacterial and viral agents that may be encountered during post-mortem examination 1

Bacteria

Group 2 Actinomyces B. Pertussis Clostridium Strep. Pneumoniae Trep. pallidum N Meningitidis 2

Group 3 B. Anthracis Mycobacterium avium + TB – – – –

Group 4 – –

Group 3 Hepatitis B and C HIV Rabies Prions

Group 4 Lassa fever Ebola Variola Congo haemorrhagic fever –

– – – –

Viruses

Group 2 Most Herpes viruses Influenza A, B and C Polio viruses Rubella – –

Tick-borne encephalitides Yellow Fever

–

The risks Assessment of the risks encountered in working with human CNS tissue depends not only on the nature of any infective agent which may be present but also the likelihood of staff being exposed to this hazard (Hall & Harrington, 1991; Grist & Emslie, 1994; HSAC, 2003a,b). The prevalence of CNS infection does vary in proportion to the demographic characteristics of the particular autopsy cohort under investigation. There is a greater risk of exposure to HIV and associated pathogens in mortuaries and laboratories that routinely handle brain tissue from drug users than in those concerned principally with ¨ examination of geriatric populations (Puschel et al., 1987; Karhunen et al., 1989; Crofts et al., 1993; Tennant & Moll, 1996; Eddleston, 1997). It is helpful to consider the likely incidence of diseases in particular environments and published data is helpful in this regard (Lucas, 1995; Murray & Lopez, 1997). The clinical history and setting are important sources of information regarding the likely risks in individual cases. For instance, although tuberculosis is uncommon in most affluent societies, the prevalence is increasing in certain subsets of the population including immigrant communities and this may include drug-resistant forms of Mycobacterium tuberculosis (Kent et al., 1994;

83

84

J. E. Bell

Fryatt, 1995). Unfortunately, the clinical history is often scanty in the context of medicolegal autopsies. In these circumstances the task of risk assessment is greatly facilitated by post-mortem testing for HIV and blood-borne viruses if these are suspected from the history or from the appearance of the patient (Karhunen et al., 1989; ACDP, 1995b). Experience with evolving patterns of disease assists clinicians and pathologists to identify high-risk patients. Thus in the UK a young patient with a dementing illness may well be suffering from variant CJD whereas this is much less likely in other countries (Will et al., 1996). Similarly, a history of intravenous drug use raises suspicions regarding ¨ the presence of HIV as well as the hepatitis viruses (Puschel et al., 1987; Tennant & Moll, 1996). Haemophiliac subjects, male homosexuals, and individuals who have been in custody are also at higher risk of HIV infection than the general population (Karhunen et al., 1989; Kennedy et al., 1991; Poznansky et al., 1994). It should be remembered that HIV infection may be asymptomatic at the time of death and there may be no AIDS-related pathology at autopsy (Baker et al., 1987; Gray et al., 1996). Post-mortem testing for HIV may prove negative in rare instances in the ¨ pre-seroconversion window of infectivity (Gurtler, 1996). Nevertheless the examination of brain tissue poses a threat from the presence of viruses in such apparently negative or asymptomatic subjects (Davis et al., 1992; Bell et al., 1993). Because the risks may be difficult to quantify in some cases, the adoption of so-called universal precautions has been advocated (CDC, 1987; Kibbler, 1997). However, there are arguments for and against such a policy (see later). Protection of the staff handling human CNS tissue in the post-mortem room and in the laboratory is based on documented risk assessments, which should cover all procedures and which should be updated regularly to ensure that no new risks have appeared and that existing precautions are adequate for protection (ACDP, 1995b; HSAC, 2003 a,b). Guidelines for best autopsy practice, including examination of the central and peripheral nervous system, have been published by the College of American Pathologists and these include commentary on universal precautions (Hutchins et al., 1994; Powers et al., 1995). Some general advice may also be found at the website for Centers for Disease Control and Prevention (www.cdc.gov/). All staff involved in these procedures should read the protocols and departments should have signed evidence that they have done so. Ongoing consultation with the hospital Infection Control Team is useful for mortuaries and for laboratories coping with a significant infective dementia caseload. Competent risk assessment of the likelihood of infection requires knowledge of the tissue distribution of the

infective agent. This is more important in the context of autopsies where the whole body is examined (Donaldson et al., 1994). Increasing knowledge of new diseases helps to establish these important facts. Thus infectivity associated with prion diseases is largely confined to CNS tissues in cases of sporadic CJD. However, in cases of variant CJD, both CNS and non-CNS tissues (such as the tonsil, spleen and other gut associated lymphoid tissue) are known to contain infective prions (Bruce et al., 2001). Autopsy-based studies have value in establishing such tissue distribution. The ability of pathogens to survive in the body after death (Nyberg et al., 1990; Ball et al., 1991; Brown & Gajdusek, 1991) even at 4 0 C, is well illustrated by the occurrence of infection transmitted during autopsy (Grist & Emslie, 1994). The persistence of prions in fresh post-mortem tissue led to tragic cases of iatrogenic transmission of disease following dural transplants and the use of pituitary extracts for the treatment of growth hormone deficiency (Brown et al., 1992). Fortunately, formalin fixation decontaminates most post-mortem tissue. The major exceptions are prions which are also able to survive most routine decontamination procedures (Table 5.2). Although infectivity Table 5.2. Cases regarded as prion disease for the purposes of autopsya 1

2

3

4 5

a

Cases of probable Creutzfeldt–Jakob disease (CJD) Defined as progressive dementia plus electroencephalogram (EEG) typical of CJD, with two or more of the following: myoclonus, visual or cerebellar disturbance, akinetic mutism and pyramidal or extrapyramidal dysfunction. Possible CJD Similar to probable CJD but lacking typical EEG changes Variant CJD Usually younger patients (but now including occasional much older subjects) with a history of up to two years, commencing with psychiatric presentation (anxiety and progressive behavioural changes) increasing cerebellar disturbance and cognitive impairment and myoclonus of late onset but lacking EEG changes of CJD, and/or with biopsy proven (cerebral or tonsillar) prion protein deposition and with pulvinar high signal on magnetic resonance imaging. Individuals treated with pituitary derived hormones or with human dura mater grafts Individuals with a family history of CJD or similar diseases including fatal familial insomnia and Gerstman–Straussler–Scheinker syndrome.

Based on information in Metters, 1992; Budka et al., 1995 a,b.


of CNS tissues from patients with prion disease is reduced by formalin fixation, it is not entirely eliminated – transmission of infection from formalin fixed tissue has been reported (Brown et al., 1986). Formic acid is effective as a post-formalin decontaminant for prion-infected tissues (Brown et al., 1990). Prevention of infection in the laboratory and mortuary is critically dependent on the use of personal protective clothing (HSAC, 2003b). The aim is to prevent transmission by hand to mouth, inhalation, splash contamination or inoculation since there is a known risk of these occurring (Henderson et al., 1990; Bull et al., 1991; Collins & Heptonstall, 1994; CDC, 1995). Risk assessment procedures should also take account of the possible vulnerability of workers for pathogens encountered in the course of daily work. Conditions such as diabetes which reduce immune competence are associated with increased risk. The dangers of radiation during pregnancy are well known. Workers with skin conditions which result in epidermal breakdown on the hands are at increased risk of contracting blood borne viruses from infected tissues. Lastly, patients with allergies for chemicals such as latex, which are encountered commonly in laboratories, require special protection (HSAC, 2003a,b). It is worth remembering that the rate of serious infection amongst patients dying of dementia in the UK is likely to be low, which places the risk to pathologists and laboratory workers in reasonable perspective (Young & Healing, 1995). Case reports of CJD occurring in pathology staff remain very rare and have not been linked unequivocally to exposure (Miller, 1988; Sitwell et al., 1988). Most of the neurodegenerative conditions, which result in dementia, such as Alzheimer’s disease, dementia with Lewy bodies and multi-infarct dementia, are not known to be associated with infective agents.

The high risk autopsy Autopsies conducted in patients known or suspected to be infected with a Category 3 pathogen are regarded as high risk. The conduct of a high-risk autopsy has been reviewed recently (Bell & Ironside, 1993; Budka et al., 1995a; Ironside & Bell, 1996). These guidelines are based on an earlier publication describing precautions to be taken in autopsies on HIV-infected persons (MacArthur & Schneiderman, 1987). Regular updates of relevant guidance are available at specialized websites which are listed in the references section. Protocols are based on the need to protect those involved from transmission of infection. Since some at least of the dangerous pathogens encountered at autopsy cause untreatable diseases (such as prion diseases)

or are treatable only with costly and difficult therapeutic regimes (such as HIV infection), prevention remains particularly important. The basic principles of a high risk autopsy are to limit the number of people exposed to likely infection, to reduce the risks to those involved in the procedure as far as is humanly possible and to minimize contamination of the environment. The use of a dedicated highrisk autopsy facility is recommended but is not essential (HSAC, 2003b). Autopsy patients infected with Category 3 pathogens can be examined in a general autopsy suite without dedicated facilities, provided that the guidelines for high-risk procedures are followed. Bodies infected with Category 4 pathogens are not subjected to autopsy except in certain specialised facilities (HSAC, 2003b). Protective clothing and planned Standard Operating Procedures are mandatory. The major aim of special clothing is to protect hands, eyes and mucous membranes. The emphasis placed on each of these depends on whether the infective agent present in the cadaver is transmitted by inhalation, inoculation or by other means. It appears that prions associated with sporadic Creutzfeldt–Jakob disease are transmitted only by ingestion or inoculation and the possibility that these are transmitted by blood or by the air borne route is considered unlikely (ACDP, 1998). There is anxiety that variant CJD prions may differ and that this infection may be transmitted by blood products (Turner & Ironside, 1998). Prior to the start of a high-risk autopsy all staff should be immunized against hepatitis B, tetanus and tuberculosis (HSAC, 2003a,b). These pathogens constitute an ongoing risk for staff involved in autopsies (Grist & Emslie, 1994, Collins & Heptonstall, 1994; Collins & Grange, 1999). There is still no effective immunization against hepatitis C. There is currently a move in the UK (HSAC, 2003b) to introduce infection status documentation to accompany each person dying in hospital, in order to facilitate safe handover to undertakers. The effect of this is to focus attention on the potential risk posed by an individual cadaver to all staff professionally in contact with the deceased. Those who will indirectly benefit from infection control notification include laboratory workers who handle fresh and fixed tissue obtained from autopsies. Risk assessments undertaken in the mortuary, based on the clinical information and the infection control notification, allow the pathologist to plan an appropriate post-mortem strategy for a particular deceased patient. Table 5.2 shows a list of cases which should be regarded as harbouring prion diseases. The conduct of a high-risk autopsy is likely to be safer when undertaken by experienced staff. Procedures should be unhurried since accidents are more likely to occur if staff feel under pressure to complete work quickly

85

86

J. E. Bell

(Ironside & Bell, 1996; HSAC, 2003b). Pathologists and mortuary technical staff in training may become familiar with high-risk protocols through practice under supervision on bodies not thought to be infected. This should ensure competence and confidence in the procedure. The high-risk autopsy is generally undertaken by an experienced pathologist accompanied by a trained mortuary technician. Ideally, a third member of staff is present who does not have direct contact with infected material and instruments but who is available for handling non-contaminated containers, labels and instruments. A selection of disposable protective clothing is available together with equipment for hands and face which is suitable for cleaning and re-use (Fig. 5.1). Hand protection is particularly important for the

staff undertaking the autopsy, particularly those with the task of making and closing incisions in the body. This is best achieved by toughened, cut-proof gloves. Stainless steel mesh gloves worn between two pairs of rubber gloves are a satisfactory protection against cutting injuries, although they do not protect against needle-stick injuries. Eyes and mucosal surfaces are protected most effectively against splash injury and inhalation by the use of a ventilated visor with fine filtration of inspired air, according to HSAC guidelines. This form of visor will also protect against aspiration of bone dust and air-borne particles generated during autopsy procedures (Kernbach-Wighton et al., 1996). Detailed guidance for the conduct of high-risk autopsies includes practical considerations, which prevent dispersal of contaminated material around the autopsy suite (Bell & Ironside, 1993; Ironside & Bell, 1996). Electrical saws with vacuum aspiration, enclosing the head within a plastic bag, dissection of organs in situ and storage tanks below the autopsy table are all devices which help to contain infective material. Opening the skull and removing the spinal cord represent points of high risk during the conduct of the autopsy. Reconstitution of the body carries the highest risk of needle-stick injury. Each of these procedures deserves special care. At the end of the autopsy the body is enclosed in a body bag and labelled ‘high risk’ for handover to the funeral directors. Relatives may view a body after a highrisk procedure but should be discouraged from touching or kissing because the risk of transmission of infection is greater after the body has been opened. At the end of the high-risk autopsy, the autopsy suite and instruments are decontaminated according to recognised protocols for the particular infective organisms. Table 5.3 lists the measures thought to be effective in prion decontamination. Chlorine based solution of high concentration is most effective for prions and is effective for all other organisms. Protocols for safe disposal of clinical waste should be adhered to (HSAC, 1999). No one should work in isolation in the mortuary while it remains contaminated (HSAC, 2003b).


Fig. 5.1. Mortuary Technical Officer clad in personal protective clothing at the commencement of a high-risk autopsy. Note that the ventilated visor accommodates spectacles without misting of the transparent section. The chain-mail gloves are used between two pairs of rubber gloves and full flexibility of the fingers is still maintained. Protection of the wrists is provided in this instance by waterproof cuffs.

Laboratories involved in diagnosis and research of dementia cases may be presented with diagnostic biopsy material as well as autopsy cases. Detailed guidance is available regarding the levels of containment for handling potentially infected human tissue (HSAC, 2003a). Laboratories are also characterized by their facilities including possession of safety cabinets. Category 3 laboratories are designated as such by control of circulated air, which is not considered essential for working with tissue infected with


Table 5.3. Measures considered effective in prion disease decontaminationa Current guidelines recommend the following procedures. It is worth noting that some doubt exists as to the efficacy of some of these measures, particularly autoclaving, and these guidelines are currently being revised. 1. Steam autoclaving (a) Porous load autoclave, 134 ◦ C for 18 minutes or six cycles for 3 minutes each. (b) Gravity displacement steam autoclaving at 134 ◦ C for 1 hour 2. Exposure to 2M sodium hydroxide for 1 hour – work surfaces should not be allowed to dry out. One hour exposure. 3. Sodium hypochlorite solution containing 20 000 parts per million available chlorine (achieved by freshly made up solution of Presept tablets (Johnson and Johnson UK) e.g. seven tablets in 500 ml distilled water. 1-hour exposure. 4. Exposure to 96% formic acid for 1 hour, particularly suitable for decontamination of formalin fixed tissues. 5. Incineration Agents which are thought to be ineffective in sterilizing CJD infected material include boiling water, formaldehyde, alcohol, phenol, gluteraldehyde, hydrogen peroxide, ionizing and UV radiation, nucleases and proteases, and dry heat up to 360 ◦ C. a

Table based on information in ACDP, 1998.

prions and blood-borne viruses. Transport to and from the laboratory of organs, particularly brain and spinal cord, and tissues infected with high risk (Category 3) pathogens should be in suitable unbreakable containers labelled ‘risk of infection’ and with specimen details (HSAC, 2003a,b). Samples should be double bagged and placed in transport containers. It is the responsibility of the pathologist to see that transport and storage of high-risk specimens is performed safely. In the case of fixed organs, infected tissues should be placed in generous quantities of formalin in order to ensure decontamination. Ideally, tissue biopsies submitted for the diagnosis of dementia should be received unfixed in the laboratory, providing the opportunity of retaining fresh material for genetic and other analyses in addition to histological preparation. Fresh tissue for PrP gene analysis and prion protein glycosylation pattern is invaluable in the diagnosis and research of prion diseases (Will et al., 1996). Culture of biopsy material may be desirable to detect viral presence such as papovavirus or Herpes simplex virus. Electron microscopy is an appropriate investigation for the

neurodegenerative diseases of childhood. Laboratories should be equipped to handle fresh tissue and samples infected with Category 3 pathogens, which may be examined within a Class 2 safety cabinet using disposable equipment. Cryostat sections should not generally be prepared for diagnosis because of subsequent decontamination problems but are frequently part of the research methodology in dedicated laboratories. The same principles apply to examination of fresh autopsy tissue, including the whole brain. Most laboratories examining dementia cases are required to slice and prepare for histology brains which have already been well fixed in formalin. Apart from prion disease cases these pose no extra risk to the pathologist or technical officer compared with an uninfected case, provided that the brain has been well fixed. The major hazard in these cases is of formalin inhalation and COSHH guidelines (1999) provide detailed guidance on minimizing exposure to formalin. Examination of fixed brains known or suspected to be infected with prions may proceed in a routine laboratory provided that care is taken for hand protection when handling sharp instruments and that surfaces are contaminated as little as possible (Budka et al., 1995a; Bell & Ironside, 1997). All neuropathology laboratories will encounter the occasional dementia case in which, unexpectedly, a diagnosis of prion disease is made only at histological examination. It is unlikely that this represents a major threat to the staff involved and guidelines have been published to cover this eventuality (Bell & Ironside, 1997). There is no need to undertake wholesale decontamination of a laboratory or autopsy suite. It is likely that several weeks, and multiple cleansing episodes, will have elapsed since the index autopsy was undertaken. Common-sense suggests that in such circumstances autopsy instruments which have been used in the case should be cleansed with sodium hydroxide, autoclaved to CJD standards, or disposed of where reasonable (Table 5.3). Laboratories in which examination of fresh brain tissue from dementia cases is undertaken regularly are likely to be operating as a brain banking facility. Careful planning should ensure safe handling of fresh unfixed brains bearing in mind the principles of prevention of infection by inoculation and by inhalation. The major hazards in these circumstances are the blood-borne viruses and prions. It may not be practical to cut a fresh brain within a safety cabinet given the detailed observation, photography and possibly imaging which are required in research protocols. Protective clothing, eye and hand protection are required. Staff should not work alone in a high-risk facility while handling infected material. Detailed guidelines for brain banking of high-risk autopsy tissue and other relevant

87

88

J. E. Bell

procedures, such as dispersal of research samples to users in other laboratories have been published (Budka et al., 1995a; Bell & Ironside, 1997). In particular, it is important to note that tissue stored at –70 ◦ C should be regarded as still infective even after a period of storage. Subsequent use of this tissue, including DNA extraction and cryostat sectioning, should be undertaken with suitable care and containment. Once DNA has been extracted it no longer represents an infective hazard and can be further examined in a routine Category 2 laboratory. Freezers should be labelled high risk, stored securely and accessed only by trained designated staff. Clinical waste should be disposed of safely (HSAC, 1999).

Protocol for accidents The UK COSHH (1999), ACDP (1995a) and HSAC (2003a,b) guidelines which set standards for good working practice in laboratories and mortuaries require that there should be a system for recording and reporting accidents. Significant events need to be notified to the Head of Department and work related infections require central reporting (RIDDOR, 1996). Injuries sustained during a high-risk autopsy require instant attention. Cuts and needlestick injuries should be carefully washed with encouragement of free blood flow from the injury. Little more can be done to prevent transmission of prion diseases (ACDP, 1998) although some injured persons have adopted heroic measures such as cleansing the wound with sodium hydroxide. If contamination of a wound with HIV is suspected, the injured person requires a blood test (serum stored for later HIV testing if necessary) and consideration by a qualified clinician as to whether anti-HIV therapy should be instituted (RCPath, 1995; Dept of Health, 1998, 2000). This attention is required immediately and certainly within one hour of injury. For this reason the line of referral for essential treatment should be clearly specified in working protocols prior to undertaking any high-risk procedures.

do represent a potential hazard to pathology staff during the autopsy and within the laboratory. Universal screening of all cadavers before autopsy, either for HIV or for prion diseases, will not solve the problem because they cannot all be detected rapidly by present methodology. For this reason, the adoption of ‘universal precautions’ has been considered seriously as a means of preventing transmission of blood borne viruses (CDC, 1987; Kibbler, 1997). Taking such precautions to their logical conclusion, all autopsies would be treated as high-risk procedures. It is important to note that this concept was developed before the escalation of concern about CJD (CDC, 1987). It is clear that pathology departments have not adopted universal precautions for the conduct of autopsies and laboratory investigation of post-mortem tissue. This working practice is based on the knowledge that the final pathology examination does not reveal the presence of infective organisms in a sufficiently high proportion of cases to justify the increased cost and time which would be entailed in using universal precautions. The further complications of treating all cases as potentially infected with prions, and the containment and decontamination implications which would stem from that, are additional disincentives to implementing universal precautions. Neuropathology departments should recognize that examination of dementia cases does entail a degree of risk for the staff involved. Standard operating procedures aim to reduce the risk sufficiently to prevent harm to health care workers in the post-mortem room and in the laboratory. At the present time, best practice is enshrined in careful risk assessment and adherence to safe working practices rather than the adoption of universal precautions.

Acknowledgements Thanks are due to Drs C. Bergeron, S. Morgello and M. Ricketts for advice on international guidelines for best autopsy practice. The help of Ms Angela Penman with preparation of the manuscript is gratefully acknowledged.

Universal precautions

REFERENCES

This chapter has stressed the risk assessment, which must be undertaken before tissue is examined or an autopsy undertaken in the context of a dementing illness. Unfortunately, even the most careful risk assessments do not allow identification of every infected case and it is likely that some are not identified until much later in the chain of examination. Despite the fact that CJD has a much higher profile nowadays, cases of prion disease are still missed clinically particularly in the ageing population. Such cases

Advisory Committee on Dangerous Pathogens (ACDP). (1995a). Categorisation of Biological Agents According to Hazard and Categories of Containment, 4th edn. London: HMSO (under revision). See also: www.doh.gov.uk/acdp. 1995b. Protection Against Blood-borne Infections in the Workplace: HIV and Hepatitis. London: HMSO. ACDP. Spongiform Encephalopathy Advisory Committee. (1998). Transmissible Spongiform Encephalopathy Agents: Safe Working and the Prevention of Infection. London: HMSO (under revision).


ACDP. (2000). Second Supplement to Categorisation of Virological Agents According to Hazard and Categories of Containment. London: HMSO. Baker, J. L., Kelen, G. D., Siverston, K. T. & Quinn, T. C. (1987). Unsuspected human immunodeficiency virus in critically ill emergency patients. Journal of the American Medical Association, 257, 2809. Ball, J., Desselberger, U. & Whitwell, H. (1991). Long-lasting viability of HIV after patient’s death. Lancet, 338, 63. Bell, J. E. & Ironside, J. W. (1993). How to tackle a possible Creutzfeldt–Jakob disease necropsy. Journal of Clinical Pathology, 46, 193–7. (1997). ‘High-risk’ brain banking: principles and practice. Neuropathology and Applied Neurobiology, 23, 281–8. Bell, J. E., Busuttil, A., Ironside, J. W. et al. (1993). Human immunodeficiency virus and the brain: investigation of virus load and neuropathologic changes in pre-AIDS subjects. Journal of Infectious Diseases, 168, 818–24. Brown, P. & Gajdusek, D. C. (1991). Survival of scrapie after three years’ interment. Lancet, 337, 269–70. Brown, P., Gibbs, C. J., Gajdusek, D. C. et al. (1986). Transmission of Creutzfeldt–Jakob disease from formalin-fixed, paraffin embedded human brain tissue. New England Journal of Medicine, 315, 1614–15. Brown, P., Wolff, A. & Gajdusek, D. C. (1990). A simple and effective method for inactivating virus infectivity in formalin fixed tissue samples from patients with Creutzfeldt–Jakob disease. Neurology, 40, 887–90. Brown, P., Preece, M. A. & Will, R. G. (1992). Friendly fire in medicine: hormones, homografts and Creutzfeldt–Jakob disease. Lancet, 340, 24–7. Bruce, M. E., McConnell, L., Will, R. G. & Ironside, J. W. (2001). Detection of variant Creutzfeldt–Jakob disease infectivity in extraneural tissues. Lancet, 358, 208–9. Budkha, H., Aguzzi, A., Brown, P. et al. (1995a). Tissue handling in suspected Creutzfeldt–Jakob disease (CJD) and other human spongiform encephalopathies (Prion disease). Brain Pathology, 5, 319–22. (1995b). Neuropathological diagnostic criteria for Creutzfeldt– Jakob disease (CJD) and other human spongiform encephalopathies (Prion diseases). Brain Pathology, 5, 459–66. Bull, A. D., Channer, J., Cross, S. S. et al. (1991). Should eye protection be worn when performing necropsies. Journal of Clinical Pathology, 44, 782. Centers for Disease Control. (1987). Recommendations for prevention of HIV transmission in health care settings. Morbidity and Mortality Weekly Report, 36, 1S–18S. (1995). Case control study of HIV seroconversion in health care workers after percutaneous exposure to HIV infected blood. Morbidity and Mortality Weekly Report, 44, 50. Cohn, J. A. (1997). HIV infection-I. British Medical Journal, 314, 487–91. Collins, G. H. & Grange, J. M. (1999). Tuberculosis acquired in laboratories and necropsy rooms. Communicable Diseases and Public Health, 2, 161–7.

Collins, M. & Heptonstall, J. (1994). Occupational acquisition of acute hepatitis B infection by health care workers: England and Wales. (1985–93). Communicable Disease Report, 4, R153–5. Control of Substances Hazardous to Health Regulations. Approved codes of practice. (1999). HSE Books. See also: www.coshhessentials.org.uk. Crofts, N., Hopper, J., Bowden, D. et al. (1993). Hepatitis-C virus infection among a cohort of Victorian injecting drug users. Medical Journal of Australia, I, 159, 237–41. Cruz-Sánchez, F. F. & Tolosa, E. (1995). How to run a brain bank. Journal of Neurological Transmission, S39, 1–242. Davis, L. E., Hjelle, B. L., Miller, V. E. et al. (1992). Early viral brain invasion in iatrogenic human immunodeficiency virus infection. Neurology, 42, 1736–9. Department of Health. (1998). Guidance for clinical health care workers: protection against infection with blood-borne viruses. UK Health Department. (2000). HIV post-exposure prophylaxis: guidance from the UK Chief Medical Officers. Expert Advisory Group on AIDS. UK Health Department. Donaldson, Y. K., Bell, J. E., Brettle, R. P. et al. (1994). Redistribution of HIV outside the lymphoid system with onset of AIDS. Lancet, 343, 382–5. Eddleston, A. L. W. (1997). Hepatitis B and health-care workers. Lancet, 349, 1339–40. Fryatt, R. J. (1995). Foreign and TB control policy in Nepal. Lancet, 346, 328. Gray, F., Scaravilli, F., Everall, I. et al. (1996). Neuropathology of early HIV infection. Brain Pathology, 6, 1–15. Grist, N. R. & Emslie, J. A. N. (1994). ACP surveys of infections in British clinical laboratories. Journal of Clinical Pathology, 47, 391–4. ¨ Gurtler, L. (1996). Difficulties and strategies of HIV diagnosis. Lancet, 348, 176–9. Hall, A. J. & Harrington, J. M. (1991). Morbidity survey of post mortem room staff. Journal of Clinical Pathology, 44, 433. Healing, T. D., Hoffman, P. N. & Young, S. E. J. (1995). The infection hazards of human cadavers. Communicable Disease Report Review, 5, 61–8. Health Services Advisory Committee (HSAC) (1999). Safe Disposal of Clinical Waste. HSE Books. (2003a). Safe Working and the Prevention of Infection in Clinical Laboratories. HSE Books. (in press). See also: www.hse.gov.uk. (2003b). Safe Working and the Prevention of Infection in the Mortuary and Post-mortem Room. HSE Books. (in press) Henderson, D. K., Fahey, B. J., Willy, M. et al. (1990). Risk for occupational transmission of human immunodeficiency virus type I associated with clinical exposures. Annals of Internal Medicine, 113, 740. Hutchins, G. M., and the Autopsy Committee of the College of America Pathologists. (1994). Practice guidelines for autopsy pathology – autopsy performance. Archives of Pathology and Laboratory Medicine, 118, 19–25. Ironside, J. W. & Bell, J. E. (1996). The ‘high-risk’ neuropathological autopsy in AIDS and Creutzfeldt–Jakob disease: principles and

89

90

J. E. Bell

practice. Neuropathology and Applied Neurobiology, 22, 388– 93. Jarvis, K. M., Davidson, F., Hanley, J. P. et al. (1996). Infection with hepatitis G virus among recipients of plasma products. Lancet, 348, 1352–5. Karhunen, P. J., Penttilä, A., Kantanen, M.-L. & Leinikki, P. (1989). Screening for HIV in medico-legal necropsies in Helsinki. British Medical Journal, 298, 1160. Kennedy, D. H., Nair, G., Elliott, L. & Ditton, J. (1991). Drug misuse and sharing of needles in Scottish prisons. British Medical Journal, 302, 1507. Kent, R. J., Uttley, H. C., Stoker, N. G. et al. (1994). Transmission of tuberculosis in British centre for patients infected with HIV. British Medical Journal, 309, 639–40. Kernbach-Wighton, G., Kuhlencord, A., Rossbach, K. & Fischer G. (1996). Bone-dust in autopsies: reduction of spreading. Forensic Science International, 83, 95–103. Kibbler, C. C. (1997). Introducing ‘universal precautions’ into clinical and laboratory environments. In Managing Biological and Chemical Risks, ed D. R. Morgan, London: Institute of Biology. Lucas, S. B. (1995). Guidelines for autopsies in the tropics. In Tropical Pathology, 2nd edn. Berlin: Springer Verlag. pp. 3–23. W. Doerr and E. Uehlinger (eds). MacArchur, S. & Schneiderman, H. (1987). Infection control and the autopsy of persons with human immunodeficiency virus. American Journal of Infection Control, 15, 172–7. Metters, J. (1992). Neuro and ophthalmic surgery procedures on patients with or suspected to have, or at risk of developing Creutzfeldt–Jakob disease (CJD), or Gerstmann–Straussler– Scheinker syndrome (GSS). PL(92)CO/4. London: Department of Health. Miller, D. C. (1988). Creutzfeldt–Jakob disease in histopathology technicians. New England Journal of Medicine, 318, 853–4. Murray, C. J. L. & Lopez, D. (1997). Alternative projections of mortality and disability by cause 1990–2020: global burden of disease study. Lancet, 349, 1498–504. Nyberg, M., Suni, J. & Haltia, M. (1990). Isolation of human immunodeficiency virus (HIV) at autopsy one to six days post mortem. American Journal of Clinical Pathology, 94, 422–5. Powers, J. M., and the Autopsy Committee of the College of American Pathologists. (1995). Practice guidelines for autopsy

pathology – autopsy procedures for brain, spinal cord and neuromuscular system. Archives of Pathology and Laboratory Medicine, 119, 777–83. Poznansky, M. C., Torkington, J., Turner, G. et al. (1994). Prevalence of HIV infection in patients attending an inner city accident and emergency department. British Medical Journal, 308, 636–7. ¨ Puschel, K., Lieske, K., Hashimoto, Y. et al. (1987). HIV infection in forensic autopsy cases. Forensic Science International, 34, 169. Reporting of Injuries, Diseases and Dangerous Occurrences Regulations (RIDDOR). (1996). HSE Books. Royal College of Pathologists Working Party Report. (1995). HIV and the Practice of Pathology. Royal College of Pathologists, London. Sitwell, L., Lach, B., Atack, E. et al. (1988). Creutzfeldt–Jakob disease in histopathology technicians. New England Journal of Medicine, 318, 854. Tennant, F. & Moll, D. (1996). Seroprevalence of hepatitis A, B, C, and D markers and liver function abnormalities in intravenous heroin addicts. Journal of Addictive Diseases, 14, 35–49. Turner, M. L. & Ironside, J. W. (1998). New-variant Creutzfeldt–Jakob disease: the risk of transmission of blood transfusion. Blood Review, 12, 255–68. Will, R. G., Ironside, J. W., Zeidler, M. et al. (1996). A new variant of Creutzfeldt–Jakob disease in the UK. Lancet, 347, 921–5. www.bns.org.uk British Neuropathology Society website. www.cdc.gov/Centres for Disease Control and Prevention website. www.cdc.gov/ncidod/diseases/cjd/cjd inf ctrl qn.htm Questions and answers regarding Creutzfeldt–Jakob disease infectioncontrol practices. www.doh.gov.uk/cjd/healthworkers.htm Creutzfeldt–Jakob disease: guidance for healthcare workers. www.official-document.co.uk/documents/doh/spongifm/ report.htm Transmissible spongiform encephalopathy agents: safe working and the prevention of infection. www.who.int/csr/resources/publications/bse/WROcdscsraph 2003.pdf WHO infection control guidelines for transmissible spongiform encephalopathies. Young, S. E. J. & Healing, T. D. (1995). Infection in the deceased: a survey of management. Communicable Disease Report Review, 5, 69–73.

6 Molecular diagnosis of dementia Vivianna M. D. Van Deerlin Center for Neurodegenerative Disease Research, Department of Pathology and Laboratory Medicine, Hospital of the University of Pennsylvania, Philadelphia, PA, USA

Introduction The identification and cloning of numerous genes associated with inherited disorders combined with the advancement and commercialization of molecular methods for the analysis of these genes in the clinical laboratory has had a significant impact on medical practice. These advancements have fuelled the rapid growth of molecular diagnostics. The field of molecular diagnostics takes the theories and principles of molecular genetics along with the technologies of molecular biology and applies them to the clinical laboratory (Amos & Patnaik, 2002). The outcome of such molecular tests can provide physicians and genetic counsellors with information that can improve their ability to offer optimal care for individuals affected with or at risk for genetic diseases. Molecular genetic diagnostic test results can be used to help confirm the diagnosis of inherited disorders in affected individuals and diagnose unaffected individuals prior to the onset of symptoms. In addition, molecular genetic testing can be used to identify individuals at increased risk for developing a disease and for prenatal or preimplantation diagnosis for couples at risk of having a child with a genetic condition. One approach to the discovery of disease-associated genes has been research studies of families in which a particular disease afflicts multiple blood relatives (Fig. 6.1). Initially genomic regions or specific genes may be found to be linked to the disease of interest followed by the identification of specific molecular abnormalities or differences associated with a clinical phenotype (Hardy & Singleton, 2000). While molecular genetic information can be used in the research laboratory to understand the pathophysiology of certain diseases and begin to develop therapies, the

challenge of clinical molecular diagnostics is to translate molecular genetic information into a clinical test that can be used to enhance the care of patients and their families physically and/or psychologically. Ultimately the identification of individuals with a genetic disease prior to the onset of symptoms may permit the institution of therapies to prevent or delay the onset or decrease the severity of disease (Cravchik et al., 2001) (Fig. 6.1). The majority of clinical laboratories that offer molecular diagnostic testing are found as part of either genetics or pathology departments, as well as in commercial laboratories, some of which may specialize in genetic testing. The subspecialty within the field of Pathology that provides molecular diagnostic testing is called molecular pathology. The advantage of molecular methods is that diagnoses can frequently be made more accurately, more rapidly, and in some cases prior to the onset of symptoms. However, heightened awareness by the lay public, the government, and professionals about the social implications of genetic testing results as well as concern over the quality of genetic testing make utilization of these tests more complicated than other diagnostic laboratory tests. The ethical, legal and social implications of genetic testing have raised an array of concerns and calls for regulation of laboratories that perform genetic testing in the United States of America (USA) that have affected and will continue to affect the practice of molecular diagnostics in the foreseeable future (Amos & Patnaik, 2002). Although the regulations and oversight organizations that are described in this chapter pertain specifically to the USA, organizations with analogous roles and responsibilities also exist internationally. The agencies and mechanisms may be different internationally, but the underlying issues are the same.


91

92

V. M. D. Van Deerlin

Diagnostic

Genome sequencing information

Develop clinical test

Prenatal/ preimplantation

Presymptomatic Study families

Identify genomic region and/or gene(s)

Identify mutation(s) or molecular mechanisms Preventative therapies

Disease with hereditary component

Basic research to understand pathophysiology

Develop specific drug therapies

Make animal models Fig. 6.1. Diagram presents some of the interactions between genetics, genomics, basic research and molecular diagnostics. While the identification of disease-associated mutations can lead to a better understanding of disease pathophysiology and to molecular diagnostics tests, the hope of the future is that elucidation of disease mechanisms will lead to preventative therapies that can be applied in conjunction with presymptomatic testing.

This chapter will introduce some of the mechanisms and terminology of human genetic variation, general issues associated with clinical genetic testing, and laboratory issues including methods for mutation detection and clinical test development. Finally, the role and application of molecular diagnostics in the diagnosis and management of familial dementia disorders will be addressed using the approach to specific disorders as examples. The research based identification of new mutations or genetic risk factors for each dementia disorder will not be addressed.

Genetic variation in humans and disease The human genome contains about 3.2 billion nucleotides encoding an estimated 30 000–35 000 genes (Lander et al., 2001; Venter et al., 2001). Amazingly, the nucleotide sequence is nearly identical in all humans. Only about 0.1% of the nucleotides are different between individuals. These differences in combination with environmental effects are in large part what makes each person unique (except for identical twins) in their appearance, behavior, and health.

Molecular diagnosis of dementia

These differences, which are referred to as polymorphisms, may or may not be associated with an outward phenotypic effect. Strictly used, the term polymorphism means the existence of two or more variants or alleles of a gene present at significant frequencies in the population, generally >1% (Cotton & Scriver, 1998). If the variation in genomic sequence is at a single nucleotide position, it is referred to as a single nucleotide polymorphism (SNP, pronounced ‘snip’). (Weiner & Hudson, 2002). About 1.4 million SNPs have been identified in the human genome. Most individual SNPs do not directly cause disease; however, some, either singly or in groups inherited together (haplotypes), have been shown to increase susceptibility for disease or modify the clinical phenotype of disease. The study of SNPs in biomedical research is an area of active investigation. In fact, the promise of pharmacogenomics is that the association of SNPs with specific responses or reactions to medications will permit medical drug therapies to be tailored for each individual to improve response and decrease side effects (Roses, 2001; Ahmadian & Lundeberg, 2002). SNPs have also become an important tool that is being used increasingly to map the locations of new disease genes. Although the term mutation is sometimes used to refer to any change in a specific DNA sequence or gene, in the medical genetics community it is reserved most commonly to imply a change that is harmful or disease causing (Cotton & Scriver 1998). Mutations can be found in either germline cells, in which case they are inherited from generation to generation, or may occur in somatic tissues during the lifetime of an individual. Somatic mutations often are associated with neoplasia. Human mutations can result from single base substitutions, insertions, deletions and dynamic mutations of DNA sequence (Antonarakis et al., 2000). Single base substitutions can result in a missense mutation, replacement of one amino acid for another, a nonsense mutation, replacement of an amino acid codon with a stop codon, and splice site mutations which can be in an intron or non-coding stretch of nucleotides which creates or destroys signals for intron/exon splicing. Variations in the length of DNA are another class of polymorphism, most commonly recognized as differences in the number of tandemly repeated nucleotide sequences that are known as either microsatellites (short, 2–7 base pair, tandem repeats) or minisatellites (larger tandem repeats) (Bennett, 2000). Most tandem repeats are polymorphic within the human population and are not associated with disease. However, microsatellite loci, in particular some trinucleotide repeats, can become unstable generally resulting in an increased number of repeats upon passage from one generation to the next. When these trinucleotide

repeat expansions exceed a certain size threshold beyond the normal range, they become pathogenic mutations that are associated with particular neurodegenerative diseases. Thus, trinucleotide repeat expansions are examples of dynamic mutations. Genetic diseases caused by mutations in one gene are referred to as single gene disorders, while complex or multigene disorders result from a combination of mutations, polymorphisms, and or environmental influences. Molecular diagnosis is most readily applied to single gene disorders. As of May 2002 approximately 10 000 genes were listed in the Online Mendelian Inheritance in Man (OMIM) database, a catalogue of human genes and genetic disorders (Cooper et al., 1998; Antonarakis et al., 2000). Of these, about 94% are inherited as either autosomal dominant or autosomal recessive disorders. Most genetic disorders associated with dementia have onset in adulthood and have an autosomal dominant pattern of inheritance. In disorders of autosomal dominant inheritance the presence of the mutation in the heterozygous state is sufficient to cause disease. Unless a de novo mutation has occurred, affected individuals have an affected parent and offspring of affected individuals have a 50% chance of being affected. In contrast, in autosomal recessive conditions, an affected individual must carry a mutation on both gene alleles (one on each chromosome). The affected individual can be homozygous for the same mutation, or in conditions in which multiple mutations are associated with disease the individual can be compound heterozygous for two different mutations in the same gene. In autosomal recessive conditions heterozygotes (or carriers) are usually unaffected. Parents of individuals affected with autosomal recessive diseases are usually carriers, unless a new mutation occurred, and the chance of the same couple having another affected offspring is 1 in 4. The new mutation rate within a gene varies depending on the size and location of the gene and occurs at varying frequencies in the male or female gametes. Once a new mutation has occurred, it will be transmitted to subsequent generations in a Mendelian fashion. Sex (X and Y) chromosome-linked diseases pose special circumstances for inheritance, which will not be discussed here. There are several factors, which can modify the inheritance of some genetic disorders and are important concepts to understand for the interpretation of genetic testing results. These are penetrance, variable expressivity and anticipation. The penetrance of a mutation is the probability that an individual carrying it will develop the associated disease phenotype at some point in life. Variable expressivity refers to the variation in phenotype or disease severity between affected individuals carrying the same mutation

93

94


(for example in the same family). Anticipation describes the tendency for a disease to manifest at an earlier age and with increased severity in successive generations, frequently associated with trinucleotide repeat expansion diseases. These factors will be discussed further in relation to particular diagnostic situations of dementia disorders. The extent of human genetic variation and the breadth of mutation mechanisms are not only what makes genetic testing important, but also what makes it an exciting challenge for the clinicians and laboratorians.

General issues in genetic testing Ethical and social issues in genetic testing Genetic testing is an important yet powerful tool. Although genetic tests are in some ways just like any other medical laboratory test, they are also inherently different because they can have tremendous social and psychological impact not only on the individual being tested, but on family members and some ethnic groups as well. For example, genetic test results may have implications for employment and insurance, although there is an effort to pass legislation in the USA aimed at protecting individuals from genetic discrimination (Carroll & Coleman, 2001). Genetic counselling is an important adjunct before and after genetic testing to provide individuals with information about the disease and the test, including the purpose of the test, its sensitivity, specificity, positive predictive value, and about the possible outcomes and implications of the test results. Although many care givers can provide such information, certified genetic counselors are specially trained to help people understand their genetic risks and the significance of genetic testing for personal risk, reproductive risk, and risk to family members. Genetic counselling is particularly important when genetic testing is performed in asymptomatic individuals for adult-onset conditions such as Huntington’s and Alzheimer’s diseases (Hedera, 2001). In these cases an intensive specially designed presymptomatic genetic counselling protocol is recommended (Burson & Markey, 2001). Given the potential impact of genetic testing, it is crucial that a signed informed consent form be obtained prior to all genetic tests. By signing an informed consent form an individual documents that he or she understands the testing, accepts the potential risks and voluntarily agrees to proceed with testing. Documentation of informed consent is necessary for both research testing and clinical testing. When the individual being tested is demented or otherwise mentally impaired the concept of informed consent becomes more difficult (Brooks, 1998). If the individual is

unable to comprehend the testing, a surrogate individual, generally someone with power of attorney, is needed to give informed consent for the testing. It is equally important that the individual providing the consent for the mentally impaired individual understand the implications that the testing results can have for at-risk family members, including potentially themselves. For laboratories performing molecular diagnostic tests careful consideration must be given to the handling and storage of genetic information to maintain the privacy and confidentiality of individuals. Additionally, the long-term storage and use of DNA samples has received a great deal of attention because stored DNA can also provide material for research into the molecular alterations associated with population diversity and genetic disorders. The legal and ethical issues posed by storage of DNA include determination of which samples to store and defining for what purpose and to whom stored samples will be made available. It is important to consider currently diagnosable disorders or disease susceptibilities as well as the potential for testing yet-to-be-identified genes (ACMG, 1995).

Indications for genetic testing Diagnostic testing Diagnostic testing is performed to confirm a specific diagnosis or narrow the differential diagnosis in a symptomatic individual. In many cases a genetic test is a cost-effective, sensitive and specific way to make a clinical diagnosis. Genetic testing for the purpose of making a diagnosis may be used to alter clinical management or for genetic counselling of relatives. Even if there is no effective therapy, simply having a diagnosis can be very important to many patients. In addition, a positive genetic testing result can save a great deal of money by preventing further unnecessary diagnostic tests. This type of genetic testing is appropriate regardless of the age of the individual. However, a negative diagnostic genetic test may not rule out a specific diagnosis if the test used does not detect all possible mutations, which is frequently the case in diseases with a large number of heterogeneous mutations.

Presymptomatic testing Presymptomatic or predictive testing is an option for asymptomatic individuals with a family history of a lateonset genetic disorder. Huntington disease is used as the classical model of presymptomatic testing; however, any untreatable adult-onset neurodegenerative disorder presents similar issues and concerns. The results, both positive and negative, can have a tremendous psychological


impact on the individual. The existence of medical intervention that reduces morbidity and/or mortality makes presymptomatic testing medically indicated. In the absence of effective therapies, the results of such predictive testing can influence an individual’s life plans, such as marriage, career, financial planning, and reproduction decisions. It is strongly recommended that special presymptomatic genetic counseling be done before proceeding with predictive testing because of the potential ramifications of such testing on psychological health, employability and insurability. As stated by a policy of the American College of Medical Genetics (ACMG), predictive testing of asymptomatic children at risk for adult-onset disorders is not appropriate unless a therapy is available or medical management would be altered by the information (ACMG/ASHG, 1995). Carrier testing Carrier testing is performed to identify whether an individual is heterozygous for a disease gene mutation inherited in a recessive manner, usually for the purpose of reproductive planning. Carriers are typically not affected. Indications for carrier testing include having: a family history of the disease, a family member who is a known carrier, or a racial or ethnic background with a high incidence of the disease. When carrier testing is done in a population or group of individuals, it is called carrier screening. Prenatal testing Prenatal genetic testing is performed on a fetal sample during pregnancy to assess whether a specific mutation is present or absent. Typically, the sample is either obtained from an amniocentesis or a chorionic villus biopsy. This is essentially a diagnostic test on a fetus that reassures the pregnant couple if it is negative, and if positive, gives the couple the opportunity to either terminate the pregnancy or plan for the birth of an affected child. Prenatal testing for adult onset disorders is controversial. Preimplantation genetic diagnosis Preimplantation genetic diagnosis (PGD) is performed in conjunction with in vitro fertilization (IVF) to identify and transfer only embryos that do not carry a specific genetic disorder (for review see Thornhill & Snow, 2002). Typically, a single cell is removed from a 6–8 cell blastomere and is used for genetic analysis using powerful DNA amplification methods. Alternatively, in other protocols PGD is performed on the first and second polar bodies of the oocyte and fertilized oocyte, respectively. PGD has been performed for more than 50 diseases to date, including early-onset Alzheimer disease. PGD is generally reserved

for couples who would be adverse to termination of an in utero pregnancy or those who are undergoing IVF for another indication. PGD is expensive and has all of the risk and complications of IVF. Nevertheless PGD is an option.

Genetic testing categories Genetic tests are categorized as being either clinical or research tests depending on the laboratory setting where the testing was performed and whether or not the results can be used for clinical decision making. Clinical genetic testing Clinical genetic tests are performed usually for a fee at the request of a physician or genetic counsellor for the purpose of obtaining a result, which can then be used for the clinical management of the individual. Laboratories that perform clinical genetic tests on human specimens in the USA must be certified under the Clinical Laboratory Improvement Amendments of 1988 (CLIA). CLIA certification ensures high quality standards and the accuracy, reliability and timeliness of patient test results regardless of where the test was performed (AMP, 1999; Schwartz, 1999; Amos & Patnaik, 2002). To become certified by CLIA, laboratories must be inspected regularly, and must follow strict guidelines for test validation and ongoing quality control, including proficiency testing. Some states, most notably New York, have developed their own accreditation systems for molecular genetic diagnostic laboratories. In the United Kingdom an independent company set up by the Royal College of Pathologists, Clinical Pathology Accreditation, is involved in inspections and accreditation of laboratories (Elles, 1997). Although clinical genetic tests are available around the world for many of the more common genetic disorders (see online resources below), this is not the case for rare or ‘orphan’ genetic diseases. The effort and cost involved in developing and performing molecular genetic tests in CLIA certified clinical laboratories is prohibitive for rare diseases. Therefore, although the human genome project is generating a large volume of data and characterizing many new disease genes, frequently the only available resource for genetic testing of these rare diseases is research laboratories whose main focus may be the study of one or two diseases. Research genetic testing Research genetic testing is generally done to advance scientific knowledge about the genetic contribution to disease and to learn more about the clinical phenotype and penetrance of gene mutations. Human subjects for genetic research studies are often highly motivated individuals who

95

96


are symptomatic or have one or more family members with an inherited disease. They are interested in helping researchers learn more about the disease, search for new therapies, and develop clinical genetic tests to help diagnose the condition regardless of whether they themselves or their family will directly benefit from the research. Participation in genetic research generally involves providing family history of physical and mental illnesses followed by the drawing of a blood sample for isolation of DNA. Human research, including genetic studies, in the USA must be approved by Institutional Review Boards (IRBs), groups of individuals that have been formally designated to review and monitor biomedical research involving human subjects. According to the recommendations of the final report of the Task Force on Genetic Testing (Holtzman & Watson, 1997) results of research genetic tests should not to be reported for clinical purposes because the quality of test performance in non-CLIA certified research laboratories cannot be assured. Special exceptions are necessary when a research laboratory is the only provider of a genetic test, which is the case for many rare diseases and when the results of such a genetic test could be useful for a patient. For example, the finding of a disease-associated mutation with a known clinical phenotype in an affected individual may prevent unnecessary additional testing. In addition, presymptomatic testing in family members is also useful for some individuals. With appropriate IRB approval, research laboratories may release research results if requested by the patient, however, in order to be used clinically the result must be confirmed in a CLIA certified clinical laboratory. A model for how analytical validation of such ‘transition’ tests can be expedited in a clinical laboratory is described under clinical validation below.

Laboratory issues in genetic testing Methods for detection of known mutations A variety of methods are available for the detection of variations in DNA sequence. For simplicity the term ‘mutation’ will be used predominantly; however, the same methods also apply to the detection of polymorphisms. A clinical laboratory interested in developing a clinical genetic test for a known mutation must select a method based on such variables as local expertise, available instrumentation, and cost. In addition, the choice of platform for mutation analysis depends on whether the known mutations are confined to a single location or a specific mechanism, or whether the mutations are scattered throughout the gene or a region of the gene. The method of choice will also depend on

the anticipated test volume. A labour-intensive assay is not suitable for a high volume test, but may be a consideration if it is the most appropriate method for an infrequently performed test. The ideal procedure for mutation detection is accurate, rapid, inexpensive, specific, and technically easy to perform. Restriction fragment length polymorphism analysis Mutations at predefined genomic locations can be identified in many ways. Most methods incorporate amplification of the genomic DNA using polymerase chain reaction (PCR) amplification in combination with a detection method. The most widely applied method is PCR amplification combined with restriction fragment length polymorphism (PCR–RFLP). This method can be applied if the mutation either creates a new restriction site in the DNA sequence or causes the loss of a pre-existing restriction site from the normal allele. After PCR amplification of the DNA sequence containing the site of interest the PCR product is digested with a restriction endonuclease, separated by electrophoresis, and the fragments visualized by staining with a dye such as ethidium bromide. The pattern of fragments generated is used to distinguish the normal allele from the variant allele in both the homozygous and heterozygous state. An example of this method applied to the detection of the ApoE gene polymorphisms is shown in Fig. 6.2. A variation of PCR–RFLP can be used even if the mutation of interest does not create or destroy a restriction endonuclease digestion site. This method is called restriction-generating PCR because a nucleotide mismatch is incorporated into one of the PCR amplification primers such that a new site is created in the amplified product in combination with the variant allele, but not with the normal allele. The advantage of PCR–RFLP is that it is an easy method to design and perform and it requires only basic molecular biology equipment. The disadvantage of PCR–RFLP is its lack of absolute specificity. There are several nucleotides that make up a restriction endonuclease recognition site and a change in any one of these nucleotides can affect the enzyme activity. Allele-specific methods for mutation detection Methods based on the specificity of PCR amplification, hybridization, or ligation for the detection of specific nucleotide changes have been gaining popularity for mutation detection (for review see Ahmadian & Lundeberg, 2002). Allele-specific amplification takes advantage of the fact that DNA polymerase will not extend a primer with a mismatch at the 3 end. In this approach, sometimes called amplification refractory mutation system, two PCR reactions are used to amplify each sample. Each reaction contains identical reagents with the exception of one of the


Hha I

A

112

fragments

158

72

E4

30

E3

30

91

E2

30

91

48

35

30, 35, 48, 72

48

35

30, 35, 48, 91

83

30, 83, 91

B

bp

125 100 75

M

1 2 3 4 5 6 Water E2/E2 E2/E3 E3/E3 E3/E4 E4/E4 E2/E4

M

50

25 Fig. 6.2. Restriction fragment length polymorphism analysis of the APOE gene. Polymorphisms in the APOE gene occur in nucleotides that code for amino acids 112 and 158 in the protein and result in three alleles: ε2, ε3 and ε4. A significant proportion of AD patients have at least one ε4 allele ( ε2/ ε4, ε3/ ε4, or ε4/ ε4). A molecular test to genotype the APOE alleles is based on the digestion pattern of the restriction endonuclease HhaI as shown in A. A portion of the APOE gene is amplified by the PCR followed by restriction digestion and gel electrophoresis. Each vertical bar in the rectangles representing the amplified portion of the gene for the three alleles represents an HhaI cut site. Some of the fragments generated are too small to be visible by gel electrophoresis. The sizes of the larger fragments are indicated. An agarose gel electrophoresis stained with ethidium bromide of the six possible combinations of the three alleles is shown in B (lanes 1–6). Size markers (M) and a negative PCR amplification control (water) are also shown.

primers, which is different in each tube: one is designed to match the variant and the other the normal allele at the 3 ends. The pattern of amplification in the two reactions indicates the genotype of the sample. Allele-specific hybridization for mutation detection can take many forms, but is based on the principle that when an oligonucleotide hybridizes to a target DNA sequence the stability of the interaction is dependent on the presence or absence of

mismatches. Therefore, under certain conditions, the perfectly matched probe will hybridize while the mismatched probe will not. Assays based on hybridization can be designed on solid supports (dot or slot blots) or in solution (real-time PCR). Ligation-based assays, such as oligonucleotide ligation assay and ligase chain reaction rely on the discriminatory properties of the enzyme ligase to join only probes that are matched perfectly to a complementary

97

98


sequence (Chen et al., 1998; Romppanen, 2001). The increase in molecular diagnostic application has driven an advance in available technologies for mutation detection. There are numerous proprietary methods that have been developed by biotechnology companies such as the r Fragment Length Polymorphism (Rossetti et al., Cleavase 1997; Oldenburg & Siebert, 2000), READIT (Bean et al., 2001; Tsongalis et al., 2001), and Invader (Kwiatkowski et al., 1999) assays, some of which require special instrumentation to perform. Detection of insertions, expansions and deletions The detection of insertions, deletions, and variations of length require different methods. Depending on the size of the deletion or expansion PCR can be used. If a large segment is deleted so that the primer binding sites on the mutant allele are missing, then the mutant allele will not be detected and only the normal allele will be amplified. Very large deletions may be seen cytogenetically; however, other large deletions are difficult to detect. Quantitative PCR can be used to detect deletions by comparing the amount of amplification product from the gene of interest to a normal two-copy gene. (Hedrich et al., 2001). The number of repeats at a trinucleotide repeat locus can be determined by PCR amplification across the repeat region followed by separation of the amplified fragments using a method that has high resolution for size determination, such as sequencing gel or capillary electrophoresis. An expansion would be detected as an increase in the size of the PCR product. In some trinucleotide repeat diseases, such as myotonic dystrophy, the repeat expansions can be so large (up to thousands of kilobases) that they are not detectable by a standard PCR amplification, thereby requiring Southern blot of genomic DNA for detection (Cummings & Zoghbi, 2000).

with real-time detection. One advantage of the capillary systems over the slab gel is that they obviate the need to prepare a gel, which can be time consuming. There are several capillary-based systems available that run either 1, or 8, 16, and even 96 or more capillaries in parallel increasing the throughput, although they increase cost as well. A disadvantage of DNA sequencing is its relatively high cost, its labour-intensive nature and the fact that in most clinical laboratories it is not amenable to high throughput. Another potential disadvantage of DNA sequencing is the risk of miscalling a heterozygous mutation as a homozygous normal. This can happen if the fluorescence intensity of the mutant allele is much lower than that of the normal allele such that the mutant allele peak is below the threshold of detection. The frequency of this type of error is decreased in part by sequencing both strands of a DNA fragment. Another interpretive concern is raised when DNA sequencing is performed as a clinical test and a novel variant of unknown significance is identified. Although mutations that affect the reading frame, cause a missense or nonsense mutation, or alter a splice site have a high likelihood of being pathogenic, definitive assessment of disease association requires additional family studies to determine whether or not the variant segregates with disease and in vitro expression and analysis of the variant gene product (Cotton & Scriver, 1998). This type of follow-up analysis of novel variants to distinguish mutations from polymorphisms is feasible and of interest for a research laboratory; however, generally not possible for a clinical laboratory. One way to decrease the cost and increase the throughput of samples is to use a screening or scanning method for the identification of variations in the target DNA sequence and then sequence only the samples that demonstrate the presence of a heterozygous nucleotide change.

Methods for heterogeneous mutations The human genome project was made a reality by tremendous advancements in automated DNA sequencing technology. DNA sequencing is the gold standard for the identification of mutations found in several locations throughout a gene because of its specificity and ability to detect many different mutations in a single amplified PCR product. DNA sequencing is now generally performed with thermostable enzymes which permit direct sequencing of PCR products in a thermal cycler, so-called cycle sequencing (Ahmadian & Lundeberg, 2002). These reactions, which typically use four-colour fluorescent dideoxy nucleotides as termination nucleotides, are separated in an automated sequencing instrument that is either slab gel or capillary electrophoresis-based

Screening for heterogeneous mutations The most commonly used screening methods are conformation-based techniques that are based on the fact that mutant molecules have aberrant electrophoretic migration under certain conditions (Cotton, 2000). These methods are reviewed elsewhere and include single strand conformation polymorphism (SSCP), conformation sensitive gel electrophoresis (CSGE), denaturing gradient electrophoresis (DGGE), heteroduplex analysis (HA), and denaturing high-performance liquid chromatography (DHPLC) (Ahmadian & Lundeberg, 2002). The advantage of these methods is that with relative ease many samples, either multiple fragments from a single large gene or the target fragment from many patient samples, can be screened for the presence of a nucleotide variation. A disadvantage of


methods such as these is that conditions for analysis must be optimized individually for each target sequence, which can be time consuming. In addition, these methods can only indicate that there is at least one nucleotide difference between two sequences, it does not specify whether the difference is a polymorphism or a mutation, nor the exact location of the sequence variation. Therefore, conformationsensitive methods are not useful in DNA fragments with many polymorphic sites as they will frequently be positive leading to a high rate of confirmatory sequencing.

Clinical test development There are many issues and hurdles that molecular diagnostic laboratories face in order to develop a clinical molecular test. The test development process begins with the establishment of clinical needs, feasibility of the testing, cost and benefits of the test (Amos & Patnaik, 2002). In principle, a molecular assay can be developed in a clinical laboratory for almost any mutation or polymorphism which has been previously associated with disease. In practice however, the effort and high cost involved in the development of a clinical molecular assay in a CLIA certified laboratory limits the development of assays to tests which have a high enough volume to offset the cost of development. Some specialized laboratories may have greater flexibility by serving as national reference laboratories for some tests, for example genetics tests for some neurological disorders may only be available from laboratories that specialize in neurogenetics. Once the decision is made to offer a clinical genetic test, an assay procedure must be developed with known positive and negative controls using one of the mutation detection methods described above depending on the types and variety of disease associated mutations. Although there are some Food and Drug Administration (FDA) approved kits for molecular testing, the majority of molecular diagnostic tests are developed in individual laboratories as so-called ‘home-brew’ assays (AMP, 1999). As required for clinical testing under CLIA these assays must be validated to ensure that the correct results are obtained and that the performance characteristics are reproducible (AMP, 1999). As new disease genes are discovered, the development of diagnostic tests may lag behind, especially for rare disease genes, or ones in which the mutations are heterogeneous necessitating more labour-intensive detection methods. Research laboratories that identify a new disease gene or new mutations in a known gene are often contacted by families or individuals that are interested in genetic testing. Until the significance of the gene is established in most cases the only thing that the research laboratory can offer

is enrolment into an IRB-approved research study. As the gene association becomes well defined and penetrance understood, some families may want results returned to them so that, if they have a detectable mutation, for example, at risk family members can consider genetic testing. Research laboratories, at least in the USA, are not certified to release patient results. On the other hand, clinical laboratories are under economic pressures and in most cases cannot afford to spend the time and money to perform a full assay development and validation for rare genetic tests. Protocols that facilitate the transfer of primarily research genetic tests to the clinical testing arena have been proposed. One such proposal suggests that a clinical laboratory can accelerate the adaptation of a research protocol for mutation detection into the clinical laboratory using methods that the laboratory is already using for other clinical assays (Van Deerlin et al., 2003). For methods that involve the amplification of genomic DNA what this would involve first is validation of the primer sequences by documentation that the PCR product generated by the clinical laboratory is of the expected size and DNA sequence compared to the research protocol. Then any method already validated in the clinical laboratory for analysis of mutations in other genes, including RFLP, SSCP or DNA sequencing could be used for the adapted clinical assay. To expedite the initial validation of the assay several split samples could be used to assess that the clinical and research laboratories obtain the same result. Although this model has not been officially approved by any regulatory agency, it is realistic from a time and cost standpoint and maintains the quality control elements necessary in genetic testing. Once the clinical laboratory has a validated protocol, for example as described above, any mutation within the same PCR product can be identified and reported as long as the size of the PCR product is verified and appropriate positive and negative controls are included in each analysis. If a research laboratory identifies a mutation of known clinical significance and has IRB approval to release the research results, then a test result report can be issued that specifically states that the result is for research use only and is not intended for clinical use. A clinical laboratory can then use this report to confirm the research laboratory result for clinical diagnosis in the affected individual. Furthermore, clinical laboratories are able to test for a specific mutation in unaffected family members if a mutation has been previously identified in the family. Clinical laboratories may also be asked to screen for mutations in affected individuals in which the diseaseassociated gene has many possible heterogeneous mutations. Although this analysis is possible using the same

99

100


Table 6.1. Genetics summary of selected familial dementia disorders Disorder

Gene name

Gene

Chromosome

Inheritance mode

OMIM#

Types of defects

Alzheimer’s disease, early-onset familial (AD1)

Amyloid Precursor Protein

APP

21q21.3–q22.05

Autosomal dominant

104760

Point mutations

Alzheimer’s disease, late-onset familial and sporadic (AD2)

Apolipoprotein E

APOE

19

Complex

107741

Polymorphisms

Alzheimer’s disease, early-onset familial (AD3) Alzheimer’s disease, early-onset familial (AD4)

Presenilin 1

PSEN 1

14q24.3

Autosomal dominant

104311

Point mutations Splice site mutation

Presenilin 2

PSEN 2

1q31–q42

Autosomal dominant

600759

Point mutations

Frontotemporal dementia with parkinsonism linked to chromosome 17 (FTDP-17)

Tau

MAPT

17q21.1

Autosomal dominant

157140

Point mutations small deletions

CADASIL

NOTCH 3

NOTCH 3

19p13.2–p13.1

Autosomal dominant

600276

Point mutations

Huntington’s disease

Huntingtin

HD

4p16.3

Autosomal dominant

143100

Trinucleotide repeat expansion

Creutzfeldt–Jakob disease Gerstmann–Sträussler– Scheinker Familial fatal insomnia

Prion protein

PRNP

20pter–p12

Autosomal dominant

176640

Point mutations, deletions, insertions

Parkinson disease with dementia

α-Synuclein

SNCA

4q21

Autosomal dominant

163890

Point mutations

generic protocol for a specific validated target sequence, it is time consuming and the sensitivity of mutation detection will vary depending on the method used and the proportion of the gene tested. Thus, screening for mutations in genes associated with rare diseases is more commonly performed in research laboratories.

Genetics of selected dementia disorders The basis of human variation and the role that genetic testing can play in clinical medicine have been described in general terms along with the social and ethical issues raised by genetic testing. The neurodegenerative disorders described briefly below are used to illustrate the role that genetic testing plays in the diagnosis and management of dementia disorders. For each disorder or category a brief synopsis of the genetic basis of disease is presented

followed by a summary of genetic testing issues. A summary of the genetics of some dementia disorders in presented in Table 6.1. For detailed description of the clinical characteristics, neuropathology and genetics of each disease the reader is referred to the specific chapters.

Trinucleotide repeat diseases Genetic basis of disease Trinucleotide repeats are members of the microsatellite (or short tandem repeat (STR)) family of tandem repeats found normally at many sites (loci) throughout the human genome. In general the numbers of repeats at each locus in the human population stays relatively constant within a certain size range that varies for each STR locus. However, at some trinucleotide repeat loci the number of repeats can become unstable resulting in an expansion


of repeats beyond the normal range. When the expansion reaches a particular size, a pathological effect results. Expansion of unstable trinucleotide repeats was discovered as a novel mechanism of gene mutation in 1991 (Fu et al., 1991; La Spada et al., 1991). Since then the pathogenesis of many neurological disorders has been attributed to trinucleotide repeat expansions, including spinobulbar muscular atrophy (SBMA, Kennedy’s disease), Huntington disease (HD), dentatorubropallidoluysian atrophy (DRPLA), Friedreich ataxia (FA), myotonic dystrophy (MD, fragile X syndrome, and the spinocerebellar ataxias (SCAs). The inheritance mode of the trinucleotide repeat disorders is typically autosomal dominant, although fragile X and SBMA are X-linked and FA is autosomal recessive. The trinucleotide repeat disorders as a group share many features including that of anticipation, the occurrence of increasing disease severity and earlier age of onset in successive generations (Cummings & Zoghbi, 2000). Trinucleotide repeat diseases can be categorized into two subclasses based on the location of the trinucleotide repeats either within or outside of the coding region of the gene. Trinucleotide repeat expansions that occur within the coding sequence of the gene are caused by CAG repeats that encode a polyglutamine tract in the respective proteins (Zoghbi & Orr, 2000). Diseases in this subclass are characterized by neurodegeneration and include SBMA, HD, DRPLA, and the SCAs (SCA1, SCA2, SCA3 (also known as Machado–Joseph disease), SCA6, and SCA7) and in two, HD and DRPLA, dementia is a prominent clinical feature. Mutant proteins containing an abnormally long polyglutamine tract cause damage to the neurons in which they are expressed in ways that are not yet completely understood. The trinucleotide repeat expansions in the second group of disorders, FA, Fragile X, and MD, are located in untranslated regions of the respective genes and the disorders are more clinically heterogeneous. HD was one of the first CAG trinculeotide repeat disorders for which direct gene analysis became available. As a result, HD serves as a model for many of the genetic counselling issues raised by genetic testing for adult-onset diseases. Genetic testing for HD The diagnosis of HD is made based on a positive family history, characteristic clinical findings, and molecular testing results that demonstrate an expansion in the number of CAG repeats in the 5 coding region of the HD gene, huntingtin. (Kremer et al., 1994; ACMG/ASHG 1998) (Fig. 6.3(a)). The number of CAG repeats in normal alleles ranges between 10 and 26. While in patients with HD the number of CAG repeats ranges from 40 to 121. Alleles of 27 to 35 repeats are considered intermediate alleles because the

individual is not at risk of developing HD, but may be at risk of having a child with an allele in the abnormal range. Individuals with repeats between 36 and 41 may or may not develop symptoms of HD over the course of a lifetime because alleles of this size have reduced penetrance. There is an inverse correlation between the number of CAG repeats in the huntigtin gene and severity of symptoms with a larger number of repeats associated with a younger age of onset. The number of repeats on both alleles can be determined by PCR amplification across the repeat locus with flanking primers followed by separation by gel electrophoresis, and sometimes followed by Southern blotting and probe hybridization to confirm the specificity of the amplified fragments (Fig. 6.3(b)). When the HD genetic test was initially developed, primers were used that amplified not only the huntingtin CAG repeats, but also included an adjacent small polymorphic CCG repeat region ranging in size from 7 to 12 repeats which does not contribute to the expression of disease. (Andrew et al., 1994). It was soon recognized that amplification of both the CAG and the CCG repeats could lead to diagnostic inaccuracy of the CAG repeat size because the CCG repeat could affect the PCR product size. Since the difference in size between affected and intermediate or normal alleles can be as little as three nucleotides, accurate size determination is critical, particularly in the range of 35 and 40 repeats. Thus, primers that amplify only the CAG repeat region should be used initially. In recognition of the need to standardize the methods and terminology used by clinical laboratories performing genetic testing for HD, practice guidelines were developed in 1998 by the American College of Medical Genetics and the American Society of Human genetics (ACMG/ASHG, 1998). When only a single CAG repeat allele is amplified it could be an indication that either the individual is homozygous for a normal-sized allele or that a very large expanded allele that could not be amplified because of its size is present. The presence of the nearby polymorphic CCG repeat can be used to help distinguish these two alternative possibilities. By using the primers that amplify both repeats there is a chance that two normal sized alleles will be identified. Alternatively, very large expansions can be identified by Southern blot hybridization of genomic DNA. Diagnostic testing for HD is straightforward using the molecular methods described. Since HD is an adult-onset disease it is possible to perform the test in an individual who is asymptomatic, but at risk of developing disease (Huggins et al., 1990; Benjamin et al., 1994). HD was the first disease for which such presymptomatic testing became available. There are a lot of reasons that individuals may seek predictive testing for HD; however, these are all social and

101


(a)

≥40 to 121 CAGn Huntington disease 36 to 39 CAGn Reduced penetrance

27 to 35 CAGn Mutable normal allele

≤26 CAGn Normal

CAGn Exon 1 (b)

Fluorescence intensity

102

Size in base pairs

Normal


Fig. 6.3. Huntington’s disease molecular diagnosis. (a) A CAG trinucleotide repeat in exon 1 of the huntingtin gene is associated with Huntington’s disease. The number of CAG repeats determines whether an individual is normal or affected. The number of CAG repeats can be determined by PCR amplification across the repeat with flanking primers (arrows). Individuals with an allele with CAG repeats in the mutable range are normal; however, the mutable allele is at risk of expanding in the next generation. Reduced penetrance alleles are sometimes associated with disease, but not always. (b) Example of a Huntington’s disease molecular diagnostic test showing a normal individual with 18 and 23 repeats, and an affected individual with a normal allele of 15 CAG repeats and an expanded allele with 41 repeats. The number of repeats was determined by PCR amplification with fluorescently labelled primers, followed separation by capillary electrophoresis. The number of CAG repeats was determined from the size of the PCR product (x-axis). The smaller secondary peaks seen adjacent to each dominant peak result from slippage (stutter) of the polymerase.


psychological since the identification of a full CAG repeat expansion is associated with virtual 100% penetrance and currently no chance for a cure or prevention. A positive predictive result can be devastating news carrying with it a risk for depression, suicide and discrimination (Almqvist et al., 1999). However, a negative result is not without risk either. It is well documented that individuals with a negative result can suffer from survivor’s guilt, an inability to focus on longterm plans, and also have an increased risk of suicide. In order to prepare individuals for either outcome of testing, extensive pre- and post-test counselling is recommended. Predictive genetic counselling usually involves pre-test interviews in which the motives for requesting the test are ascertained as well as the individual’s knowledge of HD. The pre-test counselling is given in multiple sessions over several weeks or months and includes a psychiatric evaluation. Multiple opportunities are given to the individual to change his or her mind, even after the test is performed, but before results are provided. Post-test counselling is aimed at helping the individual cope with the results of the testing. Not all at-risk individuals are interested in knowing their genetic fate, in fact overall demand for predictive testing for HD has been far below the expected rate. Yet, since symptoms of disease do not appear until after child-bearing years, some individuals who themselves do not want to be tested, also do not want to give birth to an at-risk child. In such cases it is possible to use linkage analysis to identify at-risk fetuses without determining the status of the at-risk parent (Tyler et al., 1990).

Alzheimer’s disease Genetic basis of disease Alzheimer’s disease (AD) is characterized by an adult-onset slowly progressive dementia and is the most common form of dementia. The brains of AD patients characteristically reveal the presence of neurofibrillary tangles and senile plaques. While neuropathological findings in the brain on autopsy are the gold standard for the diagnosis of AD; clinical criteria for AD diagnosis can lead to a correct diagnosis in approximately 80–90% of cases (Joachim et al., 1988). AD can be categorized as either sporadic or familial (Bird, 1998, updated 2001). About 75% of AD cases are sporadic, that is, there is no family history of AD. Sporadic cases generally occur in individuals over the age of 65. The remaining 25% of cases are familial. Familial cases are clinically and pathologically identical to the sporadic cases with the exception of a positive family history of dementia. Familial AD is further subdivided into late-onset (age of onset >60–65) and early-onset (age of onset Peri > CA1 > Amyg > NBM > 20 > Parasub > 21 > CA3/4

No correlation between degree of SP pathology and age. Rank order distribution: Temporal areas (20, 21, 22) > Amyg > EC > Peri > Presub

Tangle density increases with age: highest = EC, Peri, AON lowest = CA3, Sub., Piriform cortex, NBM, Neocortex Difference between younger and older cases = NFT in CA1 + periamygdaloid cortex in the older

Variability among ND cases: no plaques in any part of the forebrain few diffuse amyloid plaques in neocortex diffuse and neuritic plaques in neocortex + EC + Peri

NFT in EC and CA1 in all healthy ND and preclinical AD cases + density increases with age

No or a few diffuse plaques in neocortes in healthy ND cases Neuritic and diffuse plaques in neocortex in preclinical and very mild AD cases

Note: Abbreviations: ND, non-demented; NCI, no cognitive impairment; MCI, mild cognitive impairment; AD, Alzheimer’s disease; CDR, clinical dementia rating scale; EC, entorhinal cortex; Peri, perirhinal cortex; Sub, subiculum; NBM, nucleus basalis of Meynert; NFT, neurofibrillary tangles; Parasub, parasubiculum; Presub, presubiculum; Amyg, amygdala; AON, anterior olfactory nucleus.

frontal neocortex (Braak & Braak, 1990; Price et al., 1991; Arriagada et al., 1992b; Bouras et al., 1993; Giannakopoulos et al., 1994b) (Figs. 7.2, 7.3). In cases presenting with mild symptoms of cognitive decline, NFT are always present in the medial and inferior temporal cortex, including

hippocampus, entorhinal and perirhinal cortices (Figs. 7.2, 7.3). Lesions occur preferentially in layer II of entorhinal cortex, and to a lesser extent in layers III and V. Other neocortical regions display only sparse NFT (Hof et al., 1992; Bouras et al., 1993, 1994; Bierer et al., 1995). NFT

Neuropathology of the ageing brain

Fig. 7.1. Examples of NFT (a) and SP (b) from the hippocampus of an AD case. Note the flame-shape morphology of NFT and the more variable features of SP. In (b), several SP stained with an anti-amyloid antibody are visible. Scale bar (on b) = 100 m (a) and 200 m (b).

progressively spread out in the neocortex as the dementia worsens, leading in these cases to the typical neuropathologic profile of AD with a severe involvement of association neocortical areas and a relative preservation of the occipital cortex (Arnold et al., 1991; Morrison & Hof, 1997) (Fig. 7.2). SP result from the extracellular aggregation of the amyloid peptides, or A peptides. Three different enzymes, -, -, and -secretases, are involved in the proteolytic cleavage of a larger transmembrane precursor (APP, or amyloid precursor protein), giving rise to several A peptides 40 to 43 amino acids long. Also, a large number of mutations occurring in three different genes (the APP, presenilin 1, and presenilin 2 genes), are responsible for an elevated production of the most aggregative A 1–42 peptide in familial cases of early-onset AD (for review, see Selkoe, 2001). The topographic distribution of SP in non-demented elderly individuals shows a pattern different from that of NFT, as SP occur primarily in the frontal and inferior temporal neocortex (Brodmann’s area 20) and in the primary visual cortex where SP are preferentially distributed in layer IV (Rogers & Morrison, 1985; Braak et al., 1989). The hippocampal formation, the entorhinal and perirhinal cortices, and the frontal cortex are less often affected in cases presenting with mild dementia. In normal ageing, mostly primitive plaques, devoid of amyloid dense core or crown of dystrophic neurites, are usually found. In very mildly or mildly demented cases, higher densities of deposits may be observed in neocortical areas, as well as in subcortical structures such as the basal ganglia. These amyloid deposits are mainly primitive plaques that in the neocortex are preferentially localized in the supragranular layers. While the neurodegenerative

process is progressing, amyloid deposits affect all cortical areas and many subcortical structures, and consist mainly of mature plaques containing a dense amyloid core and dystrophic neurites (Fig. 7.1). The visual areas of the occipital cortex generally have high SP densities (Rogers & Morrison, 1985; Price et al., 1991; Arriagada et al., 1992b; Dickson et al., 1992; Bouras et al., 1994). It should be noted that SP distribution in the brain of elderly people is not as consistent as the regional and laminar patterns described above for NFT (Bussière et al., 2002). Finally, amyloid angiopathy can also be observed in pathological ageing, as well as in cases classified as normal ageing with no amyloid deposition in the brain parenchyma (Dickson et al., 1992).

Neuronal loss in normal ageing Neuronal loss, together with NFT formation and amyloid deposition, is a characteristic feature of AD and several neurodegenerative disorders. It has also been mentioned as a possible signature of the ageing brain. In recent years, the unbiased counting methods of stereology have been used to investigate this parameter more precisely, and have led to a reconsideration of the idea of age-related neuronal loss (Morrison & Hof, 1997). Evidence of neuronal loss has been shown in the brain of patients with very mild cognitive dysfunction, in parallel with the incidence of NFT formation in selectively vulnerable regions such as the entorhinal cortex and the hippocampal formation. However, several studies have failed to find any significant decrease in the total number

115

116

` and J. H. Morrison P. R. Hof, T. Bussiere

Fig. 7.2. Regional and laminar NFT distribution and neuronal loss in normal ageing and AD. The flame-shaped structures represent a semi-quantitative assessment of NFT densities. An estimate of the percent of neuronal loss is shown by the grayscale (at bottom). In normal ageing (CDR 0), a few NFT are consistently observed in layer II of the entorhinal cortex (EC) and the superior frontal cortex (SFC) remains devoid of NFT. There is no neuronal loss in normal ageing. In contrast, mild cognitive impairment (CDR 0.5) is characterized by higher NFT densities in the entorhinal cortex. Very rare NFT are observed in superior frontal cortex. A significant degree of neuronal loss is present in layer II of the entorhinal cortex, whereas the neocortical areas are not affected. In definite AD (CDR 2), NFT are found in very high densities in layer II of the entorhinal cortex, but in moderately high densities in the superior frontal cortex. The degree of neuronal loss parallels NFT densities in these regions and the size of boxes reflects cortical shrinkage.

of neurons in elderly individuals with no cognitive impairments (Fig. 7.2). Gómez-Isla and collaborators have shown that there is no difference in the total number of neurons or neuronal densities in any of the layers of the entorhinal cortex between the sixth and ninth decades of life in a group of subjects with normal cognitive functions (clinical dementia rating) (CDR score of 0) (Gómez-Isla et al., 1996). A significant neuronal loss was described in the entorhinal cortex of patients with very mild cognitive dysfunction compared to control subjects, layers II and IV being particularly affected with decreases of 57% and 41%, respectively. Similarly, Price and colleagues have described a substantial neuronal loss in very mild AD, but little or none in the entorhinal cortex and CA1 field in healthy ageing. They also reported that preclinical AD cases resemble healthy nondemented cases in the volume and number of neurons in the entorhinal cortex and the CA1 field (Price et al., 2001). Kordower and colleagues have described a dramatic decrease in the number and volume of neurons in layer II of the entorhinal cortex in elderly patients with mild cognitive impairment compared to individuals with no cognitive impairment (decrease of 63.5% and 24.1%, respectively). These authors found a comparable loss and atrophy of neurons in layer II of the entorhinal cortex in cases with mild cognitive impairment and AD patients, and a progressive loss of layer II volume. Also, they showed that the morphological changes in the entorhinal cortex are clearly associated with impairments in measures of the delayed recall performance, which reflects the hippocampal function (Kordower et al., 2001). West and colleagues have shown that, in a population of 45 normal ageing subjects aged 13 to 101 years, a considerable neuronal loss restricted to the hilus and the subiculum over the ages studied, whereas the other fields of the hippocampus were not affected (West et al., 1994) (Fig. 7.4). This study also demonstrated a major difference between normal cases and AD cases, the AD cases exhibiting the most important neuronal loss in the CA1 field of the hippocampus. Moreover, our own stereological determinations of the number of NFT-free cells and NFT-bearing cells in the entorhinal cortex and its relationship to the cognitive status of the cases demonstrate that CDR 2 represents a pivotal stage in the neurodegenerative process and subsequent neuronal loss. With respect to the CDR 0 and 0.5 cases, a moderate number of intracellular NFT are observed, while the extracellular NFT (or ghost tangles) are rare. The total number of neurons is not significantly decreased in these cases (Fig. 7.5). The neurofibrillary pathology becomes significant in the CDR 2 cases, with a severe increase of intracellular and extracellular NFT, whereas the total number of neurons is decreased consistently. Interestingly, such changes are not observed in the neocortex prior to the CDR 3 stages


Fig. 7.3. Neurofibrillary tangle and SP densities in layer II of the entorhinal cortex in a series of 61 non-demented elderly individuals (left panel). The bars on the right side of the graph represent NFT counts expressed in % of the total number of neurons on Nissl-counterstained sections. The bars on the left side of each panel represent SP counts per mm2 . Note that NFT are present in all of the cases in the entorhinal cortex, and that SP counts display no correlation with the presence and density of NFT. Some of these cases (bottom of the graph) are characterized by consistently high NFT densities which may represent a neuropathologic correlate of very early stages of AD. The top right panel shows a few NFT in layer II of the entorhinal cortex in a non-demented 60 years old individual. The bottom right panel shows the distribution of lesions in the neocortex of a severe AD case.

further demonstrating that progression of neurofibrillary degeneration to the neocortex is a characteristic of dementia but not of normal ageing (Bussière et al., 2003a,b). Altogether, these studies indicate that large populations are essential to investigate accurately the age-related events leading to the formation of NFT, and SP and to neuronal loss in cognitively intact elderly people, and to define the differences in distribution of the lesions between elderly without cognitive impairments and patients with age-related memory deficits and AD. To date, the results from few studies that have reviewed large case series (Giannakopoulos et al., 1994a, 1997a), suggest that select neuronal subpopulations and select cortical circuits are susceptible to neuropathological changes (Hof et al., 1990; Hof & Morrison, 1990). It appears that the time-course and distribution pattern of these changes are different in normal brain ageing and AD (Fig. 7.5), and that extensive hippocampal alterations are correlated with age-related

cognitive deficits, whereas NFT formation in neocortical association areas of the temporal cortex is a prerequisite for the development of AD (Bouras et al., 1993; Morrison & Hof, 1997). Finally, there is no clear correlation throughout cortical regions between the presence of NFT and SP. Thus, the current evidence emphasizes a differential distribution of NFT and SP, and suggests that the involvement of neocortical areas is a key factor to distinguish between normal cerebral ageing and a neurodegenerative mechanism leading to AD (Hof et al., 1992; Bouras et al., 1993; Morrison & Hof, 1997).

Centenarians: a special case in pathology of brain ageing Because there appears to be an age-related progression in the severity of neuropathological changes in brain ageing,

117

118


Fig. 7.4. Percentage of neuronal loss in the various subfields of the hippocampal formation in normal ageing and AD based on the data of West et al. (1994). This stereologic study revealed the neurons in the hilus of the dentate gyrus and the pyramidal neurons of the subiculum are affected as severely in cognitively normal patients as in AD cases.

Fig. 7.5. Total number of neurons in layer II of the entorhinal cortex in CDR 0.5 and CDR 2 cases. Neuron counts were obtained stereologically and have been subdivided into normal neurons revealed by Nissl stain, neurons containing an intracellular NFT (i.e., a Nissl-stained cell labelled with an antibody to hyperphosphorylated tau protein), and extracellular ghost NFT. Note that a significant proportion of the neurons demonstrate transitional pathology in the CDR 0.5 cases, but that the majority remains unaffected by NFT formation. However, in the CDR 2 cases, the majority of neurons has progressed to NFT formation as revealed by the higher proportion of intra- and extracellular NFT.


AD may be considered an end-point of the normal senescence process. It is thus relevant to consider what would be the upper limit of normal brain ageing and whether the neuropathological characteristics the ‘oldest-old’ differ from younger populations. Several reports on populations of centenarians and nonagenarians demonstrate that overall the regional and laminar patterns of NFT and SP distribution in non-demented very old people show subtle differences compared to that in younger non-demented cases (Bouras et al., 2001). However, in cognitively unimpaired centenarians, the occurrence of cerebral lesions is high in areas 22 (superior temporal cortex), 24 and 23 (anterior and posterior cingulate cortex), which contrasts with observations in nondemented younger cases where the pathology is mostly confined to the inferior temporal gyrus, suggesting the existence of a certain degree of differential cortical vulnerability after 90 years of age (Giannakopoulos et al., 1996). Unlike in younger cases, the anterior CA1 field of the hippocampus shows much higher NFT densities in demented compared to non-demented centenarians, suggesting that massive NFT formation in the anterior portion of the hippocampus is typical of AD development in centenarians. In sharp contrast to findings in younger AD cases, NFT densities in the entorhinal cortex are comparable between demented and non-demented centenarians. In Brodmann’s areas 7b and 39 (postero-inferior parietal and angular cortex), 22, 23 and 24, demented centenarians have significantly higher NFT densities compared to nondemented cases, whereas there is no significant difference in areas 9 and 20, as well as in areas 17 and 18. In addition, when a distinction is made between cognitively intact centenarians and centenarians presenting with very mild cognitive impairment, a clear association is observed between increased NFT formation in layers V and VI of area 20 and the cognitive performance (Giannakopoulos et al., 1996). This differs from younger cases with age-associated memory impairment who are characterized by increases in NFT counts in layer III of area 20 (Hof et al., 1992; Bouras et al., 1993, 1994), implying that different subsets of corticocortical connections from the inferior temporal cortex are affected in the early stages of the dementing process in very old people. Thus, NFT formation in the temporal neocortex may play a less prominent role in the clinical expression of AD in centenarians compared to younger patients. Although some involvement of area 20 may account for very mild symptoms of cognitive impairment in this age group, overt clinical signs of AD in oldest-old individuals requires both high NFT densities in the CA1 field and progression of the alterations to areas 7, 22, 23 and 24, which are not affected consistently in most of the other cases at

the early stages of the dementing process (Giannakopoulos et al., 1996). However, the exact clinical repercussion of these differences is still unclear and detailed neuropsychological testing as well as functional imaging studies will be necessary to assess whether cortical functions depending on cingulate and parietal areas are compromised preferentially in AD centenarians. Evaluations of centenarians have revealed a clear correlation between SP densities in areas 7, 9, 20, 22, 23 and 24 and the severity of dementia (Fig. 7.8, Giannakopoulos et al., 1996; Bouras et al., 2001). These observations differ from studies of younger demented and non-demented cases, where the relationship between cognitive status and amyloid burden remains controversial (Arriagada et al., 1992a,b; Näslund et al., 2000; Parvathy et al., 2001; Bussière et al., 2002). It implies that SP formation may represent a pathological hallmark of AD severity in centenarians. In addition to particularities in NFT and SP distribution, centenarians show a differential pattern of neuronal loss in the cerebral cortex compared to younger patients (Giannakopoulos et al., 1996). No statistically significant difference in neuron densities is observed in the CA fields, hilus of the dentate gyrus, and subiculum between demented and non-demented centenarians. In contrast, demented centenarians show significant neuronal loss in layers II and V of the entorhinal cortex, in area 20, and to some degree in area 9, compared to non-demented cases. The absence of neuronal loss in the CA fields and subiculum of demented centenarians contrasts with recent data demonstrating up to 68% neuron loss in the CA1 field and 47% in the subiculum in younger AD cases (West et al., 1994). Thus, centenarians display no significant neuronal loss overall in the CA1 field in spite of the presence of high NFT densities, indicating that these NFT may be at the early stages of their formation and that neuron death has not yet occurred (Giannakopoulos et al., 1996). Neuronal loss in the entorhinal cortex of centenarians is, however, compatible with estimates of total neuron numbers in the entorhinal cortex of patients younger than 95 years (Gómez-Isla et al., 1996), yet it is lower than in younger AD cases, suggesting that a mild decrease in neuron numbers in the entorhinal cortex, as well as in areas 9 and 20 may be sufficient to induce a significant cognitive failure after 95 years of age. From a clinical and epidemiological standpoint, it is important to note that these data from very old individuals strengthen the hypothesis that AD is not on a continuum with normal brain ageing, and that in centenarians the symptoms of dementia may differ considerably from younger patients (Bouras et al., 2001). For example, the prevalence of dementia remains unchanged after 95 years of age, and there is no association in centenarians between

119

120


AD and the presence of apolipoprotein E4 isoform (Bouras et al., 2001). These observations suggest that very old individuals are characterized by differential brain ageing and susceptibility to the dementing process (Giannakopoulos et al., 1996; Bouras et al., 2001).

Cerebrovascular changes in normal ageing The cerebral vasculature comprises two main components, the macrovasculature, and the microvasculature or parenchymal capillary system. The macrovascular system includes the large arteries forming the circle of Willis and the vertebrobasilar arteries, the penetrating arterioles, and the veins draining the blood into the brain sinuses. These arteries are involved in the regulation of the cerebral blood flow and cerebral perfusion. The microvascular system consists of precapillaries and capillaries, devoid of smooth muscle, forming a three-dimensional network within the brain parenchyma. This system exhibits a characteristic histological organization, made of tight junctions between the adjacent endothelial cells and a basement membrane. This specific morphology confers part of the function of the blood–brain barrier, which is to control the passage of water, nutrients and other molecules such as certain hormones between the systemic circulation and the brain (for review, see Kalaria, 1996; Farkas & Luiten, 2001). Numerous studies have investigated the effect of ageing on the cerebrovascular system, and its functional consequences. However, most of them have been performed in animal models, and few data are available from humans. Nevertheless, some characteristic anatomical modifications affecting both the macro- and the microvasculature have been described in the ageing human brain. They could contribute to the impairment of the hemodynamic parameters, such as a decrease of both cerebral blood and flow velocity, resulting in a modification of the brain perfusion. Consequently, such functional alterations may be responsible to some degree for the deficits in cognitive abilities observed in the elderly (Farkas & Luiten, 2001; de la Torre, 2000). The vertebral and basilar arteries have been known to be already rather rigid at birth, and to become stiffer with ageing (Nagasawa et al., 1979). Modifications of elastin and collagen, the two histological components contributing to the passive elastic properties of a blood vessel, have been described that may contribute to the marked thickening of the vessel walls in old subjects. As a consequence, vessel walls are less distensible in old compared to young subjects (Nagasawa et al., 1979). Similar morphological changes have also been described in smaller arteries (Nagasawa

et al., 1979). A thickening of the basement membrane has also been shown to be a characteristic feature of aged cerebral arteries. The capillary density and the ultrastructure of the capillary walls are both subject to modifications in the ageing brain. A decreased capillary density has been described in the human hypothalamus with age (Abernethy et al., 1993), as well as in both cortical and hippocampal regions in old rats (24 month-old and older), compared to adult rats (12 month-old) (Amenta et al., 1995a,b). By treating the old rats with a calcium channel blocker for 6 months, Amenta and collaborators were able to counter the ageing-related microanatomical modifications occurring in the brain. They demonstrated an increase in the number and average length of alkaline-phosphatase-reactive capillaries, as well as an increase of the capillary density, as reflected by the reduction of the intercapillary distance. Moreover, Sonntag and colleagues have shown that the vascular density on the surface of the cortex is correlated with the plasma levels of insulin-like growth factor 1. They also demonstrated an influence of growth hormone on the cortical vascular density in older animals. Finally, the density of the microvascular system in ageing might be under the influence of the calcium homeostasis as well as circulating hormones (Amenta et al., 1995a,b; Sonntag et al., 1997). The modified ultrastructure of the capillary walls has been well characterized in ageing rats, whereas it has not been studied extensively in human because of the rapid degeneration of the capillary network that takes place after death. Nonetheless, it has been shown that a substantial number of brain capillaries exhibit altered morphologies during ageing and that this phenomenon is enhanced further in AD (Buée et al., 1994). In particular, the formation of collapsed capillaries and precapillaries (string vessels; Fig. 7.6), and coiling or twisting of small arterioles appears to be a typical sign of ageing of the cerebral vasculature. These changes also frequently occur in close association with amyloid deposition, particularly in the primary visual cortex (Fig. 7.6). In addition, an extensive loss of blood vessels that parallels cortical shrinkage accompanies the progression and severity of AD (Buée et al., 1994). The presence of perivascular collagen deposits (microvascular fibrosis) and the modification of the basement membrane, known as basement membrane thickening, have been shown in the motor cortex, the entorhinal cortex and the CA1 field of the hippocampus in old rats, as well as in the CA1 and CA3 fields of the hippocampus of ageing rhesus monkeys (Keuker et al., 2000). Alterations of the basement membrane occur as localized thickening, splitting or duplication (Farkas et al., 2000). The exact biochemical correlates


of basement membrane thickening are still unclear, and the involvement of all of the different histological components of the basement membrane (collagen, proteoglycans, laminin, fibronectin) has been evoked. Ultimately, this morphological alteration of the microvasculature in the ageing brain and its functional consequence, the perturbation of the blood–brain barrier function, are likely to affect cognitive function. The cellular components of the microvascular system, namely pericytes and endothelial cells, can also be affected by the cerebral ageing process. A loss of pericytes has been described in the ageing brain (Stewart et al., 1987), as well as more subtle intracellular modifications such as increased number of inclusions and enlargement of mitochondria. Interestingly, comparable changes have been observed in the subcortical white matter of very old macaque monkeys and were shown to correlate with the degree of cognitive impairment in these animals (Peters, 1999). However, the functional significance of these features in context of the normal ageing process remains uncertain, as they could reflect an activation of the immune function of the pericytes. Finally, an elongation and a loss of endothelial cells are classically accepted as age-related phenomena. However, a disruption of the blood–brain barrier due to a modification of the endothelial layer has not been classically reported in the normal aged brain.

Evidence for cortical disconnection The ‘cortical disconnection’ theory was first evoked in 1965 to explain some of the cognitive disturbances described in elderly persons (Geschwind, 1965). Morphometric studies have described an atrophy of the white matter in the normal ageing brain, which is probably caused by a loss of myelinated fibres of small diameter (Meier-Ruge et al., 1992; Tang et al., 1997). More recently, magnetic resonance imaging (MRI) has lent considerable support to the ‘cortical disconnection’ hypothesis (Esposito et al., 1999; O’Sullivan et al., 2001). By using positron emission tomography in a group of healthy volunteers aged from 18 to 80 years, Esposito and collaborators have investigated variations in absolute regional cerebral blood flow during two cognitive tasks involving the working memory in two different functional areas of the brain. Their study provided indirect evidence that the functional link between the left dorsolateral prefrontal cortex and the right hippocampus normally activated in young subjects during the two tasks might be altered in older subjects (Esposito et al., 1999). Recently, O’Sullivan and collaborators (2001) have used diffusion tensor imaging to investigate the cerebral white

Fig. 7.6. Pathology of the microvasculature in ageing. Note the association of the amyloid deposits with many capillaries ((a), (b), curved arrows) and the presence of pathological string vessels (b). The straight arrow points to a NFT. Scale Bar on (a) = 100 m and (b) 50 m.

matter and provide a correlation to the cognitive impairment occurring in normal ageing. Considering diffusional anisotropy of water molecules as a marker of white matter tract integrity, they compared a group of healthy, elderly volunteers with a group of younger subjects. Their findings are in favour of an alteration of the white matter ultrastructure that is more prominent in frontal regions, and which increases with age (O’Sullivan et al., 2001). Altogether, these data suggest that the early disruption of the white matter integrity in the ageing brain may be an important factor to explain the impairment of the cognitive function. In this context, furthermore, it is interesting to note that these white matter changes parallel progressive alterations in specific neocortical neuronal populations which may be at the origin of long corticocortical pathways and that are exquisitely vulnerable to NFT formation in AD (Hof et al., 1990; Hof & Morrison, 1990; Morrison & Hof, 1997).

Non-human primates as a model for human brain ageing Monkeys and great apes do not suffer from age-related neuropsychiatric disorders which occur only in humans, such as AD, and therefore provide a model in which the effects of ageing can be studied independently from the confounding effects of a concomitant dementia (Lacreuse et al., 1998). The vast majority of morphologic studies of brains of aged non-human primates indicate that very few changes take place during ageing (for review see Hof & Duan, 2001; Erwin et al., 2001; Hof et al., 2002b). Recent data on amyloid deposition, possible occurrence of AD-type neurofibrillary degeneration, and other neuropathological endpoints have provided valuable insight on etiopathogenesis of changes

121

122


that occur during ageing in the brain of old non-human primates. The frequent occurrence of amyloid protein in the brain of aged prosimians, monkeys, and great apes represents an interesting model of the dynamics of amyloid deposition in AD. In macaques, SP are present predominantly in prefrontal and primary somatosensory cortices, whereas lower densities of deposits are found in the hippocampus, insula, cingulate, motor, auditory, visual, temporal and parietal cortices and amygdala (Heilbroner & Kemper, 1990). This distribution differs considerably from that observed in AD patients. In fact, amyloid is seen consistently only in a few animals older than 25 years (Sloane et al., 1997), and the density of amyloid deposits is not correlated with the cognitive deficits in such animals. The earliest change associated with the formation of SP is the development of electron dense bodies in presynaptic terminals and dendrites and the accumulation of APP in astrocytes and microglial cells (Martin et al., 1994). In old monkeys, amyloid deposits containing non-fibrillar A have been documented in neurites, dendrites, and glial elements. Neuritic plaques and filamentous neuronal changes featuring 10 nm parallel helical filaments occur in old monkeys, although they are much rarer than in old human brains or in AD cases (Wisniewski et al., 1973). Some NFT have also been reported in the brains of cognitively impaired monkeys (Wisniewski et al., 1973). Filamentous hyperphosphorylated tau pathology, morphologically and biochemically comparable to the fibrillary lesions characteristic of a variety of human neurodegenerative disorders, have been described in aged baboons (Papio hamadryas cynocephalus), Campbell’s guenons (Cercopithecus campbelli), one brown lemur (Eulemur fulvus) and in a single albino rhesus monkey (Macaca mulatta) (Schultz et al., 1998, 2000a,b; Härtig et al., 2000; Kiatipattanasakul et al., 2000). These lesions occur in neurons, astrocytes and oligodendrocytes, preferentially in the hippocampus. They display, at least in a few baboons, an age-dependent progression in severity (Schultz et al., 1998; 2000a,b). The fact that these lesions are found not only in nerve cells but also in glial cells, extends the non-human primate spectrum of age-related pathologies to a larger group of human neurodegenerative diseases that differ from AD by their profile of tau proteins and characteristic lesion distributions (Buée et al., 2000). Moreover, recent data indicate that in certain species such as the prosimian grey mouse lemur (Microcebus murinus), numerous tau protein aggregates are found in the vicinity of amyloid deposits. Contrasting sharply with old humans, neocortical areas are frequently affected even in relatively young mouse lemurs, whereas the subiculum and entorhinal cortex are involved only occasionally in animals older than 8 years (Giannakopoulos et al., 1997b). This indicates

that tau-related pathology can exhibit species-specific patterns of regional and cellular distribution that do not correspond to the age-related human pathology. These species may represent useful models to study the relationships between the formation and distribution of tau and amyloid lesions in animals whose biology and behavior are well documented and that can be subjected to extensive cognitive testing (for review see Erwin et al., 2001; Gilissen et al., 2001; Hof & Duan, 2001). A substantial ultrastructural pathology has also been described in the ageing macaque monkey brain, particularly an age-related cortical and subcortical axonal pathology. In some myelinated fibres, dense oligodendrocytic materials accumulate and lead to splitting of the major dense line of the myelin sheaths (Wenk et al., 1991; Nielsen & Peters, 2000; Peters et al., 2000), while other fibres show doubling of the myelin sheaths with layers of compact myelin surrounding each other. Their consistent presence in aged macaque monkeys indicates a generalized agerelated dysfunction of oligodendrocytes. In addition, oligodendrocytes commonly are seen as clusters or arranged in rows in old animals, whereas they occur singly in young animals (Peters, 1996). These observations suggest abnormal oligodendrocytic networks, which could interfere with the function of the blood–brain barrier in old monkeys. In fact, oligodendrocytes have been observed in close contact with brain microvessels, where they modify the limiting glial membrane normally formed by astrocytes. These results point to the fact that glial cells may represent a crucial target of normal ageing in the primate brain, even though their numbers does not change in old animals (Peters et al., 1994). Importantly, all current studies agree that neuronal loss does not occur in the cerebral cortex and in subcortical structures in non-human primates. Stereological studies of the total numbers of cells in the entorhinal cortex and the CA1 field of the hippocampus in macaque monkeys and in great apes do not reveal any change in old individuals compared to young subjects (Gazzaley et al., 1997; Erwin et al., 2001; Perl et al., 2000) (Fig. 7.7). It is, however, important to relate the changes observed in the white matter in these animals to subtle alterations in neuronal morphology, that may have considerable repercussions on the function of certain cortical circuits. They may be related directly to deficits in visual and spatial recognition tasks commonly seen in aged animals (Peters et al., 1994; Peters, 1999). Changes in the morphology of the terminal dendrites of pyramidal neurons occur during ageing, and stereologic estimates have revealed region-specific reduction of the numerical densities of synapses per unit volume in old monkeys (Peters et al., 1998). These data suggest that ageing involves substantial damage


Fig. 7.7. Stereologic analysis of synapse densities in layer I of prefrontal cortex area 46 (a) and the dentate gyrus outer molecular layer (b) of old macaque monkeys compared to young animals. There is a decrease in the density of synapses in the prefrontal cortex, but they are unchanged in the dentate gyrus of aged animals (data modified from Peters et al., 1998, and Tigges et al., 1996). The left panel on (c) shows layer II of the entorhinal cortex in a young adult rhesus monkey, and the right panel in an old animal. No cell loss is observed histologically as well as quantitatively in old monkeys (d ). Quantitative data are shown as means ± S.D. Scale bar on (c) = 100 m.

of terminal dendritic arbours of neocortical neurons (Tigges et al., 1995; Peters et al., 1998). A significant impoverishment in the complexity of apical and basal dendritic trees of neurons in the prefrontal cortex, as well as a decrease in the spine densities at all levels of the dendritic arbors occur (Figs. 7.7, 7.8) (Hof & Duan, 2001; Duan et al., 2003). Also of note, NMDAR1 receptor expression in aged macaques, detected by quantitative immunohistochemistry, decreases specifically in the dentate gyrus in absence of neuronal loss in layer II of the entorhinal cortex, the site of origin of the projection to the dentate gyrus (Gazzaley et al., 1996, 1997). This is consistent with minor age-related changes in synapse counts reported in the dentate gyrus (Tigges et al., 1995), and with receptor binding assays in aged monkeys (Rosene & Nicholson, 1999). Similarly, a considerable downregulation in the expression of the glutamate receptor subunits NMDAR1 and AMPA GluR2 has been demonstrated in old monkeys in corticocortical projection

neurons identified by tract-tracing, with 20–40% fewer neurons expressing these subunit proteins in old animals (Hof et al., 2002a). Interestingly, the level of downregulation was more marked in neurons participating in long corticocortical pathways, which are prone particularly to neurofibrillary degeneration in AD (Morrison & Hof, 1997). Such changes imply that a reduction in synaptic function is likely to occur during normal brain ageing in primates and may represent a fundamental mechanism leading to the mild cognitive impairment observed not only in experimental animals but in elderly patients as well.

Conclusions Even in elderly subjects with clinically intact intellectual functions, the AD-related neuropathologic changes are a common finding. In particular, layer II of the entorhinal cortex is always affected by neurofibrillary degeneration.

123

124


Fig. 7.8. Three-dimensional reconstruction of neurons furnishing identified projections from the superior temporal cortex to prefrontal area 46 in a young (a) and an old (b) macaque monkeys. Note the less complex dendritic arborization patterns in the old animal. Spines densities are also decreased in the old animal (d) compared to the young one (c).

Severe involvement of neocortical association areas results in degeneration of corticocortical circuits and is required for the clinical expression of the dementia. Thus, the key to dementia may reside in the mechanisms by which pathological changes that are limited relatively to the hippocampal formation at early stages of the disease begin to involve neocortical circuits (Hof et al., 2003; Bussière et al., 2003a,b; Morrison & Hof, 1997), thereby differentiating the memory defects associated with normal ageing (also known as ‘benign senescent forgetfulness’), which may be related to pathological alterations restricted to the hippocampal formation, from the more generalized memory and cognitive deficits characterizing dementing disorders. In any case, the neuropathological evidence from human brains and non-human primate models suggests that, in addition to the topographically restricted accumulation of NFT and SP, the presence of a rather subtle pathology, affecting not neurons solely but the white matter and the microvascular system as well, is likely to contribute to the mild cognitive changes that are typical of ageing. Importantly, these observations imply that elderly individuals can maintain a

high level of cognitive performance while sustaining significant compromise of hippocampal circuits, and that they may rely more on neocortical than on hippocampal circuits for memories essential for daily activities. It is worth mentioning that, in terms of cognitive performance, healthy elders may describe difficulties in learning and retrieving new information although in reality, the problem lies in the amount of information that one such individual can learn within a given period of time in comparison to younger subjects. The major difference with early AD patients is that after a delay, healthy elders will have retained the new information, whereas patients with very mild impairment retain little of the new information. In contrast, AD is characterized by the involvement of neocortical circuits, leading to a global neocortical disconnection syndrome that presents clinically as dementia (Hof et al., 1990; Morrison & Hof, 1997). This interpretation of the pathological features of ageing and AD suggest that dementia results from changes in association neocortex, whereas extensive hippocampal alterations can exist in absence of neocortical involvement and with only minor disruptions in activities of daily living.


Acknowledgements We thank Drs C. Bouras, P. Giannakopoulos, J. M. Erwin, L. Buée, S. L. Wearne, and H. Duan for their contributions to our research, and W. G. M. Janssen and A. P. Leonard for expert technical assistance. Supported by grants from the NIH (AG02219, AG05138, and AG06647), and by the Brookdale Foundation, The Philippe Foundation, the Howard Hughes Medical Institute, and Mount Sinai School of Medicine.

REFERENCES The National Institute on Aging, and Reagan Institute Working Group on Diagnostic Criteria for the Neuropathological Assessment of Alzheimer’s Disease (1997). Consensus recommendations for the postmortem diagnosis of Alzheimer’s disease. Neurobiol Aging, 18, S1–S2. Abernethy, W. B., Bell, M. A., Morris, M. & Moody, D. M. (1993). Microvascular density of the human paraventricular nucleus decreases with ageing but not hypertension. Exp. Neurol., 121, 270–4. Amenta, F., Cavallotti, D., Del Valle, M. et al. (1995a). Age-related changes in brain microanatomy: sensitivity to treatment with the dihydropyridine calcium channel blocker darodipine (PY 108-068). Brain Res. Bull., 36, 453–60. Amenta, F., Ferrante, F., Mancini, M., Sabbatini, M., Vega, J. A. & Zaccheo, D. (1995b). Effect of long-term treatment with the dihydropyridine-type calcium channel blocker darodipine (PY 108-068) on the cerebral capillary network in aged rats. Mech. Ageing Dev., 78, 27–37. Arnold, S. E., Hyman, B. T., Flory, J., Damasio, A. R. & Van Hoesen, G. W. (1991). The topographical and neuroanatomical distribution of neurofibrillary tangles and neuritic plaques in the cerebral cortex of patients with Alzheimer’s disease. Cereb. Cortex, 1, 103–16. Arriagada, P. V., Growdon, J. H., Hedley-Whyte, E. T. & Hyman, B. T. (1992a). Neurofibrillary tangles but not senile plaques parallel duration and severity of Alzheimer’s disease. Neurology, 42, 631–9. Arriagada, P. V., Marzloff, K. & Hyman, B. T. (1992b). Distribution of Alzheimer-type pathologic changes in nondemented elderly individuals matches the pattern in Alzheimer’s disease. Neurology, 42, 1681–8. Bierer, L. M., Hof, P. R., Purohit, D. P., Carlin, L., Schmeidler, J., Davis, K. L. & Perl, D. P. (1995). Neocortical neurofibrillary tangles correlate with dementia severity in Alzheimer’s disease. Arch. Neurol., 52, 81–8. Bouras, C., Hof, P. R. & Morrison, J. H. (1993). Neurofibrillary tangle densities in the hippocampal formation in a non-demented population define subgroups of patients with differential early pathologic changes. Neurosci, Lett., 153, 131–5. Bouras, C., Hof, P. R., Giannakopoulos, P., Michel, J. P. & Morrison, J. H. (1994). Regional distribution of neurofibrillary tangles and

senile plaques in the cerebral cortex of elderly patients: a quantitative evaluation of a one-year autopsy population from a geriatric hospital. Cereb. Cortex, 4, 138–50. Bouras, C., Vallet, P. G., Kövari, E. et al. (2001). Longevity and brain ageing: the paradigm of centenarians. In Functional Neurobiology of Aging, ed. P. R. Hof & C. V. Mobbs, pp. 82–90. San Diego: Academic Press. Braak, H. & Braak, E. (1990). Neurofibrillary changes confined to the entorhinal region and an abundance of cortical amyloid in cases of presenile and senile dementia. Acta Neuropathol., 80, 479–86. Braak, H., Braak, E. & Kalus, P. (1989). Alzheimer’s disease: areal and laminar pathology in the occipital isocortex. Acta Neuropathol., 77, 494–506. Buée, L., Hof, P. R., Bouras, C. et al. (1994). Pathological alterations of the cerebral microvasculature in Alzheimer’s disease and related dementing disorders. Acta Neuropathol, 87, 469– 80. Buée, L., Bussière, T., Buée-Scherrer, V., Delacourte, A. & Hof, P. R. (2000). Tau protein isoforms, phosphorylation and role in neurodegenerative disorders. Brain Res. Rev., 33, 95–130. Bussière, T., Gold, G., Kövari, E. et al. (2003a). Stereologic analysis of neurofibrillary tangle formation in prefrontal cortex area 9 in aging and Alzheimer’s disease. Neuroscience, 117, 577–92. Bussie`re, T., Giannakopoulos, P., Bouras, C. et al. (2003b). Progressive degeneration of nonphosphorylated neurofilament protein-enriched pyramidal neurons predicts cognitive impairment in Alzheimer’s disease: a stereologic analysis of prefrontal cortex area 9. J. Comp. Neurol., 463, 281–302. Bussière, T., Friend, P. D., Sadeghi, N. et al. (2002). Stereologic assessment of the total cortical volume occupied by amyloid deposits and its relationship with cognitive status in aging and Alzheimer’s disease. Neuroscience, 112, 75–91. Crystal, H., Dickson, D., Fuld, P. et al. (1988). Clinico-pathologic studies in dementia: nondemented subjects with pathologically confirmed Alzheimer’s disease. Neurology, 38, 1682–7. de la Torre, J. C. (2000). Impaired cerebromicrovascular perfusion. Summary of evidence in support of its causality in Alzheimer’s disease. Ann. N.Y. Acad. Sci., 924, 136–52. Dickson, D. W., Crystal, H. A., Mattiace, L. A. et al. (1992). Identification of normal and pathological ageing in prospectively studied nondemented elderly humans. Neurobiol. Aging, 13, 179–89. Duan, H., Wearne, S. L., Rocher, A. B. et al. (2003). Age-related morphologic alterations in dendrites and spine densities of corticocortically projecting neurons in macaque monkeys. Cereb. Cortex, 13, 950–61. Erwin, J. M., Nimchinsky, E. A., Gannon, P. J., Perl, D. P. & Hof, P. R. (2001). The study of brain ageing in great apes. In Functional Neurobiology of Aging, ed. P. R. Hof & C. V. Mobbs, pp. 447–56. San Diego: Academic Press. Esposito, G., Kirkby, B. S., Van Horn, J. D., Ellmore, T. M. & Berman, K. F. (1999). Context-dependent, neural system-specific neurophysiological concomitants of ageing: mapping PET correlates during cognitive activation. Brain, 122, 963–79.

125

126


Farkas, E. & Luiten, P. G. (2001). Cerebral microvascular pathology in ageing and Alzheimer’s disease. Prog. Neurobiol., 64, 575– 611. Farkas, E., De Jong, G. I., de Vos, R. A., Jansen Steur, E. N. & Luiten, P. G. (2000). Pathological features of cerebral cortical capillaries are doubled in Alzheimer’s disease and Parkinson’s disease. Acta Neuropathol., 100, 395–402. Gazzaley, A. H., Siegel, S. J., Kordower, J. H., Mufson, E. J. & Morrison, J. H. (1996). Circuit-specific alterations of N-methyl-D-aspartate subunit 1 in the dentate gyrus of aged monkeys. Proc. Natl Acad. Sci. USA, 93, 3121–5. Gazzaley, A. H., Thakker, M. M., Hof, P. R. & Morrison, J. H. (1997). Preserved number of entorhinal cortex layer II neurons in aged macaque monkeys. Neurobiol. Aging, 18, 549–53. Geschwind, N. (1965). Disconnexion syndromes in animals and man. I. Brain, 88, 237–94. Giannakopoulos, P., Hof, P. R., Giannakopoulos, A. S. et al. (1994a). Dementia in the oldest-old: quantitative analysis of 12 cases from a psychiatric hospital. Dementia, 5, 348–56. Giannakopoulos, P., Hof, P. R., Mottier, S., Michel, J. P. & Bouras, C. (1994b). Neuropathological changes in the cerebral cortex of 1258 cases from a geriatric hospital: retrospective clinicopathological evaluation of a 10-year autopsy population. Acta Neuropathol., 87, 456–68. Giannakopoulos, P., Hof, P. R., Kövari, E., Vallet, P. G., Herrmann, F. R. & Bouras, C. (1996). Distinct patterns of neuronal loss and Alzheimer’s disease lesion distribution in elderly individuals older than 90 years. J. Neuropathol. Exp. Neurol., 55, 1210–20. Giannakopoulos, P., Hof, P. R., Michel, J. P., Guimon, J. & Bouras, C. (1997a). Cerebral cortex pathology in ageing and Alzheimer’s disease: a quantitative survey of large hospital-based geriatric and psychiatric cohorts. Brain Res. Rev., 25, 217–45. Giannakopoulos, P., Silhol, S., Jallageas, V. et al. (1997b). Quantitative analysis of tau protein-immunoreactive accumulations and beta amyloid protein deposits in the cerebral cortex of the mouse lemur, Microcebus murinus. Acta Neuropathol., 94, 131–9. Gilissen, E. P., Dhenain, M. & Allman, J. M. (2001). Brain ageing in strepsirhine primates. In Functional Neurobiology of Aging, ed. P. R. Hof & C. V. Mobbs, pp. 421–33. San Diego: Academic Press. Gómez-Isla, T., Price, J. L., McKeel, Jr., D. W., Morris, J. C., Growdon, J. H. & Hyman, B. T. (1996). Profound loss of layer II entorhinal cortex neurons occurs in very mild Alzheimer’s disease. J. Neurosci., 16, 4491–500. Härtig, W., Klein, C., Brauer, K. et al. (2000). Abnormally phosphorylated protein tau in the cortex of aged individuals of various mammalian orders. Acta Neuropathol., 100, 305–12. Heilbroner, P. L. & Kemper, T. L. (1990). The cytoarchitectonic distribution of senile plaques in three aged monkeys. Acta Neuropathol., 81, 60–5. Hof, P. R. & Duan, H. (2001). Age-related morphologic alterations in the brain of Old World and New World anthropoid monkeys. In Functional Neurobiology of Aging, ed. P. R. Hof & C. V. Mobbs, pp. 435–46, San Diego: Academic Press.

Hof, P. R. & Morrison, J. H. (1990). Quantitative analysis of a vulnerable subset of pyramidal neurons in Alzheimer’s disease: II. Primary and secondary visual cortex. J. Comp. Neurol., 301, 55–64. Hof, P. R., Cox, K. & Morrison, J. H. (1990). Quantitative analysis of a vulnerable subset of pyramidal neurons in Alzheimer’s disease: I. Superior frontal and inferior temporal cortex. J. Comp. Neurol., 301, 44–54. Hof, P. R., Bierer, L. M., Perl, D. P. et al. (1992). Evidence for early vulnerability of the medial and inferior aspects of the temporal lobe in an 82-year-old patient with preclinical signs of dementia. Regional and laminar distribution of neurofibrillary tangles and senile plaques. Arch. Neurol., 49, 946–53. Hof, P. R., Bussière, T., Gold, G. et al. (2003). Stereologic evidence for persistence of viable neurons in layer II of the entorhinal cortex and the CA1 field in Alzheimer disease. J. Neuropathol. Exp. Neurol., 62, 55–67. Hof, P. R., Duan, H., Page, T. L. et al. (2002a). Age-related changes in GluR2 and NMDAR1 glutamate receptor subunit protein immunoreactivity in corticocortically projecting neurons in macaque and patas monkeys. Brain Res., 928, 175–86. Hof, P. R., Gilissen, E. P., Sherwood, C. C. et al. (2002b). Comparative neuropathology of brain ageing in primates. In Aging in Nonhuman Primates, Interdisciplinary Topics in Gerontology Vol. 31, ed. J. M. Erwin & P. R. Hof. Basel, Karger, pp. 130–54. Kalaria, R. N. (1996). Cerebral vessels in ageing and Alzheimer’s disease. Pharmacol. Ther., 72, 193–214. Keuker, J. I., Luiten, P. G. & Fuchs, E. (2000). Capillary changes in hippocampal CA1 and CA3 areas of the ageing rhesus monkey. Acta Neuropathol., 100, 665–72. Kiatipattanasakul, W., Nakayama, H., Yongsiri, S. et al. (2000). Abnormal neuronal and glial argyrophilic fibrillary structures in the brain of an aged albino cynomolgus monkey (Macaca fascicularis). Acta Neuropathol., 100, 580–6. Kordower, J. H., Chu, Y., Stebbins, G. T. et al. (2001). Loss and atrophy of layer II entorhinal cortex neurons in elderly people with mild cognitive impairment. Ann. Neurol., 49, 202–13. Lacreuse, A., Herndon, J. & Moss, M. (1998). Vieillissement des fonctions cognitives chez l’homme et le macaque rhésus (Macaca mulatta). Primatologie, 1, 333–77. Martin, L. J., Pardo, C. A., Cork, L. C. & Price, D. L. (1994). Synaptic pathology and glial responses to neuronal injury precede the formation of senile plaques and amyloid deposits in the ageing cerebral cortex. Am. J. Pathol., 145, 1358–81. Meier-Ruge, W., Ulrich, J., Bruhlmann, M. & Meier, E. (1992). Agerelated white matter atrophy in the human brain. Ann. N.Y. Acad. Sci., 673, 260–9. Mirra, S. S., Hart, M. N. & Terry, R. D. (1993). Making the diagnosis of Alzheimer’s disease. A primer for practicing pathologists. Arch. Pathol. Lab. Med., 117, 132–44. Morrison, J. H. & Hof, P. R. (1997). Life and death of neurons in the ageing brain. Science, 278, 412–19. Nagasawa, S., Handa, H., Okumura, A., Naruo, Y., Moritake, K. & Hayashi, K. (1979). Mechanical properties of human cerebral


arteries. Part 1, Effects of age and vascular smooth muscle activation. Surg. Neurol., 12, 297–304. Näslund, J., Haroutunian, V., Mohs, R. et al. (2000). Correlation between elevated levels of amyloid beta-peptide in the brain and cognitive decline. J. Am. Med. Assoc., 283, 1571–7. Nielsen, K. & Peters, A. (2000). The effects of ageing on the frequency of nerve fibers in rhesus monkey striate cortex. Neurobiol. Aging, 21, 621–8. O’Sullivan, M., Jones, D. K., Summers, P. E., Morris, R. G., Williams, S. C. & Markus, H. S. (2001). Evidence for cortical ‘disconnection’ as a mechanism of age-related cognitive decline. Neurology, 57, 632–8. Parvathy, S., Davies, P., Haroutunian, V. et al. (2001). Correlation between Ax-40-, Ax-42-, and Ax-43-containing amyloid plaques and cognitive decline. Arch. Neurol., 58, 2025–32. Perl, D. P., Good, P. F., Bussière, T., Morrison, J. H., Erwin, J. M. & Hof, P. R. (2000). Practical approaches to stereology in the setting of ageing- and disease-related brain banks. J. Chem. Neuroanat., 20, 7–19. Peters, A. (1996). Age-related changes in oligodendrocytes in monkey cerebral cortex. J. Comp. Neurol., 371, 153–63. Peters, A. (1999). Normal ageing in the cerebral cortex of primates. In Cerebral Cortex, Vol. 14, ed. A. Peters & J. H. Morrison, pp. 49–80. New York: Kluwer Academic-Plenum. Peters, A., Leahu, D., Moss, M. B. & McNally, K. J. (1994). The effects of ageing on area 46 of the frontal cortex of the rhesus monkey. Cereb. Cortex, 4, 621–35. Peters, A., Morrison, J. H., Rosene, D. L. & Hyman, B. T. (1998). Feature article: are neurons lost from the primate cerebral cortex during normal ageing? Cereb. Cortex, 8, 295–300. Peters, A., Moss, M. B. & Sethares, C. (2000). Effects of ageing on myelinated nerve fibers in monkey primary visual cortex. J. Comp. Neurol., 419, 364–76. Price, J. L. & Morris, J. C. (1999). Tangles and plaques in nondemented ageing and ‘preclinical’ Alzheimer’s disease. Ann. Neurol., 45, 358–68. Price, J. L., Davis, P. B., Morris, J. C. & White, D. L. (1991). The distribution of tangles, plaques and related immunohistochemical markers in healthy ageing and Alzheimer’s disease. Neurobiol. Aging, 12, 295–312. Price, J. L., Ko, A. I., Wade, M. J., Tsou, S. K., McKeel, D. W. & Morris, J. C. (2001). Neuron number in the entorhinal cortex and CA1 in preclinical Alzheimer disease. Arch. Neurol., 58, 1395–402.

Rogers, J. & Morrison, J. H. (1985). Quantitative morphology and regional and laminar distributions of senile plaques in Alzheimer’s disease. J. Neurosci., 5, 2801–8. Rosene, D. & Nicholson, T. (1999). Neurotransmitter receptor changes in the hippocampus and cerebral cortex in normal ageing. In Cerebral Cortex, Vol. 14, ed. A. Peters & J. H. Morrison, pp. 111–28, New York: Kluwer Academic-Plenum. Schultz, C., Braak, E. & Braak, H. (1998). A nonhuman primate model for tau-positive cytoskeletal pathology affecting neurons, astrocytes, and oligodedrocytes. Clin. Neuropathol., 17, 251. Schultz, C., Dehghani, F., Hubbard, G. B. et al. (2000a). Filamentous tau pathology in nerve cells, astrocytes, and oligodendrocytes of aged baboons. J. Neuropathol. Exp. Neurol., 59, 39–52. ¨ U., Braak, E. & Braak, H. (2000b). Schultz, C., Hubbard, G. B., Rub, Age-related progression of tau pathology in brains of baboons. Neurobiol. Aging, 21, 905–12. Selkoe, D. J. (2001). Alzheimer’s disease: genes, proteins, and therapy. Physiol. Rev., 81, 741–66. Sloane, J. A., Pietropaolo, M. F., Rosene, D. L. et al. (1997). Lack of correlation between plaque burden and cognition in the aged monkey. Acta Neuropathol., 94, 471–8. Sonntag, W. E., Lynch, C. D., Cooney, P. T. & Hutchins, P. M. (1997). Decreases in cerebral microvasculature with age are associated with the decline in growth hormone and insulin-like growth factor 1. Endocrinology, 138, 3515–20. Stewart, P. A., Magliocco, M., Hayakawa, K. et al. (1987). A quantitative analysis of blood-brain barrier ultrastructure in the ageing human. Microvasc. Res. 33, 270–82. Tang, Y., Nyengaard, J. R., Pakkenberg, B. & Gundersen, H. J. (1997). Age-induced white matter changes in the human brain: a stereological investigation. Neurobiol. Aging, 18, 609–15. Tigges, J., Herndon, J. G. & Rosene, D. L. (1995). Mild age-related changes in the dentate gyrus of adult rhesus monkeys. Acta Anat., 153, 39–48. Wenk, G. L., Walker, L. C., Price, D. L. & Cork, L. C. (1991). Loss of NMDA, but not GABA-A, binding in the brains of aged rats and monkeys. Neurobiol. Aging, 12, 93–8. West, M. J., Coleman, P. D., Flood, D. G. & Troncoso, J. C. (1994). Differences in the pattern of hippocampal neuronal loss in normal ageing and Alzheimer’s disease. Lancet, 344, 769–72. Wisniewski, H. M., Ghetti, B. & Terry, R. D. (1973). Neuritic (senile) plaques and filamentous changes in aged rhesus monkeys. J. Neuropathol. Exp. Neurol., 32, 566–84.

127

8 Neuroimaging Alzheimer’s disease Arthur W. Toga, Michael S. Mega, and Paul M. Thompson Laboratory of Neuro imaging, UCLA School of Medicine, Los Angeles, CA, USA

Challenges in population-based brain mapping Imaging studies of clinical populations continue to uncover new patterns of altered structure and function, and novel algorithms are being applied to relate these patterns to cognitive and genetic parameters. Post-mortem brain maps are also beginning to clarify the molecular substrates of disease. As imaging studies expand into ever-larger patient populations, population-based brain atlases (Mazziotta et al., 1995; Thompson et al., 2000a,b) offer a powerful framework to synthesize results from disparate imaging studies. These atlases use novel analytical tools to fuse data across subjects, modalities, and time. They detect group-specific features not apparent in individual patients’ scans. Once built, these atlases can be stratified into subpopulations to reflect a particular clinical group, such as individuals at genetic risk for AD, patients with mild cognitive impairment (MCI) or different dementia subtypes (frontotemporal dementia/semantic dementia), or patients undergoing different drug treatments. The disease-specific features these atlases resolve can then be linked with demographic factors such as age, gender, handedness, as well as specific clinical or genetic parameters (Mazziotta et al., 1995; Toga & Mazziotta, 1996; Thompson et al., 2001a–e). New brain atlases are also being built to incorporate dynamic data (Thompson et al., 2002). Despite the significant challenges in expanding the atlas concept to the time dimension, dynamic brain atlases are beginning to include probabilistic information on growth rates that may assist research into pediatric disorders (Thompson et al., 2000a,b) as well as revealing patterns of degenerative rates in Alzheimer’s disease (Fox et al., 1996; Thompson et al.,

2001a–e, 2002; Chan et al., 2001). Imaging algorithms are also significantly improving the flexibility of digital brain templates. Deformable brain atlases are adaptable brain templates that can be individualized to reflect the anatomy of new subjects. These atlases may be used for automated parcellation of new brain scans (Collins et al., 1995; Iosifescu et al., 1997), to define regions of interest in functional and metabolic studies (Dinov et al., 2000), and for anatomical shape assessment (Thompson et al., 1997; Csernansky et al., 2000). Probabilistic atlases are research tools that retain information on cross-subject variations in brain structure and function. These atlases are powerful new tools with broad clinical and research applications (Mazziotta et al., 1995, 2001; Kikinis et al., 1996; Toga & Thompson, 1998b; Thompson et al., 2001a–e, 2002).

Disease-specific atlases This chapter introduces the topic of a disease-specific brain atlas (Fig. 8.1). This type of atlas is designed to reflect the unique anatomy and physiology of a particular clinical subpopulation (Thompson et al., 1997, 2001a–e; Mega et al., 1997, 1999, 2000a–c; Narr et al., 2001a; Cannon et al., 2002). Based on well-characterized patient groups, these atlases contain thousands of structure models, as well as composite maps, average templates, and visualizations of structural variability, asymmetry and group-specific differences. They act as a quantitative framework that correlates the structural, metabolic, molecular and histologic hallmarks of the disease (Mega et al., 1997, 2000a–c). Additional algorithms use information stored in the atlas to recognize anomalies and label structures in new patients. Because they retain information on group anatomical variability,


128

Cambridge Books Online © Cambridge University Press, 2009

Neuroimaging Alzheimer’s disease

Fig. 8.1. Elements of a disease-specific atlas. This schematic shows the types of maps and models contained in a disease-specific brain atlas (Thompson et al., 2000, 2002a,b; Mega et al., 2000a–c). A diverse range of computational anatomical tools are required to generate these average brain image templates (continuum-mechanical atlas), models and maps. Disease-specific brain atlases, such as this one based on patients with Alzheimer’s disease (AD), allow imaging data from diverse modalities to be compared and correlated in a common 3D coordinate space. 3D anatomical models (e.g. cortical surfaces, bottom row), were extracted from a database of structural MRI data from AD patients. Models of these and other structures were digitally averaged and used to synthesize an average brain template (Continuum-Mechanical Atlas, middle) with well-resolved anatomical features in the mean shape and size for the population (see text for details). By rotating and scaling new images to occupy the same space as this template, models of subcortical, ventricular and deep nuclear structures can be built (lower left). Average models for patients and controls can then be used to compute average patterns and statistics of cortical variability and asymmetry (top left), to chart average profiles of grey matter loss in a group, and to detect atrophy in a group or individual (probability maps; left column). Mega et al. (1997, 1999) also fused histologic maps of post mortem neurofibrillary tangle (NFT) staining density, biochemical maps of beta-amyloid distribution, and 3D metabolic FDG-PET data obtained 8 hours before death, in the same patient with AD (top middle panels). By classifying grey and white matter (tissue classification) and unfolding the topography of the hippocampus (right panels), Zeineh et al. (2001) revealed the fine-scale anatomy and dynamics of brain activation during memory tasks, using high-resolution functional MRI (time course shown for activation in right parahippocampal cortex, PHC). Atlasing techniques can represent and compare these diverse datasets in a common coordinate space, enabling novel multi-subject and cross-modality comparisons.

disease-specific atlases are a type of probabilistic atlas specialized to represent a particular clinical group. The resulting atlases can identify patterns of altered structure or function, and can guide algorithms for knowledge-based image analysis, automated image labelling (Collins et al., 1995; Pitiotet al., 2002), tissue classification(Zijdenbos &Dawant, 1994) and functional image analysis (Dinov et al., 2000). We present data from several on-going projects, whose goal is to create disease-specific maps and atlases of the

brain in Alzheimer’s disease. Since current brain templates, typically based on young normal subjects, poorly represent the anatomy of these clinical populations, the resulting atlases offer a promising framework to investigate the disease. Pathological change can be tracked over time, and disease-specific features resolved. Rather than simply fusing information from multiple subjects and sources, new mathematical strategies are introduced to resolve groupspecific features not apparent in individual scans.


129

130

A. W. Toga, Michael S. Mega and Paul M. Thompson

In AD, early neuronal loss occurs in the entorhinal, parahippocampal and temporo-parietal cortices, consistent with the spatial pattern of early perfusion deficits and metabolic change. These deficits mirror the time course of cognitive impairment, proceeding from the entorhinal, temporal and perisylvian association cortices into more anterior regions as the disease progresses. In principle, volumetric magnetic resonance imaging (MRI) scans have sufficient resolution and tissue contrast to track cortical grey matter loss in a living individual. Yet, gyral patterns are extremely variable across subjects, making it difficult to calibrate individual patterns of grey matter loss against a normative population. It is also hard to determine the average profile of early tissue loss in a group. If 3D profiles of grey matter could be compared, this could be useful (i) for early diagnosis and assessing disease modification in an individual or group and (ii) for understanding how cortical changes relate to the fundamental anatomy of the cortex. This chapter addresses these problems.

and epilepsy (Kikinis et al., 1994; Cook et al., 1994; Fuh et al., 1997), and diffuse cortical atrophy is typical of Alzheimer’s disease, Pick’s disease and other dementias (Schmidt, 1992). Gyral anomalies, such as cortical dysplasias, have also been linked with neurodevelopmental delay (Sobire et al., 1995). Nonetheless, ratings of structural change in the cortex are still largely based on qualitative assessment (Berg et al., 1993). To clarify how diseases affect the cortex, specialized approaches are described for averaging cortical anatomy across subjects. Gyral pattern matching (Thompson & Toga, 1996; Thompson et al., 2000a,b, 2001a–e) is used to create average cortical models, to measure cross-subject differences, and to encode the magnitude and principal directions of anatomical variation at the cortex. In the resulting cortical templates, subtle features emerge. Regional asymmetries appear that are not apparent in individual anatomies. Population-based maps of cortical variation reveal a mosaic of variability patterns that are characteristic of each cortical region.

Statistical brain templates Central to the construction of a disease-specific atlas is the creation of averages, templates and models to describe how the brain and its component parts are organized, and how they are altered in disease. Statistical models are created to reveal how major anatomic systems are affected, how the pathology progresses, and how these changes relate to demographic or genetic factors. To create templates that reflect the morphology of a diseased group, specialized strategies are required for population-based averaging of anatomy (Thompson et al., 1996a,b, 2002; Grenander & Miller, 1998). In one approach (Thompson et al., 1999), sets of high-dimensional elastic mappings, based on the principles of continuum mechanics, reconfigure the anatomy of a large number of subjects in an anatomical image database. These three-dimensional deformation fields are used to create a crisp anatomical image template to represent the brain in Alzheimer’s disease, with highly resolved structures in their mean spatial location. The mappings also generate a richly detailed local encoding of anatomic variability, with up to a billion parameters (Thompson & Toga, 1996; Grenander & Miller, 1998; Miller et al., 2002). The resulting variability parameters are stored as a tensor field (p. 150) and leveraged by pattern recognition strategies that automatically identify anatomical structures in new patients’ scans, and identify disease-specific characteristics (Thompson et al., 2001a–e; Pitiot et al., 2002).

Cortical patterns Cortical patterns are altered in a variety of diseases. Sulcal pattern anomalies have been identified in schizophrenia

Pathology detection Normal anatomic complexity makes it difficult to design automated strategies that detect abnormal brain structure. Considerable research has focused on uncovering specific patterns of anatomic alterations in Alzheimer’s disease and other dementias (Friedland & Luxenberg, 1988). At the same time, brain structure is so variable that group-specific patterns of anatomy and function are often obscured.

Demographic factors Reports of structural differences in the brain linked to gender and handedness are still controversial, and these factors may also interact with disease-specific abnormalities (Narr et al., 2001a,b; Thompson et al., 2002). Other factors that interfere with analysis include educational level, premorbid adjustment, treatment history and response, and the duration and course of illness (Carpenter et al., 1993). Interactions of these variables make it harder to detect diseasespecific patterns and relate them to clinical and genetic parameters (Laakso et al., 2000a,b). The importance of these linkages has propelled computational anatomy to the forefront of brain imaging investigations. To distinguish abnormalities from normal variants, a realistically complex mathematical framework is required to encode information on anatomic variability in homogeneous populations (Grenander & Miller, 1998; Thompson & Toga, 2002). As we shall see, elastic registration or warping algorithms offer substantial advantages in creating brain atlases that encode patterns of anatomic variation and detect pathology.



Dynamic brain atlases Brain atlases have traditionally relied on static representations of anatomy, but many of the diseases that affect the human brain are progressive (e.g. dementia, neoplastic tumours). The progression of a disease may also be modulated by therapy, ranging from drug treatment to surgery. In response to these challenges, dynamic brain atlases retain spatio-temporal information on patterns of neuroanatomic change. They offer a means to analyze the dynamics of disease. Later in the chapter, we describe how atlases can be expanded to incorporate quantitative (4D) maps of growth or degenerative patterns in the brain. These maps characterize local growth or atrophy rates in development or disease. Atlases that incorporate confidence limits on growth rates, in particular, offer a new type of normative framework to analyse aberrant brain degeneration and its modification by drug treatment or demographic factors (Thompson et al., 2001).

Types of brain atlases

Perhaps surprisingly, few atlases of neuropathology use a standardized three-dimensional coordinate system to integrate data across patients, techniques, and acquisitions. Digital templates placed in a well-defined coordinate space (Evans et al., 1991; Friston et al., 1995; Drury & Van Essen, 1997), together with algorithms to align data with them (Toga, 1998), have enabled the pooling of brain mapping data from multiple subjects and sources, including large patient populations. As we shall see, standardized coordinate systems also allow parameterized, anatomical models to be statistically combined (Thompson et al., 1996a,b). By combining anatomical models, the results of morphometric studies can be leveraged to create disease-specific brain templates. Automated algorithms can then capitalize on atlas descriptions of anatomical variance to guide image segmentation (Le Goualher, 1999; Pitiot et al., 2002), tissue classification (Zijdenbos and Dawant, 1994), functional image analysis (Dinov et al., 2000; Zeineh et al., 2001) and pathology detection (Thompson et al., 1997, 2001a–e).

Early anatomic templates

Coordinate systems Rapid progress has been made by research groups developing standardized three-dimensional brain atlases (Talairach & Tournoux, 1988; Greitz et al., 1991; Höhne et al., 1992; Thurfjell et al.., 1993; Kikinis et al., 1996; Nowinski et al., 2000; Mazziotta et al., 2001). While few of these atlases aim to represent anatomy and function in disease, several commercially-available atlases of pathology combine histologic data with illustrative metabolic or structural images. The Harvard Brain Atlas (Johnson, 1996) is a rich source of annotated CT, MRI, SPECT and PET (single photon/positron emission computed tomography) images from a number of clinical populations. Cerebrovascular, neoplastic and degenerative diseases are represented (including stroke, vascular dementia, and Alzheimer’s disease), as are inflammatory, autoimmune and infectious diseases (multiple sclerosis and AIDS). In a similar effort, the Atlas of Brain Perfusion SPECT has been produced by Brigham and Women’s Hospital (Holman et al., 1994). This atlas presents 21 SPECT images with co-registered scans (SPECT merged with CT or MRI), and all scans are annotated with relevant clinical information and case histories. Other collections focus on post mortem data. The On-line Neuropathology Atlas developed by ¨ & the University of Debrecen Medical School (Hegedus Molnár, 1996) includes labelled images of the normal brain, with an extensive collection of pathological images from patients with cerebrovascular disease, and neoplasms, as well as inflammatory and degenerative disorders.

Research on brain atlases was originally based on the premise that brain structure and function imaged in any modality can be better localized by correlation with higher resolution anatomic data placed in an appropriate spatial coordinate system. Because of their detailed characterization of anatomy, most early brain atlases were derived from one, or a few, post-mortem specimens (Brodmann, 1909; Van Buren & Maccubin, 1962; Talairach & Szikla, 1967; Van Buren & Borke, 1972; Schaltenbrand & Wahren, 1977; Matsui & Hirano, 1978; Talairach & Tournoux, 1988; Ono et al., 1990). These anatomical references typically represent a particular feature of the brain, such as a neurochemical distribution (Mansour et al., 1995), myelination patterns (Smith, 1907; Mai et al., 1997), or the cellular architecture of the cortex (Brodmann, 1909).

Multi-modality atlases Because of the superior anatomic resolution, several digital atlases have been created using cryosection imaging. This technique allows the serial collection of photographic images from a cryoplaned specimen blockface (Bohm et al., 1983; Greitz et al., 1991; Toga et al., 1994; Mega et al., 1997, 1999). Using 10242 , 24-bits/pixel digital colour cameras, cryosection imaging offers a spatial resolution as high as 100 microns/voxel for whole human head cadaver preparations, or higher for isolated brain regions (Toga et al., 1997). Unlike paper atlases, digital cryosection volumes are amenable to a variety of resampling and repositioning schemes. Structures can therefore be rendered and


131

132


Fig. 8.2. Elastic registration of brain maps and molecular assays. Post-mortem tissue sections, from patients with Alzheimer’s disease, are gridded (left) to produce tissue elements for biochemical assays. These assays provide detailed quantitative measures of the major hallmarks of AD, including beta-amyloid and synaptophysin density. To pool this data in a common coordinate space, tissue elements are elastically warped back into their original configuration in the cryosection blockface (middle panel). Image data acquired from the same patient in vivo can then be correlated with regional biochemistry (Mega et al., 1997). When tissue sections are warped to the blockface, continuum-mechanical models are used to make the deformations reflect how real physical tissues deform. The complexity of the required deformation vector field in a small tissue region (magnified vector map, right) demonstrates that very flexible, high-dimensional transformations are essential (Thompson & Toga, 1996; Schiemann et al., 1996; Christensen et al., 1996). These deformation vector fields project histologic and biochemical data back into their in vivo configuration, populating a growing Alzheimer’s disease atlas with maps of molecular content and histology.

visualized from any angle. In the Visible Human Project (Ackerman et al., 2001), two (male and female) cadavers were cryoplaned and imaged at 1.0 mm intervals (0.33 mm for the female data), and the entire bodies were also reconstructed via 5000 post-mortem CT and MRI images. The resulting digital datasets consist of over 15 gigabytes of image data. While not an atlas per se, the Visible Human data has served as the foundation for developing related atlases of regions of the cerebral cortex (Drury & Van Essen, 1997), and high-quality brain models and visualizations (Schiemann et al., 1996; Stewart et al., 1996). Using multimodality data from a patient with a localized pathology, and more recently the Visible Human data, Höhne and coworkers developed a commercially available brain atlas designed for teaching neuroanatomy (VOXEL-MAN; Höhne et al., 1990, 1992; Tiede et al., 1993).

Post-mortem data fusion Fusion of metabolic and functional images acquired in vivo with post-mortem biochemical maps provides a unique view of the relationship between brain function and pathology. Mega et al. (1997) scanned Alzheimer’s patients in the terminal stages of their disease using both MRI and PET. Using elastic registration techniques (Thompson et al., 1996a,b), these data were combined with post mortem histologic images showing the gross anatomy (Toga et al., 1994), a Gallyas stain of neurofibrillary tangles, and a vari-

ety of spatially indexed biochemical assays (Fig. 8.2). The resulting multimodality maps of the Alzheimer’s disease brain relate the anatomic and histopathologic underpinnings of the disease in a standardized coordinate space. These data are further correlated with in vivo metabolic and perfusion maps of this disease. The resulting maps are key components of a growing disease-specific atlas (Mega et al., 2000a,b,c).

Analysing brain data A driving force that made anatomical templates important in brain imaging was the need to perform brain to brain comparisons. Anatomic variations severely hamper the integration and comparison of data across subjects, and can lead to misleading results (Meltzer & Frost, 1994; Woods, 1996; Bookstein, 2001; Ashburner & Friston, 2001). Motivated by the need to standardize and pool data across subjects, and compare results across laboratories, several registration methods have been developed to align brain mapping data with an atlas. The simplest registration techniques are linear, removing global differences in brain size. Non-linear approaches, however, can eliminate even the most local size and shape differences that distinguish one brain from another. Transforming individual datasets into the shape of a single reference anatomy, or onto a 3D digital brain atlas, allows subsequent comparison of brain function across individuals (Christensen et al., 1993;



Fig. 8.3. Computing anatomical differences with a deformable brain atlas. Structure boundaries from a patient with Alzheimer’s disease (b), here imaged with 3D MRI, are overlaid on a cryosection atlas (a), which has been registered to it using a simple linear transformation. A surface-based image warping algorithm then drives the atlas into the configuration of the patient’s anatomy (c). When the atlas is deformed, there are two useful products. The first is a high-resolution anatomical template that is customized to reflect the individual’s anatomy, and the second is a mathematical record of the shape differences between the atlas and the individual (warped grid, (d )). The amount of deformation required is displayed as a tensor map (only 2 components of the fully 3D transformation are shown). Tensor maps, and derived vector or scalar fields, can be analysed in a statistical setting to examine anatomic variation, detect pathology, or track structural changes over time (Thompson et al., 2001a–e). Histological neurochemical maps, accessible only post-mortem, can be transferred onto a living subject’s scan with a similar warping technique (Mega et al., 1997, 1999).

Ashburner et al., 1997; Zeineh et al., 2001). Interestingly, the transformations required to remove individual differences in anatomy are themselves a rich source of morphometric data (Thompson et al., 1997, 2001a–e; Grenander & Miller, 1998; Ashburner et al., 1998). As we shall see later, this data can be used to create disease-specific atlases.

Individualized brain atlases Anatomic variations No two brains are the same, which presents a challenge in creating standardized atlases. Even without pathology, brain structures vary between individuals in every metric; shape, size, position and orientation relative to each other (Steinmetz et al., 1989, 1990). Due to the obvious limita-

tions of a fixed atlas, new algorithms were developed to elastically re-shape an atlas to the anatomy of new individuals (Fig. 8.3). The resulting deformable brain atlases more accurately project atlas data into new scans. Their uses include surgical planning (Warfield et al., 1998; St-Jean et al., 1998), anatomical labelling (Iosifescu et al., 1997) and shape measurement (Thompson et al., 1997; Haller et al., 1997; Csernansky et al., 1998; Subsol et al., 1997). The shape of the digital atlas is adapted using local warping transformations (dilations, contractions and shearing) producing an individualized brain atlas (Evans et al., 1991; Miller et al., 1993; Christensen et al., 1993; Sandor & Leahy, 1994; 1995; Rizzo et al., 1995). Pioneered by Bajcsy & colleagues at the University of Pennsylvania (Broit, 1981; Bajcsy & Kovacic, 1989; Gee et al., 1993, 1995; Gee & Bajscy, 1998), this approach was adopted by the Karolinska Brain


133

134


Atlas Program (Seitz et al., 1990; Thurfjell et al., 1993; Ingvar et al., 1994). Warping algorithms can transfer maps of cytoarchitecture, biochemistry, functional and vascular territories into the coordinate system of different subjects (for a review, see Toga, 1998). Intricate patterns of structural variation in anatomy can be accommodated. These transformations must allow any segment of the atlas anatomy, however small, to grow, shrink, twist and even rotate, to produce a transformation that encodes local differences in topography from one individual to another. Non-linear mapping of raster volumes or 3D geometric atlases onto individual datasets has empowered many studies of disease. These include brain structure labeling for hippocampal morphometry in dementia (Haller et al., 1997), analysis of subcortical structure volumes in schizophrenia (Iosifescu et al., 1997; Csernansky et al., 1998), estimation of structural variation in normal and diseased populations (Collins et al., 1995; Thompson et al., 1997), and segmentation and classification of multiple sclerosis lesions (Warfield et al., 1995). Digital anatomic models can also be projected into PET data to define regions of interest for quantitative calculations of regional cerebral blood flow (Ingvar et al., 1994; Dinov et al., 2000; Mega et al., 2000a,b,c). These template-driven segmentations require extensive validation relative to more labour-intensive manual delineation of structures, but show considerable promise in studies of disease (see Mega et al., 2000a,b,c).

Analyzing brain data with an atlas The ability to relate atlas data to a new subject’s brain images also operates in reverse. By inverting the deformation field that reconfigures an atlas to match an individual, an individual’s data can be non-linearly registered with the atlas, removing subject-specific anatomical differences. Functional data can then be compared and integrated across subjects, with confounding anatomical effects factored out. Since they transfer multi-subject data more accurately into a stereotaxic framework, non-linear registration algorithms are now increasingly used in functional image analysis packages (Seitz et al., 1990; Friston et al., 1995; Ashburner et al., 1997; Woods et al., 1998). Because variations in structure and function are so great, and both are altered in disease, non-linear registration approaches become relevant in creating disease-specific templates. These algorithms eliminate the anatomic component of functional variation, and are required to separate variations in structure and function. They are also vital in creating deformable atlases, which offer a means to represent, and measure, variations in structure.

When extended to accommodate more subjects, deformations that match an atlas with each patient in a population can be used to create statistical maps of anatomy, revealing patterns of variability, asymmetry or abnormality in a group (Thompson et al., 1996a,b, 1997). With a modeldriven approach, graphical surface models represent each major anatomic system, so a comprehensive geometric atlas can be built. Average representations can be created for each anatomical element, along with statistical maps that can be visualized directly or used to guide subsequent image analysis.

Continuum-mechanical atlases Many brain atlases have been developed to deform according to the principles of continuum mechanics (Broit, 1981; Bajcsy & Kovacic, 1989; Christensen et al., 1996; Davatzikos, 1996; Gee & Bajscy, 1998; Thompson et al., 2001a–e). This feature is relevant to understanding how variations in structure can be encoded. In modeling the atlas deformations, differential equations are used to make the deforming atlas conform to the behavior of elastic or fluid materials. An advantage of this approach is that the well-understood mathematics enforces several desirable characteristics in the mappings. For instance, atlas-to-patient mappings should be one-to-one (i.e. the deformed atlas should not tear or self-intersect). This is surprisingly difficult to guarantee, unless continuum-mechanical or variational methods are applied (Christensen et al., 1995; Dupuis et al., 1998). The continuum-mechanical operators that govern the atlas deformations also have a spectral (or eigenfunction) representation that helps calculate the mappings rapidly (Miller et al., 1993; Ashburner et al., 1997).

Statistical templates The deformable template framework has also been widely tested in computer vision applications where shape variability needs to be accommodated, such as written digit identification or face recognition. This makes it easier to build a statistical theory of shape for encoding brain variation, using Gaussian fields (Thompson et al., 1996a,b, 1997; Davatzikos et al., 1996; Ashburner et al., 1997; Gee & Bajscy, 1998; Dupuis et al., 1998; Thirion et al., 1998; Cao & Worsley, 1999; Le Goualher et al., 1999) or Riemannian shape manifolds (Bookstein, 1997). Probabilistic transformations can then be applied to deformable anatomic templates to create a type of probabilistic atlas that measures variability and detects pathology (Thompson et al., 1997, 1998a,b; Joshi et al., 1998; Grenander & Miller, 1998).



Fig. 8.4. Continuum-mechanical warping. (a) The complex transformation required to reconfigure one brain into the shape of another can be determined using continuum-mechanical models, which describe how real physical materials deform. In this illustration, two line elements embedded in a linearly elastic 3D block (lower left) are slightly perturbed (arrows). The goal is to find how the rest of the material deforms in response to this displacement. The Cauchy-Navier equations (shown in discrete form, top) are solved to determine the values of the displacement field vectors, u(x), throughout the 3D volume. (b) Lamé Elasticity Coefficients. Different choices of elasticity coefficients, ␭ and ␮, in the Cauchy–Navier equations (shown in continuous form, top) result in different deformations, even if the applied internal displacements are the same. In histologic applications where an elastic tissue deformation is estimated, values of the elasticity coefficients can be chosen which limit the amount of curl (lower right) in the deformation field. Stiffer material models (top left) may better reflect the deformational behaviour of tissue during histologic staining procedures. Note: To emphasize differences, the displacement vector fields shown in this figure have been multiplied by a factor of 10. The Cauchy–Navier equations, derived using an assumption of small displacements, are valid only when the magnitude of the deformation field is small.

Individualizing an atlas To understand how deformable atlases work, consider the deforming atlas to be embedded in a 3D elastic or fluid medium (see Figs. 8.3, 8.4). The medium is subjected to distributed internal forces, which reconfigure it, and lead the image to match the target.

Model-driven deformable atlases The extreme difficulty of finding structures in new patients based on intensity criteria alone has led several groups to develop model-driven deformable atlases (Thompson et al., 1997; Toga & Thompson, 1997). Anatomic models provide an explicit geometry for individual structures in each scan, such as landmark points, curves or surfaces.

Because the digital models reside in the same stereotaxic space as the atlas data, surface and volume models stored as lists of vector coordinates are amenable to digital transformation, as well as geometric and statistical measurement (Thompson et al., 1996a,b). The underlying 3D coordinate system is central to all atlas systems, since it supports the linkage of structure models and associated image data with spatially-indexed neuroanatomic labels, preserving spatial information and adding anatomical knowledge. In the following sections, we show how anatomical models can create probabilistic atlases and disease-specific templates. Statistical averaging of models provides a means to analyze brain structure in morphometric projects, localizing disease-specific differences with statistical and visual power. We first describe how models can drive deformable atlases and make average models of anatomy, measuring patient-specific differences in considerable detail.


135

136


When deforming an atlas to match a patient’s anatomy, mesh-based models of anatomic systems help guide the mapping of one brain to another (Figs. 8.3, 8.4). Anatomically driven algorithms guarantee biological as well as computational validity, generating meaningful object-to-object correspondences, especially at the cortex. Ultimately, accurate warping of brain data requires: (i) matching entire systems of anatomic surface boundaries, both external and internal, and (ii) matching relevant curved and point landmarks, including ones within the surfaces being matched (e.g. primary sulci at the cortex, tissue type boundaries at the ventricular surface).

Anatomical models

surfaces. Recent success of sulcal extraction approaches based on deformable surfaces (Vaillant & Davatzikos, 1997) led us to combine a 3D skeletonization algorithm with deformable curve and surface governing equations to automatically produce parameterized models of cingulate, parieto-occipital, and calcarine sulci, without manual initialization (Zhou et al., 1999). Additional, manuallysegmented surfaces can also be given a uniform rectilinear parameterization using algorithms described in Thompson et al. (1996a,b), and used to drive the warping algorithm. Each resultant surface mesh is analogous in form to a uniform rectangular grid, drawn on a rubber sheet, which is subsequently stretched to match all data points.

Probabilistic atlases

Since much of the functional territory of the human cortex is buried in cortical folds or sulci, a generic structure is built to model them (Fig. 8.5; Thompson & Toga, 1996). The underlying data structure is a connected system of surface meshes, in which the individual meshes are parametric. These surfaces are 3D sheets that divide and join at curved junctions to form a connected network of models. With the help of these meshes, each patient’s anatomy is modelled in sufficient detail to be sensitive to subtle differences in disease. Separate surfaces model the deep internal trajectories of features such as the parieto-occipital sulcus, the anterior and posterior calcarine sulcus, the Sylvian fissure, and the cingulate, marginal and supracallosal sulci in both hemispheres. Additional gyral boundaries are represented by parameterized curves lying in the cortical surface. The ventricular system is modelled as a closed system of 14 connected surface elements whose junctions reflect cytoarchitectonic boundaries of the adjacent tissue (Fig. 8.6; Thompson & Toga, 1998). Information on the meshes’ spatial relations, including their surface topology (closed or open), anatomical names, mutual connections, directions of parameterization, and common 3D junctions and boundaries is stored in a hierarchical graph structure. This ensures the continuity of displacement vector fields defined at mesh junctions.

Surface parameterization After imposing an identical regular grid structure on anatomic surfaces from different subjects (Fig. 8.5), the explicit geometry can be exploited to drive and constrain correspondence maps that associate anatomic points in different subjects. Structures that can be extracted automatically in parametric form include the external cortical surface, ventricular surfaces, and several deep sulcal

Encoding anatomic variability Many morphometric studies focus on identifying systematic alterations in anatomy in a variety of diseases. These studies are complicated by the substantial overlap between measures of normal and diseased anatomy. Normal anatomic complexity makes group specific patterns hard to discern. However, disease-specific variants may be easier to localize by creating average models of anatomy, rather than deriving volumetric descriptors. In response to these challenges, probabilistic atlases are research tools that retain information on anatomic and functional variability (Mazziotta et al., 1995, 2001; Thompson et al., 2000a,b, 2002). A probabilistic atlas solves many of the limitations of a fixed atlas in representing highly variable anatomy. As the subject database increases in size and content, the digital form of the atlas allows efficient statistical comparisons of individuals or groups. In addition, the population that an atlas represents can be stratified into subpopulations to represent specific disease types, and subsequently by age, gender, handedness, or genetic factors.

Parametric mesh modelling Parametric meshes (Thompson et al., 1996a,b) offer a means to create average models of anatomy. Once anatomical data is transformed to a standardized coordinate space, such as the Talairach space, a computational grid structure can be imposed on anatomical surface boundaries. These mesh models represent boundary point locations in stereotaxic coordinates (Figs. 8.5, 8.6). Averaging of corresponding grid points across subjects results in an average surface model for each structure (Fig. 8.6 shows an example of



Fig. 8.5. Modelling anatomy with surface meshes. The derivation of a standard surface representation for a range of brain structures makes it easier to compare anatomical models from multiple subjects. An algorithm converts a set of digitized points on an anatomical structure boundary (e.g., deep sulci (a)) into a parametric grid of uniformly spaced points in a regular rectangular mesh stretched over the surface ((b); Thompson et al., 1996). By averaging nodes with the same grid coordinates across subjects (c), an average surface is produced for the group. However, information on each subject’s individual differences is retained as a vector-valued displacement map (d,e). This map indicates how that subject deviates locally from the average anatomy. The root mean square magnitude (e) of these deviations provides a variability measure whose values can be visualized using a colour code ( f ). These maps can be stored to measure variability in different anatomic systems, including ventricular and deep sulcal (Thompson et al., 1998a,b) surfaces. A more complex method measures cross-subject variations in gyral patterns, with a surface matching procedure that better reflects anatomical variations at the cortex (see main text). These maps can be stored to measure variability ( f ) and detect typical (or abnormal) patterns of brain structure in different anatomic systems.

average ventricular anatomy). At the same time, knowledge of each subject’s deviations from the group average anatomy can be retained as a vector displacement map (Fig. 8.5). After storing these maps from large numbers of subjects, local biases in the magnitude and direction of

anatomic variability can be displayed as a map. Variability maps for deep sulcal surfaces are shown in Fig. 8.7. In these maps, the colour shows the root mean square magnitude of the displacement vectors that map individuals to the group mean. Separate maps are displayed for elderly normals


137

Fig. 8.6. Population-based maps of average ventricular anatomy in normal ageing and Alzheimer’s disease. In patients and controls, 3D parametric surface meshes (Thompson et al., 1996) were used to model 14 ventricular elements (a), and meshes representing each surface element were averaged by hemisphere in each group. (b) An average model for Alzheimer’s patients (red; AD) is superimposed on an average model for matched normal controls (blue; NC). Mesh averaging reveals enlarged occipital horns in the Alzheimer’s patients, and high stereotaxic variability (c) in both groups. Extreme variability at the occipital horn tips also contrasts sharply with the stability of septal and temporal ventricular regions. A top view of these averaged surface meshes reveals localized asymmetry, variability, and displacement within and between groups. These subcortical asymmetries emerge only after averaging of anatomical maps in large groups of subjects.

Fig. 8.7. Population-based maps of 3D structural variation and asymmetry. Statistics of 3D deformation maps help define confidence limits on normal anatomic variation. 3D maps of anatomic variability and asymmetry are shown for 10 subjects with Alzheimer’s disease (AD; age: 71.9 ± 10.9 yrs.), and 10 normal elderly subjects matched for age (72.9 ± 5.6 yrs.), gender, handedness and educational level (Thompson et al., 1998a,b). Normal Sylvian fissure asymmetries (right higher than left; p < 0.0005) were significantly greater in AD than in controls ( p < 0.0002;top panels). In the 3D variability maps derived for each group (lower panels), the colour encodes the root mean square magnitude of the displacement vectors that map surfaces from each of the ten patients’ brains onto the average. 3D cortical variability (lower right panel ) increased in AD from 2–4 mm at the corpus callosum to a peak standard deviation of 19.6 mm at the posterior left Sylvian fissure.



(mean age: 72.9 ± 5.6 yrs.; all 10 right-handed), and demographically matched Alzheimer’s patients (age: 71.9 ± 10.7 yrs.; all 10 right-handed; mean Mini-Mental State Exam score: 19.7 ± 5.7, out of 30). As expected, there is an extraordinary increase in anatomical variability from deep structures (0–5 mm at the corpus callosum) to peak r.m.s. values of 12–13 mm at the posterior Sylvian fissures (Thompson et al., 1998a,b). In AD however, Sylvian fissure variability rose extremely sharply from an SD of 6.0 mm rostrally on the left to 19.6 mm caudally. Underlying atrophy and possible left greater than right degeneration of perisylvian gyri (Loewenstein et al., 1989; Siegelet al., 1996) may widen the Sylvian fissure, superimposing additional individual variation and asymmetry on that seen in normal aging.

the approach was used to detect pre-clinical hippocampal atrophy in patients with minimal cognitive impairment (Mega et al., 2000a,b,c). To identify more focal effects, we attempted to identify regionally selective patterns of callosal change in patient groups with Alzheimer’s disease (Thompson et al., 1998a,b). The mid-sagittal callosum was first partitioned into five sectors (Fig. 8.8; Duara et al.

Brain asymmetry A third feature observable from the average anatomical models (Figs. 8.6 and 8.7) is that consistent patterns of brain asymmetry can be mapped, despite wide variations in asymmetry in individual subjects. In dementia, the increased cortical asymmetry probably reflects asymmetric progression of the disease. Fig. 8.6 shows average maps of the lateral ventricles, in Alzheimer’s disease and matched elderly normal populations. As expected, the ventricles are significantly enlarged in dementia. Notice, however, that a pronounced asymmetry is observed in both groups (left volume larger than right, P < 0.05). This is an example of an effect that becomes clear after group averaging of anatomy, and is not universally apparent in individual subjects. It is, however, consistent with prior volumetric measurements. Anatomical averaging can also be cross-validated with a traditional volumetric approach. Occipital horns were on average 17.1% larger on the left in the normal group (4070.1 ± 479.9 mm3 ) than on the right (3475.3 ± 334.0 mm3 ; P < 0.05), but no significant asymmetry was found for the superior horns (left: 8658.0 ± 976.7 mm3 ; right: 8086.4 ± 1068.2 mm3 ; p > 0.19) or for the inferior horns (left: 620.6 ± 102.6 mm3 ; right: 573.7 ± 85.2 mm3 ; P > 0.37). The asymmetry is clearly localized in the 3D group average anatomic representations. In particular, the occipital horn extends (on average) 5.1 mm more posteriorly on the left than the right. The capacity to resolve asymmetries in a group atlas can assist in studies of disease-specific cortical organization (Thompson et al., 1997, 2001a–e; Mega et al., 1998a,b; Zoumalan et al., 1999; Narr et al., 1998; Sowell et al., 2001; Cannon et al., 2002).

Fig. 8.8. Corpus callosum in Alzheimer’s disease. Midsagittal corpus callosum boundaries were averaged from patients with Alzheimer’s disease and from elderly controls matched for age, educational level, gender and handedness. The average representations show a focal shape inflection in the Alzheimer’s patients relative to normal elderly subjects of the same age. A statistically significant tissue loss is also found at the isthmus (2nd sector, when the structure is partitioned into fifths). The isthmus connects regions of temporo-parietal cortex that exhibit early neuronal loss and perfusion deficits in AD (Thompson et al., 1998a,b).

Corpus callosum differences We also tested the ability of anatomical averaging to identify disease-specific patterns in clinical populations. First,


139

140


(1991)). This roughly segregates callosal fibres from distinct cortical regions. In AD, focal fibre loss was expected at the callosal isthmus (sector 2) whose fibres selectively innervate the temporo-parietal regions with early neuronal loss and perfusion deficits (Brun & Englund, 1981). Consistent with this hypothesis, a significant area reduction at the isthmus was found, reflecting a dramatic 24.5% decrease from 98.0 ± 8.6 mm2 in controls to 74.0 ± 5.3 mm2 in AD (P < 0.025). Terminal sectors (1 and 5) were not significantly atrophied, and the central midbody sector showed only a trend toward significance (16.6% mean area loss; p < 0.1), due to substantial inter-group overlap. Average boundary representations, however, localized these findings directly. At the isthmus, average models in AD showed a pronounced shape inflection at stereotaxic location (0.0,−25.0,19.0) (see Fig. 8.8).

Atlas-based pathology detection Deformable probabilistic atlases As noted earlier, warping algorithms create deformation maps (Fig. 8.7) that indicate 3D patterns of anatomic differences between any pair of subjects. By defining probability distributions on the space of deformation transformations which drive the anatomy of different subjects into correspondence (Grenander, 1976; Amit et al., 1991; Grenander Miller, 1994; Thompson et al., 1997; Thompson et al., 1997), statistical parameters of these distributions can be estimated from databased anatomic data to determine the magnitude and directional biases of anatomic variation. Encoding of local variation can then be used to assess the severity of structural variants outside of the normal range, which, in brain data, may be a sign of disease (Thompson et al., 1997). The encoding of anatomical shape and gyral pattern variation can also assist in resolving additional disease-specific features, such as average patterns of cortical grey matter loss, as described next.

Mapping grey matter loss in Alzheimer’s disease In a recent report (Thompson et al., 2001a–e) we measured cortical grey matter distribution and disease-related grey matter loss in Alzheimer’s disease. Figure 8.9 shows a surface-based probability field that indicates the regional significance of grey matter loss across the cortex in the entire AD cohort. Red (P < 0.005) denotes brain regions where the average grey matter index is significantly less in the AD cohort than in the control group. All

averages and comparisons are made across corresponding areas of cortex, defined by gyral pattern matching (a procedure described in the next sections). Given these statistics, two types of inference are possible. First, the a priori hypothesis of grey matter loss in the temporal and parietal cortices was confirmed. There was also evidence for a region of maximal loss throughout the lateral temporal surface and the parietal operculum bilaterally (P < 0.0001–0.001).

Deficit maps Percentages may also be plotted on the cortex (Fig. 8.10) to visualize the average deficit in patients relative to healthy controls. A pervasive left greater than right hemisphere reduction in grey matter was found (with up to 20–30% loss locally). The pattern of grey matter loss is consistent with findings from metabolic studies (e.g. Loewenstein et al., 1989) that the left hemisphere is, on average, more severely affected at this stage of the disease. The occipital cortices were comparatively spared bilaterally, as were the sensorimotor cortices (0–5% loss, P > 0.05). There was also severe grey matter loss (20–30%, P < 0.001 − 0.0001) in the middle frontal gyrus, in the vicinity of areas 9 and 46 (Rajkowska & Goldman-Rakic, 1995). We investigated further whether the regions of more significant grey matter loss (Fig. 8.9) reflected a correspondingly greater average reduction in the local grey matter index (Fig. 8.10). This was important, as a greater significance value can result either from (i) a genuinely greater percentage reduction in the mean grey matter in AD or (ii) a local reduction in the variance of the grey matter index across the group, which translates into a greater detection sensitivity. Interestingly, a map of the percentage reduction in average grey matter (Fig. 8.10) followed approximately the same anatomical pattern, suggesting that there is indeed a hierarchy in the severity of grey matter loss at this stage of the disease, rather than a fluctuation in the local power of the statistical model to detect it. Temporal and temporo-parietal cortices exhibited severe (10–30%) reductions in grey matter. This contrasted with a comparative sparing of the superior margins of the central and post-central gyri and occipital poles (0–5% loss). In mild to moderate AD, diffuse grey matter loss occurs across the majority of the cortex, but it is interesting that the superior central and post-central gyri and occipital poles show very little reduction in grey matter when adjacent posterior temporal cortex and the parietal operculum are severely affected, in both the percentage loss and statistical anatomical maps.



Fig. 8.9. Statistical map of average grey matter loss in Alzheimer’s disease (N = 46). Based on averaging and comparing grey matter measurements across equivalent regions of cortex in all 46 subjects, a statistical field can be generated that reflects whether the average grey matter is reduced in patients (average of 26 subjects) relative to controls (average of 20 subjects), and the significance of this reduction at each cortical location. As hypothesized, pervasive left-greater-than-right reductions are found, with severe, more localized reductions in temporal lobe and temporo-parietal cortex. This profile of grey matter loss mirrors the anatomical distribution of early perfusion deficits and metabolic change in mild to moderate AD.

By averaging cortical features in an AD population and matched elderly controls, mean profiles of grey matter loss (Fig. 8.10), as well as local patterns of anatomical variation and cerebral asymmetry (Fig. 8.7) can be identified. Severe reductions in grey matter (up to 30% loss) were observed across the lateral temporal surfaces and temporoparietal cortices bilaterally. Patterns of left greater than right grey matter loss also emerged, with severe grey matter loss observed bilaterally in the vicinity of Brodmann areas 9 and 46, regions of increased synaptic loss and ␤-amyloid protein deposition in AD (Clinton et al., 1994). The comparative sparing of the superior post-central and central gyri and the occipital poles (0–5% loss, P < 0.05)

is consistent with preservation of sensorimotor and visual function at this stage of the disease, at the same time as perfusion and metabolic deficits are prevalent in association cortices.

Hemispheric differences Patterns of greater grey matter loss in the left hemisphere corroborate earlier reports (Loewenstein et al., 1989) of predominant left hemisphere metabolic dysfunction in mild to moderate AD, when cerebral glucose utilization is measured by positron emission tomography (PET). Structural, perfusion and metabolic studies suggest that the left


141

142


Fig. 8.10. Map of average grey matter loss in Alzheimer’s disease expressed as a percentage of average control values (N = 46). This map expresses the same data as Fig. 8.9 as a percentage reduction in the measurement of grey matter, when equivalent cortical regions are averaged and compared between AD patients and controls. The percentage reduction in average grey matter followed approximately the same anatomical pattern as the significance map, suggesting that there is indeed a hierarchy in the severity of grey matter loss at this stage of the disease, rather than a fluctuation in the local power of the statistical model to detect it. Again, temporal and temporo-parietal cortex exhibited severe (10%–30%) reductions in grey matter. This contrasted with a comparative sparing of the superior margins of the central and postcentral gyri and the occipital poles (0%–5% loss).

hemisphere may be more susceptible to neuronal loss, instead of the alternative explanation that equivalent neuronal loss may result in greater functional deficits on one side, due to asymmetrical cortical organization. Greatest grey matter loss in the temporo-parietal cortex may underlie the prominent temporal-parietal hypometabolism found consistently in early AD, often asymmetrically (Friedland and Luxenberg, 1988; Johnson et al., 1998). Although the focus of this study (Fig. 8.10) was to determine patterns of grey matter loss in vivo, immunocytochemical studies have reported between 11 and 50% synaptic loss in the superior temporal and inferior parietal cortices, with a comparative sparing of occipital cortices. Relatively greater atrophy is often reported in the temporal lobe relative to overall cerebral volume (Murphy et al., 1993). The early progression of AD pathology into the parietal and frontal association cortices suggests a degeneration of synaptically linked cortical pathways, and this pattern correlates

with symptoms of memory impairment, aphasias, apraxias, personality changes and spatial deficits (Roberts et al., 1993). Interestingly, grey matter loss at autopsy is predominantly cortical in Alzheimer’s patients under 80 years of age (Hubbard and Anderson, 1981), when volumes of subcortical nuclei are not significantly different between patients and controls (De La Monte, 1989). Nonetheless, atrophy of the amygdala and basal nuclei (Cuénod et al., 1993) may ultimately be followed by alterations in thalamic nuclei (Jernigan et al., 1991), induced perhaps by degeneration of their cortical projection areas. At this stage, the pathological burden of AD may be greater in terms of functional deficits, and synaptic loss, in the heteromodal cortex than in the idiotypic cortex. In our prior studies AD patients exhibited significantly greater asymmetry and structural variability in the deep perisylvian cortex, relative to controls matched for age, gender, educational level and handedness (P < 0.05; Fig. 8.7;



Fig. 8.11. Mapping brain asymmetry in a population. The average magnitude of brain asymmetry in a group (N = 20, elderly normals) can be assessed based on warping fields that map the cortical pattern of one hemisphere onto a reflected version of the other, and then flow the observations again so that corresponding measures can be averaged across subjects (Thompson et al., 2001; see main text for methods). Variations in asymmetry are also non-stationary across the cortex (lower left), and a Hotelling’s T2 statistical field can be computed to map the significance of the asymmetry (lower right) relative to normal anatomic variations.

Thompson et al., 1998a,b). Clear differences in both AD cortical variation and grey matter distribution suggest the need for disease-specific brain atlases that better reflect the disease-related anatomy of patients and calibrate individual loss against statistical data from normative populations.

Mapping asymmetries

a group atlas presents opportunities to analyse diseases with asymmetrical progression, including different stages of AD, and to map hypothesized alterations in cortical and hippocampal asymmetry in diseases such as schizophrenia (Falkai et al., 1992; Kikinis et al., 1994; Kulynych et al., 1996; Csernansky et al., 1998; Narr et al., 2001a,b).

Cortical modelling

There is a substantial literature on Sylvian fissure cortical surface asymmetries (Eberstaller, 1884; Cunningham, 1892; Geschwind & Levitsky, 1968; Davidson & Hugdahl, 1994) and their relation to functional lateralization (Strauss et al., 1983), handedness (Witelson & Kigar, 1992), language function (Davidson & Hugdahl, 1994), asymmetries of associated cytoarchitectonic fields (Galaburda & Geschwind, 1981) and their thalamic projection areas (Eidelberg and Galaburda, 1982). The improved ability to localize asymmetries of cortical organization (Fig. 8.11) or tissue loss in

Cortical morphology is notoriously complex, and presents unique challenges in anatomic modelling. The cortex is also severely affected in disorders such as Alzheimer’s disease, Pick’s disease and other dementias, by tumour growth, and in cases of epilepsy, cortical dysplasias, and schizophrenia. A major challenge in investigations of disease is to determine (i) whether cortical organization is altered, and if so, which cortical systems are implicated, and (ii) whether normal features of cortical organization are lost, such as sulcal


143

144


Fig. 8.12. Average and probabilistic brain templates. Direct averaging of imaging data after a simple affine transform into stereotaxic space washes cortical features away ((a); Evans et al., 1994; N = 305 normals; (b) shows a similar approach with N = 9 Alzheimer’s patients). By first averaging a set of vector-based 3D geometric models, and warping each subject’s scan into the average configuration, a well-resolved average brain template is produced, which reflects the anatomy of patients with mild to moderate Alzheimer’s disease (c). Deformation vector maps (e) store individual deviations (brown mesh) from a group average (white surface, (d )), and their covariance fields ( f ) store information on the preferred directions and magnitude (g) of anatomic variability (pink colours, large variation; blue colours, less).

pattern asymmetries. This requires methods to create (i) a well-resolved average model of the cortex specific for a diseased group, and (ii) a statistical framework to compare these average models with normative data.

Averaging images or averaging geometry In an atlasing context, it would be ideal to create a diseasespecific template for a clinical group with well-resolved anatomical features in their mean anatomical configuration. Unfortunately this cannot be achieved by averaging together structural images in the traditional way, after a simple linear transformation to a standard space (Evans et al., 1994). If images are averaged in this way, cortical patterns are washed away due to wide variations in gyral organization (Fig. 8.12, top left). We describe a way to avoid this. First, an average cortical surface model is created with wellresolved gyral features in the group mean configuration.

Continuum-mechanical mappings are then used to bring each subject’s gyral pattern into correspondence with the average cortex. Maps of cortical variation are created as a by-product (colour maps, Fig. 8.12). Finally, a highdimensional mapping (driven by 84 structures per brain) elastically deforms each brain into the group mean geometric configuration. Once elastically reconfigured, the scan intensities are averaged on a voxel-by-voxel basis to produce a group-specific atlas template with a wellresolved cortex (Fig. 8.12, lower panels). A disease-specific brain imaging template, created to represent patients with Alzheimer’s disease, will be used to illustrate this method.

Cortical matching Cortical anatomy can be compared, between a pair of subjects, by computing the warped mapping that elastically transforms one cortex into the shape of the other. These



transformations can also match large networks of gyral and sulcal features with their counterparts in the target brain (Thompson & Toga, 1996; Thompson et al., 1997; Davatzikos, 1996; Van Essen et al., 1997; Fischl et al., 1999). Differences in cortical organization prevent exact gyrus-bygyrus matching of one cortex with another. Nonetheless, an important intermediate goal has been to match a comprehensive network of sulcal and gyral elements that are consistent in their incidence and topology across subjects (Fig. 8.13; Ono et al., 1990; Leonard, 1996; Kennedy et al., 1998; Thompson et al., 2001a–e).

Cortical averaging The warping field deforming one cortex into gyral correspondence with another can be used to create an average model of the cortex in a patient group. To do this, all 3D curves representing gyral curves in a group of subjects are first transferred to a flattened (Fig. 8.13), or spherical (Fig. 8.14), parameter space (see e.g. Thompson et al., 2001a–e, for details of the method). This procedure represents the unfolded topography of the cortex on a 2D surface, so that features in the cortex can be more easily compared from one subject to another. Next, each curve is uniformly re-parameterized to produce a regular curve of 100 points in the flattened space whose corresponding 3D locations are uniformly spaced. A set of 36 average gyral curves for the group is created by vector averaging all point locations on each curve. This average curve template (curves in Fig. 8.15(2)) serves as the target for alignment of individual cortical patterns (cf. Fischl et al., 1999, for a similar approach). Each individual cortical pattern is transformed into the average curve configuration using a flow field within the flattened space (Fig. 8.15(1), 8.15(2)).

Cortical variability By using a color code (Fig. 8.15(4)) to identify original cortical locations in 3D space (Fig. 8.15(5 )), displacement fields can be recovered mapping each patient into gyrus-bygyrus correspondence with the average cortex. Anatomic variability is then defined at each point on the average cortex as the root mean square (r.m.s.) magnitude of the 3D displacement vectors, assigned to each point, in the surface maps driving individuals onto the group average (Thompson et al., 1996a,b, 1997, 1999). This variability pattern shows the magnitude of cortical pattern variation in an elderly population, and is visualized as a colour-coded map (Fig. 8.16).

Tensor maps of directional variation Structures do not vary to the same degree in every coordinate direction (Thompson et al., 1996a,b), and even these directional biases vary by cortical system. The principal directions of anatomic variability in a group can be shown in a tensor map (Fig. 8.17). The maps have two uses. First, they make it easier to detect cortical atrophy in an individual patient, which may be small in magnitude but in an unusual direction. Second, they significantly increase the information content of Bayesian priors used for automated structure extraction and identification (Gee et al., 1995; Mangin et al., 1995; Royackkers et al., 1996; Pitiot et al., 2002). Fig. 8.17 shows a tensor map of variability for normal subjects, after linearly mapping 20 elderly subjects’ data into Talairach space (all right handed, 10 males, 10 females). Ellipsoidal glyphs indicate the principal directions of variation – they are most elongated along directions where anatomic variation is greatest across subjects. Each glyph represents the covariance tensor of the vector fields that map individual subjects onto their group average. Because gyral patterns constrain the mappings, the fields reflect variations in cortical organization at a more local level than can be achieved by only matching global cortical geometry. Note the elongated glyphs in anterior temporal cortex, and the very low variability (in any direction) in entorhinal and inferior frontal areas. By better defining the parameters of allowable normal variations, the resulting information can be leveraged to distinguish normal from abnormal anatomical variants, and can map patterns of atrophy in Alzheimer’s disease (Thompson et al., 1997).

Brain averaging Average image templates So far we have described a scheme to create average anatomical models for specific patient groups. By assembling these average models for a wide range of systems (cortex, hippocampus, ventricles, deep sulci, and basal ganglia), an annotated atlas of structures can be built. Nonetheless, before new data can be pooled into the atlas, an average intensity image template is also required that reflects the unique morphology of the diseased population. This makes it easier for automated, intensity-based registration algorithms (e.g. Woods et al., 1993, 1998) to align new data with the atlas. To create a mean image template for a group, several approaches are possible. Which one is used depends on the application objectives. We describe a particular approach,


145

146


Fig. 8.13. Measuring differences in cortical anatomy. Based on an individual’s 3D MRI scan (a), detailed surface models of the cerebral cortex can be generated (b),(c). A template of 3D curved lines is delineated on these surfaces, capturing the morphology of the sulcal pattern. On the lateral brain surface, important functional landmarks include the central (CENT), pre- and post-central (preCENT, poCENT), superior and inferior frontal sulci (SFS, IFS), intraparietal sulcus (IP), Sylvian fissure (SF) and superior temporal sulcus (STS). Medial surface landmarks include the corpus callosum (CC), anterior and posterior calcarine (CALCa/p), parieto-occipital, subparietal, paracentral, paracingulate, and cingulate sulci, and the superior and inferior rostral sulci. A spherically parameterized, triangulated 3D mesh represents the cortical surface; (d ) shows the grid structure around the anterior corpus callosum. When the parameter space of the surface is flattened out (e), landmarks in the folded brain surface can be reidentified (e.g. IRS, SRS, etc.). (The white patch by the corpus callosum is where the surface model cuts across the white matter of the brainstem). To avoid loss of 3D information in the flattening, a colour code is used to store where each flat map location came from in 3D, with red colours brighter where the lateral (X) coordinate is larger, green colours brighter where the posterior-to-anterior co-ordinate (Y) is larger, etc. The warping of these colour maps, and the averaging of the resulting images, provides a surprising strategy for creating average cortical models for a group of subjects, and for exploring cortical pattern variation.



Fig. 8.14. Cortical pattern matching. Cortical anatomy can be compared, for any pair of subjects (3D models; top left), by computing a 3D deformation field that reconfigures one subject’s cortex onto the other (3D matching field, top right). In this mapping, gyral patterns can be constrained to match their counterparts in the target brain. To do this, flattening or inflation of the extracted cortical surface provides a continuous inverse mapping from each subject’s cortex to a sphere or plane. A vector field u(r) in the parameter space can then drive the gyral pattern elements into register on the sphere (see spherical flow). The full mapping (top middle) is recovered in 3D space as a displacement vector field matching cortical regions in one brain into precise structural registration with their counterparts in the other brain. Tensor Maps (middle and lower left): Different amounts of local dilation and contraction (encoded in the metric tensor of the mapping, gjk (r)) are required to transform the cortex into a simpler 2-parameter surface. These variations complicate the direct application of 2D warping equations for matching their features. Using a covariant tensor approach (red box) the regularization operator L is replaced by its covariant form L‡ . Correction terms i ) compensate for fluctuations in the metric tensor of the inflation and flattening procedures. (Christoffel symbols jk This (1) makes the matching field independent of the underlying gridding of the surface (spherical or planar), and (2) eliminates effects of metric distortions that occur in the inflation or flattening procedure.

which guarantees that the average template has (i) well-resolved cortical features (Thompson et al., 1999), and (ii) the average size and shape for a subject group (Woods et al., 1998). To create an atlas template that is consistent with an average set of anatomical models, highdimensional model-based registration is required. If scans are mutually aligned with only a linear transformation, the resulting average brain is blurred in the more variable anatomical regions. The resulting average brain also tends to exceed the average dimensions of the component brain images.

By averaging geometric and intensity features separately (cf. Ge et al., 1995; Bookstein, 1997; Grenander & Miller, 1998; Thompson et al., 1999), a template can be made with the mean intensity and geometry for a patient population. We illustrate this approach by using the cortical transformations defined above (Figs. 8.13–8.15) to create a wellresolved disease-specific image template for an Alzheimer’s disease population (Fig. 8.12, lower panels). First, a group of well-characterized Alzheimer’s patients was selected, for whom a range of anatomical surface models (84 per brain) had been created in prior morphometric


147

Fig. 8.15. Average cortex in Alzheimer’s disease. An average cortical surface model is shown for a group of patients (here N = 9 Alzheimer’s patients) with well-defined sulcal features appearing in their average geometric configuration. If sulcal position vectors are averaged without mathematically aligning the intervening gyral patterns (5 ), sulcal features are not reinforced across subjects: a smooth average cortex is produced. By matching gyral patterns across subjects before averaging, a crisper average cortex is produced (6). This type of average cortical model can be created for a group of patients by first flattening each subject’s cortical model to a 2D square (panel 1; see also Figs. 8.13 and 8.14). A colour-coded map (3) stores a unique colour triplet (RGB) at each location in the 2D parameter space encoding the (x,y,z) coordinate of the 3D cortical point mapped to that 2D location. By averaging these colour maps pixel-by-pixel across subjects, and then decoding the 3D colours into a surface model, a smooth cortical model (5) is produced. However, a well-resolved average model (6) is produced, with cortical features in their group mean location, if each subject’s colour map is first flowed (4) so that sulcal features are driven into the configuration of a 2D average sulcal template (2). The average curve set is defined by 2D vector averaging of many subjects’ flattened curves. In this flow (4), codes indexing similar 3D anatomical features are placed at corresponding locations in the parameter space, and are thus reinforced in the group average (6).



Fig. 8.16. Mapping 3D cortical variability in an Alzheimer’s disease brain atlas. The profile of variability across the cortex is shown (N = 26 Alzheimer’s patients), after differences in brain orientation and size are removed by transforming individual data into Talairach stereotaxic space. The following views are shown: oblique frontal, frontal, right, left, top, bottom. Extreme variability in posterior perisylvian zones and superior frontal association cortex (16–18 mm; red colours) contrasts sharply with the comparative invariance of primary sensory, motor, and orbitofrontal cortex (2–5 mm, blue colours).

projects (Thompson et al., 1998a,b). An initial image template for the group was constructed by (i) using automated linear transformations (Woods et al., 1993) to align the MRI data with a randomly selected image, (ii) intensity-averaging the aligned scans, and then (iii) recursively re-registering the scans to the resulting average affine image. The resulting average image was adjusted to have the mean affine shape for the group (Woods et al., 1998). Images and surface models were then linearly aligned to this template, and an average surface set was created for the group (Thompson et al., 1997). Displacement maps (Fig. 8.12) driving the surface anatomy of each subject into correspondence with the average surface set were then computed, and were extended to the full volume with surface-based elastic warping (see Figs. 8.3, 8.4; Thompson & Toga, 1996, 1998).

These warping fields reconfigured each subject’s 3D image into the average anatomic configuration for the group. By averaging the reconfigured images (after intensity normalization), a crisp image template was created to represent the group (Fig. 8.12, lower panels). Note the better-resolved cortical features in the average images after high-dimensional cortical registration. If desired, this AD-specific atlas can retain the coordinate matrix of the Talairach system (with the anterior commissure at (0,0,0)) while refining the gyral map of the Talairach atlas to encode the unique anatomy of the AD population. By explicitly computing matching fields that relate gyral patterns across subjects, a well-resolved and spatially consistent set of probabilistic anatomical models and average images can be made to represent the average anatomy and its variation in a subpopulation.


149

150


disease (MMSE score: 19.3 + / − 2.0). At this stage, patients often present for initial evaluation, and MR, PET and SPECT scans have maximal diagnostic value. Nonetheless, by expanding the underlying patient database, atlases are under construction to represent the more advanced stages of Alzheimer’s disease, and MCI patients, for whom neuroimaging may be maximally informative. By stratifying the population according to different criteria, different atlases can be synthesized to represent other clinically defined groups.

Image distortion and registration accuracy

Fig. 8.17. Tensor maps reveal directional biases of cortical variation. Tensor maps can be used to visualize these complex patterns of gyral pattern variation at the cortex. The maps are based on a group of 20 elderly normal subjects. Colour distinguishes regions of high variability (pink colours) from areas of low variability (blue). Ellipsoidal glyphs indicate the principal directions of variation – they are most elongated along directions where there is greatest anatomic variation across subjects. Each glyph represents the covariance tensor of the vector fields that map individual subjects onto their group average anatomic representation. The resulting information can be leveraged to distinguish normal from abnormal anatomical variants using random field algorithms, and can define statistical distributions for feature labelling at the cortex (cf. Le Goualher et al., 1999; Vaillant & Davatzikos, 1997).

Disease progression The anatomical templates so far described, for demented and healthy elderly populations, have been based on homogeneous patient groups, matched for age, gender, handedness, and educational level. Since AD, in particular, is a progressive disease (see next Section), the initial atlas template was created to reflect a particular stage in the

Since the anatomy of a dementia population is poorly reflected by current imaging templates, substantially less distortion is applied by mapping multi-modality brain data into an atlas that reflects AD morphology (Mega et al., 1997; Thompson et al., 2000a,b). Incoming subjects deviate least from the mean template in terms of both image intensity and anatomy. Registration of their imaging data to this template therefore requires minimal image distortion. Since the template has the average affine shape for the group (Woods et al., 1998), least distortion is applied when either linear or non-linear, approaches are used. Interestingly, automated registration approaches were able to reduce anatomic variability to a greater degree if a specially prepared image template was used as a registration target (Thompson et al., 2000a,b).

Other average templates Several approaches are under active development to create average brain templates. Many of them are based on high-dimensional image transformations. Average templates have been made for the Macaque brain (Grenander & Miller, 1998), and for individual structures such as the corpus callosum, (Davatzikos, 1996; Gee & Bajcsy, 1998), central sulcus (Manceaux-Demiau et al., 1998), cingulate and paracingulate sulci (Paus et al., 1996; Thompson et al., 1997), hippocampus (Haller et al., 1997; Joshi et al., 1998; Csernansky et al., 1998; Thompson et al., 1999) and for transformed representations of the human and Macaque cortex (Drury & Van Essen, 1997; Grenander & Miller, 1998; Thompson et al., 1999; Fischl et al., 1999). Under various metrics, incoming subjects deviate least from these mean brain templates in terms of both image intensity and anatomy. Registration of new data to these templates not only requires minimal image distortion, but also allows faster algorithm convergence. This is because with smaller deformations, non-global minima of



Fig. 8.18. Tensor maps of brain change: visualizing growth and atrophy. If follow-up (longitudinal) images are available, the dynamics of brain change can be measured with tensor mapping approaches (Thompson et al., 2000a,b). These map volumetric change at a local level, and show local rates of tissue growth or loss. Fastest growth is detected in the isthmus of the corpus callosum in two young girls identically scanned at ages 6 and 7 (a), and at ages 9 and 13 (b). Maps of loss rates in tissue can be generated for the developing caudate ((c), here in a 7–11 year old child), and for the degenerating hippocampus (d ),(e). In (e), a female patient with mild Alzheimer’s disease was imaged at the beginning and end of a 19-month interval with high-resolution MRI. The patient, aged 74.5 years at first scan, exhibits faster tissue loss rates in the hippocampal head (10% per year, during this interval) than in the fornix. These maps may ultimately help elucidate the dynamics of therapeutic response in an individual or a population (Thompson et al., 2000a,b, 2001a–e; Haney et al., 2001).

the registration measure may be avoided, as the parameter space is searched for an optimal match. For these reasons, templates that reflect the mean geometry and intensity of a group are a topic of active research (Grenander & Miller, 1998; Woods et al., 1998; Thompson et al., 1999).

Dynamic (4D) brain atlases 4D coordinate systems Atlasing of data from the developing or degenerating brain presents unique challenges (Thompson et al., 2001a–e). Serial scanning of human subjects (Fox et al., 1996; Subsol et al., 1997; Freeborough et al., 1998; Thompson et al.,

1998a,b) or experimental animals (Jacobs & Fraser, 1994) in a dynamic state of disease or development offers the potential to create 4D models of brain structure. Warping algorithms can then be applied to serial scan data to track disease and growth processes in their full spatial and temporal complexity. Maps of anatomical change can be generated by warping scans acquired from the same subject over time (Thirion & Calmon, 1997; Thompson et al., 2000a,b). These algorithms can generate dynamic descriptors of how the brain changes during normal aging and Alzheimer’s disease (Figs. 8.18 and 8.19). They are also of interest for investigating and staging brain development in childhood and adolescence, and detecting aberrant tissue loss (Thompson et al., 2000a,b, 2001a–e; Sowell et al., 2001). In an atlas setting, these


151

152


Fig. 8.19. Tensor maps of local volumetric loss in normal elderly individuals. Local volume loss patterns in the hippocampus of an elderly subject (here, over a 6-month interval) are hard to appreciate from raw MRI data (left). They can be localized by using 3D surface models to drive a 3D continuum-mechanical partial differential equation (PDE; see Fig. 8.4) from which dynamic statistics of loss are derived. Comparison and averaging of this loss rate data across subjects requires a second PDE to convect the attribute data onto an average neuroanatomical atlas ( final 4 panels; see Thompson et al., 1997, 2000a,b, 2001a–e for methods and applications).

4-dimensional maps can provide normative criteria for early brain change in patients with dementia (Jernigan et al., 1991; DeCarli et al., 1992; Janke et al., 2001; Thompson et al., 2001a–e), with mild cognitive impairment (Studholme et al., 2001), or in those at genetic risk for Alzheimer’s disease (Small et al., 2000). An interesting application is the compilation of dynamic maps to characterize brain change in individual patients, which we illustrate next.

The growth maps obtained in these studies exhibit several striking characteristics. First, foci of rapid growth at the callosal isthmus appeared consistently across puberty, and attenuated as subjects progressed into adolescence. Meanwhile, rapid rates of tissue loss were also revealed at the head of the caudate, in an earlier phase of development.

Mapping brain development and degeneration

Tissue loss in dementia

In our initial human studies (Thompson et al., 2000a,b, 2001a–e, 2002), we developed several algorithms to create 4D quantitative maps of growth patterns in development, as well as degeneration in dementia. In a pediatric neuroimaging project, time-series of high-resolution MRI scans from healthy children were analysed. The resulting tensor maps of growth provided spatially detailed information on local growth patterns, quantifying rates of tissue maturation, atrophy, shearing and dilation in the dynamically changing brain architecture (Fig. 8.18). Pairs of scans were selected to determine patterns of structural change across the inter-scan interval. Deformation processes recovered by a high-dimensional warping algorithm were then analysed using vector field operators to produce a variety of tensor maps (Figs. 8.18 and 8.19). These maps were designed to reflect the magnitude and principal directions of dilation or contraction, the rate of strain, and the local curl, divergence and gradient of flow fields representing the growth processes recovered by the transformation.

In a pilot dementia study (Thompson et al., 2001a–e, 2002), dynamic maps of atrophic rates were generated for 17 AD patients and 14 demographically matched controls scanned repeatedly over a 4-year period (interscan interval: 2.6 ± 0.3 yrs.; final age: 71.3 ± 1.8 yrs.). 4D maps of annual atrophic rates were aligned elastically across subjects, averaged, and confidence limits were computed for tissue loss at each anatomical point throughout the brain. Profiles of local atrophic rates were visualized. Left faster than right hippocampal tissue loss was detected in controls (L: −3.8 ± 1.6%/yr.; R:−0.5 ± 1.2%/yr.; P < 0.05). Significantly faster loss rates were found bilaterally in AD (L:−5.9% ± 1.7%/yr.; R:−7.1 ± 3.2%/yr.; P < 0.03), and their 3D profiles were visualized. In controls, these loss rates peaked at a localized region of the medial surface of the left hippocampal head. In AD, an anterior to posterior shift was detected in the region of peak loss, which broadened to encompass the entire hippocampus, bilaterally. Local atrophic rates were significantly linked to the rate of



Fig. 8.20. Creating brain maps and anatomical models. An image analysis pipeline (Thompson et al., 2001a–e) is shown here. It can be used to create maps that reveal how brain structure varies in dementia populations, and how it is modulated by genetic factors or drug treatment. This approach aligns new 3D MRI scans from patients and controls (1) with an average brain template based on a population (in young normal studies the ICBM template is used, developed by the International Consortium for Brain Mapping; in dementia studies an AD-specific template is used; see Fig. 8.12). Tissue classification algorithms then generate maps of grey matter, white matter and CSF (2). To help compare cortical features from subjects whose anatomy differs, individual gyral patterns are flattened (3) and aligned with a group average gyral pattern (4). If a colour code indexing 3D cortical locations is flowed along with the same deformation field (5), a crisp group average model of the cortex can be made (6), relative to which individual gyral pattern differences (7), group variability (8) and cortical asymmetry (9) can be computed. Once individual gyral patterns are aligned to the mean template, differences in grey matter distribution or thickness (10) can be mapped, pooling data from homologous regions of cortex. Correlations can be mapped between disease-related deficits and genetic risk factors (11). Maps may also be generated visualizing linkages between deficits and clinical symptoms, cognitive scores, and medication effects.

cognitive decline, as measured by MMSE scores (r = 0.7; P < 0.05). In capturing brain change, deformation-based methods can be complementary to voxel-based morphometric methods (Ashburner & Friston, 2000; Good et al., 2001a,b), and methods that estimate whole brain atrophic rates (Subsol et al., 1997; Calmon & Roberts, 2000; Collins et al., 1995; Smith et al., 2001). Voxel-based methods typically

use a simple pixel-by-pixel subtraction of scan intensities registered rigidly across time. Tensor-based methods (Thompson et al., 2000a,b), however, can distinguish local from global effects, and true tissue loss from shifts in anatomy. These can confound image subtraction methods. These dynamic maps show promise in charting the dynamic progress of Alzheimer’s disease, and reveal a changing pattern of deficits. Applications of these dynamic


153

154


mapping approaches include measuring the statistics of brain growth in development (Thompson et al., 2000a,b), and measuring tumor response to novel chemotherapy agents (Haney et al., 2001). By building probability densities on registered tensor fields (e.g. Thompson et al., 2001a–e), a quantitative framework can also be established to detect normal and aberrant brain change in dementia (Thompson et al., 2002), and its modulation in clinical trials. Efforts are currently underway to use these tools to map brain regions where deficit patterns are modified by drug treatment and known risk genotypes.

Conclusion Encoding patterns of anatomical variation in disease presents significant challenges. By describing an atlasing scheme that treats intensity and geometric variation separately, we described the creation of well-resolved image templates and probabilistic models of anatomy that reflect the average morphology of a group. The continual refinement of anatomic templates is likely to be leveraged by algorithms for population-based morphometry in large image databases (Fig. 8.20), and by next-generation probabilistic atlases. Atlas data on anatomic variability can also act as Bayesian prior information to guide algorithms for automated image registration and labelling. The resulting atlases are expandable in every respect, and may be stratified into subpopulations according to clinical, demographic or genetic criteria. We also described approaches for creating and averaging brain models. These techniques produce statistical maps of group differences, abnormalities, and patterns of variation and asymmetry (Fig. 8.20). These maps and models are key components of disease-specific brain atlases. We also described registration algorithms that transfer post-mortem maps into an atlas, to correlate them with functional and metabolic data. The result is a multi-modality atlas that relates cognitive and functional measures with the cellular and pathologic hallmarks of the disease. Accurate mapping of grey matter changes in a living population with AD holds significant promise for genetic, longitudinal and interventional studies of dementia. In any study where staging of the disease is required, the ability to calibrate grey matter integrity against a reference population is essential. As well as disease-specific atlases reflecting brain structure in dementia, research is under way to build dynamic brain atlases that retain probabilistic information on growth rates in development and degeneration. Refinement of these atlas systems to support dynamic and

disease-specific data should generate an exciting framework to investigate variations in brain structure and function in large human populations.

Acknowledgements This work was supported by research grants from the National Center for Research Resources (P41 RR13642), the National Library of Medicine (LM/MH05639), National Institute of Neurological Disorders and Stroke and the National Institute of Mental Health (NINDS/NIMH NS38753), and by a Human Brain Project grant to the International Consortium for Brain Mapping, funded jointly by NIMH and NIDA (P20 MH/DA52176). Additional support provided by NIH R21 grants EB01561 and RR19771 (to P. T.).

REFERENCES Ackerman, M. J., Yoo, T & Jenkins, D. (2001). From data to knowledge – the Visible Human Project continues. Medinfo, 10(2): 887–90. Amit, Y., Grenander, U., Piccioni, M. (1991). Structural image restoration through deformable templates, J Am Stat Assoc, 86(414):376–86. Ashburner, J. & Friston, K. J. (2000). Voxel-based morphometry – the methods. Neuroimage, 11, (6):805–21. Ashburner, J. & Friston, K. J. (2001). Why voxel–based morphometry should be used. Neuroimage, 14(6):1238–43. Ashburner, J., Neelin, P., Collins, D. L., Evans, A. C. & Friston, K. J. (1997). Incorporating prior knowledge into image registration, Neuroimage, 6(4):344–52. Ashburner, J., Hutton, C., Frackowiak, R., Johnsrude, I., Price, C. & Friston, K. (1998). Identifying global anatomical differences: deformation-based morphometry. Hum Brain Mapping, 6 (5–6):348–57. Bajcsy, R. & Kovacic, S. (1989). Multiresolution Elastic Matching. Comp Vision, Graphics Image Proc, 46:1–21. Berg, L., McKeel, D. W. Jr, Miller, J. P., Baty, J. & Morris, J. C. (1993). Neuropathological indexes of Alzheimer’s disease in demented and nondemented persons aged 80 years and older. Arch Neurol, 50(4):349–58. Bohm, C., Greitz, T., Kingsley, D., Berggren, B. M. & Olsson, L. (1983). Adjustable computerized brain atlas for transmission and emission tomography. Am J Neuroradiol, 4:731–3. Bookstein, F. L. (1997). Landmark methods for forms without landmarks: morphometrics of group differences in outline shape, Med Image Anal, 1(3):225–43. Bookstein, F. L. (2001). ‘Voxel-based morphometry’ should not be used with imperfectly registered images. Neuroimage, 14(6):1454–62. Brodmann, K. (1909). Vergleichende Lokalisationslehre der Grosshirnrinde in ihren Prinzipien dargestellt auf Grund des



Zellenbaues, Barth, Leipzig. In: Some papers on the cerebral cortex. [Translated as: Brodmann, K. (1960) On the comparative localization of the cortex, pp. 201–230. Springfield, IL: Thomas.] Broit, C. (1981). Optimal registration of deformed images, PhD dissertation, Univ Pennsylvania, USA. Brun, A. & Englund, E. (1981). Regional pattern of degeneration in Alzheimer’s disease: neuronal loss and histopathologic grading. Histopathology, 5:549–64. Calmon, G. & Roberts, N. (2000). Automatic measurement of changes in brain volume on consecutive 3D MR images by segmentation propagation. Magn Reson Imaging, 18(4): 439–53. Cannon, T. D., Thompson, P. M., van Erp, T. et al. (2002). Cortex Mapping Reveals Regionally Specific Patterns of Genetic and Disease-Specific Gray-Matter Deficits in Twins Discordant for Schizophrenia, Proc Nat Acad Sci USA, 99 (5): 3228– 33. Cao, J. & Worsley, K. J. (1999). The geometry of the hotelling’s T-squared random field with applications to the detection of shape changes, Ann Statist, 27: 925–42. Carpenter, W. T., Buchanan, R. W., Kirkpatrick, B., Tamminga, C. & Wood, F. (1993). Strong inference, theory testing, and the neuroanatomy of schizophrenia. Arch Gen Psychiatry, 50: 825–31. Chan, D., Fox, N. C., Jenkins, R., Scahill, R. I., Crum, W. R. & Rossor, M. N. (2001). Rates of global and regional cerebral atrophy in AD and frontotemporal dementia. Neurology, 57(10): 1756–63. Christensen, G. E., Rabbitt, R. D. & Miller, M. I. (1993). A deformable neuroanatomy textbook based on viscous fluid mechanics, 27th Ann. Conf. on Inf. Sciences and Systems, 211–216. Christensen, G. E., Miller, M. I., Marsh, J. L. & Vannier, M. W. (1995). Automatic analysis of medical images using a deformable textbook, Proc Comp Assist Radiol, Springer, Berlin, 152–7. Christensen, G. E., Rabbitt, R. D. & Miller, M. I. (1996). Deformable templates using large deformation kinematics. IEEE Trans. on Image Processing, 5(10):1435–47. Clinton, J., Blackman, S. E. A., Royston, M. C. & Robert, G. W. (1994). Differential synaptic loss in the cortex in AD: a study using archival material. NeuroReport, 5:497–500. Collins, D. L., Holmes, C. J., Peters, T. M. & Evans, A. C. (1995). Automatic 3D model-based neuroanatomical segmentation. Human Brain Mapping, 3:190–208. Cook, M. J., Free, S. L., Fish, D. R., Shorvon, S. D., Straughan, K. & Stevens, J. M. (1994). Analysis of cortical patterns, In: Magnetic Resonance Scanning and Epilepsy. (Shorvon SD, ed), 263–74. New York: Plenum. Csernansky, J. G, Joshi, S., Wang, L. et al. (1998). Hippocampal morphometry in schizophrenia by high dimensional brain mapping. Proc Natl Acad Sci USA, 95(19):11406–11. Csernansky, J. G., Wang, L., Joshi, S. et al. (2000). Early DAT is distinguished from aging by high-dimensional mapping of the hippocampus. Dementia of the Alzheimer type. Neurology, 55(11):1636–43.

Cuénod, C. A., Denys, A., Michot, J. L. et al. (1993). Amygdala atrophy in AD: an in vivo magnetic resonance study. Arch Neurol, 50:941–5. Cunningham, D. J. (1892). Contribution to the surface anatomy of the cerebral hemispheres, Cunningham Memoirs (R Irish Acad.), 7:372. Davatzikos, C. (1996). Spatial normalization of 3D brain images using deformable models, J Comp Assist Tomog, 20(4):656– 65. Davatzikos, C., Vaillant, M., Resnick, S. M., Prince, J. L., Letovsky, S. & Bryan, R. N. (1996). A computerized approach for morphological analysis of the corpus callosum, J Comp Assist Tomogr, 20(1):88–97. Davidson, R. J. & Hugdahl, K. (1994). Brain Asymmetry. Cambridge, MA: MIT Press. De La Monte, S. M. (1989). Quantification of cerebral atrophy in pre-clinical and end-stage AD. Ann Neurol, 25:450–9. DeCarli, C., Haxby, J. V., Gillette, J. A., Teichberg, D., Rapoport, S. I. & Schapiro, M. B. (1992). Longitudinal changes in lateral ventricular volume in patients with dementia of the Alzheimer type. Neurology, 42(10):2029–36. Dinov, I. D., Mega, M. S., Thompson, P. M. et al. (2000). Analyzing Functional Brain Images in a Probabilistic Atlas: A Validation of Sub-Volume Thresholding, J Comp Assist Tomog, 24(1): 128–38. Drury, H. A. & Van Essen, D. C. (1997). Analysis of functional specialization in human cerebral cortex using the visible man surface based atlas. Hum. Brain Mapping, 5:233–7. Duara, R., Kushch, A., Gross-Glenn, K. et al. (1991). Neuroanatomic differences between dyslexic and normal readers on magnetic resonance imaging scans. Arch Neurol., 48(4):410–16. Dupuis, P., Grenander, U. & Miller, M. I. (1998). Variational problems on flows of diffeomorphisms for image matching. QUART APPL MATHEMATICS, 56 (3):587–600. ¨ Oberflachen Anatomie der Grosshirn Eberstaller, O. (1884). Zur Hemisphaeren. Wien Med Bl, 7:479,642,644. Eidelberg, D. & Galaburda, A. M. (1982). Symmetry and asymmetry in the human posterior thalamus: I. Cytoarchitectonic analysis in normal persons. Arch Neurol, 39:325–32. Evans, A. C., Dai, W., Collins, D. L., Neelin, P. & Marrett, S. (1991). Warping of a computerized 3D atlas to match brain image volumes for quantitative neuroanatomical and functional analysis SPIE Med. Imaging, 1445:236–47. Evans, A. C., Collins, D. L., Neelin, P., MacDonald, D., Kamber, M. & Marrett, T. S. (1994). Three-dimensional correlative imaging: applications in human brain mapping, in: Functional Neuroimaging: Technical Foundations, R. W. Thatcher, M. Hallett, T. Zeffiro, E. R. John, M. Huerta, eds., 145–62. Falkai, P., Bogerts, B., Greve, B. et al. (1992). Loss of sylvian fissure asymmetry in schizophrenia. A quantitative post mortem study. Schizophr Res, 7(1):23–32. Fischl, B., Sereno, M. I., Tootell, R. B. H. & Dale, A. M. (1999). High-resolution inter-subject averaging and a coordinate system for the cortical surface. Hum Brain Mapp, 8(4):272– 84.


155

156


Fox, N. C., Freeborough, P. A. & Rossor, M. N. (1996). Visualisation and quantification of rates of atrophy in AD. Lancet, 348: 94–7. Freeborough, P. A. & Fox, N. C. (1998). Modeling brain deformations in Alzheimer disease by fluid registration of serial 3D MR images. J Comp Assist Tomog, 22(5), 838–43. Friedland, R. P. & Luxenberg, J. (1988). Neuroimaging and dementia. In Clinical Neuroimaging: Frontiers in Clinical Neuroscience, Vol. 4, ed, W. H. Theodore, pp. 139–63. New York: Allan Liss. Friston, K. J., Holmes, A. P., Worsley, K. J., Poline, J. P., Frith, C. D. & Frackowiak, R. S. J. (1995). Statistical parametric maps in functional imaging: a general linear approach. Human Brain Mapping, 2: 189–210. Fuh, J. L., Mega, M. S., Thompson, P. M., Cummings, J. L. & Toga, A. W. (1997). Cortical complexity maps and cognition in Alzheimer’s disease. Proc. 122nd Annual Meeting of the American Neurological Association, 1997. Galaburda, A. M. & Geschwind, N. (1981). Anatomical asymmetries in the adult and developing brain and their implications for function, Adv Pediatr, 28:271–92. Ge, Y., Fitzpatrick, J. M., Kessler, R. M. & Jeske-Janicka, M. (1995). Intersubject brain image registration using both cortical and subcortical landmarks. SPIE Image Proc, 2434:81–95. Gee, J. C. & Bajscy, R. K. (1998). Elastic matching: Continuummechanical and probabilistic analysis In A. W. Toga, ed., Brain Warping, Academic Press, San Diego. Gee, J. C., Reivich, M. & Bajcsy, R. (1993). Elastically deforming an atlas to match anatomical brain images. J Comp Assist Tomogr, 17(2):225–36. Gee, J. C., LeBriquer, L., Barillot, C., Haynor, D. R. & Bajcsy, R. (1995). Bayesian approach to the brain image matching problem, Institute for Research in Cognitive Science Technical Report 95–08, April 1995. Geschwind, N. & Levitsky, W. (1968). Human brain: left–right asymmetries in temporal speech region. Science, 161: 186. Good, C. D., Ashburner, J. & Frackowiak, R. S. J. (2001a). Computational neuroanatomy: new perspectives for neuroradiology. Rev Neurol (Paris), 157(8–9 Pt 1):797–806. Good, C. D., Johnsrude, I. S., Ashburner, J., Henson, R. N., Friston, K. J. & Frackowiak, R. S. J. (2001b). A voxel–based morphometric study of ageing in 465 normal adult human brains. Neuroimage, 14(1 Pt 1):21–36. Greitz, T., Bohm, C., Holte, S. & Eriksson, L. (1991). A computerized brain atlas: construction, anatomical content and application. J Comp Assist Tomogr, 15(1):26–38. Grenander, U. (1976). Pattern synthesis: lectures in pattern theory. Appl Math Sci 13, Springer-Verlag. Grenander, U. & Miller, M. I. (1994). Representations of knowledge in complex systems, J Roy Statist Soci B, 56(4):549–603. (1998). Computational anatomy: an emerging discipline, Technical Report, Dept. of Mathematics, Brown University. Haller, J. W., Banerjee, A., Christensen, G. E. et al. (1997). Threedimensional hippocampal MR morphometry with highdimensional transformation of a neuroanatomic atlas. Radiology, 202(2):504–10.

Haney, S., Thompson, P. M., Cloughesy, T. F., Alger, J. R. & Toga, A. W. (2001). Tracking tumor growth rates in patients with malignant gliomas: a test of two algorithms. Am J Neuroradiol, 22(1): 73–82. ¨ K. & Molnár, P. (1996). Neuroanatomy and neuropatholHegedus, ogy on the Internet. Department of Neurology, University Medical School, Debrecen, Hungary, http://www.neuropat. dote.hu/. Höhne, K. H., Bomans, M., Pommert, A. et al. (1990). 3D Visualization of tomographic volume data using the generalized voxel model, Vis Comput, 6:28–36. Höhne, K. H., Bomans, M., Riemer, M., Schubert, R., Tiede, U. & Lierse, W. (1992). A 3D anatomical atlas based on a volume model. IEEE Comp Graphics Appl, 12:72–8. Holman, L. B., Chandak, P. K. & Garada, B. M. (1994). Atlas of Brain Perfusion SPECT, http://brighamrad.harvard.edu/education/ online/BrainSPECT/BrSPECT.html Hubbard, B. M. & Anderson, J. M. (1981). A quantitative study of cerebral atrophy in old age and senile dementia. J Neurol Sci, 50:135–45. Ingvar, M., Eriksson, L., Greitz, T. et al. (1994). Methodological aspects of brain activation studies – cerebral blood flow determined with [0–15]-Butanol and positron emission tomography. J Cereb Blood Flow and Metab, 14(4):628–638. Iosifescu, D. V., Shenton, M. E., Warfield, S. K. et al. (1997). An automated registration algorithm for measuring MRI subcortical brain structures. NeuroImage, 6(1):13–25. Jack, C. R. Jr, Petersen, R. C., Xu, Y., O’Brien, P. C., Smith, G. E., Ivnik, R. J., Tangalos, E. G. & Kokmen, E. (1998). Rate of medial temporal lobe atrophy in typical aging and Alzheimer’s disease. Neurology, 51:993–999.[Abstract]. Jacobs, R. E. & Fraser, S. E. (1994). Magnetic resonance microscopy of embryonic cell lineages and movements. Science, 263(5147):681–4. Janke, A. L., Zubicaray, Gd., Rose, S. E., Griffin, M., Chalk, J. B. & Galloway, G. J. (2001). 4D deformation modeling of cortical disease progression in Alzheimer’s dementia. Magn Reson Med, 46(4):661–6. Jernigan, T. L, Salmon, D., Butter, N. et al. (1991). Cerebral structure on MRI, Part II: specific changes in Alzheimer’s and Huntington’s diseases. Biol Psychiat, 29:68–81. Johnson, K. A. (1996). The Harvard Whole Brain Atlas, http://www.med.harvard.edu/AANLIB/home.html Johnson, K. A., Jones, K., Holman, B. L. et al. (1998). Preclinical prediction of Alzheimer’s disease using SPECT. Neurology, 50:1563–71.[Abstract] Joshi, S., Miller, M. I. & Grenander, U. (1998). On the geometry and shape of brain sub-manifolds. Int J Pattern Recogn Artif Intell, 11:1317–43. Kennedy, D. N., Lange, N., Makris, N., Bates, J., Meyer, J. & Caviness, V. S. Jr. (1998). Gyri of the human neocortex: an MRI-based analysis of volume and variance. Cereb Cortex, 8(4): 372–84. Kikinis, R., Shenton, M. E., Gerig, G. et al. (1994). Temporal lobe sulco-gyral pattern anomalies in schizophrenia: an in vivo



MR three-dimensional surface rendering study, Neurosci Lett, 182:7–12. Kikinis, R., Shenton, M. E., Iosifescu, D. V. et al. (1996). A digital brain atlas for surgical planning, model-driven segmentation, and teaching, IEEE Trans. Visualization and Comp Graphics, 2(3):232–41. Kulynych, J. J., Vladar, K., Jones, D. W. & Weinberger, D. R. (1996). Superior temporal gyrus volume in schizophrenia: a study using MRI morphometry assisted by surface rendering. Am J Psychiat, 153:50–6. Laakso, M. P., Frisoni, G. B., Kononen, M. et al. (2000a). Hippocampus and entorhinal cortex in frontotemporal dementia and Alzheimer’s disease: a morphometric MRI study. Biol Psychiatry, 47(12):1056–63. Laakso, M. P., Lehtovirta, M., Partanen, K., Riekkinen, P. J. & Soininen, H. (2000b). Hippocampus in AD: a 3-year follow-up MRI study. Biol Psychiat, 47:557–61. Le Goualher, G., Procyk, E., Collins, D. L., Venugopal, R., Barillot, C. & Evans, A. C. (1999). Automated extraction and variability analysis of sulcal neuroanatomy, IEEE Trans Med Imaging, 18(3):206–17. Leonard, C. M. (1996). Structural variation in the developing and mature cerebral cortex: noise or signal? In Developmental Neuroimaging: Mapping the Development of Brain and Behavior, eds, R. W. Thatcher, G. Reid Lyon, J. Rumsey & N. Krasnegor, Academic Press, San Diego 207–31. Loewenstein, D. A., Barker, W. W., Chang, J. Y. et al. (1989). Predominant left hemisphere metabolic dysfunction in dementia. Arch Neurol, 46:146–52. Mai, J., Assheuer, J. & Paxinos, G. (1997). Atlas of the Human Brain, Academic Press, San Diego. Manceaux-Demiau, A., Bryan, R. N. & Davatzikos, C. (1998). A probabilistic ribbon model for shape analysis of the cerebral sulci: application to the central sulcus. J Comput Assist Tomogr, 22:962–71. Mangin. J.-F., Frouin, V., Bloch, I., Regis, J. & López–Krahe, J. (1995). From 3D magnetic resonance images to structural representations of the cortex topography using topology-preserving deformations. J Math Imaging Vision, 5: 297–318. Mansour, A., Fox, C. A., Burke, S., Akil, H. & Watson, S. J. (1995). Immunohistochemical localization of the cloned Mu opioid receptor in the rat CNS, J Chem Neuroanat, 8(4):283–305. Matsui, T. & Hirano, A. (1978). An Atlas of the Human Brain for Computerized Tomography. Igako-Shoin, Japan. Mazziotta, J. C., Toga, A. W., Evans, A. C., Fox, P. & Lancaster, J. (1995). A probabilistic atlas of the human brain: theory and rationale for its development, NeuroImage, 2: 89–101. Mazziotta, J. C. et al. (2001). A probabilistic atlas and reference system for the human brain [Invited Paper], J Roy Soc [in press]. Mega, M. S., Chen, S., Thompson, P. M. et al. (1997). Mapping pathology to metabolism: coregistration of stained whole brain sections to PET in Alzheimer’s disease. NeuroImage, 5: 147–153.

Mega, M. S., Dinov, I. D., Lee, L. et al. (1998a). Dissecting neural networks underlying the retrieval deficit from the amnestic memory disorder using [99mTc]-HMPAO-SPECT. Proc Am Behav Neurol Soc. Mega, M. S., Thompson, P. M., Cummings, J. L. et al. (1998b). Sulcal variability in the Alzheimer’s brain: correlations with cognition. Neurology, 50:145–51. Mega, M. S., Chu, T., Mazziotta, J. C. et al. (1999). Mapping biochemistry to metabolism: FDG-PET and beta-amyloid burden in Alzheimer’s disease. NeuroReport, 10(14):2911–17. Mega, M. S., Fennema-Notestine, C., Dinov, I. D. et al. (2000a). Construction, testing, and validation of a sub-volume probabilistic human brain atlas for the elderly and demented populations. 6th International Conference on Functional Mapping of the Human Brain, San Antonio, Texas. Mega, M. S., Thompson, P. M., Dinov, I. D., Toga, A. W. & Cummings, J. L. (2000b). The UCLA Alzheimer brain atlas project: structural and functional applications. In Proceedings of the 2000 World. Alzheimer’s Congress, Washington, DC. Mega, M. S., Thompson, P. M., Toga, A. W., Cummings, J. L. (2000c). Brain mapping in dementia. In Brain Mapping: The Disorders, A. W. Toga, J. C. Mazziotta eds., Academic Press, San Diego. Meltzer, C. C. & Frost, J. J. (1994). Partial volume correction in emission-computed tomography: focus on Alzheimer disease. In Functional Neuroimaging: Technical Foundations, R. W. Thatcher, M. Hallett, T. Zeffiro, E. R. John & M. Huerta, eds., Academic Press, San Diego, pp. 163–70. Miller, M. I., Christensen, G. E., Amit, Y. & Grenander, U. (1993). Mathematical textbook of deformable neuroanatomies. Proc Natl Acad Sci USA, 90:11944–8. Miller, M. I., Younes, L. & Trouve, A. (2002). On the Metrics and Euler– Lagrange Equations of Computational Anatomy, [in press]. Murphy, D. G. M., DeCarli, C. D., Daly, E. et al. (1993). Volumetric magnetic resonance imaging in men with dementia of the Alzheimer type: correlations with disease severity. Biol Psychiat, 34:612–21. Narr, K. L., Cannestra, A. F., Thompson, P. M., Sharma, T. & Toga, A. W. (1998). Morphological variability maps of the corpus callosum and fornix in schizophrenia. 4th Int. Conf. on Human Brain Mapping, NeuroImage, 7(4):S506. Narr, K. L., Thompson, P. M., Sharma, T. et al. (2001a). Threedimensional mapping of temporo-limbic regions and the lateral ventricles in schizophrenia: gender effects. Biol Psychiatry, 50(2):84–97. Narr, K. L., Thompson, P. M., Sharma, T. et al. (2001b). 3D mapping of gyral shape and cortical surface asymmetries in schizophrenia: gender effects. Am J Psychiatry 1, 158(2):244– 55. Nowinski, W. L., Yang, G. L. & Yeo, T. T. (2000). Computer-aided stereotactic functional neurosurgery enhanced by the use of the multiple brain atlas database. IEEE Trans Med Imaging, 19(1):62–9. Ono, M., Kubik, S. & Abernathey, C. D. (1990). Atlas of the Cerebral Sulci. Stuttgart: Thieme.


157

158


Paus, T., Tomaioulo, F., Otaky, N. et al. (1996). Human cingulate and paracingulate sulci: pattern, variability, asymmetry and probabilistic map, Cereb Cortex, 6:207–14. Pitiot, A., Thompson, P. M. & Toga, A. W. (2002). Spatially and temporally adaptive elastic template matching. IEEE-TMI. Rajkowska, G. & Goldman-Rakic, P. (1995). Cytoarchitectonic definition of pre-frontal areas in the normal human cortex: II. Variability in locations of areas 9 and 46 and relationship to the Talairach coordinate system. Cereb Cortex, 5:323–37. [Abstract] Rizzo, G., Gilardi, M. C., Prinster, A., Grassi, F., Scotti, G., Cerutti, S. & Fazio, F. (1995). An elastic computerized brain atlas for the analysis of clinical PET/SPET data. Eur J Nucl Med, 22(11): 1313–18. Roberts, G. W., Leigh, P. N. & Weinberger, D. R. (1993). Neuropsychiatric Disorders, Ch 2(1). Gower Medical Publishers. Royackkers, N., Desvignes, M. & Revenu, M. (1996). Construction automatique d’un atlas adaptatif des sillons corticaux, ORASIS 96, Clermont-Ferrand, pp. 187–192. Sandor, S. R. & Leahy, R. M. (1994). Matching deformable atlas models to pre-processed magnetic resonance brain images. Proc IEEE Conf. on Image Processing, 3:686–690. (1995). Towards automated labeling of the cerebral cortex using a deformable atlas, In Y. Bizais, C. Barillot, R. Di Paola, [eds.]. Info Proc in Med Imag, 127–38. Schaltenbrand, G. & Wahren, W. (1977). Atlas for Stereotaxy of the Human Brain, 2nd edn., Stuttgart: Thieme. Schiemann, T., Nuthmann, J., Tiede, U. & Höhne, K. H. (1996). Segmentation of the visible human for high-quality volume-based visualization, Vis Biomed Comp, 4:13–22. Schmidt, R. (1992). Comparison of magnetic resonance imaging in Alzheimer’s disease, vascular dementia and normal aging. Eur Neurol, 32(3):164–9. Seitz, R. J., Bohm, C., Greitz, T. et al. (1990). Accuracy and precision of the computerized brain atlas programme for localization and quantification in positron emission tomography. J Cereb Blood Flow Metab, 10:443–57. Siegel, B. V. Jr, Shihabuddin, L., Buchsbaum, M. S., Starr, A., Haier, R. J., Valladares & Neto, D. C. (1996). Gender differences in cortical glucose metabolism in Alzheimer’s disease and normal aging. J Neuropsychiatry Clin Neurosci, 8(2):211–14. Small, G. W., Ercoli, L. M., Silverman, D. H. et al. (2000). Cerebral metabolic and cognitive decline in persons at genetic risk for Alzheimer’s disease. Proc Natl Acad Sci USA, 97(11): 6037–42. Smith, G. E. (1907). A new topographical survey of the human cerebral cortex, being an account of the distribution of the anatomically distinct cortical areas and their relationship to the cerebral sulci. J Anat, 41:237–54. Smith, S. M., De Stefano, N., Jenkinson, M. & Matthews, P. M. (2001). Normalized accurate measurement of longitudinal brain change. J Comput Assist Tomogr, 25(3):466–75. Sobire, G., Goutieres, F., Tardieu, M., Landrieu, P. & Aicardi, J. (1995). Extensive macrogyri or no visible gyri: distinct clinical, electroencephalographic, and genetic features accord-

ing to different imaging patterns. Neurology, 45:1105–11. [Abstract] Sowell, E. R., Thompson, P. M., Tessner, K. D. & Toga, A. W. (2001). Accelerated brain growth and cortical gray matter thinning are inversely related during post-adolescent frontal lobe maturation. J Neurosci, 21(22):8819–29. Steinmetz, H., Furst, G. & Freund, H.-J. (1989). Cerebral cortical localization: application and validation of the proportional grid system in MR imaging, J Comp Assist Tomogr, 13(1): 10–19. (1990). Variation of perisylvian and calcarine anatomic landmarks within stereotaxic proportional coordinates, Am J Neuroradiol, 11(6):1123–30. Stewart, J. E., Broaddus, W. C. & Johnson, J. H. (1996). Rebuilding the visible man. Vis Biomed Comp, 4:81–6. St-Jean, P., Sadikot, A. F., Collins, L. et al. (1998). Automated atlas integration and interactive three-dimensional visualization tools for planning and guidance in functional neurosurgery. IEEE Trans Med Imaging, 17(5):672–80. Strauss, E., Kosaka, B. & Wada, J. (1983). The neurobiological basis of lateralized cerebral function, a review. Hum neurobiol, 2(3):115–27. Studholme, C., Cardenas, V., Schuff, N., Rosen, H., Miller, B. & Weiner, M. W. (2001). Detecting spatially consistent structural differences in Alzheimer’s and fronto temporal dementia using deformation morphometry. MICCAI, 41–8. Subsol, G., Roberts, N., Doran, M., Thirion, J. P. & Whitehouse, G. H. (1997). Automatic analysis of cerebral atrophy, Magn Reson Imaging, 15(8):917–27. Talairach, J. & Szikla, G. (1967). Atlas d’Anatomie Stereotaxique du Telencephale: Etudes Anatomo-Radiologiques. Paris: Masson & Cie. Talairach, J. & Tournoux, P. (1988). Co-planar Stereotaxic Atlas of the Human Brain. New York: Thieme. Thirion, J.-P. & Calmon, G. (1997). Deformation analysis to detect and quantify active lesions in 3D medical image sequences, INRIA Technical Report 3101. Thirion, J.-P., Prima, S. & Subsol, S. (1998). Statistical analysis of dissymmetry in volumetric medical images. Med image Anal [in press]. Thompson, P. M. & Toga, A. W. (1996). A surface-based technique for warping 3-dimensional images of the brain. IEEE Trans Med Imaging, 15(4):1–16. (1998). Anatomically-driven strategies for high-dimensional brain image warping and pathology detection. In Brain Warping, (A. W. Toga), ed., Academic Press, 311–36. (2002). A Framework for Computational Anatomy [Invited Paper]. Comp Visualization Sci 5:1–12. Thompson, P. M., Schwartz, C., Lin. R. T., Khan, A. A. & Toga, A. W. (1996a). 3D statistical analysis of sulcal variability in the human brain. J Neurosci, 16(13):4261–74. Thompson, P. M., Schwartz, C. & Toga, A. W. (1996b). Highresolution random mesh algorithms for creating a probabilistic 3D surface atlas of the human brain, NeuroImage, 3: 19–34.



Thompson, P. M., MacDonald, D., Mega, M. S., Holmes, C. J., Evans, A. C. & Toga, A. W. (1997). Detection and mapping of abnormal brain structure with a probabilistic atlas of cortical surfaces. J Comp Assist Tomogr, 21(4):567–81. Thompson, P. M., Giedd, J. N., Blanton, R. E. et al. (1998a). Growth patterns in the developing human brain detected using continuum-mechanical tensor maps and serial MRI. 5th Int. Conf. on Human Brain Mapping, Montreal, Canada. Thompson, P. M., Moussai, J., Khan, A. A. et al. (1998b). Cortical variability and asymmetry in normal aging and Alzheimer’s disease. Cereb Cortex, 8(6):492–509. Thompson, P. M., Woods, R. P., Mega, M. S. & Toga, A. W. (1999). Mathematical/computational challenges in creating population-based brain atlases. Hum Brain Mapping, 8(2). Thompson, P. M., Mega, M. S., Narr, K. L., Sowell, E. R., Blanton, R. E. & Toga, A. W. (2000a). Brain image analysis and atlas construction, Invited Chapter, in: Fitzpatrick M [ed.], SPIE Handbook on Medical Image Analysis, Society of Photo-Optical Instrumentation Engineers (SPIE) Press. Thompson, P. M., Mega, M. S. & Toga, A. W. (2000b). Disease-specific brain atlases, Invited Chapter. In Brain Mapping: The Disorders, A. W. Toga, J. C. Mazziotta, [eds.], Academic Press. Thompson, P. M., Cannon, T. D., Narr, K. L. et al. (2001a). Genetic influences on brain structure. Nat Neurosci, 4(12): 1253–8. Thompson, P. M., de Zubicaray, G., Janke, A. L. et al. (2001b). Detecting dynamic (4D) profiles of degenerative rates in Alzheimer’s disease patients, using high-resolution tensor mapping and a brain atlas encoding atrophic rates in a population. 7th Annual Meeting of the Organization for Human Brain Mapping, Brighton, England. Thompson, P. M., Mega, M. S., Vidal, C., Rapoport, J. L. & Toga, A. W. (2001c). Detecting disease-specific patterns of brain structure using cortical pattern matching and a populationbased probabilistic brain atlas. IEEE Conference on Information Processing in Medical Imaging (IPMI), UC Davis, 2001. In Lecture Notes in Computer Science (LNCS), 2082:488–501, M. Insana, R. Leahy [eds.], Springer-Verlag. Thompson, P. M., Mega, M. S., Woods, R. P. et al. (2001d). Early cortical change in Alzheimer’s disease detected with a diseasespecific population-based brain atlas. Cereb Cortex, 11(1): 1–16. Thompson, P. M., Vidal, C. N., Giedd, J. N. et al. (2001e). Mapping adolescent brain change reveals dynamic wave of accelerated gray matter loss in very early-onset schizophrenia. Proc Nat Acad Sci USA, 98(20):11650–5. Thompson, P. M., Narr, K. L., Blanton, R. E. & Toga, A. W. (2002). Mapping structural alterations of the corpus callosum during brain development and degeneration. In M. Iacoboni, E. Zaidel, eds., The Corpus Callosum, Kluwer Academic Press [in press]. Thurfjell, L., Bohm, C., Greitz, T. & Eriksson, L. (1993). Transformations and algorithms in a computerized brain atlas. IEEE Trans Nucl Sci, 40(4), pt. 1:1167–91.

Tiede, U., Bomans, M., Hohne, K. H. et al. (1993). A computerized three-dimensional atlas of the human skull and brain. Am J Neuroradiol, 14(3):551–9; discussion 560–1. Toga, A. W. (1998). Brain Warping, Academic Press, San Diego. Toga, A. W. & Mazziotta, J. C. (1996). Brain Mapping: The Methods, 1st Edition. Academic Press. Toga, A. W. & Thompson, P. M. (1997). Measuring, mapping, and modeling brain structure and function, SPIE Medical Imaging Symposium, Newport Beach, CA, USA; SPIE Lecture Notes Volume 3033. (1998a). An introduction to brain warping. In Brain Warping, A. W. Toga, ed., Academic Press, San Diego. (1998b). Multimodal brain atlases. In Advances in Biomedical Image Databases, S. Wong, ed., pp. 53–88. Kluwer Academic Press. Toga, A. W., Ambach, K., Quinn, B., Hutchin, M. & Burton, J. S. (1994). Postmortem anatomy from cryosectioned whole human brain. J Neurosci Methods, 54(2):239–52. Toga, A. W., Goldkorn, A., Ambach, K., Chao, K., Quinn, B. C. & Yao, P. (1997). Postmortem cryosectioning as an anatomic reference for human brain mapping. Comput Med Imaging Graph, 21(2):131–41. Vaillant, M. & Davatzikos, C. (1997). Finding parametric representations of the cortical sulci using an active contour model. Med Image Anal, 1(4):295–315. Van Buren, J. M. & Borke, R. C. (1972). Variations and Connections of the Human Thalamus, Vols. 1 & 2. New York: Springer. Van Buren, J. M. & Maccubin, D. (1962). An outline atlas of human basal ganglia and estimation of anatomic variants. J Neurosurg, 19:811–39. Van Essen, D. C., Drury, H. A., Joshi, S. C. & Miller, M. I. (1997). Comparisons between human and macaque using shapebased deformation algorithms applied to cortical flat maps. 3rd Int. Conference on Functional Mapping of the Human Brain, Copenhagen, NeuroImage, 5(4):S41. Warfield, S., Dengler, J., Zaers, J. et al. (1995). Automatic identification of gray matter structures form MRI to improve the segmentation of white matter lesions. Proc Med Robotics & Comp Assist Surg (MRCAS), 55–62. Warfield, S., Robatino, A., Dengler, J., Jolesz, F. & Kikinis, R. (1998). Nonlinear registration and template driven segmentation. In Brain Warping, ed. A. W. Toga, AcademicPress, Ch.4: 67–84. Witelson, S. F. & Kigar, D. L. (1992). Sylvian fissure morphology and asymmetry in men and women: bilateral differences in relation to handedness in men. J Comp Neurol, 323:326–40. Woods, R. P. (1996). Modeling for intergroup comparisons of imaging data. NeuroImage, 4(3):84–94. Woods, R. P., Mazziotta, J. C. & Cherry, S. R. (1993). MRI–PET registration with automated algorithm. J Comput Assist Tomogr, 17:536–46. Woods, R. P., Grafton, S. T., Watson, J. D. G., Sicotte, N. L. & Mazziotta, J. C. (1998). Automated image registration: II. Intersubject validation of linear and nonlinear models. J Comput Assist Tomogr, 22:153–65.


159

160


Zeineh, M. M., Engel, S. A., Thompson, P. M. & Bookheimer, S. (2001). Unfolding the human hippocampus with highresolution structural and functional MRI, New Anat (Anat Rec.), 265(2):111–20. Zhou, Y., Thompson, P. M. & Toga, A. W. (1999). Automatic extraction and parametric representations of cortical sulci. Comp Graphics Appl, 19(3):49–55.

Zijdenbos, A. P. & Dawant, B. M. (1994). Brain segmentation and white matter lesion detection in MR images. Crit Rev Biomed Eng, 22:401–65. Zoumalan, C. I., Mega, M. S., Thompson, P. M., Fuh, J. L., Lindshield, C. & Toga, A. W. (1999). Mapping 3D patterns of cortical variability in normal aging and Alzheimer’s disease. Annual Conference of the American Academy of Neurology.


9 Alzheimer’s disease James H. Morris1 and Zsuzsanna Nagy2 1

Historical Alzheimer’s disease (AD) is the sole cause of about half of cases of dementia in later life and a significant contributor to cognitive decline in a further quarter (Mirra & Hyman, 2002). It is therefore overwhelmingly the most important factor in what has been called the silent epidemic of dementia that is occurring in societies with ageing populations. Since this includes virtually the entire currently developed world, and because the incidence of AD can be confidently expected to rise in all societies where economic progress leads to increased life expectancy, AD ranks with tuberculosis and malaria in economic and social importance. In part for this reason, an enormous amount of research work has been performed in the last two decades and significant progress has been made in our understanding of the pathophysiology of Alzheimer’s disease. It is true that no cure is yet in sight, but progress has been substantial and there are several avenues of research that might generate significant therapeutic advances within the next few years. No account of the disease would be complete without some mention of the man himself. Amaducci (Amaducci et al., 1986) in an interesting article has described some aspects of the early history of the definition of the disease. Alois Alzheimer in 1907, when he was working in the Laboratory of Anatomy at the Psychiatric and Neurologic Clinic in Munich, described the clinical and pathologic features of the disease, which came to be named for him in a demented patient who had become symptomatic in her early 50s (Alzheimer, 1907). It was Alzheimer, along with his coworkers, Bonfiglio and Perusini, who first described neurofibrillary tangles.

John Radcliffe Hospital Oxford, UK 2 Radcliffe Infirmary Oxford, UK

Senile plaques had been identified much earlier (Bloq & Marinesco, 1892). In the same year that Alzheimer described his case, Fischer, who was working with Pick in Prague, reported the presence of senile plaques, then called miliary necrosis or miliary foci, in 12 of 16 cases of what was called senile dementia (Fischer, 1907). Amaducci suggests that it is ‘inexplicable’ that the natural connection between the plaques in Fischer’s cases and those in Alzheimer’s case was not made, but allow that it is just possible that professional and institutional rivalry may have played a role (Amaducci et al., 1986)! Whatever the original reason, from this point on until relatively recently a distinction was made between pre-senile ‘Alzheimer’s disease’ occurring in patients under the age of 65, and brains of those of greater age with similar pathological findings that were at first called ‘senile dementia’. This distinction resulted, over the succeeding decades, in the publication of a considerable body of work directed at demonstrating similarities or differences between ‘Alzheimer’s disease’ and ‘senile dementia’. With the progressive recognition of the impossibility of separating them pathologically, ‘senile dementia’ metamorphosed first into the rather cumbersome, ‘senile dementia of the Alzheimer type’ (SDAT) and finally was incorporated into AD proper. This is one of the (few) circumstances when the current vogue for ‘political correctness’ would be entirely justified in condemning the arbitrary distinction between pre-senile and senile dementia as ‘ageist’!! Ironically, this abolition of an essentially arbitrary distinction based entirely on age, has coincided with the recognition of a genuine genetic etiologic heterogeneity in AD that may ultimately be manifested in some real pathological differences among the different genetic types.


161

162

J. H. Morris and Z. Nagy

Although in the decades following the initial descriptions, there was a considerable body of work published on both AD and senile dementia, for most people of this generation the modern history of AD begins with the series of papers published by Tomlinson and his colleagues starting in 1966 (Blessed et al., 1968; Roth et al., 1966; Tomlinson et al., 1968, 1970). Part of the impetus behind these studies was the continuing reverberations resulting from the separation of AD from senile dementia that was made on such ‘inexplicable’ grounds in 1907. In the intervening decades there had been an ongoing debate on the validity of the distinction between the two, the later stages of which are reflected in summaries of work to date in the first two editions of Greenfield’s Neuropathology (McMenemey, 1958, 1963). This debate effectively received its quietus in the third edition (Corsellis, 1976) in part as a result of the expansion of work on AD that occurred in the late 1960s and early 1970s following Tomlinson’s publications. However, in the early 1960s there was still a widely held, though by no means universal (Corsellis, 1962), belief that there was a poor correlation between the presence of dementia in older people and the concomitant presence of Alzheimer changes. In an interesting pre-echo of the recently kindled debate on the importance of the degree of education in the development of dementia, some of the reported lack of correlation was attributed to factors in the individual’s response and the implicit concept of ‘reserve capacity’ expressed in the thought that a degree of cerebral degeneration that produced dementia in one person might be tolerated in another. The studies of Tomlinson and his co-workers were centred around the comparison of the neuropathological findings in 50 brains from demented elderly (Tomlinson et al., 1970) to those of 28 undemented controls (Tomlinson et al., 1968). In the paper where they recorded the findings in the brains of 50 demented old people, they made the diagnosis of what was then called senile dementia in 25 of the 50 brains and concluded that the same process played a significant role in approximately a further quarter of the cases. In the 28 control brains, they found markedly lower frequency and density of Alzheimer changes and concluded that there was in fact a very significant correlation between the presence of Alzheimer changes and the occurrence of dementia in older patients (Blessed et al., 1968, Roth et al., 1966). The nature and results of their studies effectively set the agenda for much of the work on AD in succeeding decades. Their prescience in identifying so many of the significant problems of dementia in general and AD in particular can be illustrated most effectively by listing some of the major findings and discussion points of these seminal papers.

r The essentially quantitative nature of the differences be-

tween the degree of Alzheimer change in the demented and non-demented aged; r The relative significance of plaques and tangles in the causation of the dementia; r The confusion generated by the use of the term senile dementia in a pathologically imprecise way and the validity or not of its distinction from pre-senile AD; r The problems of getting a genuinely representative sample of the aged population to estimate the frequency of disease in the community; r The possibility that Alzheimer change is more frequent in women; r The difficulty of adequate demonstration that controls have normal mental function; r The degree and significance of cerebrovascular disease and infarction in the aetiology of dementia; r Their suggestion that the diagnosis of arteriosclerotic dementia is made more frequently on clinical grounds than can be confirmed pathologically. As will be seen in subsequent discussion, many of these problems, in some cases hardly changed, are still with us.

Epidemiology Clinical Studies As will be seen again in relation to dementia in association with cerebrovascular disease, disarmingly simple appearing questions such as ‘how common is a disease?’ are difficult to answer precisely. AD is no exception to this rule although it needs to be acknowledged that the purely epidemiological difficulties in establishing the incidence and prevalence of a genetically heterogeneous disease with a large sporadic component that is predominantly a condition of later life are very considerable (Brayne, 1993). Another of the major reasons for the difficulty in establishing the frequency of AD is that, in most cases, the diagnosis of AD cannot be made unequivocally by clinical means alone and hence a clinical diagnosis of AD has to be confirmed by (usually) post-mortem neuropathological examination (Galasko et al., 1994; Kukull et al., 1990; Lopez et al., 1990). Clinical methods are particularly unreliable in elderly patients with mixed dementias where there is a combination of Alzheimer change with a significant burden of cerebrovascular disease or Parkinsonism (Bowler et al., 1998; Gilleard et al., 1992; Klatka et al., 1996; Nagy et al., 1997). In terms of the incidence and prevalence of AD in the community there is currently no large-scale community study of cognitive decline where the clinical estimates

Alzheimer’s disease

of the frequency of AD are substantiated by subsequent neuropathological examination of the brain in the study patients who die (Holmes et al., 1999; Lim et al., 1999; Lopez et al., 2000; Pearl, 1997). One recent study that approaches this ideal is the UK Cognitive Function and Ageing Study, a community study with neuropathology on a representative sample of cases (Esiri et al., 2001). However, this study did not include full clinical assessment. In the absence of a pathologically confirmed study of the frequency of AD in an unselected community population some information can certainly be gathered from the various general surveys of patients with cognitive decline or dementia. There is a considerable body of literature relating to clinical estimates of the incidence and prevalence of cognitive decline and dementia within community populations including populations in France (Dartigues et al., 1991), China (Zhang et al., 1990), Italy (Rocca et al., 1990) and the central (Kokmen et al., 1989) and northeast United States (Framingham) (Bachman et al., 1992). Although there is some variation in individual percentages, meta-analyses (Hofman et al., 1991; Jorm et al., 1987) of these and other surveys are united in their agreement that the prevalence and age-specific incidence rates of cognitive decline in the general population approximately doubles with each decade over the age of 60 and reaches a maximum of around 25–35% of the population in those over 85 years of age (Skoog et al., 1993). One study that produces somewhat different results is the East Boston study (Evans et al., 1989) where more than 3600 people were surveyed (just over 80% of all the age eligible persons in the community population). When fractionated into age cohorts, of those 65–74 years old 3.0% (95% CL 0.8–5.2%) had probable AD, while in the 75–84 year cohort the figures were 18.7% (95% CL 13.2–24.2%) and in the 85–94-year-old group a staggering 47.2%(95% confidence limits 37.0–63.2%) were considered to have probable AD. Katzman (Katzman et al., 1989) discussed the possible reasons for the significantly larger fraction of demented patients in this study, which probably relate, at least in part, to the lack of functional assessment in the estimation of dementia. Although the estimate of an approximately 50% rate for dementia in the over-85s can perhaps be discounted, a rate of 25–35% is still, even given current longevity, a very formidable amount of disease. There is still controversy over whether the increasing incidence continues into the tenth decade but some studies suggest that above the age of 95 years the rate of dementia stabilizes at about 45% (Wernicke & Reischies, 1994). The difficulty in arriving at a definite conclusion is at least partly a result of the small numbers of nonagenarians and centenarians in most surveys of cognitive decline and partly be-

cause of differences in diagnostic criteria adopted, particularly the way in which mildly affected persons are classified. Although quite a wide range of ethnic diversity is included in these studies, they seem in general to suggest that different ethnic groups tend to have a generally similar pattern of increase in age-related cognitive decline and the incidence of Alzheimer’s disease. An exception to this rule may be African populations. In examining aged populations from Africa there are considerable methodological difficulties in establishing chronological age, particularly in very old people, although historical event recording and the use of birth records have been used with claimed success. Studies seem to indicate a low incidence of clinically estimated AD in elderly Nigerians (Ogunniyi et al., 1992) and this has been supported by findings of small numbers of neurofibrillary tangles and plaques (Osuntokun et al., 1992) and amounts of A4 deposition (Osuntokun et al., 1994) in non-demented individuals that are proportionately markedly lower than found in equivalent aged Australian populations (Davies et al., 1988). Interestingly, this reported low incidence in Africans living in Africa is not replicated in those of African descent living in America (Gorelick et al., 1994; Heyman et al., 1991; Schoenberg et al., 1985).

Clinico-pathologic studies Given these estimates of the clinical prevalence of cognitive decline or clinically diagnosed AD, the question is whether and to what extent these clinical rates translate simply into pathological AD. As reported by Mendez et al. (1992) clinico-pathological studies have produced a rather wide range of diagnostic accuracy that is not greatly improved by the use of more systematic diagnostic criteria. A significant consideration is at what point in the disease process the clinical diagnosis is made. In patients with late stage advanced dementia that has been progressing for some time it might be reasonable to expect a considerably greater degree of accuracy in clinical diagnosis than early in the course of the disease, but diagnosis at this time in the disease is significantly less useful to the patient and his family (Nagy et al., 1998a). In studies of small groups of carefully selected patients with clinically typical AD the accuracy of the clinical diagnosis of AD is usually between 80 and 90% (Sulkava et al., 1983; Wade et al., 1987), and can reach 100% (Martin et al., 1987; Nagy et al., 1998a). However, with less well-characterized patients and earlier dementia diagnostic accuracy can fall as low as 50% (Homer et al., 1988; Nagy et al., 1998a). The study by Joachim and colleagues (Joachim et al., 1988), analysed 150 consecutive brains from a wide variety

163

164


of community and hospital sources in New England but all had a clinical diagnosis of AD. Although not precisely specified, the sample represented the whole spectrum of investigation from tertiary hospital to community nursing home, but probably excluded most patients with severe cerebrovascular disease. As such it probably represents a broad spectrum of what might be called ‘uncharacterized degenerative dementia’. Their findings were that almost 90% of patients with the clinical diagnosis of AD at the time of death met what were then the accepted neuropathological criteria (Khachaturian, 1985) for the diagnosis. Two other large-scale studies (Jellinger et al., 1990; Mendez et al., 1992) suggest that in studies of generalized dementia and clinically diagnosed AD, approximately 60% of the patient population will have pure AD with a further 15% a combination of AD with either Parkinson’s disease or cerebrovascular disease. A few other rarer combinations with AD were also seen. Of the remaining 25%, a significant fraction can be expected to have either Parkinson’s disease alone or cerebrovascular disease alone, and the remainder a fairly long list of rarer causes of dementia, including, of course some cases of Creutzfeldt–Jakob disease. Interestingly, in these large studies, only between 1 and 2% had no reported pathology that could account for the dementia. The study of the Consortium to Establish a Registry for Alzheimer’s disease (CERAD) (Gearing et al., 1995) in a survey of 106 patients attained a diagnostic accuracy of 87%. These patients were assessed by standardized clinical neuropsychological and imaging tools (Davis et al., 1992; Morris et al., 1989) and were therefore a rather thoroughly investigated group of patients. In this study, those found not to have AD were very heterogeneous in terms of their pathology, which tends to suggest that the clinical diagnostic batteries used for these research studies are becoming quite refined. Although, as recent studies (Hogervorst et al., 2000; Nagy et al., 1998a) indicate, the accuracy of the clinical diagnosis of AD is as high as 90% or more, the negative predictive value of the current diagnostic criteria is still poor (Nagy et al., 1998a). The major cause of this is due to the fact that most of the diagnostic protocols rely on the detection of fairly advanced functional loss or neuronal fallout. The accuracy of more sensitive tests, that detect subtle functional changes, typical of the early stages of AD, are still to be confirmed by large, post-mortem confirmed studies (Corkin, 1998; Deweer et al., 1995; Fabrigoule et al., 1998; Kato et al., 2001; Saykin et al., 1999). To what extent do pathological studies of this type provide information about the incidence and prevalence of AD in the population at large? As has been described by Brayne (1993) in her review of clinico-pathological studies

of the dementias the interpretation of these studies is complicated by issues relating to the criteria used for the pathological diagnosis of Alzheimer’s disease and the clinical diagnosis of AD (Burns et al., 1991; Burns et al., 1990; Byrne, 1991; Kellett et al., 1991; Lindley & Dennis, 1991; Manning, 1991; Tozer, 1991). Further, as Brayne notes, almost all of the studies where neuropathological verification is included also have some degree of institutional bias (Blessed et al., 1968; Katzman et al., 1988; Roth et al., 1966; Wilcock et al., 1982) built into them, which makes them unable definitively to answer the simple question we started with. With all these reservations, it is safe to conclude that in the dementia of older age, when easily diagnosable cerebrovascular disease has been excluded as the cause of the dementia, the large majority of the remaining cases will be examples of AD. However, in a specific patient there is at present, in most cases, no substitute for pathologic examination to establish a diagnosis of AD. This is particularly so if the task is to differentiate AD from other degenerative causes of dementia such as diffuse Lewy body disease and mixed dementia. This is not to say that pathologic examination is problem free, the controversies in diagnostic criteria for AD are considered in a later section.

Risk factors In classical epidemiological terms there are a number of risk factors that are associated with the development of AD and these have been examined in a number of studies (Heyman et al., 1984; Katzman, 1993a; Prince et al., 1994; Rocca et al., 1986). One of the most important of the risk factors, age, has already been discussed, but genetic factors, history of head trauma, and plasma homocysteine levels (Clarke et al., 1998, 2000) have also been shown to be important. The specific genetic factors that operate in the pathogenesis of AD will be discussed in more detail later but in general terms it has been calculated that genetic factors could play a significant part in at least a quarter of cases (van Duijn et al., 1991), and probably more. It is likely that this is an underestimate of the total genetic influence given that in families with onset of AD after the age of 65 years family history will often be negative for parents who have died before they were old enough to have developed the disease. In practical terms, there is a, relatively small, group of patients with early onset familial AD where a specific gene defect predisposing to the development of AD can be made by molecular genetic techniques (p. 103). It is likely that the size of this group will increase as more molecular genetic information becomes available although the utility of


molecular genetics as a diagnostic tool in specific individual cases is likely to be most conspicuous in early onset cases. In the, numerically much larger, group of late onset cases of AD, the polygenic nature of the genetic component of most cases suggests that it is unlikely that molecular biologic techniques will be the path to routine clinical diagnosis. Thus, for the present, neuropathological examination of the brain, either by biopsy (which is not usually justified), or post-mortem examination, is required to confirm a clinical diagnosis in the overwhelming majority of cases. In the consideration of genetic influences it must be remembered that, as revealed by identical twin studies that are discordant for AD (Cook et al., 1981; Jarvik et al., 1980; Kumar et al., 1991; Rapoport et al., 1991) there are certainly a large number of genuinely sporadic cases of AD. Another recognised risk factor for the development of AD is head trauma. An episode of head trauma which causes loss of consciousness or results in admission to hospital is reported to increase the chance of subsequently developing AD by a factor of two (Gentleman & Roberts, 1991; Mortimer et al., 1991). Head injury has been shown to be associated with the acute deposition of -amyloid protein (AP) in the cortical ribbon and increased expression of the -amyloid precursor protein (APP) in adjacent cortical neurons (Roberts et al., 1990). It has also been convincingly demonstrated that the repeated head trauma associated with a career in boxing leads to a well-characterized progressive dementing syndrome and the formation of diffuse plaques and tangles within the brain (Corsellis et al., 1973; Dale et al., 1991; Roberts et al., 1990). Roberts et al. (1994) also report anecdotally that head trauma can seemingly advance the development of dementia in patients with a APP gene mutation that predisposes to dementia. However, although there are tantalizing and very suggestive links between acute head injury, increased APP expression in cells and acute deposition of AP in the cortex, it is not at all clear how this could predispose to the development of AD years or decades later. One suggestion is that the release of acute-phase reactant proteins in trauma could facilitate the conversion of non-fibrillar -amyloid peptide to a more stable form and hastens the formation of the compact amyloid that is more likely to stimulate a neuritic reaction (Katzman, 1993a). If such a mechanism was operative, it could be argued that the deposition of microscopic quantities of persisting fibrillar amyloid following trauma could act as an inciting nidus to further amyloid deposition many years later. This might effectively advance a process that would otherwise have occurred at a considerably later time. As eloquently expressed by Katzman (Katzman, 1993b), AD has been considered to be a democratic process and no respector of rank or position, affecting University

Presidents and physicians, mathematicians and musicians as well as those without memorial. However, there are recent studies that seem to indicate that the Alzheimer process is not quite the populist democrat it has been portrayed to be, and that degree of education has a significant influence in the development of symptomatic dementia in AD. Studies from geographic and cultural locations as disparate as Western Europe (Bonaiuto et al., 1995; Dartigues et al., 1991; Fratiglioni et al., 1991; Sulkava et al., 1985), Israel (Korczyn et al., 1991) and Shanghai (Hill et al., 1993; Zhang et al., 1990) indicate that there is a protective effect of education, although it should be noted that no such protective effect was observed in Rochester Minnesota (Beard et al., 1992) or Cambridge England (O’Connor et al., 1991). An idea of the possible extent of this effect is given in the studies of elderly Catholic nuns where those with a bachelor’s degree were more than twice as likely to be functionally independent as those without a college education (Snowden et al., 1989). Katzman (1993b) and Friedland (1993), in their reviews discuss a number of possible explanations for this effect. It has been demonstrated that synaptic density declines in association neocortex in AD (Masliah et al., 1989) and it has been shown that there is a good correlation between the density of synaptic terminals and MMSE scores (DeKosky & Scheff, 1990; Terry et al., 1991). Katzman advances the possibility that in those with higher education there is a greater synaptic density that would provide a reserve synaptic capacity. He considers that this effect might be sufficient to delay the onset of symptoms by up to 5 years and thereby reduce the prevalence of dementia at a given age by up to 50%. The effects of education described in the previous paragraph fuel the suspicion that, far from being democratic, the Alzheimer process is actually rather elitist, and its democratic credentials are further undermined by indications that it might be sexist as well. In a number of the epidemiological studies of cognitive decline and clinically diagnosed Alzheimer’s disease there is a considerably higher age specific prevalence in women (Bachman et al., 1992; Kokmen et al., 1989; Molsa et al., 1982; Sulkava et al., 1983; Zhang et al., 1990), although other studies do not find a significant difference in prevalence (Fratiglioni et al., 1991; Hachinski et al., 1975; Hofman et al., 1991; Wernicke & Reischies, 1994). There is rather more uncertainty as to whether there is a gender difference in age specific incidence rates (Hagnell et al., 1992; Katzman et al., 1989; Prince et al., 1994; Schoenberg et al., 1987). As discussed by Henderson and Buckwalter the interpretation of such differences is complicated by the fact that there are at least some differences in ways in which men and women perform in cognitive tasks in AD and the

165

166


possibility that the observed gender differences in prevalence might also reflect some subtle ascertainment biases that favour men (Henderson & Buckwalter, 1994). A systematic difference in prevalence could be explained by differences in rate of progression or mortality, although there is no evidence to suggest that this is the case. A confounding factor here may also be the apparent importance of duration of education in the development of dementia. It has been suggested that many of the studies that show an excess of women contain large numbers of elderly women with little formal education and that this might be sufficient to account for the greater prevalence of dementia in women in some studies. However, the possibility that there may be a real gender difference in susceptibility to the Alzheimer process has been given some indirect support from studies that report improved cognitive performance in women with a clinical diagnosis of AD taking hormone replacement therapy (Henderson et al., 1994). At a pathological level, there has been a report that in Down’s syndrome the severity of Alzheimer pathological change is greater in women than in men (Raghavan et al., 1994). These reports, and the presence of oestrogen receptors on cholinergic neurons of the basal forebrain (Toran-Allerand et al., 1992) have spawned a tentative hypothesis that links lack of oestrogen to reduced cholinergic activity and the possibility of impaired postmenopausal cognitive function (Fillit et al., 1986; Honjo et al., 1989). Age, genetic factors and head trauma are probably the best recognized risk factors for the development of AD but there are a number of other candidates that have been advanced. These include myocardial infarction (Aronson et al., 1990) and coronary artery disease (Sparks et al., 1990) and it has been suggested that cerebrovascular disease may also be a factor (Jellinger, 1976). Maternal age and hypothyroidism (Kalmijn et al., 2000; van Duijn et al., 1991) have also been canvassed as risk factors, but there is little evidence to support them as having a significant role. The role of aluminium is a subject of continuing controversy (Chapter 21). Recent studies also indicate that elevated plasma homocysteine levels, linked to vitamin B12 and folate deficiency, also represents a risk factor for the development of AD and dementia (Clarke et al., 1998, 2000). Whether elevated homocysteine levels are directly responsible for the causation of AD or are merely a surrogate marker for an underlying process remains to be established. Given the depressing number of adverse factors for the development of AD it may come as some relief that there is a factor that appears to decrease the risk of developing AD. However, any relief is immediately tempered by the revelation that the factor in question is cigarette smoking

(Brenner et al., 1993) since any protective effect of cigarette smoking is bought at a very high price in lung cancer, heart disease (which may increase the risk again!), stroke, and all the other ills rightly condemned by James I. The protective effect against AD is, however, not negligible with unmatched odds ratios of between 0.47 and 0.8. A pooled re-analysis has generated a statistically significant negative association with an odds ratio of 0.78 (CI: 0.62–0.98). To date, the small-scale clinical trials conducted in the efficacy of nicotine for the treatment of cognitive symptoms of dementia in AD are inconclusive, but promising (LopezArrieta et al., 2001).

Neuroimaging Outside the standard clinical methods of history and examination, the other important clinical tool in attempting to make the diagnosis of AD in life is neuroimaging (Smith, 2002) and there have been CT studies that have indicated significant medial temporal lobe atrophy in dementia (de Leon et al., 1989; Kido et al., 1989) (Fig. 9.1). Clinico-pathologic studies have suggested a high correlation between early medial temporal lobe atrophy as demonstrated by temporal lobe oriented CT scanning with a subsequent pathologic diagnosis of AD (Jobst et al., 1992, 1994b) (Fig. 9.1), and there has been some subsequent confirmation (Nagy et al., 1999a,b; Pasquier et al., 1994). Similar findings have also been reported using MRI scanning (Erkinjuntti et al., 1993; Scahill et al., 2002). On serial scans, patients with a subsequent post-mortem diagnosis of AD can be seen to undergo a rapidly progressive atrophy of the medial temporal lobe structures and detectable atrophy is present quite early in the progression of cognitive decline. This being so, medial temporal atrophy is an early phenomenon and offers a potentially useful diagnostic examination in patients with cognitive decline. However, although helpful in defining the location of disease, techniques of imaging the temporal lobe are not likely always to be able to distinguish AD from some other possible causes of medial temporal lobe atrophy such as Pick’s disease or frontal lobe degeneration (see also Chapter 8). The other imaging modality that might have a role to play in the premortem diagnosis of AD is single-photon emission computed tomography (SPECT), which measures regional blood flow within the brain. There have been a number of studies of dementia and AD using this scanning modality (Claus et al., 1994), and it is reported that AD often gives a rather characteristic temporo-parietal deficit (Fig. 9.2). As with many modalities, its sensitivity significantly improved with increasing severity of cognitive


Fig. 9.1. Temporal lobe orientated computerized tomography (CT) in AD: CT scans of a normal control (left) and a patient with AD (right) with the patient positioned in the scanner to obtain a view of the medial temporal lobe. In the patient with AD, the brain is generally atrophic, but the atrophy is particularly marked in the medial temporal lobes.

Fig. 9.2. Single photon emission computerized tomography (SPECT) in AD: SPECT scans of the same control and patient as in Fig. 9.1. the patient with AD (right) shows the bilateral parietal perfusion defects typically seen in AD.

decline. In their study (Claus et al., 1994) with specificity set at 90%, sensitivity rose from 42% in mild clinically diagnosed AD to 79% in severe cases. One attractive possibility that might enable AD to be distinguished from other

causes of temporal lobe atrophy might be to combine the use of temporal lobe oriented CT/MRI scanning to define focal atrophy with SPECT to see the changes in blood flow (Jobst et al., 1994a, 1992). Preliminary results on the

167

168


OPTIMA study group of patients suggest that this combination offers a significantly higher degree of sensitivity and specificity than either modality used alone. The development of magnetic resonance imaging (MRI) techniques also confirm the usefulness of imaging techniques, especially the measurement of progressive brain atrophy in the diagnosis of dementia (Rizzo et al., 1997; Strijers et al., 1997; Zakzanis, 1998). Additionally functional MRI studies indicate that this method might be useful in the early identification of learning difficulties in patients with incipient Alzheimer’s disease (Strijers et al., 1997). The main handicap of these studies, however, remains the lack of post-mortem confirmation of diagnosis.

Gross findings Brain weight As with other diseases, gross brain weight is not, in individual cases, a very helpful quantity except at the extremes of the weight curve. It is the general experience that in younger patients with AD, there is likely to be considerably greater atrophy and loss of brain weight than in older patients where both atrophy and brain weight often fall well within accepted norms in the presence of AD. Most brains that are less than 950 grams in weight will prove on neuropathologic examination to have some pathologic process, but on its own the brain weight is of very little value in identifying the nature of the process.

External appearance The external appearance of the brain in AD, although sometimes characteristic, is not diagnostic. Particularly in younger patients, there may be obvious atrophy affecting the frontal, temporal and parietal lobes (Fig. 9.3). In some cases, the degree of atrophy can be sufficiently focal and severe to resemble that seen in Pick’s disease (Joachim et al., 1986). However, many cases that, on microscopic examination prove to have AD, have a most undistinguished external appearance with either mild generalized frontotemporal atrophy not greater than that seen in a number of aged brains to no detectable atrophy at all. This impression has received numerical support from the study of Hubbard and Anderson (1981a) who showed that in patients with AD above the age of 80 years cerebral volume and atrophy were often within the normal range. One of the reasons for younger patients to show more marked atrophy is probably that they tend to have a longer survival and hence a greater chance for atrophy to develop. Although a fronto-temporal bias in the atrophy is very much the usual finding, there

Fig. 9.3. The most advanced degree of cerebral atrophy, rarely seen in Alzheimer’s disease, shows very marked gyral atrophy involving the frontal, temporal and parietal lobes. With this degree of atrophy, even the primary cortex around the Rolandic sulcus shows marked atrophy.

are rare but well described cases of AD where the principal burden of disease falls on the occipital regions and this is reflected in a predominantly parieto-occipital pattern of atrophy (Berthier et al., 1991; Brun & Englund, 1981; Levine et al., 1993). However, as is the case with the primary progressive aphasias (Chapters 1 and 11), this localization of disease, while it has many common features to the associated clinical phenomenology, is pathologically heterogeneous. Victoroff et al. (1994) report three cases of what has come to be called posterior cortical atrophy with diagnoses of Alzheimer’s disease, Creutzfeldt–Jakob disease and subcortical gliosis.

Brain slices As with the gross appearance, there are no pathognomonic changes on brain slices. There may be some degree of macroscopically apparent atrophy with widening of the cortical sulci. The cortical ribbon is rarely visibly attenuated, although there may be some narrowing in the most severely affected cases, particularly in the temporal lobe (Hubbard et al., 1981b). Conversely, the hippocampal formation is very often visibly atrophic, a feature that is reflected in the frequent expansion of the temporal horns of the lateral ventricle (Fig. 9.4). It is this frequent grossly apparent atrophy of the hippocampus that is presumably the anatomic basis for the radiological finding of medial temporal lobe atrophy in AD. The appearance of the amygdala is very variable. Sometimes it appears to have an almost normal volume while on other occasions it is severely atrophic to naked eye inspection. Given its location at the anterior


Fig. 9.4. Part of a brain slice from a case of Alzheimer’s disease showing marked hippocampal atrophy with accompanying dilatation of the temporal horns of the lateral ventricles. The parietal lateral ventricles are also considerably dilated, although the dilatation is often most conspicuous in the occipital poles.

end of the temporal horns of the lateral ventricles, gross atrophy of the amygdala is usually found in those cases with very marked dilatation of the temporal horns of the lateral ventricles. As with the external appearance, very occasionally the cortical changes are sufficiently fronto-temporally localized as to resemble the changes in Pick’s disease even to the extent of some visible cortical attenuation in the most affected cortical areas (Fig. 9.5). Dilatation of the frontal and occipital poles of the lateral ventricles is usually apparent, but, again particularly in older persons, is not invariably obvious and in some cases the size can be within the normal limits for age (Berg et al., 1993; Hubbard et al., 1981b; Tomlinson et al., 1970). Not infrequently there may be a discrepancy between apparent

Fig. 9.5. In some cases of Alzheimer’s disease there is very marked atrophy of the temporal lobe as severe as that seen in Pick’s disease. In this example there is no sparing of the superior temporal gyrus that is typically seen in Pick’s disease.

Fig. 9.6. Dilatation of the occipital poles of the lateral ventricles. In some cases of Alzheimer’s disease this is the location of the most conspicuous ventricular dilatation.

cortical atrophy and the degree of ventricular dilatation. In some cases, the ventricles may be quite dilated where the cortical sulci are not strikingly widened. When there is significant atrophy and ventricular dilatation, there is almost always notable expansion of the occipital poles of the lateral ventricles (Fig. 9.6). Some expansion of the third ventricle is usually evident, and in severe cases it becomes barrel shaped. In contrast, the aqueduct and fourth ventricle are normal sized. Except in those cases with concomitant cerebrovascular disease or severe amyloid angiopathy, the white matter is not usually visibly abnormal in character. As evidenced by the dilatation of the cerebral ventricles that usually accompanies widening of the cortical sulci, there is invariably some reduction in overall volume of white matter, which has been reported to vary between 3 and 19% in a quantitative study (de la Monte, 1989). In the deep grey nuclei of the cerebral hemispheres, atrophy, when it is detectable, is restricted to the caudate and putamen. On CT scans atrophy of the caudate with proportional expansion of the frontal poles of the lateral ventricles, can occasionally be sufficiently marked to be confusable with the appearance of Huntington’s disease, although the clinical history will usually be sufficient to give the lie to this. On brain slices, although there may be considerable reduction in the bulk of the caudate (less conspicuous in the putamen), some degree of convex curvature in relation to the lateral ventricle is almost always retained to distinguish the appearance from that of severe Huntington’s disease (Chapter 16). The degree of atrophy is almost always symmetric, and gross asymmetry of cortical atrophy or ventricular dilatation should suggest a search for an alternative diagnosis.

169

170


No changes are apparent in the macroscopic appearance of the globus pallidus and thalamus or in the substantia innominata (location of the nucleus basalis of Meynert) even though it is markedly affected in AD. Elsewhere in the brain, the only other consistent macroscopic finding in Alzheimer’s disease is loss of pigmentation in the locus ceruleus. The three major causes of macroscopically visible loss of pigmentation in the locus ceruleus are idiopathic Parkinson’s disease (Chapter 15), progressive supranuclear palsy (PSP)(Chapter 11) and AD. In idiopathic Parkinson’s disease the depigmentation of the locus ceruleus is accompanied by depigmentation of the substantia nigra. In PSP there is depigmentation of the substantia nigra and also usually a detectable dilatation of the fourth ventricle that is particularly noticeable superiorly where it merges with the aqueduct. This pattern of dilatation is a result of the degeneration of the dentate nucleus of the cerebellum and atrophy of the superior cerebellar peduncle. Both of the other major causes of depigmentation of the locus ceruleus are accompanied by pallor of the substantia nigra. Consequently, the finding of pallor of the locus ceruleus without depigmentation of the substantia nigra in a patient with dementia is one of the most suggestive gross findings that support a diagnosis of AD. However the converse, i.e. depigmentation of both the substantia nigra and the locus ceruleus does not exclude the diagnosis of AD since the combination of AD and idiopathic Parkinson’s disease in the same patient is not at all uncommon. There has been quite a wide variation in estimates of the frequency of this combination, but in the series of 150 consecutive cases published by Joachim et al. (1998) this combination was present in 11% of cases of AD.

Microscopic appearance The major microscopic findings in AD are well known and have been frequently described and recently reviewed (Mirra & Hyman, 2002). The major pathological features are senile plaques and the various manifestations of neuritic pathology, principally neurofibrillary tangles but also neuropil threads and neuritic plaques. There are also other microscopic features, notably Hirano bodies and granulovacuolar degeneration as mentioned in Chapter 1.

Senile plaques Senile plaques are best visualized by silver stains and/or immunocytochemistry for amyloid- protein. However they are visualized, they have a wide range of morphological appearances and with the application of different

Fig. 9.7. A low power illustration of cortex containing numerous diffuse plaques showing their characteristic irregular outline. Bielschowsky ×50.

Fig. 9.8. Typical diffuse plaque with irregular outline and no discernable neuritic reaction. Immunohystochemistry for -amyloid ×250.

antibodies and computerised reconstructions the descriptions have become increasingly sophisticated. However, for most practical and diagnostic purposes it is still probably most useful to divide them into the so-called ‘diffuse’ or ‘pre-amyloid’ plaques and the ‘classical’ or ‘neuritic’ plaques (Figs. 9.7, 9.8 and 9.9). Diffuse plaques, although they stain with Bielschowsky and MS silver stains and are immunochemically reactive for the amyloid- protein, have an amorphous irregular configuration and do not stain with Congo red. This pattern of staining implies that in diffuse plaques the A is not in the -pleated sheet conformation that is characteristic of amyloids of all compositions. Immunochemically, there is also little or no evidence of reactive astrocytic or microglial


Fig. 9.9. Mature plaque showing the characteristic central core of amyloid surrounded by a neuritic reaction. Neuropil threads are seen in the surrounding neuropil (arrows). Double immunohystochemistry for -amyloid (brown) and hyperphosphorylated tau (red).

changes and no neuronal damage. When examined with the electron microscope, the neuropil has a normal appearance, and there are few, if any of the characteristic amyloid filaments. Morphologically, the neuritic plaque is composed of a central amyloid core surrounded by a halo of distorted neurites that contain the characteristic paired helical filaments found in neurofibrillary tangles and that stain for the tau protein. The amyloid in the core of these plaques is composed of the protein that is characteristic of AD, and is similar to the amyloid found in the blood vessels in congophilic angiopathy of AD. Ultrastructurally, the plaque cores show a characteristic appearance with fingers of fibrillar amyloid extending into the surrounding neuropil (Fig. 9.10). An additional feature of neuritic plaques is their association

Fig. 9.10. Electron micrograph of a mature plaque core showing the characteristic fibrillar organisation of amyloid. Uranyl acetate/lead citrate ×2500.

with glial elements (Mackenzie et al., 1995; Oka et al., 1998; Sasaki et al., 1997), which raises the possible pathogenic role of microglia in the formation of senile plaques (Sheng et al., 1998). Alternatively, it is possible that the aggregated -amyloid leads to the secondary activation of glial cells (Hu et al., 1998) which in turn may have tertiary deleterious effects on neuronal survival (Combs et al., 2000). The feature common to these two types of plaque is the amyloid protein which is a 39 to 43 amino acid protein (A) derived from a much larger amyloid precursor protein (APP), which occurs in various lengths between 695 to 770 amino acids. The APP is a transmembrane protein of currently unknown function and the A portion is part of the transmembrane region of this protein. The relationship between these two types of plaque is not entirely clear. The simple view would be that the diffuse plaque is an antecedent of the neuritic plaque, but it is clearly not that straightforward. Studies of patients with Down’s syndrome have shown that the earliest change in the development of Alzheimer pathology is the development of diffuse plaques in the entorhinal regions and that the formation of neuritic plaques is a later event, but there is no convincing evidence that individual diffuse plaques actually evolve into the neuritic variety. One major difference between the two types of plaque is the nature of the amyloid protein that is present. Studies using end specific monoclonal antibodies to A in AD have shown diffuse plaques in cortex and basal ganglia and cerebellum are A1–42(43) positive but A1–40 negative, and that the appearance of plaques positive for both A1–42(43) and A1–40 was strongly correlated with the presence of mature neuritic plaques (Iwatsubo et al., 1994, Murphy et al., 1994). In patients with Down’s syndrome it has been shown that the deposition of A1–42(43) precedes that of A1–40 by about a decade (Iwatsubo et al., 1995). It is not known what factors influence this change or whether it represents in situ processing of A1–42(43) into the shorter A1–40 or an aggregation of A1–40 on to existing deposits of A1–42(43). It also seems very likely that tissue factors have a considerable influence on the conformation of plaques since there are a number of regions in the brain where diffuse plaques are commonly found but neuritic plaques are invariably absent. The best examples of this are in the caudate/putamen and the cerebellum (Joachim et al., 1989) (Fig. 9.11). In both these structures diffuse plaques are common in Alzheimer’s disease, but even when there are abundant neuritic plaques in the neocortex, they are not found in these structures. Whatever is spurring the neuritic reaction in the cortex is not present in the caudate/putamen or cerebellum. In the amygdala, there is a marked tendency for there to be neuritic plaques in the more medial nuclei,

171

172


(a)

(b) Fig. 9.11. (a) Diffuse plaques in the putamen. Methenamine silver ×125 (b) Diffuse plaques in the molecular layer of the cerebellum. Bielschowsky.

while those in the lateral nuclei have much less neuritic content. Senile plaques may be found in the hippocampal formation and throughout the cortex. In the neocortex they are most frequent in the association areas and there is marked relative sparing of the primary cortical areas (Esiri et al., 1986). This is most easy to see in the primary optic cortex, probably because it is very easy to identify primary cortex in this situation. In almost all cases of AD there is a mixture of diffuse and neuritic plaques, and in severe cases very large numbers of plaques are present.

Neurofibrillary changes As described in Chapter 1 the three expressions of neurofibrillary pathology are neurofibrillary tangles, the dystrophic neurites of neuritic plaques and the so-called neuropil threads, which are found, scattered in the neuropil (Fig. 9.12). All represent intracellular accumulations

Fig. 9.12. Typical appearance at medium power of cortex in Alzheimer’s disease showing the combination of mature plaques with prominent neuritic reaction, neurofibrillary tangles and neuropil threads scattered in the intervening neuropil. Bielschowsky ×50.

of paired helical filaments (PHF) the basic component of which is hyperphosphorylated tau protein. In AD neurofibrillary tangle formation is particularly conspicuous in Sommer’s sector of the hippocampus, entorhinal cortex and in the amygdala. However, it has long been recognized that tangles are present in these regions in many older people who are not known to be demented. Ball (1977) has shown that, although neurofibrillary tangles are present in the hippocampus of intellectually intact older people, they are at a lower density than in patients with AD. This observation is also implicit in the staging system of Braak and Braak (1991)(see below) where tangles may be present in hippocampus and entorhinal cortex in ‘presymptomatic’ disease. For this reason, it is difficult to rely solely on hippocampal and adjacent temporal lobe sections for the diagnosis of AD. Within the neocortical mantle, neurofibrillary tangles are found widely distributed in the cortex, though as with senile plaques, there is a marked tendency to spare the primary cortical areas. One important practical point in the microscopic examination of the brain in AD is that visualisation of neurofibrillary pathology requires the use of special stains which may be one of a number of silver stains, immunocytochemical techniques for amyloid or neurofibrillary epitopes, or thioflavine S. It is effectively impossible accurately to estimate the degree of neurofibrillary pathology with haematoxylin and eosin or other routine histological stains. Although individual tangles may be quite easy to identify in the hippocampal pyramidal cell layer using H&E, it is much more difficult to see them in the neocortex. A feature that can be evaluated with H&E is the degree of attenuation of the neuropil that occurs as a result of neuronal loss


Fig. 9.13. Haematoxylin and eosin stain of the upper layers of parietal cortex in severe Alzheimer’s disease showing marked neuronal loss and neuropil pruning. The faint nodularity seen in this photograph is a reflection of the presence of numerous neuritic plaques. H&E ×50.

in AD (Fig. 9.13). The attenuation is most conspicuous in layers one and two of the cortex and usually most severe in the temporal lobe. This can range from a change in the density of the neuropil that just allows the individual processes to begin to be distinguished to a severe microvacuolation and collapse of the upper cortical layers (Brun et al., 1981). This vacuolation has sometimes been confused with the vacuolar change that occurs in Creutzfeldt–Jakob disease, although its location in the upper layers of the cortex is uncharacteristic for this disease. Smith et al. showed that some degree of this vacuolar change could be found in the temporal lobe in a majority of cases of AD (Fig. 9.14) (Smith et al., 1987). The conformation of neurofibrillary tangles is very dependent on the neuron in which they accumulate so that the archetypal ‘flame-shaped’ tangle is found in larger pyramidal neurones such as those in the hippocampal pyramidal cell layer while, globose tangles are more typically seen in neurones such as those in the basal nucleus of Meynert and the locus ceruleus (Fig. 9.15). Ultrastructually, neurofibrillary tangles are composed of swaths of

Fig. 9.14. Haematoxylin and eosin-stained section showing the ‘spongy’ vacuolation of the upper layers of the cortex seen in some cases of Alzheimer’s disease. This change resembles that seen in the fronto-temporal dementias. In Creutzfeldt–Jakob disease more pronounced in the lower layers of the cortex. H&E ×125.

intracytoplasmic paired helical filaments with a diameter of about 20 nm and a periodicity of 80 nm (Fig. 9.16) which displace the cytoplasmic contents and the nucleus (Fig. 9.17). The paired helical filaments of the neurofibrillary tangle are very insoluble, so insoluble in fact that they remain within the neuropil as ‘ghost’ or ‘tombstone’ tangles after the death and degeneration of the cell in which they developed. In this condition they can still be stained by silver stains such as Bielschowsky and can be easily differentiated from intracellular tangles by their different staining intensity. Free in the neuropil the tangles become coated with other molecules and develop different immunochemical reactivity, for example to antibodies directed against glial fibrillary acidic protein (GFAP), ubiquitin and apolipoprotein E (Yamaguchi et al., 1994). A small fraction is also immunoreactive to the amyloid protein (Tabaton et al., 1991; Yamaguchi et al., 1991). This process is reflected ultrastructurally, in that ‘ghost tangles’ lose the characteristic paired helical conformation of the intracellular tangle and become thicker straight filaments (Figs. 9.18, 9.19). As described in the section on plaques, the development of dystrophic neurites, which is associated with the

173

174


Fig. 9.15. The general shape of the neuron defines the configuration of tangles that form within the cell. In this illustration of two neurofibrillary tangles from the basal nucleus of Meynert, one has a ‘pyramidal’ configuration while the other is globose. ‘Gallyas’ ×250.

Fig. 9.18. Low power electron micrograph of an extracellular, ‘ghost’ or ‘tombstone’ tangle. No cell membrane or nucleus is present, and degenerate organelles are present forming vacuoles between the tangle material. Uranyl acetate/lead citrate ×2500.

Fig. 9.16. High power illustration of the characteristic paried helical filaments that are the major constituent of neurofibrillary tangles. Uranyl acetate/lead citrate ×60 000.

Fig. 9.19. Higher power view of the fibrils shows that the paired helical configuration has been lost and replaced by straight filaments. The degenerate organelles between the fibrils are clearly seen at this power. Uranyl acetate/lead citrate ×30 000.

Fig. 9.17. Low power electron micrograph of an intracellular neurofibrillary tangle. Uranyl acetate/lead citrate ×3000.

coalescence of dispersed protein into focal deposits of fibrillary protein amyloid, is one of the stages in the evolution of the mature plaque. The third expression of neuritic pathology, the neuropil threads (Braak et al., 1986) are found between the plaques and tangles in the neocortex, entorhinal cortex and amygdala (Fig. 9.12) and also in subcortical areas such as the periaqueductal grey. As with the dystrophic neurites, their staining and immunochemical characteristics resemble neurofibrillary tangles, and ultrastructurally they contain paired helical filaments. Their presence is thought to reflect dendritic sprouting by neurons affected by tangle formation and they are a manifestation of neuritic pathology (Braak & Braak, 1988; Ihara, 1988).


Subcortical pathology in AD

Congophilic angiopathy

The cortical pathology in AD is so dramatic that it tends to overwhelm the considerable subcortical changes that are also found. In terms of senile plaque distribution, as well as the cortex, neuritic plaques can easily be found in the hypothalamus and mammillary bodies, the olfactory bulbs (Esiri & Wilcock, 1984) and the midbrain tegmentum. Smaller numbers may also be found in the region of the basal forebrain and floor of the fourth ventricle. The subcortical neuritic pathology is also conspicuous. The nucleus basalis of Meynert in the basal forebrain is a conspicuous site of neurofibrillary tangle formation (Rogers et al., 1985; Whitehouse et al., 1981, 1982) and has been the subject of a large number of studies (Mirra & Hyman, 2002). The importance of this region is that the large neurons of the nucleus basalis of Meynert are the source of cholinergic input to the cortex and loss of cholinergic supply and choline acetyl transferase in the cerebral cortex is perhaps the most well established neurochemical change in AD (Davies & Maloney, 1976). The consensus of the numerous studies of the nucleus basalis is that there is loss of between 40 and 70% of the neurons in this nucleus in AD (Arendt et al., 1983; Wilcock et al., 1988) and a generally good correlation with the neurochemical change in the cortex, the severity of the dementia and the degree of Alzheimer change seen histologically (Wilcock et al., 1982). There is some evidence that there is a greater degree of nucleus basalis neuron loss in younger patients (Tagliavini & Pilleri, 1983; Whitehouse et al., 1983). It is also reported that the part of the nucleus that supplies the temporal lobe is more severely affected (Wilcock et al., 1988). The other major sites of subcortical tangle formation and neuronal loss are the periaqueductal grey matter and the dorsal raphe which are the source of serotonergic supply to the cortex (Chen et al., 2000; Tomlinson, 1989) and the locus ceruleus (Mann et al., 1980; Matthews et al., 2002), which is the main source of noradrenergic supply. In some of these regions there is a tendency for a degree of neuronal loss to occur with age and this age related neuronal loss may be accompanied by a degree of tangle formation. The age related changes in these regions are, however, almost always considerably less than that seen in patients with AD. In line with the reduction in noradrenergic and serotonergic neurons there are also changes in cortical transmitter content and associated metabolites and marker enzymes although the correlations among neuronal loss, tangle formation, neurochemistry and clinical symptomatology are less straightforward than for the cholinergic system.

Amyloid deposition (congophilic angiopathy) in the cerebral arteries has been recognized since the beginning of the 20th century and its association with Alzheimer’s disease was also recognised early in the study of the disease (Vinters, 1987). Amyloid can be formed by molecules that have the capacity to aggregate in a conformation called a -pleated sheet so that amyloid is derived from a number of different sources. Most of the systemic causes of amyloid deposition are not associated with any amyloid deposition in the cerebral vasculature. Cerebral vascular amyloid deposition, although particularly associated with Alzheimer’s disease, also occurs in a number of other disease states including prion disease, where the amyloid is derived from the prion protein, and in the Icelandic form of hereditary cerebral haemorrhage with amyloidosis (HCHWA) where the genetic defect is located in the gene for cystatin C (Cohen et al., 1983; Jensson et al., 1987) (Chapter 14). Focal vascular amyloid deposition has also been described in vascular malformations (Hart et al., 1988) and after radiation therapy (Mandybur & Gore, 1969). In some respects, the amyloid hypothesis of AD started with congophilic angiopathy when Glenner and Wong first sequenced the cerebrovascular amyloid and showed that it had a previously undescribed amino acid sequence (Glenner & Wong, 1984). Using protein chemical and immunocytochemical methods this protein subunit was subsequently shown to be shared between cerebrovascular amyloid and diffuse and neuritic plaques, although the latter appear to contain a slightly shortened form of the protein with 40 amino acids (A1–40) rather than the A1–42(43) form that appears to predominate in diffuse plaques (Iwatsubo et al., 1994). Both forms of the A polypeptide are present in cerebrovascular amyloid (Joachim et al., 1988; Roher et al., 1993). Congophilic angiopathy is an almost invariable finding in AD, the frequency being to some extent dependent on how hard it is looked for, (Joachim et al., 1988; Mandybur, 1975), but is also common in elderly patients without AD (Esiri et al., 1986; Vinters & Gilbert, 1983). As with AD, the incidence rises with increasing age. The distribution and severity of amyloid angiopathy is unconnected to the severity of the Alzheimer changes or the duration of the disease and as such is of no diagnostic value although it is reported to be related to the apolipoprotein E genotype (see below). Congophilic (amyloid) angiopathy characteristically affects the smaller branches of the sulcal arteries and the penetrating arteries in the cerebral cortex. Within the cerebral cortex (Fig. 9.20) both the long and short penetrating arteries are affected, but, with the long penetrating arteries that supply the white matter, amyloid is found only in the

175

176


Fig. 9.20. Bielschowsky stain of amyloid angiopathy in a small artery entering the cortex from the subarachnoid space. Amyloid can be seen surrounding the smooth muscle cells in the wall of the artery. Amyloid deposits are also present in the parenchyma adjacent to the artery but separated from it by the Virchow–Robin space.

section of the artery passing through the cortex. When the artery enters the subcortical white matter, the amyloid deposition stops, although there may be other vascular wall changes. In general, when there is significant amyloid angiopathy the deposition tends to be most pronounced in the occipital lobe, and least conspicuous in the temporal lobe (Tomonaga, 1981; Vinters & Gilbert, 1983). The other major site at which amyloid angiopathy is found is in the cerebellum, where most of the deposition is in the small subarachnoid arteries. Some regions of the cerebral hemisphere, specifically the white matter, the basal ganglia, most of the thalamus and often the hippocampus are notably spared amyloid deposition, although the amygdala is often affected and the dorso-medial nucleus of the thalamus may also show pericapillary vascular amyloid deposition. Microscopically, the amyloid is deposited in the walls of the smaller arteries, first tending to surround the smooth

Fig. 9.21. Amyloid angiopathy affecting a small arteriole. Bielschowsky stain.

Fig. 9.22. Electron micrograph of a relatively early stage of amyloid deposition with amyloid fibril deposition in the spaces between the smooth muscle cells. Uranyl acetate/lead citrate ×7000.

Fig. 9.23. Electron micrograph of a later stage of amyloid deposition where the smooth muscle cells are more closely invested by amyloid. In some cases the smooth muscle cells are entirely surrounded by amyloid fibrils. This is analogous to the situation seen in the light micrograph Fig. 9.21. Uranyl acetate/lead citrate ×5000.

muscle cells that comprise the muscular wall of the vessel and eventually replacing them (Figs. 9.21, 9.22 and 9.23). This process can be followed ultrastructurally, with smooth muscle cells becoming progressively surrounded by a thickening band of amyloid fibrils. At a later stage the smooth muscle cells undergo degeneration and at the final stage the entire wall of the vessel is composed of a dense meshwork of amyloid fibrils with no surviving cellular components except for the endothelial cells lining the lumen of the artery (Fig. 9.24). At the intermediate stage, where the smooth muscle cells are surrounded, but have not yet degenerated, the amyloid can be seen on light microscopy as congophilic rings in the wall of the vessel. A recognized complication


Fig. 9.24. In the most advanced stage of amyloid deposition, the smooth muscle cells, as in this example, degenerate and the wall of the vessel is composed almost entirely of amyloid. Even at this stage the endothelial cells remain intact. Uranyl acetate/lead citrate ×1500.

of amyloid angiopathy is an increased incidence of lobar haemorrhage, which can be recurrent. Curiously, the haemorrhages are typically subcortical rather than intracortical and so occur from vessels that are not themselves affected by amyloid deposition. A further peculiarity of these haemorrhages is that they tend to occur in the frontal lobes, and not in the parieto-occipital lobes where the angiopathy is most severe (Gilbert & Vinters, 1983). Single lobar haemorrhages are usually associated with a focal neurological syndrome, but if they are multiple, they can cause, or contribute to cognitive decline. Since many of the patients have concomitant AD it can be hard to assess the significance of amyloid associated haemorrhages to cognitive decline, but cases of recurrent haemorrhage with progressive dementia in the absence of marked plaque and tangle formation have been described (Griffiths et al., 1982; Okazaki et al., 1979). It is not known whether there is a close relationship between the mechanisms of amyloid formation and deposition in blood vessel walls and the amyloid deposition that occurs in plaques and plaque cores. The disconnection between the severity of amyloid angiopathy and the severity of Alzheimer change in the parenchyma suggests that different cell biological mechanisms may govern the formation of amyloid in the two circumstances. Conceivably, for example, there could be a different pathway leading to the formation of the amyloid protein, and certainly, if tissue factors play a significant part in the conversion of the protein from the soluble form to the fibrillar form of amyloid pleated sheets they can be expected to be different in the cerebral parenchyma and in the wall of a blood vessel. However, there are clearly some common factors in the increased production of protein amyloid as there

Fig. 9.25. -amyloid labelling of a perivascular plaque showing amyloid and neuritic reaction in the parenchyma around a small arteriole with amyloid angiopathy. Double immunohystochemistry for -amyloid (brown) and hyperphosphorylated tau (red).

are examples of early Alzheimer changes in Down’s syndrome that are accompanied by congophilic angiopathy. This finding indicates that the additional chromosome 21, perhaps merely by increasing the production of the protein, is affecting the formation and deposition of amyloid in both the cerebral parenchyma and the cerebral vessels. One interesting microscopic feature that demonstrates the significance of amyloid deposition in the generation of plaque neuritic reactions is the phenomenon of perivascular plaque formation (Fig. 9.25). Perivascular plaques are often present in cases with severe amyloid angiopathy and the wall of an intracortical vessel is completely replaced by amyloid. When this occurs there can be extension of amyloid deposition across the Virchow-Robin space and into the adjacent cerebral parenchyma. It should be noted that perivascular plaques are only seen when there is massive deposition of amyloid in the vessel so that it is reasonable to conclude that the amyloid being deposited is arising from the blood vessel, rather than arising in the parenchyma and being deposited secondarily in the blood vessel. Ultrastructurally, this deposition very much resembles that seen in ordinary plaque cores with ‘fingers’ of amyloid fibrils extending into the tissue. This may be accompanied by a marked neuritic reaction associated with this amyloid. It is not credible to assume that there is any particular tissue selectivity about the location of this extension of amyloid

177

178


Fig. 9.28. Subpial deposition of amyloid in a case with severe amyloid angiopathy. Immunohystochemistry for -amyloid.

White matter changes in Alzheimer’s disease Fig. 9.26. Pericapillary plaque formation. A further smaller reaction is also present further along the same capillary. Double immunohystochemistry for -amyloid (brown) and hyperphosphorylated tau (red).

Fig. 9.27. Pericapillary amyloid showing the very characteristic fibrillar appearance of amyloid deposits. Bielschowsky ×250.

into the cortex. This suggests that it is some aspect of the presence of the amyloid that generates the neuritic reaction in otherwise unaffected neuropil. On the subject of vascular deposition of amyloid, there are occasional cases where there is very marked pericapillary deposition of amyloid. This almost always occurs in the context of AD, although there are occasional cases of marked pericapillary amyloid deposition with very little in the way of plaque and tangle formation (Figs. 9.26, 9.27). A further feature sometimes seen in cases with severe disease is the subpial deposition of amyloid (Fig. 9.28).

The development of CT and subsequently MRI opened new horizons in the examination of the brain in life and, not surprisingly, revealed some unexpected aspects of disease. A striking finding is that many older people have periventricular and diffuse white matter regions of low density, a group of changes that have been called leukoaraiosis (Hachinski et al., 1987). In general MR is more sensitive than CT in detecting this type of change which can be seen in patients with cerebrovascular disease (Aharon-Peretz et al., 1988; Erkinjuntti et al., 1987) but also asymptomatic patients and in patients with AD (Awad et al., 1986; Erkinjuntti et al., 1987, 1989; George et al., 1986a, b; Mirsen et al., 1991). Most studies suggest that the strongest correlates of the more severe grades of radiological change are advancing age and cerebrovascular disease (Awad et al., 1986; Brilliant et al., 1995). In terms of cognitive function, results have been ambiguous, but it is probably fair to conclude that there is no clear and consistent relationship between the severity of leukoaraiosis and any degree of cognitive decline (Brilliant et al., 1995). Pathological studies of the white matter in AD (Brun & Englund, 1986) have shown some degree of white matter change in approximately 60% of patients, with severe changes being present in about 20% of patients. The pathological change consists of a diffuse incomplete demyelination with some associated loss of axons and a mild reactive astrocytosis. Occasional macrophages are also found. The blood vessels show a variable hyaline thickening. These changes are similar to those described in Binswanger’s disease, but lack the cavitating and frankly necrotic lesions that are usually present in this disease (Caplan & Schoene, 1987; De Reuck et al., 1980). The severity of the white matter changes was unrelated to either the distribution or the


severity of the Alzheimer process in the grey matter, so that the changes were not likely to be the result of wallerian degeneration of white matter axons secondary to cortical neuronal loss (Brilliant et al., 1995; Brun & Englund, 1986; Englund et al., 1988). The absence of any degree of change in 45% of patients with AD also supports this conclusion. There is also no correlation with distribution or severity of amyloid angiopathy (Englund et al., 1988; Janota et al., 1989). The absence of significant associations with the severity of AD or AD associated structural changes in blood vessels suggest that, although a frequent finding in AD, the white matter lesion is not specifically related to the Alzheimer process. However, recent studies indicate that white matter lucencies are associated with accelerated cognitive decline and may predict the appearance of cognitive deficit in patients with mild cognitive impairment (Meyer et al., 2000; Wolf et al., 2000).

Pattern of development of AD changes and their relation to the symptoms of dementia This discussion can be broken into three areas of interest: (i) spatial and temporal distribution of AD changes (see Chapter 1) (ii) clinical correlates of regional and laminar specific pathology. (see Chapter 1) (iii) significance of plaques and tangles in the symptoms of dementia As has been described in Chapter 1 there have been a number of correlative studies of the overall severity of the dementia of AD with a morphologic indicator of the overall burden of disease. The balance of scientific opinion is that, in terms of Alzheimer pathology, neurofibrillary tangle numbers or some other reflection of neurofibrillary pathology are the best predictor of overall cognitive decline (Arriagada et al., 1992; Berg et al., 1993; Crystal et al., 1988; McKee et al., 1991; Morris et al., 1991; Nagy et al., 1995b; Neary et al., 1986; Terry et al., 1991; Wilcock et al., 1982). A concept that is at variance with these studies is that of ‘plaque only’ AD (Terry et al., 1987) where it was proposed that, in the elderly population (over 80 years), there was a group of demented patients with AD where the only pathological manifestation was the presence of senile plaques. Closer examination of this group of patients suggests that some of the patients with ‘plaque only’ AD are cases where there is mixed disease and, in particular, a high prevalence of cortical Lewy bodies. The dementia in these patients is then a reflection of the effects of the early stage of two dementing diseases with additive effects. Thus, in cases of suspected AD, when only plaques are found a search should

be made for another disease, very often Parkinson’s disease or cerebrovascular disease, that might be contributing to the symptoms of dementia. Progressing from this correlation between the symptoms of dementia and the overall burden of neuritic pathology, recent studies have begun to look more closely at the different regions of the cortex and have found that dementia scores are correlated most closely with the burden of neurofibrillary pathology in the frontal and parietal lobes (Nagy et al., 1995b). In a further refinement, some investigators have looked beyond plaques and tangles to parameters such as neuronal loss and reductions in synaptic density as more direct reflections of functional impairment of the different regions of the brain. These studies suggest that the strongest correlate with pre-mortem dementia is neocortical synapse density in the mid-frontal lobe (DeKosky & Scheff, 1990; Masliah et al., 1991; Terry et al., 1991). In a parallel development, there has also been increasing attention paid to different aspects of cognitive decline so that for example, in one study, Samuel et al. using a combination of neuropsychological tests have found that changes in higher cortical functions such as conceptualization, attention and initiation were most strongly correlated with alterations in midfrontal synapse density while for the more memory-orientated tests the strongest predictor was neurofibrillary tangle formation in the nucleus basalis of Meynert (Samuel et al., 1994). In another study (Nagy et al., 1996) memory deficits correlated closely with the severity of hippocampal neurofibrillary tangle formation, while occipital lobe pathology was associated with apraxia (Smith et al., 2001). This chapter has as its major focus the dementia associated with AD and much published work has been directed towards correlating pathological and neurochemical features of the disease with quantitative measures of cognitive decline. However, many of the most distressing features of the disease, particularly for family members and other carers, are actually the behavioural changes such as incessant wandering, aggression, paranoia, sexual disinhibition and depression that often accompany the dementia and are most frequently the symptoms that precipitate the need for institutional care (Wilcock & Jacoby, 1991). Psychiatric symptoms may also be among the earliest presenting symptoms of AD (Becker et al., 1994) and there is a recent study that shows that their presence can be a predictor of a faster cognitive decline (Chui et al., 1994). In this important field there has been much less systematic work than in relation to the dementia of AD but it is an area that is beginning to generate significant interest. Recent studies indicate that in demented patients, psychotic symptoms are related to more severe neurofibrillary pathology (Farber et al., 1998;

179

180


Sweet et al., 2000). Additionally the presence of Lewy bodies also increases the likelihood of psychotic episodes in AD patients (Ballard et al., 1995, 2001; Forstl, 1999, Snow & Arnold, 1996). There are some studies that suggest that the depression is associated with neuronal loss in the locus ceruleus and reduced levels of noradrenaline in the cortex (Hope & Fairburn, 1992; Zubenko et al., 1990; Zweig et al., 1988) although there is not a good correlation between the severity of the neuronal loss and the severity of the depressive symptoms. Another study indicates that there is a significant decrease in serotonin uptake sites in the temporal cortex in AD associated with depressive symptoms (Chen et al., 1996). Similarly, other studies have suggested that aggressive behaviour is correlated with evidence of diminished 5HT function in frontal cortex (Chan-Palay & Asan, 1989; Palmer et al., 1988) or cholinergic deficits (Minger et al., 2000). However, in what may become a recurring theme of clinico-pathologic correlative studies of psychiatric features, Chen et al. (2000) have shown that, although there is 40% loss of 5HT neurons in the dorsal raphe nucleus in AD cases with aggressive behaviour compared to controls, there is no correlation between the severity of neuronal loss in this region and the severity of cognitive or behavioural changes during life. This whole area of neurochemical and receptor changes in the behavioural changes in AD is only at the beginning of investigation and the correlations, given the complexity introduced by the existence of multiple different receptor types for transmitters such as 5HT and dopamine, the cholinergic system are likely to be complex and interdependent. What may be a harbinger of the conceptual problems that are likely to result from this complexity is the current difficulty in clarifying the relationship between dopamine deficiencies, a possible cholinergic-monoaminergic imbalance, and the presence of hallucinations (Perry et al., 1990, 1991) in patients with dementia. At a morphological level, this debate takes the form of whether a clinical history of hallucinations is associated with the finding of Lewy bodies in the cerebral cortex and substantia nigra (Perry et al., 1990) and to what extent the concomitant presence of AD changes, particularly the presence of senile plaques, is a necessary accompaniment of cortical Lewy bodies or a coincidental finding. The suggested association between cortical Lewy bodies and senile plaques raises the question of a more than incidental relationship between AD, Parkinson’s disease and what has come variously to be called diffuse Lewy body disease (Burkhardt et al., 1988; Crystal et al., 1990; Dickson et al., 1989; Gibbs et al., 1985; Kosaka, 1990; Kosaka et al., 1984; Lennox et al., 1989; Sugiyama et al., 1994; Yoshimura, 1983), senile dementia of the Lewy body type

(Perry et al., 1990), the Lewy body variant of Alzheimer’s disease (Hansen et al., 1990), and dementia with Lewy bodies (McKeith et al., 1996). This whole rather convoluted and controversial topic is considered in more detail in Chapter 15. Another aspect of the clinical presentation of AD that has begun to be addressed is the possible differences arising from the genetic heterogeneity of AD. Although the different genetic subtypes and sporadic late onset disease seem to have a very similar pattern of progressive cognitive decline (Haltia et al., 1994), a study of Finnish chromosome 14 encoded AD suggests that this group of patients have a high frequency of myoclonus and epilepsy. This is a combination of symptoms and signs that is otherwise very rare in AD, and has led many of these patients to be considered to have prion disease prior to definitive diagnosis. Another difference in this group is the generally earlier age of onset of symptoms that is usually before the age of 50 years, in contrast to the families with genetic abnormalities in the amyloid precursor protein where onset tends to be between the ages of 50 and 60 years.

Diagnostic criteria One of the most vexed questions in the pathological study of Alzheimer’s disease is that of the criteria that should be used for the diagnosis. In their seminal studies of the disease, Tomlinson and colleagues found what they construed to be a relationship between the number of senile plaques and the severity of cognitive decline. For a long time this formulation was considered to be acceptable and found its most concrete expression in Khatchaturian’s article reporting the results of a workshop sponsored by the National Institute of Aging (NIA), the American Association of Retired Persons (AARP), the National Institute of Neurological and Communicative Disorders and Stroke (NINCDS) and the National Institute of Mental Health (NIMH) (Khachaturian, 1985). The consensus of the neuropathological panel at this workshop was that the diagnosis could be made based on the mean number of senile plaques in three regions of neocortex taken from the temporal, frontal and parietal lobes (areas not specified) and graduated for age, so that with increasing age a greater number of plaques was necessary for the diagnosis to be considered secure. The panel considered that tangles would often be present, but that the diagnosis was not conditional on their presence. The insecurity of the criteria adopted was highlighted by a later paragraph of the report suggesting that a reduction (possibly of up to 50%) in the number of plaques required for


the diagnosis would be acceptable in patients with a clinical history of dementia. The Khatchaturian criteria specifically enshrined the possibility of making the diagnosis of AD in the absence of a history of dementia and without the necessity of being able to find neurofibrillary tangles, particularly in patients over the age of 75 years. In use, there are practical problems particularly in relation to achieving a ‘mean’ involvement with only three areas, and the nature of the plaques that count, the age effect and the ‘possible’ allowance for the presence of clinical dementia. A problem with these criteria is their dependence on uncharacterized plaque counts. There has always been a contrary position that has emphasised the importance of neuritic pathology in relation to the presence of dementia. This position has been reinforced over the years by reports of individuals or small series of unequivocally non-demented patients who were well examined in life, but who, on post-mortem examination were found to fulfil Khatchaturian criteria for Alzheimer’s disease in terms of plaque counts (Crystal et al., 1988). Tierney et al. (1988) applied a range of diagnostic criteria for AD but included the presence of neurofibrillary tangles in the requirement for diagnosis and by varying the criteria obtained clinico-pathologic agreement ranging from 64 to 86%. Our experience with the Tierney criteria (Nagy et al., 1995b) is that they are rather too strict and lead to the exclusion of cases that have significant disease burden and cognitive decline to be really useful. This concentration on senile plaques as a diagnostic indicator was also adopted by the Consortium to Establish a Registry for Alzheimer’s Disease (CERAD), but with a number of very significant differences (Mirra et al., 1991). First, they introduced the formulation of possible, probable and definite AD on the basis of age-related numbers of plaques estimated in a semiquantitative way and the presence or absence of a clinical history of dementia. Using CERAD criteria, in the absence of a clinical history of dementia, no matter how severe the Alzheimer changes only a diagnosis of possible AD can be made, so that this constitutes acceptance of the proposition that there are no absolute morphological criteria for the diagnosis of AD. Second, the CERAD criteria as revised require the semiquantitation of neuritic plaques and ignore diffuse ones altogether. More recent updates of the CERAD protocol (Mirra, 1997) include some refinement of the definitions of possible AD and Parkinson’s disease and make some specific suggestions about sections to take for the diagnosis of cerebrovascular dementia. The use of the CERAD protocol became the established tool for the diagnosis of Alzheimer’s disease and related disorders for both clinical as well as for research purposes.

Technique: Bielschowsky recommended (Litchfield & Nagy, 2001), but alternatives acceptable. Sample: (i) Middle frontal gyrus (ii) Superior and middle temporal gyrus (iii) Inferior parietal lobule (iv) Anterior cingulate gyrus (v) Hippocampus (vi) Entorhinal cortex and amygdala (vii) Midbrain including substantia nigra. Additional sections possible Histologic criteria: (i) Neuritic plaque assessment (ii) Semiquantitative assessment of area of maximum involvement matching this with illustrations of neuritic plaque densities representing each of three scores. (iii) In frontal, temporal or parietal cortex (iv) Age related plaque score. Ranks: Possible, probable or definite The final diagnosis requires history of dementia and takes into account the age of the patient.

Staging the severity of AD pathology A set of criteria has been advanced by Braak and Braak as a means of semi-quantitatively staging the severity of Alzheimer pathological change (Braak & Braak, 1991). The stages that are defined by this system span the pre-clinical and symptomatic phases of what is presumed to be the progressive accumulation of plaques and tangles over time and their progressively wider dissemination through the brain. Braak staging: Techniques: 100 m sections/Gallyas Samples Anterior hippocampus Posterior hippocampus Occipital Histologic Criteria Semi-quantitative (+ − + + +) of NFT and neuropil threads Comments: The use of this system has a significant technical problem in that the nature of the section preparation (100 m) and staining protocols are probably too difficult and restrictive to be used on a routine basis. Although

181

182


efforts have been made to adopt this diagnostic system for routine use utilizing thin paraffin sections (Nagy et al., 1998d) it is still mainly used for research purposes rather than routine diagnosis.

Choice of histopathological criteria There are a number of issues that are relevant to the choice of histopathologic criteria for the diagnosis of AD (Mirra et al., 1993). An important initial point is that the selection of diagnostic criteria is influenced by their intended use. For example, criteria for general use or multi-centre brain banking might be different from those adopted in an individual laboratory for a specific research purpose. In this chapter we are concerned with diagnostic criteria for general rather than research use. Starting from the first principles, whatever diagnostic criteria are advocated they should meet four practical tests. These are that the criteria should be: (i) Simple: Since whatever is suggested must be capable of being performed in a variety of laboratories and used by a wide range of people trained to different standards it must be very straightforward. In practice, this means that it should not require either elaborate sampling or complex staining. (ii) Transferable: Any set of criteria intended for general use should be able to be applied in different centres and give the same result. (iii) Validated: What is adopted should, if possible, already have wide international experience of use. It would not be sensible to propose an entirely novel protocol for the diagnosis of AD. (iv) Versatile: The practical reality of the neuropathological examination of patients with dementia is that some of the cases that get submitted with a clinical diagnosis of Alzheimer’s disease will have some other dementing disease. This means that, in selecting a diagnostic protocol for general use, the diagnosis of Alzheimer’s disease should not be considered in isolation from that of other dementing diseases that will be encountered by any laboratory that accepts cases with a clinical diagnosis of Alzheimer’s disease. Some assessment of the merits of the available diagnostic criteria for AD can be gained by examining them in using the criteria described above: (i) Simple: In technical terms the CERAD is within the compass of standard neuropathological techniques while the Braak staging is more difficult to accommodate within the standard laboratory procedures. In procedural terms, CERAD calls for assessment of maximum involvement. Our experience has been that

it is quite easy in practice to go to the area of maximum involvement. This is a significant practical advantage in using a system, such as CERAD, that requires assessment of the maximum extent of disease. One of the practical problems in using CERAD is in the definition of a neuritic plaque, about which there is an inevitable imprecision that sometimes causes problems of interpretation and assessment of severity of disease. (ii) Transferable: This is a very important requirement, particularly since the diagnosis of AD rests on a form of quantitative estimation of the amount of pathological change. In relation to this, a European multicentre study supported by EURAGE was published by Duyckaerts et al. (1990) which showed that there was marked variation in absolute numerical values in plaque and tangle counts from different centres, although there was much better agreement among the different centres about the ranking of the cases submitted. More recently, similar results have been obtained by Dr Mirra and her collaborators in CERAD (Mirra, 1997; Mirra et al., 1994). In a study of differences among laboratories using the same staining techniques as in the EURAGE study, although the specific numbers were very variable among the different centres, when a semiquantitative rather than a quantitative method was used (i.e. the methodology suggested by CERAD) the different laboratories were very close in the order in which they ranked the cases for severity. These findings demonstrate that numerical methods cannot at present be recommended for use as the basis of a diagnostic standard where different laboratories will be using a common assessment methodology. In individual laboratories, where there is a specific research purpose and a much more detailed methodology, strictly quantitative methods are more appropriate. (iii) Validated: As has been described, the major currently existing diagnostic standard for AD is the CERAD. These criteria have had very wide exposure. In relation to the various diagnostic criteria for AD, we in Oxford have been collecting prospectively assessed cases for the last few years in the OPTIMA (Oxford Project to Investigate Memory and Aging) project and have performed a comparative assessment of the Khatchaturian, CERAD, and Tierney A3 staging in relation to the cognitive status prior to death (Nagy et al., 1995b, 1998b). Overall, CERAD offers the best correlation. This is true even though the best correlation in terms of individual microscopic findings is with tangle numbers in the frontal and parietal lobes. As already mentioned, this reflects the fact that CERAD requires


the age adjusted semi-quantitative assessment of neuritic plaques and not just total plaques and is therefore really more a measure of neuritic pathology than plaque number. Validation requires that known nondemented as well as demented cases are analysed to discover the ‘false positive’ rate that a given set of criteria may produce. This has unfortunately not been carried out as yet. (iv) Versatile: In this area CERAD explicitly envisages the possibility of diagnoses other than AD. Overall, this review of the currently available diagnostic criteria for AD clearly brings us to the conclusion that CERAD (or a modification) should be recommended as the standard diagnostic tool for AD. No single diagnostic protocol is likely to be completely satisfactory for all purposes, but in our view CERAD has the merits of being: (a) relatively simple (b) has been successfully applied in a number of different centres. (c) has obtained wide national and international recognition and is regularly updated. (d) a diagnostic protocol that can be used for most if not all cases that are submitted as dementia (a significant fraction of which will not be AD) Beyond the selection of suitable diagnostic criteria there are two other practical aspects to the diagnosis of AD. The use of clinical information in the diagnosis of AD This is not really considered in most of the pathologic diagnostic criteria for AD, except to the extent that ‘dementia’ may or may not be present clinically, although the CERAD criteria require some clinical information to complete the assessment. Any discussion of the role of clinical information raises the question of whether the diagnosis of AD can be made in a patient with no history of dementia because the patient has never been examined. CERAD allows only the diagnosis of possible AD in the absence of a history of dementia, no matter how severe the pathological changes. A related question is what sort of clinical information is desirable in the study of patients with AD. At an MRC sponsored conference on brain banking in dementia it was suggested that minimum clinical information in dementia should consist of: (a) a clinical vignette (b) a clinical diagnosis made on operation criteria (DSM III) (c) a standardized rating of mental function (e.g. MMSE) (d) a dementia severity rating (e) preferably recurrent testing at 6-month intervals if possible.

Differential diagnosis There are three aspects to this question. The first is that of the differential diagnosis of patients with dementia who are submitted for neuropathological examination with a clinical diagnosis of ‘Alzheimer’s disease’. This is adequately covered in the requirement for versatility in the diagnostic protocol. The second is the rather shorter list of conditions that could be mistaken histopathologically for AD. About the only disease that could be seriously confused with AD would be Gerstmann–Straussler syndrome (Chapter 17) where, as is not infrequent, the neocortical amyloid deposits are accompanied by marked tangle formation. While this can certainly occur and make the neocortical appearance possible to confuse with AD, the microscopic appearance of the cerebellum is usually quite characteristic in GSS and markedly different from AD. If there is any doubt, immunocytochemistry could be used to define the prion origin for the amyloid deposits in GSS. A second differential involves cortical Lewy body disease in which plaques occur in cortex more abundantly than might be expected by chance. Most cases of cortical Lewy body disease (Chapter 15) have almost exclusively non-neuritic plaques in cortex but some cases also have neuritic plaques. Thus, it is necessary to examine all cases of suspected AD for the possible presence of cortical Lewy bodies. Such a search should include Lewy bodies in the substantia nigra (see further discussion in Chapter 15). A third differential that may be mentioned is progressive supranuclear palsy (Chapter 11) in which quite extensive neurofibrillary pathology may be present in the cortex. However, in this condition there is usually macroscopic dilatation of the fourth ventricle, not seen in AD, and abundant microscopic subcortical pathology, as described in Chapter 11, also not seen in AD. Even in the cortex neurofibrillary pathology shows a subtly different distribution in progressive supranuclear palsy than in AD with, for example, neurofibrillary tangles in the dentate gyrus of the hippocampus and in glial cells. The third aspect in the differential diagnosis of AD is the question of how to classify the non-demented aged with some Alzheimer changes. This is where, as a practical matter, we need to distinguish between a staging system such as the Braak classification and diagnostic criteria for AD. The question that is being asked here is what can and can not be included in the diagnosis of AD, and is there such a thing as ‘pre-Alzheimer’s disease’ (i.e. patients that are going to go on to develop dementia) that can be identified neuropathologically before the patient has any symptoms of dementia? What view is taken on this depends very much on how the pathophysiology of AD is envisaged. If protein

183

184


amyloid deposition is considered to be an essential and early stage in the process that leads to the development of neuritic pathology in AD, then restricted neurofibrillary change or amyloid deposition without cortical neuritic pathology might represent a presymptomatic stage of Alzheimer’s disease. This is a point made by Braak in his stages 1–4 where tangles are confined to the hippocampus and entorhinal cortex. Our view on this is that, although it is in fact a very important practical issue, there is currently insufficient information to offer useful guidance. We regard this as very much an area for active research that will require the examination of a large number of non-demented prospectively assessed controls to be able to formulate an answer to this question.

A minimalist approach to the pathological diagnosis of AD The pathologic scenario that we have been discussing in the last few paragraphs, and which reflects the whole thrust of this volume, has been geared towards the needs of those with a specific interest in dementia in general and AD in particular. Somewhat reluctantly, we are forced to acknowledge that there is also another world where a more ruthlessly pragmatic approach to diagnosis is frequently taken. The needs of the pathologist working in this milieu can be encapsulated in the question ‘what is the minimum sampling and staining required to make the diagnosis of AD?’. While we have had to suppress all our neuropathological instincts to write this, we have concluded that the short answer is 2 sections, each of which can be accommodated on a standard slide. Our recommended sections are: (i) frontal or parietal convexity (ii) lateral temporal lobe, which should include the middle temporal gyrus. The sections should be stained with H&E and a silver stain, preferably one such as Bielschowsky that shows both plaques and tangles. An alternative would be immunostaining adjacent sections for -amyloid and hyperphosphorylated tau. It is perhaps worth mentioning that in a minimalist approach we specifically do not recommend sampling the hippocampus. Ball and colleagues have shown that the appearance of the hippocampus in AD is unreliable from a diagnostic standpoint (Ball, 1977). In this situation the appearance of the hippocampus can be quite misleading since there are frequently moderate and even large numbers of neurofibrillary tangles present in aged people with no significant cognitive decline. Conversely, in AD the severity of the neuronal loss can sometimes produce the appearance of very little neuritic pathology.

In the frontal or parietal lobe section, the presence of significant numbers of tangles, which will invariably be accompanied by neuritic plaques, will be sufficient to make the diagnosis of AD. The rationale for also taking a section of temporal lobe is that, particularly in very old patients, it is our judgement that AD can be present with significant lateral temporal lobe tangle formation but relatively few, if any, tangles being found in the frontal or parietal lobes. It is possible however that many of these patients may have other processes such as diffuse Lewy body formation that might be contributing to their dementia. The reason for not relying only on a temporal lobe section is that in younger patients with only moderate temporal lobe Alzheimer pathology, the absence of frontal or parietal lobe involvement would dissuade us from making the diagnosis of AD. As might be expected, the minimalist approach comes with the neuropathological equivalent of a government health warning to the effect that, although these two sections will allow the diagnosis of AD to be made, they are not adequate for the diagnosis of almost all other causes of dementia. Hence, dementia as a result of another process will not be diagnosable, and, further, other processes contributing to the dementia will also probably be missed.

Pathogenesis of Alzheimer’s disease Particularly in the past decade there has been an enormous expansion in the volume of published work on AD. This reflects both the wide recognition of the social and financial impact of AD in an ageing population and the major advances that have occurred in our understanding of the pathogenesis of the Alzheimer process over the last decade. What follows is only the briefest of surveys that covers only the major highlights of the story as it is currently understood and refers directly to only a small fraction of the published literature. As with all fields of scientific endeavour, one of the features of the progress of ideas is their unpredictability; what seems like an obvious line of advance can be diverted into quite another channel by an unexpected finding. To maintain some sort of organisation it has been divided into sections on molecular genetics, amyloid and neuritic pathology. There is necessarily considerable overlap among these areas and this inevitably generates some repetition, for which reader toleration is requested. For those wishing to delve further into the labyrinth of AD pathogenesis, there are a number of excellent introductory reviews by recognized experts in their field that provide an initial guide down the various avenues of current research into the molecular


genetics of AD (El-Agnaf & Irvine, 2000; Feany & Dickson, 1996; Gillmore et al., 1997; Levy-Lahad & Bird, 1996; Lippa, 1999; McLaurin et al., 2000; Selkoe, 2000, 2001; Swartz et al., 1999; Tolnay & Probst, 1999; Varadarajan et al., 2000).

Molecular genetics As has been described, epidemiological studies of AD looking for risk factors for the development of the disease have consistently shown the importance of the presence of a positive family history for the disease. Although most cases of AD are not clearly familial there are a minority of families, most with early onset disease, in which AD appears to be inherited in the manner of an autosomal dominant gene. In those families without a clear inheritance pattern, although first-degree relatives are at increased risk of developing the disease there is wide divergence of opinion as to the extent of the risk (Breitner et al., 1988; Heyman et al., 1984; Huff et al., 1988; Rocca et al., 1986; van Duijn et al., 1991). The study of the molecular genetics of AD (see also Chapters 6 and 24) is relatively young but its earliest results (St George-Hyslop et al., 1990) confirmed earlier suggestions that what we call AD is aetiologically heterogeneous, and later findings unambiguously have confirmed this supposition (Hyman & Tanzi, 1995). In overall terms, what has been shown is that in early onset AD, there are familial cases that are inherited in a mendelian pattern that have been linked to mutations of genes on chromosomes 1, 14 and 21 and that, together, they probably account for around half of these inherited cases. In addition to these cases inherited as a result of a genetic defect, the allelic inheritance of the apolipoprotein E (APOE) gene on chromosome 19 has a large influence on the susceptibility to the development of AD in older age groups. It is also the case that there are large numbers of late onset cases that are familial, but that in these cases no clear inheritance pattern is evident suggesting that, although there is a major genetic component, it is polygenic. Early onset familial AD Chromosome 21 Molecular genetic interest initially focussed on chromosome 21 because of the long recognized association between Down’s syndrome (trisomy 21) and Alzheimer change in that virtually all Down’s syndrome patients who survive into their fourth decade develop the pathological features of AD (Heston & Mastri, 1977) (Chapter 10). The attraction of chromosome 21 was significantly increased when it was recognised that the gene for the -amyloid precursor protein, which is the source of the specific A protein that forms the amyloid deposits at the centre of the

plaques in AD, and is also deposited in the cerebral vessels, was located on the long arm of this chromosome close to the Down’s syndrome critical region. Not surprisingly, attention was concentrated on the Amyloid Precursor Protein (APP) gene, as the potential source of the genetic defect responsible for familial AD (St George-Hyslop et al., 1987). Although the search for a specific linkage with the APP gene was initially unsuccessful (Goate et al., 1991; Podlisny et al., 1987; Schellenberg et al., 1988; St George-Hyslop et al., 1987; Tanzi et al., 1987; Van Broeckhoven et al., 1987) evidence was found of linkage of AD to chromosome 21 in a single family. This was subsequently shown to be the result of a valine-to-isoleucine substitution at codon 717 of the APP770, which is in the transmembrane portion of the APP and just outside the A region of the protein. By late 1993 a total of 11 families had been identified with this mutation that segregates with AD (Fidani et al., 1992; Karlinsky et al., 1992; Naruse et al., 1991; Sorbi et al., 1993; Yoshioka et al., 1991). Conversely, the mutation has not been found to be present in normal individuals or in a large number of other families who do not have early onset AD. As has been indicated above, this sort of mutation is associated with early onset disease, and in the affected families, mean age of onset has been in the fifth and sixth decades (Hardy et al., 1991). A number of other mutations have been described in the APP gene on chromosome 21. Two are at the same location as that described by Goate et al. (APP717) but substitute glycine and phenylalanine for the valine normally present at this location (Chartier-Harlin et al., 1991; Farlow et al., 1994; Kennedy et al., 1993; Murrell et al., 1991). A double point mutation has also been described at APP670/671 in a Swedish familial AD kindred (Mullen et al., 1992a,b). All the families with mutations in these two locations show early onset AD, but there are other mutations in the APP gene that manifest not as a progressive dementia, but as hereditary cerebral haemorrhage with amyloidosis (Chapter 14). In the affected families there is large-scale deposition of amyloid in cerebral vessels and the formation of numerous diffuse plaques in the cerebral parenchyma. Affected individuals tend to suffer from recurrent cerebral haemorrhages starting in their forties. The first family with this disease was found to have a mutation at APP693 that results in the substitution of glutamine for glutamic acid (Levy et al., 1990) although in a subsequent family with a similar phenotype, the affected individuals have an amino acid substitution at codon 692 (Hendriks et al., 1992). This second family is particularly interesting in that the family members with the gene defect may present either with cerebral haemorrhages or with dementia although it is not known what factors influence the mode of presentation.

185

186


In addition to these symptomatic gene defects, genetic studies of the APP gene have revealed a small number of other point mutations that have no clear association with AD either because they were found in normal individuals or in single patients where there was no evidence of segregation with disease (Clark & Goate, 1993). Although the findings on chromosome 21 are very interesting and indicate that there are circumstances where modification of the APP gene can lead to the development of AD, it must also be remembered that linkage to genetic abnormalities on chromosome 21 is present in only a very small fraction (probably only 2–3%) of the early onset familial cases of AD (Chartier-Harlin et al., 1991; Van Broeckhoven et al., 1987). The recognition of this resulted in the search for linkage to early onset familial disease on other chromosomes. Chromosome 14 Several studies of familial groups have indicated linkage of early onset familial AD to a locus on the long arm of chromosome 14 (Haltia et al., 1994; Mullen et al., 1992a,b; Schellenberg et al., 1992a; St George-Hyslop et al., 1992; Van Broeckhoven et al., 1992). In contrast to the rather small fraction of early onset familial AD cases linked to abnormalities on chromosome 21, as many as 70% of cases of early onset FAD may link to the locus on chromosome 14. Many of these cases also have an unusually early onset (approximately a decade earlier those with chromosome 21 mutations) suggesting that the defect has a fundamental effect on the processes leading to the development of AD. In initial studies of this chromosome the region flanking the disease locus was first narrowed to an 8.9 megabase region at 14q24.3. Candidate genes for the AD disease locus included promising contenders such HSPA-2, a 70kD heat shock protein, the proto-oncogene c-Fos, which is a DNA binding regulatory protein and -1-antichymotrypsin. However, these seemingly promising contenders appear all to have been eliminated by Sherrington et al. who have cloned a gene that they have designated S182 from six linked pedigrees where there are five different missense mutations found only in family members with AD (Sherrington et al., 1995). The molecular configuration of this 467 amino acid protein, presenilin 1 (PS1), is indicative of an integral membrane protein. Immunocytochemical studies of PS1 suggest an intracellular location in both Golgi and endoplasmic reticulum. Several defects have already been located in PS1 and new mutations are described almost on a daily basis (Lippa, 1999). The nature of the metabolic disturbance introduced by the genetic defects is not clearly established as yet although there are suggestions that it can control -secretase processing of APP and

that PS1 mutations lead to increased amounts of -amyloid (Wolfe et al., 1999). It has also been shown in vitro that PS1 is involved in the trafficking and generation of APP fragments (Naruse et al., 1998). Recent animal studies however, show that loss of PS1 function changes APP processing (Palacino et al., 2000; Steiner et al., 1999). It also appears that PS1 mutations sensitize neurones to apoptosis due to interactions with several members of the Bcl-2 protein family (Lippa, 1999; Passer et al., 1999; Rohan de Silva & Patel, 1997) and increase neuronal vulnerability to glutamate and oxidative stress in a calcium-dependent manner (Czech et al., 2000; Guo et al., 1999; Mattson et al., 2000). Chromosome 1 Chromosomes 21 and 14 do not exhaust the possible locations for genetic abnormalities associated with early onset familial AD. In the course of their studies of genetically related Volga German families Schellenberg et al. have demonstrated that the disease in these families with early onset disease is not linked to abnormalities on either chromosome 14 or 21 (Schellenberg et al., 1992a). The genetic defect in this group has recently been identified in a gene at chromosome 1q31–42 as a molecule with a very high degree of structural and amino acid sequence homology to the PS1 protein on chromosome 14, including the characteristic 7 transmembrane regions (Levy-Lahad & Bird, 1996; Levy-Lahad et al., 1995a,b). Because of the structural homology, the gene product has been initially referred to as PS2. Within the transmembrane regions of the molecule 84% of the amino acid sequence is identical in PS1 and PS2. This high degree of homology was in fact a significant element in the identification of this particular gene as the most probable candidate gene for the cause of AD in the Volga German families. Another source of similarity between PS2 and PS1 is that the genetic defect in the Volga German families on PS2 has been located close to the beginning of one of the transmembrane regions, as also have several of the reported mutations in the PS1 gene. Recent studies indicate that, as expected from the structural similarities, mutations of the gene on chromosome 1 will have generally similar metabolic effects to the PS1 gene on chromosome 14. It has been found that PS2 is a signalling molecule in the apoptotic pathway (Sun et al., 2001) and its mutations result in overproduction of -amyloid (Sato et al., 2001). The mechanisms by which these mutations in the APP gene, on chromosome 14 and chromosome 1 produce AD are not addressed by molecular genetic techniques. However, in relation to APP gene defects on chromosome 21, one highly suggestive finding is that cells in vitro transfected with APP containing the 670/671 mutation have been found to have a 5–8 fold increase in A production


(Cai et al., 1993). Increased A production could plausibly be associated with an increased likelihood of amyloid deposition, thereby initiating the pathophysiologic cascade leading to the development of AD. The consequences to APP metabolism of the other mutations of the APP gene have not yet been established. Clearly, mutations on chromosomes 1 and 14 affect A production and metabolism. However, the possibility that some of the effects of chromosome mutations, especially those that do not directly affect the APP gene, operate remotely or on selected cell populations such as for example, microglial cells, may make them difficult to investigate in in vitro systems. Late-onset AD Most of the genetic linkage and molecular studies have concentrated on families with early disease, and there has been relatively little work on the genetics of late onset disease. This is largely because of the inherent and ineradicable difficulties of working with older age groups. However, AD has frequently been observed to cluster in families with late onset although, because the disease is so common in the elderly population, some degree of familial clustering can be expected for non-genetic reasons. Chromosome 19 The prevailing view is that most of the late onset forms of AD are multifactorial in origin with contributions from genetic and environmental influences. In this group the predominant genetic influences are presumed to be polygenic and large-scale segregation analyses in this group of patients has suggested that a major gene effect was responsible for only about 4% of the variance (Farrar et al., 1991). However, linkage analysis in families with late onset AD suggested an association with markers on the proximal part of the long arm of chromosome 19 at position q13.2 (Pericak-Vance et al., 1991). Alzheimer’s disease and the apolipoproteins Two genes that map to the appropriate region on chromosome 19q13.2 are those for apolipoprotein C2 and apolipoprotein E and association studies have been carried out for both of these genes. Apolipoprotein C2 (ApoC2) was found to be significantly increased in affected family members in 23 families (Schellenberg et al., 1987, 1992b). Perhaps more significantly, in the initial report, the frequency of the E4 allele of Apolipoprotein E was increased from 13% to 48% in a group of 30 AD patients from different families (Strittmatter et al., 1993). These results were effectively duplicated in numerous subsequent studies of familial and sporadic late onset cases which show a significantly increased frequency of the E4 allele compared to age

and gender matched controls (Chartier-Harlin et al., 1994; Corder et al., 1993; Mayeux et al., 1993; Nagy et al., 1995a; Noguchi et al., 1993; Payami et al., 1993; Saunders et al., 1993; Smith et al., 1996) (Chapter 6). The, already considerable, interest in the apolipoprotein E gene was greatly augmented when it was shown by the Duke University group that, while the possession of the ε4 allele is associated with an increase in the frequency of AD, the frequency of the ε2 allele in patients with AD is markedly lower than in the normal population (Corder et al., 1994), a result that has subsequently been confirmed (Benjamin et al., 1994b). Possession of the ε2 allele therefore confers protection against the development of Alzheimer’s disease. This result has been strengthened by studies in our own OPTIMA population of prospectively assessed postmortem confirmed cases of AD (Smith et al., 1994). Interestingly, ApoE genes do not seem to have any effect on the incidence of either Parkinson’s disease or diffuse Lewy body disease (Benjamin et al., 1994a), and it has been reported that there is no association between apolipoprotein Eε4 and AD in Nigerians (Osuntokun et al., 1995), although interestingly there is a reported association in African Americans (Hendrie et al., 1995). It has also been shown that possession of the ε2 allele also protects against heart disease and leads to an increase in the frequency of the ε2 allele in centenarians compared to younger age groups (Menzel et al., 1983). This finding demonstrates that the protective effect cannot be an artefact of early death from other causes of patients with the ε2 allele who, had they survived, would subsequently have gone on to develop AD. It is not at all clear how the known functions of these proteins could be relevant to the development of AD. It is possible to envisage ways that they might influence the processing of APP, and apoE has been detected in amyloid deposits (of both Alzheimer and non-Alzheimer type), neurofibrillary tangles and congophilic angiopathy (Namba et al., 1991). There are also some indications that the different isoforms of apoE bind amyloid to different degrees, with the E4 isoform having a more rapid binding (Strittmatter et al., 1993). This has led to the suggestion that apoE might in some way enhance the sequestration of soluble A into plaques and thus hasten the development of one of the significant morphologic manifestations of AD. It has been reported that patients with the ε4 allele have significantly greater vascular and plaque A deposition in the cerebral cortex (Schmechel et al., 1993). At the molecular level there has been great interest in the interaction between the different apolipoprotein alleles and A. Ma et al. (1994) have shown that Apo E4 produces a greater acceleration in amyloid fibril formation than either Apo E2 or Apo E3. In an interesting correlation with the recent evidence

187

188


that the predominant A species in plaques is A1-42(43) it was also shown that the effect of Apo E4 was significantly more pronounced on A1-42(43) than on A1-40. While not conclusive these results tend to suggest that the presence of the Apo Eε4 allele might be expected to enhance any tendency to fibrillar amyloid deposition and plaque formation and might therefore be expected to favour the development of AD (Younkin, 1995). A study from Finland (Polvikoski et al., 1995) has shown that there is a correlation between apolipoprotein E genotype and the presence of dementia, the frequency and degree of cortical amyloid deposition, and the number of neurofibrillary tangles in a community-based autopsy study of patients over the age of 85 years. These features were all found to the greatest degree in patients with the ε4/ε4 or ε4/ε3 genotype, and to a progressively lesser degree in patients with the ε3/ε3 and ε3/ε2 genotypes. In terms of the dementia, 28 of 33 patients carrying the ε4 allele were demented, although the presence of five patients carrying the ε4 allele who were not demented shows that other factors must be involved in the development of dementia and confirms the status of the apolipoprotein genotype as a major risk factor for the development of dementia. A rather more heterodox view of the significance of the E4 isoform in AD that has recently been put forward by Roses and Strittmatter is that the association might take the form of preferential protection of patients with the apoEε2 genes making such patients less likely to develop AD. Their suggestion is that this might occur by differences in the binding to tau protein which would prevent, or at least influence, the abnormal phosphorylation of tau that occurs in the formation of neurofibrillary pathology (Roses, 1994). In line with this it has been found that generally the ApoE3 isoform seems to encourage neurite outgrowth and synaptic remodelling, thus promoting repair mechanisms in neurons (Poirier, 1994; Poirier et al., 1995). In one significant respect the debate about the significance of the apolipoproteins on chromosome 19 differs from the other genetic associations of AD in that there is no suggestion of a genetic defect on chromosome 19, but rather that the different normal apolipoprotein alleles confer a different predisposition to develop AD. The apolipoproteins are more properly considered as normally occurring genetically determined risk factors for AD not genetic defects that initiate or accelerate the pathophysiology of the Alzheimer process. By introducing the topic of the amyloid precursor protein, the discussion of the molecular biology of AD has, to a degree, anticipated the examination of the cellular pathophysiology. The two major neuropathologic features of

Alzheimer’s disease are senile plaques and neurofibrillary pathology. There is, as indicated, clear evidence to suggest that the development of neuritic pathology in the form of neurofibrillary tangles and other neuritic components is the key determinant of the development of the dementia seen in AD. Neuritic pathology however, is not unique to Alzheimer’s disease, occurring for example in other degenerative diseases such as progressive supranuclear palsy, ALS/Parkinsonism/Dementia of Guam, and GerstmannStaussler-Scheinker syndrome.

Amyloid in Alzheimer’s disease Unlike neurofibrillary pathology, which occurs in a number of quite disparate pathologic processes, the amyloid that is deposited in AD is rather more specific. To claim that it is unique to AD would not be quite correct, since amyloid angiopathy and diffuse amyloid plaques are quite frequently seen in the brains of older people who do not have AD. As mentioned above in the discussion of the molecular biology of chromosome 21, there is a hereditary disease characterised by deposition of vascular amyloid that results in recurrent cerebral haemorrhage without the necessary presence of dementia or other features of AD. However, the existence of some families with early onset AD associated with a genetic abnormality in the APP gene establishes amyloid deposition as one of the central processes in the development of AD. Hence, an understanding of the pathophysiology and consequences of amyloid production and deposition are central to an understanding of the pathophysiology of AD itself, and over the past decade an enormous volume of published work has focused on the process of amyloid production and deposition in aged brains and AD. Metabolism of APP and production of A As has already been mentioned, the amyloid in AD is produced from the APP located on chromosome 21. This protein is actually a small family of transmembrane glycoproteins that exists in at least four forms that arise as a result of variable exon splicing of a single gene. Two forms, APP751 and APP770 contain a Kunitz-type protease inhibitor (KPI) domain, while the other two, APP695 and APP714 do not (Neve et al., 1988). The function of the molecule is not known, although its primary structure with an intracellular C-terminus, a transmembrane region and an extracellular N-terminus have led to the suggestion that it may function as a cell surface receptor (Kang et al., 1987). More recent studies indicate that APP has a role in the development of the nervous system (Kirazov et al., 2001), and has a role in the maintenance of synaptic connections and in stressresponse and injury repair (Panegyres, 2001).


α Secretase β Secretase

γ Secretase

β APP

N

C

α Secretase

β Secretase sAPP β

sAPP α

γ Secretase

γ Secretase

Aβ Aggregated A β Fig. 9.29. The processing of APP in normal circumstances and in Alzheimer’s disease. The processing of the APP protein by the and secretases (normal conditions) leads to the formation of the soluble APP fragment (sAPP). The processing of APP by the and secretases leads to the formation of the -amyloid (A) fragment that aggregates and becomes the main component of amyloid plaques.

APP is expressed in almost all mammalian cells and shows a high degree of evolutionary conservation. Although the function of APP is not known, there are several circumstances that alter its cellular expression. These include metabolic insults such as ischaemia and head trauma, and also exposure to excitotoxins, all of which upregulate expression of APP. APP levels also vary during development and in response to neurotrophic factors and cytokines. None of these influences give any very specific indication of APP function, but have suggested that it might have a role in neuronal plasticity and the responses of the brain to trauma or other damage. In AD there is deposition of what is called -amyloid that is derived from a fragment of this molecule. Conceptually, the process of amyloid formation could be a result of excess production of amyloid- as a result of overexpression of the APP gene, the production of an abnormal gene product or altered processing of normally derived -amyloid fragment. The heterogeneity that we have already seen in the genetic influences on the development of AD make it probable that all three of these mechanisms operate in the pathogenesis of AD. The A deposited in AD is composed of a hydrophobic peptide derived the APP that is 40–42 amino acids long derived from a part of the APP molecule 12–14 amino

acids of which are intramembranous with a 28 amino acid extracellular segment (Kang et al., 1987). There is now a considerable body of literature on the possible mechanisms of production of A from APP (Fig. 9.29). At least three different cellular mechanisms of APP processing have been discovered. One mechanism is the action of the so-called -secretase, which cleaves intact in situ APP just outside the transmembrane region. This mode of processing precludes A production because the cleavage site is within the A region. The action of this enzyme releases a soluble product (APPs ) into the medium (Esch et al., 1990). However, studies have shown that this mechanism probably accounts for a minority fraction of cellular APP processing and this finding, together with the obvious fact of A deposition in many aged brains, prompted the search for alternative methods of APP processing that would release the A fragment and permit the deposition of amyloid. This search rapidly led to the discovery of a lysosomal pathway processing APP, which is capable of producing A. Protein chemical studies demonstrating the accumulation of A-containing fragments in cells in which lysosomal function has been inhibited have been supported by the direct immunocytochemical demonstration of A epitopes in lysosomes (Golde et al., 1992). Further evidence for the existence of a lysosomal pathway was the demonstration

189

190


Plaque Soluble portein

Activated astrocytes

Insoluble peptide

Neuritic elements

Microglia

Fig. 9.30. The formation of plaques. The excreted -amyloid forms insoluble aggregates in the extracellular space. Neuritic elements grow into these aggregates leading to the formation of neuritic plaques. Activated astrocytes and microglial invade these amyloid deposits and contribute to the formation of the senile plaque.

that APP labelled with antibodies or biotin is incorporated into lysosomes (Benowitz et al., 1989; Haass et al., 1992a). To these two pathways, a third has been added with the discovery that A is normally secreted as a soluble peptide into biological media (Haass et al., 1992b; Seubert et al., 1992; Shoji et al., 1992). This process seems to occur in the absence of any membrane injury and to be a normal part of the metabolism of the cell, and its occurrence suggests that A production is not necessarily a manifestation of aberrant intermediary metabolism in a damaged or defective cell. The A so produced has been detected using A antibodies to immunoprecipitate the conditioned media. Sequencing of the immunoprecipitated material shows it to be A32−34 . This form of A secretion has been found in a variety of cells types in culture including human neurons, astrocytes and endothelial cells, but, in these experiments, is not detectable in lysosomes. The site of production of the A in this processing pathway is not certain, although the Golgi apparatus has been suggested as one possibility. The finding of A secretion as a result of normal intermediary metabolism suggests that it might have a normal physiologic function. Currently this is a speculative suggestion that is made at least credible by its action to produce trophic effects on neurons when added to culture medium and the demonstration of its ability to interact with extracellular matrix components to promote neurite outgrowth (Selkoe, 1994).

Neurotoxicity of amyloid A more conventional role for A is that of a neurotoxin, this being a more obvious way in which the deposition of amyloid could be a harbinger for the subsequent develop-

ment of neuritic pathology. One of the difficulties in accepting a simple model of A as a significant neurotoxin is the existence of the diffuse plaque. As described earlier, diffuse plaques are ‘amorphous’ deposits of A that occur in grey matter and that are not apparently associated with any disturbance in the tissue. There is no detectable sign of damage to neurons or glia in association with these deposits, a finding that indicates that the mere presence of A alone does not produce detectable parenchymal damage. This has been seen by some as a major objection to the ‘amyloid cascade hypothesis’ of AD, but the deposition of A and its evolution into protein amyloid is envisaged as a multistage process of which the deposition of A in the tissue is only the first. The evolution of the mature plaque, from the initial deposition of A to a senile plaque with a central core of fibrillar protein amyloid surrounded by a halo of distorted paired helical filament (PHF)-containing neurites, is a process that offers several possible avenues that could lead to local cell damage (Smith et al., 1994). In diffuse plaques, although A is present, ultrastructural examination shows few or none of the fibrils characteristic of the amyloid seen in mature plaque cores and amyloid angiopathy. The development of the more conventional plaque core involves a molecular conformational change of the A peptide to a -pleated sheet and the aggregation of these changed molecules into fibrils (Fig. 9.30). This change in physico-chemical state of A could of itself substantially alter its local effects. In support of this suggestion there is some experimental evidence that the use of ‘in vitro aged’ A, where prolonged incubation of synthetic peptides in buffer leads to conformational change and aggregation, can increase the toxicity of A for cultured neurones.


Neurofibrillary pathology Although the role of -amyloid protein in the pathogenesis of AD has attracted a great deal of attention the mere deposition of the protein in the tissues is insufficient to produce cognitive decline. There are a number of good studies of protein amyloid deposition and non-neuritic senile plaques that show significant overlap between normals and patients with symptomatic AD (Crystal et al., 1988, 1993; Davies et al., 1988). The best correlations of morphologic change with cognitive decline are with the formation of neuritic pathology. As has been described earlier, neuritic pathology occurs in three basic forms; neurofibrillary tangles that develop in the cell bodies and apical dendrites of neurones, the so-called neuropil threads which form in distal dendrites and the abnormal dilated tortuous neurites seen in association with some amyloid deposits (neuritic plaques) (see Chapter 1). All these structures have in common the presence of PHF as their major component and a smaller component of straight filaments. PHF are relatively insoluble so that the degeneration of the nerve cell does not result in the immediate dissolution of the PHF, which remain as extracellular ‘ghost’ tangles that can be seen in the neuropil. Ghost tangles are most prominent in the entorhinal cortex and the hippocampus where they may be present in large numbers, but can usually be seen in other areas such as the nucleus basalis of Meynert, the periaqueductal grey, the raphe and the locus ceruleus. Curiously, they are rarely conspicuous in the neocortex, even when there are large numbers of intracellular tangles. Composition of tangles There has been a large amount of work relating to the molecular biology and chemistry of neurofibrillary pathology and tau and other tangle-related proteins (Arvanitakis & Wszolek, 2001; Delacourte, 2001; Garcia & Cleveland, 2001; Goedert & Spillantini, 2001). PHF are composed almost entirely of an abnormally phosphorylated variant of a normally occurring cellular protein associated with microtubules called tau. When combined into tangles, PHF are extremely insoluble and correspondingly hard to study, but there is a phase where PHF have formed within the cytoplasm but not yet aggregated into tangles and in this dispersed form PHF are much more soluble. Work on tau and neurofibrillary pathology has been greatly aided by the discovery in 1991 of extraction techniques for the dispersed PHF. In the normal adult cell tau is produced in six isoforms ranging in size from 352 to 441 amino acids with three or four sets of tandem repeats and lettered A–F in order of increasing size (Goedert et al., 1989; Himmler, 1989). Fetal cells produce only isoform A. Much of the work done on

the function of this protein has been done in vitro but it is almost certain that the normal function of this protein is to promote the formation and increase the stability of the cytoplasmic microtubules. Experiments in which anti-sense oligonucleotides to block the function of tau have been inserted in developing neurons show that this prevents differentiation of neurites into the axon and suggest that tau is important in the development of axonal morphology. This is reflected in the presence of microtubules within the normal axon and it is reasonable to suppose that any process that seriously disrupts the formation and stability of microtubules would have a significant effect on axonal morphology and functions such as rapid axonal transport in which microtubules play a critical role. Phosphorylation of tau and the formation of PHF Normal adult tau has a number of phosphorylation sites that play a key role in its normal physiological function. Increasing the phosphorylation state reduces the degree of binding of tau to microtubules (Bramblett et al., 1993) and regulation of the processes of phosphorylation and dephosphorylation of tau is probably critical to its normal function. In tau isolated from material obtained at postmortem, a major difference between normal functioning tau and the tau protein found in neurofibrillary tangles and PHF is that the tau in PHF is much more heavily phosphorylated (Hasegawa et al., 1992; Lee et al., 1991). Curiously, as was suggested by the fact that antibodies to fetal tau label neurofibrillary tangles while antibodies to normal adult tau do not, fetal tau is much more heavily phosphorylated than the normal adult isoforms isolated from post-mortem brain. In this respect it more closely resembles the degree of phorphorylation seen in PHF-tau. These results tended to suggest that abnormal hyperphosphorylation was an essential feature of the conversion of normal tau into PHFtau (Goedert, 1993; Trojanowski et al., 1993a,b). However, an aspect of the difficulties in working with human disease for which there is not a good animal model is illustrated by the recent finding that soluble tau isolated from brain biopsies where there is no post-mortem delay has a similar degree of phosphorylation to PHF tau (Matsuo et al., 1994). The difference between the phosphorylation state of post-mortem and biopsy derived tau is attributed to the action of dephosphorylating enzymes in the post-mortem interval. This suggests that the phosphorylation of tau seen in PHF is not simply the result of excessive phosphorylation, but rather a result of either a change in the balance of phosphorylating and dephosphorylating enzymes in the AD brain or an increase in the stability of the phosphorylation state of tau in AD (Fig. 9.31). Studies on tau extracted from PHF have demonstrated that the phosphorylation is

191

192


MT

NPt

MT-τ τ

τ τ-P

NFT PHF- τ

SPs ABNORMAL PHOSPHORYLATION •overactive kinases •hypo-active phosphatases

Fig. 9.31. Tau and Alzheimer-type neuritic pathology. The tau protein (τ ) released from the microtubules (MT) is phosphorylated due to the overactive kinases and hypo-active phosphatase activity. The phospho-tau (τ -P) has an affinity to form paired helical filaments (PHF-t) that are the main components of the neuritic elements of senile plaques (SPs), neurofibrillary tangles (NFT) and neuropil threads (NPt).

not specific to any of the particular isoforms of the protein since all the six adult isoforms are present in the PHF in the same proportions as in normal brain (Goedert et al., 1992). As would be expected the major consequence of this change in phosphorylation state is that phosphorylated tau has a much lower affinity for microtubules than normal tau. That this loss of affinity is a result of the phosphorylation can be shown by the fact that tau extracted from dispersed PHF and subsequently dephosphorylated has a normal affinity for microtubules. As with the formation of protein amyloid from non fibrillar A, the steps in the assembly of dispersed PHF from phosphorylated tau and their subsequent aggregation into neurofibrillary tangles are not clearly defined but it has been shown that PHFlike filaments can self-assemble in vitro from bacterially expressed fragments of tau. A difference between the tau in dispersed PHF and the PHF aggregated into neurofibrillary tangles is that in dispersed PHF the whole tau molecule is present, while in the tangle fragments only the carboxy terminus half of the tau molecule is present and the protein is also ubiquitinated. From these studies of the biochemistry of tau it is clear that regulation of the phosphorylation state of the tau protein is a key event in the process of the formation of PHF and neurofibrillary pathology from tau protein. Attention

is now shifting to looking for the candidate enzymes that could perform the phosphorylation and dephosphorylation and the events that could control or modify their activity. Examination of the sequence of PHF tau has shown that there are between six and eight sites on the molecule that are phosphorylated in PHF tau, and most of them are next to a proline residue. Hence, the most obvious candidates for the enzymes that perform the phosphorylation are what are called proline directed kinases. If the abnormal phosphorylation of tau seen in PHF is a consequence of hyperactivity of the normal mechanism of phosphorylation there are a number of candidate proline directed enzymes that are capable of phosphorylating tau such as microtubule associated protein (MAP) kinases, glycogen synthase kinase (GSK) 3 and other proline directed kinases. Protein phosphatases 2A (PP2A) and 2B (PP2B) (Matsuo et al., 1994) have been identified as being able to dephosphorylate tau. The abnormal phosphorylation of tau that occurs in AD implies that the normal regulatory mechanism that controls the phosphorylation state of tau has broken down but currently it is not known where the regulatory defect, or defects, are located. The precise effects of the reduction in binding of tau to microtubules on the formation, stability and function of the microtubular apparatus of the cell are not known, but


Microtubules

Tau protein

Dendrites Neuropil threads

Tangles

Axon Fig. 9.32. The formation of neuritic pathology in Alzheimer’s disease. In healthy neurones the microtubules form a dynamically changing but stable cytoskeleton. In AD the tau is dissociated from the microtubules and gives rise to the formation of tangles, which occupy the cell body of the neurone, while the neurites filled with PHFs appear as neuropil threads. These features, due to the insoluble nature of PHFs, persist even after the death of the neurone.

in the presence of a significant accumulation of PHF there is a reduction in the normal cytoskeletal content of microtubules and neurofilaments (Fig. 9.32). Given the wide distribution of PHF in cell bodies and neuropil threads it is not difficult to surmise that the extensive cytoskeletal disruption that this implies would be associated with a severe disruption of axonal transport and cell metabolism and could ultimately produce accelerated cell death (Lee et al., 1992; Trojanowski et al., 1993a,b). Another possible mechanism of cell damage is that the hyperphosphorylation of tau is a reflection of a more generalized activation of kinases in the affected neurones that might be a manifestation of disordered regulation of intracellular calcium. There is certainly evidence of abnormalities of calcium regulating proteins including protein kinases that are altered in AD and a suggestion of increased protease activity including calcium activated proteases. It is at this level that a connection between tissue deposition of A and neurofibrillary pathology is beginning to emerge (Landfield et al., 1992; Mattson et al., 1993). Whatever the precise mechanism, as an empiric observation, it is abundantly clear from the presence within tissue of ‘ghost’ extracellular tangles that the presence of tangles within cells is associated with cell death and that

the development of neurofibrillary pathology is seriously detrimental to cell health. Regulation of tau and its association with amyloid The amyloid hypothesis depends on the demonstration that there is a causative relation between the production and/or deposition of extracellular A and the later development of pathological changes among which are intracellular neurofibrillary pathology, loss of synapses and neuronal death. However, the study of the toxic effects of A has been complicated by inconsistent results with A being reported to have both neurotrophic and neurotoxic effects on neurones in culture. What is not clear is the mechanism of this toxic effect, although a number of possibilities have been advanced, and whether A acts alone or in concert with other influences such as excitotoxins, free radicals, ischaemia or hormonal effects. A possible direct connection between amyloid, and hyperphosphorylated tau, and therefore with neurofibrillary pathology comes from reports that A can induce the formation of calcium channels in artificial membranes (Arispe et al., 1993) and can amplify Ca++ signalling in neurones (Hartmann et al., 1993). At least in principle this could prove

193

194


a mechanism to permit increased or unregulated entry of Ca++ into affected cells leading to inappropriate activation of calcium activated kinases, including those that could hyperphosphorylate the tau protein. It is exceedingly unlikely that amyloid cytotoxicity will be shown to be as simple as this mechanism suggests. This is particularly the case bearing in mind the slowness of the process of development of neurofibrillary pathology and, as indicated from the study of patients with Down’s syndrome, the number of years that elapse from the first deposition of non-fibrillar A to the final flowering (if that is the appropriate word) of Alzheimer pathology that follows over the following two decades. This very tardiness argues for a small, but perhaps persistent, perturbation in neuronal metabolism, that results in the slow but relentless accumulation of damage that is eventually sufficient to cause neuronal death, having first progressively compromised neuronal function. Additionally, the evidence suggesting that the aggregated form of the A molecule is neurotoxic in any way is hampered by the fact that these in vitro studies tend to use extremely large concentrations of the peptide (Neve et al., 1988). Although there is a considerable body of scientific opinion that regards amyloid production and deposition as the starting point of a pathophysiological process that ultimately leads to the development of the neuritic changes that are closely correlated with the dementia of AD the ‘amyloid cascade hypothesis’ is far from complete. While the chemical nature of the amyloid and the identity of the APP have already been identified and the metabolic pathways by which A (the polypeptide that forms the amyloid) is derived from the APP, are in the process of solution, there still remain major questions to be answered. Most importantly, the ‘amyloid cascade hypothesis’ does not provide a precise cell biological connection between the extracellular deposition of amyloid and the intracellular development of neuritic pathology. Although several hypotheses have been formulated to discern the association between amyloid and cellular damage and the possible interactions between calcium influx, protease activation and hyperphosphorylation of tau, none of the existing theories provide a satisfactory answer to the question. The most exiting recent development in the amyloid field is the discovery that vaccination with soluble A generates toxicity neutralizing antibodies (Lambert et al., 2001), reduces the number of amyloid plaques in transgenic animals (Schenk et al., 1999) and may even reduce age-related cognitive impairment in animal models (Ingram, 2001). The possibility of long-term vaccination therefore became a possible therapeutic avenue for the reduction of cortical amyloid deposits and associated cognitive deficit in

Alzheimer’s disease patients. In the light of recent events, however, the clinical value of such intervention became questionable (Munch & Robinson, 2002).

The cell division cycle and Alzheimer’s disease The latest, and most comprehensive, theory for the formation of AD-related amyloid and neuritic pathology is the ‘cell cycle hypothesis’. Although it has been a long enduring dogma of neuroscience that neurons of the adult central nervous system are terminally differentiated cells, the finding of cell division-related proteins in neurons leads to the formulation of the hypothesis that cell cycle re-entry may be at the root of the pathogenesis of AD. The extensive search for answers into the involvement of cell division related events into the pathogenesis of AD, despite (or maybe fuelled by) strong opposition, lead to the accumulation of supporting evidence from several independent groups (Arendt, 2000; Arendt et al., 1996; Gartner et al., 1999; Nagy, 2000; Nagy et al., 1998c; Vincent et al., 1998; Yang et al., 2001). These studies indicate that while cell cycle re-entry on its own may not be harmful to neurons, since it could be followed by re-differentiation and synaptic remodelling, the progression of the cycle into its late stages will inadvertently lead to neuronal death or the accumulation of hyperphosphorylated tau and -amyloid (Nagy, 2000). From early studies it was apparent that the areas affected by Alzheimer’s disease are the ones undergoing extensive synaptic remodelling even at an advanced age (Arendt et al., 1995a,b,c, 1998; Braak et al., 2000). It looks possible that the temporary loss of synapses, during this remodelling phenomenon, may trigger cell cycle-related events in neurons (Arendt et al., 1995a,b,c; Nagy et al., 1998c). This in turn, in patients where the regulatory mechanisms are not fully functional, allows neurones to progress through DNA replication (Yang et al., 2001) and into the late stages of cell division. Cytochinesis in adult neurons does not seem possible, and the cells that reached the mitotic phase will either die via a programmed cell death mechanism (reminiscent of apoptosis) or, if they are protected from cell death, may survive, and since all cellular mechanisms necessary are activated, produce hyperphosphorylated tau, destabilize their microtubular system and accumulate -amyloid. This hypothesis, although providing an explanation for the concomitant development of both amyloid plaques and neurofibrillary pathology, provokes more questions than it answers. Why would these phenomena affect mainly the specific areas of the brain that are involved in AD? How do the known risk factors and protective factors affect neuronal cell cycle regulation? And last but not least, what are


the factors that lead to the de-regulation of the cell cycle in neurones in some elderly, causing AD, but not in others? Probably the search for answers will be a long one and it will take years until all the pieces fall into place to let us understand fully the development of AD. For the clinical scientist who interacts with patients, perhaps the most attractive feature of any hypotheses regarding the pathogenesis of AD is that, if it turns out to be substantially correct, it may open up a number of potential therapeutic options. This would be of the greatest benefit to the very large numbers of people at risk from AD who otherwise face the certain prospect of a progressive and irreversible dementia with all the personal, family and social consequences that this implies.

REFERENCES Aharon-Peretz, J., Cummings, J. L. & Hill, M. A. (1988). Vascular dementia and dementia of the Alzheimer type. Cognition, ventricular size, and leuko-araiosis. Arch Neurol, 45, 719–21. ¨ Alzheimer, A. (1907). Uber eine eigenartige Erkränkung der Hirnrinde. Allgemeine Zeitschrife fur ¨ Psychiatrie, 64, 146–8. Amaducci, L. A., Rocca, W. A. & Schoenberg, B. S. (1986). Origin of the distinction between Alzheimer’s disease and senile dementia: how history can clarify nosology. Neurology, 36, 1497–9. Arendt, T. (2000). Alzheimer’s disease as a loss of differentiation control in a subset of neurons that retain immature features in the adult brain. Neurobiol Aging, 21, 783–96. Arendt, T., Bigl, V., Arendt, A. & Tennstedt, A. (1983). Loss of neurons in the nucleus basalis of Meynert in Alzheimer’s disease, paralysis agitans and Korsakoff’s Disease. Acta Neuropathol, 61, 101–8. Arendt, T., Bruckner, M. K., Bigl, V. & Marcova, L. (1995a). Dendritic reorganisation in the basal forebrain under degenerative conditions and its defects in Alzheimer’s disease. II. Ageing, Korsakoff’s disease, Parkinson’s disease, and Alzheimer’s disease. J Comp Neurol, 351, 189–222. Arendt, T., Bruckner, M. K., Bigl, V. & Marcova, L. (1995b). Dendritic reorganisation in the basal forebrain under degenerative conditions and its defects in Alzheimer’s disease. III. The basal forebrain compared with other subcortical areas. J Comp Neurol, 351, 223–46. Arendt, T., Marcova, L., Bigl, V. & Bruckner, M. K. (1995c). Dendritic reorganisation in the basal forebrain under degenerative conditions and its defects in Alzheimer’s disease. I. Dendritic organisation of the normal human basal forebrain. J Comp Neurol, 351, 169–88. Arendt, T., Rodel, L., Gartner, U. & Holzer, M. (1996). Expression of the cyclin-dependent kinase inhibitor p16 in Alzheimer’s disease. Neuroreport, 7, 3047–9. Arendt, T., Bruckner, M. K., Gertz, H. J. & Marcova, L. (1998). Cortical distribution of neurofibrillary tangles in Alzheimer’s disease

matches the pattern of neurons that retain their capacity of plastic remodelling in the adult brain. Neuroscience, 83, 991– 1002. Arispe, N., Pollard, H. B. & Rojas, E. (1993). Giant multilevel cation channels formed by Alzheimer disease amyloid beta-protein [A beta P-(1–40)] in bilayer membranes. Proc Natl Acad Sci USA, 90, 10573–7. Aronson, M. K., Ooi, W. L., Morgenstern, H. et al. (1990). Women, myocardial infarction, and dementia in the very old. Neurology, 40, 1102–6. Arriagada, P. V., Growdon, J. H., Hedley-Whyte, E. T. & Hyman, B. T. (1992). Neurofibrillary tangles but not senile plaques parallel duration and severity of Alzheimer’s disease. Neurology, 42, 631–9. Arvanitakis, Z. & Wszolek, Z. K. (2001). Recent advances in the understanding of tau protein and movement disorders. Curr Opin Neurol, 14, 491–7. Awad, I. A., Spetzler, R. F., Hodak, J. A., Awad, C. A. & Carey, R. (1986). Incidental subcortical lesions identified on magnetic resonance imaging in the elderly. I. Correlation with age and cerebrovascular risk factors. Stroke, 17, 1084– 9. Bachman, D. L., Wolf, P. A., Linn, R. et al. (1992). Prevalence of dementia and probable senile dementia of the Alzheimer type in the Framingham Study. Neurology, 42, 115–19. Ball, M. J. (1977). Neuronal loss, neurofibrillary tangles and granulovacuolar degeneration in the hippocampus with ageing and dementia. A quantitative study. Acta Neuropathol (Berl), 37, 111–18. Ballard, C., Bannister, C., Graham, C., Oyebode, F. & Wilcock, G. (1995). Associations of psychotic symptoms in dementia sufferers. Br J Psychiatry, 167, 537–40. Ballard, C. G., O’Brien, J. T., Swann, A. G., Thompson, P., Neill, D. & McKeith, I. G. (2001). The natural history of psychosis and depression in dementia with Lewy bodies and Alzheimer’s disease: persistence and new cases over 1 year of follow-up. J Clin Psychiatry, 62, 46–9. Beard, C. M., Kokmen, E., Offord, K. P. & Kurland, L. T. (1992). Lack of association between Alzheimer’s disease and education, occupation, marital status, or living arrangement. Neurology, 42, 2063–8. Becker, J. T., Boller, F., Lopez, O. L., Saxton, J. & McGonigle, K. L. (1994). The natural history of Alzheimer’s disease. Description of study cohort and accuracy of diagnosis. Arch Neurol, 51, 585–94. Benjamin, R., Leake, A., Edwardson, J. A. et al. (1994a). Apolipoprotein E genes in Lewy body and Parkinson’s disease. Lancet, 343, 1565. Benjamin, R., Leake, A., McArthur, F. K. et al. (1994b). Protective effect of apoE epsilon 2 in Alzheimer’s disease. Lancet, 344, 473. Benowitz, L. I., Rodriguez, W., Paskevich, P., Mufson, E. J., Schenk, D. & Neve, R. L. (1989). The amyloid precursor protein is concentrated in neuronal lysosomes in normal and Alzheimer disease subjects. Exp Neurol, 106, 237–50.

195

196


Berg, L., McKeel, D. W., Jr., Miller, J. P., Baty, J. & Morris, J. C. (1993). Neuropathological indexes of Alzheimer’s disease in demented and nondemented persons aged 80 years and older. Arch Neurol, 50, 349–58. Berthier, M. L., Leiguarda, R., Starkstein, S. E., Sevlever, G. & Taratuto, A. L. (1991). Alzheimer’s disease in a patient with posterior cortical atrophy. J Neurol Neurosurg Psychiatry, 54, 1110–11. Blessed, G., Tomlinson, B. E. & Roth, M. (1968). The association between quantitative measures of dementia and of senile change in the cerebral grey matter of elderly subjects. Br J Psychiatry, 114, 797–811. Bloq, P. & Marinesco, G. (1892). Sur les lesions et la pathogenie de l’epilepsie dit essentielle. Semain Medical, 12, 445–6. Bonaiuto, S., Rocca, W. A., Lippi, A. et al. (1995). Education and occupation as risk factors for dementia: a population-based case-control study. Neuroepidemiology, 14, 101–9. Bowler, J. V., Munoz, D. G., Merskey, H. & Hachinski, V. (1998). Fallacies in the pathological confirmation of the diagnosis of Alzheimer’s disease. J Neurol Neurosurg Psychiatry, 64, 18–24. Braak, H. & Braak, E. (1988). Neuropil threads occur in dendrites of tangle-bearing nerve cells. Neuropathol Appl Neurobiol, 14, 39–44. (1991). Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol, 82, 239–59. Braak, H., Braak, E., Grundke-Iqbal, I. & Iqbal, K. (1986). Occurrence of neuropil threads in the senile human brain and in Alzheimer’s disease: a third location of paired helical filaments outside of neurofibrillary tangles and neuritic plaques. Neurosci Lett, 65, 351–5. Braak, H., Del Tredici, K., Schultz, C. & Braak, E. (2000). Vulnerability of select neuronal types to Alzheimer’s disease. Ann NY Acad Sci, 924, 53–61. Bramblett, G. T., Goedert, M., Jakes, R., Merrick, S. E., Trojanowski, J. Q. & Lee, V. M. (1993). Abnormal tau phosphorylation at Ser396 in Alzheimer’s disease recapitulates development and contributes to reduced microtubule binding. Neuron, 10, 1089–99. Brayne, C. (1993). Clinicopathological studies of the dementias from an epidemiological viewpoint. Br J Psychiatry, 162, 439– 46. Breitner, J. C., Silverman, J. M., Mohs, R. C. & Davis, K. L. (1988). Familial aggregation in Alzheimer’s disease: comparison of risk among relatives of early- and late-onset cases, and among male and female relatives in successive generations. Neurology, 38, 207–12. Brenner, D. E, Kukull, W. A., van Belle, G. et al. (1993). Relationship between cigarette smoking and Alzheimer’s disease in a population-based case-control study. Neurology, 43, 293–300. Brilliant, M., Hughes, L., Anderson, D., Ghobrial, M. & Elble, R. (1995). Rarefied white matter in patients with Alzheimer disease. Alzheimer Dis Assoc Disord, 9, 39–46. Brun, A. & Englund, E. (1981). Regional pattern of degeneration in Alzheimer’s disease: neuronal loss and histopathological grading. Histopathology, 5, 549–64.

Brun, A. & Englund, E. (1986). A white matter disorder in dementia of the Alzheimer type: a pathoanatomical study. Ann Neurol, 19, 253–62. Burkhardt, C. R., Filley, C. M., Kleinschmidt-DeMasters, B. K., de la Monte, S., Norenberg, M. D. & Schneck, S. A. (1988). Diffuse Lewy body disease and progressive dementia. Neurology, 38, 1520–8. Burns, A., Luthert, P., Levy, R., Jacoby, R. & Lantos, P. (1990). Accuracy of clinical diagnosis of Alzheimer’s disease. Br Med J, 301, 1026. Burns, A., Levy, R. & Jacoby, R. (1991). Diagnosis of Alzheimer’s disease. Br Med J, 302, 48. Byrne, E. (1991). Diagnosis of Alzheimer’s disease. Br Med J, 302, 48. Cai, X. D., Golde, T. E. & Younkin, S. G. (1993). Release of excess amyloid beta protein from a mutant amyloid beta protein precursor. Science, 259, 514–16. Caplan, L. & Schoene, W. (1987). Clinical features of subcortical arteriosclerotic encephalopathy (Binswanger’s disease). Neurology, 28, 1206–15. Chan-Palay, V. & Asan, E. (1989). Alterations in catecholamine neurons of the locus coeruleus in senile dementia of the Alzheimer type and in Parkinson’s disease with and without dementia and depression. J Comp Neurol, 287, 373–92. Chartier-Harlin, M. C., Crawford, F., Houlden, H. et al. (1991). Earlyonset Alzheimer’s disease caused by mutations at codon 717 of the beta-amyloid precursor protein gene. Nature, 353, 844–6. Chartier-Harlin, M. C., Parfitt, M., Legrain, S. et al. (1994). Apolipoprotein E, epsilon 4 allele as a major risk factor for sporadic early and late-onset forms of Alzheimer’s disease: analysis of the 19q13.2 chromosomal region. Hum Mol Genet, 3, 569–74. Chen, C. P., Alder, J. T., Bowen, D. M. et al. (1996). Presynaptic serotonergic markers in community-acquired cases of Alzheimer’s disease: correlations with depression and neuroleptic medication. J Neurochem, 66, 1592–8. Chen, C. P., Eastwood, S. L., Hope, T., McDonald, B., Francis, P. T. & Esiri, M. M. (2000). Immunocytochemical study of the dorsal and median raphe nuclei in patients with Alzheimer’s disease prospectively assessed for behavioural changes. Neuropathol Appl Neurobiol, 26, 347–55. Chui, H. C., Lyness, S. A., Sobel, E. & Schneider, L. S. (1994). Extrapyramidal signs and psychiatric symptoms predict faster cognitive decline in Alzheimer’s disease. Arch Neurol, 51, 676– 81. Clark, R. F. & Goate, A. M. (1993). Molecular genetics of Alzheimer’s disease. Arch Neurol, 50, 1164–72. Clarke, R., Smith, A. D., Jobst, K. A., Refsum, H., Sutton, L. & Ueland, P. M. (1998). Folate, vitamin B12, and serum total homocysteine levels in confirmed Alzheimer disease. Arch Neurol, 55, 1449– 55. Clarke, R., Joachim, C., Esiri, M. et al. (2000). Leukoaraiosis at presentation and disease progression during follow-up in histologically confirmed cases of dementia. Ann NY Acad Sci, 903, 497–500.


Claus, J. J., van Harskamp, F., Breteler, M. M. et al. (1994). The diagnostic value of SPECT with Tc 99m HMPAO in Alzheimer’s disease: a population-based study. Neurology, 44, 454–61. Cohen, D. H., Feiner, H., Jensson, O. & Frangione, B. (1983). Amyloid fibril in hereditary cerebral hemorrhage with amyloidosis (HCHWA) is related to the gastroentero-pancreatic neuroendocrine protein, gamma trace. J Exp Med, 158, 623–8. Combs, C. K., Johnson, D. E, Karlo, J. C., Cannady, S. B. & Landreth, G. E. (2000). Inflammatory mechanisms in Alzheimer’s disease: inhibition of beta-amyloid-stimulated proinflammatory responses and neurotoxicity by PPARgamma agonists. J Neurosci, 20, 558–67. Cook, R. H., Schneck, S. A. & Clark, D. B. (1981). Twins with Alzheimer’s disease. Arch Neurol, 38, 300–1. Corder, E. H., Saunders, A. M., Strittmatter, W. J. et al. (1993). Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer’s disease in late onset families. Science, 261, 921–3. Corder, E. H., Saunders, A. M., Risch, N. J. et al. (1994). Protective effect of apolipoprotein E type 2 allele for late onset Alzheimer disease. Nat Genet, 7, 180–4. Corkin, S. (1998). Functional MRI for studying episodic memory in ageing and Alzheimer’s disease. Geriatrics, 53 (Suppl 1), S13– 15. Corsellis, J. A. N. (1962). Mental Illness and the Aging Brain. The Distribution of Pathological Change in a Mental Hospital Population. London: Oxford University Press. Corsellis, J. (1976). Aging and the Dementias. In Greenfield’s Neuropathology, ed. W. Blackwood, & J. Corsellis, pp. 796–848. Edward Arnold. Corsellis, J., Bruton, C. & Freeman-Browne, D. (1973). The aftermath of boxing. Psychol. Med., 3, 270–3. Crystal, H., Dickson, D., Fuld, P. et al. (1988). Clinico-pathologic studies in dementia: nondemented subjects with pathologically confirmed Alzheimer’s disease. Neurology, 38, 1682–7. Crystal, H. A., Dickson, D. W., Lizardi, J. E., Davies, P. & Wolfson, L. I. (1990). Antemortem diagnosis of diffuse Lewy body disease. Neurology, 40, 1523–8. Crystal, H. A., Dickson, D. W., Sliwinski, M. J. et al. (1993). Pathological markers associated with normal ageing and dementia in the elderly. Ann Neurol, 34, 566–73. Czech, C., Tremp, G. & Pradier, L. (2000). Presenilins and Alzheimer’s disease: biological functions and pathogenic mechanisms. Prog Neurobiol, 60, 363–84. Dale, G. E., Leigh, P. N., Luthert, P., Anderton, B. H. & Roberts, G. W. (1991). Neurofibrillary tangles in dementia pugilistica are ubiquitinated. J Neurol Neurosurg Psychiatry, 54, 116–18. Dartigues, J. F., Gagnon, M., Michel, P. et al. (1991). [The Paquid research program on the epidemiology of dementia. Methods and initial results]. Rev Neurol, 147, 225–30. Davies, L., Wolska, B., Hilbich, C. et al. (1988). A4 amyloid protein deposition and the diagnosis of Alzheimer’s disease: prevalence in aged brains determined by immunocytochemistry compared with conventional neuropathologic techniques. Neurology, 38, 1688–93.

Davies, P. & Maloney, A. J. (1976). Selective loss of central cholinergic neurons in Alzheimer’s disease. Lancet, 2, 1403. Davis, P. C., Gray, L., Albert, M. et al. (1992). The Consortium to Establish a Registry for Alzheimer’s Disease (CERAD). Part III. Reliability of a standardized MRI evaluation of Alzheimer’s disease. Neurology, 42, 1676–80. de la Monte, S. M. (1989). Quantitation of cerebral atrophy in preclinical and end-stage Alzheimer’s disease. Ann Neurol, 25, 450–9. de Leon, M. J., George, A. E., Stylopoulos, L. A., Smith, G. & Miller, D. C. (1989). Early marker for Alzheimer’s disease: the atrophic hippocampus. Lancet, 2, 672–3. De Reuck, J., Crevits, L., De Coster, W., Sieben, G. & vander Eecken, H. (1980). Pathogenesis of Binswanger chronic progressive subcortical encephalopathy. Neurology, 30, 920–8. DeKosky, S. T. & Scheff, S. W. (1990). Synapse loss in frontal cortex biopsies in Alzheimer’s disease: correlation with cognitive severity. Ann Neurol, 27, 457–64. Delacourte, A. (2001). The molecular parameters of tau pathology. Tau as a killer and a witness. Adv Exp Med Biol, 487, 5–19. Deweer, B., Lehericy, S., Pillon, B. et al. (1995). Memory disorders in probable Alzheimer’s disease: the role of hippocampal atrophy as shown with MRI. J Neurol Neurosurg Psychiatry, 58, 590–7. Dickson, D. W., Crystal, H., Mattiace, L. A. et al. (1989). Diffuse Lewy body disease: light and electron microscopic immunocytochemistry of senile plaques. Acta Neuropathol, 78, 572–84. Duyckaerts, C., Delaere, P., Hauw, J. J. et al. (1990). Rating of the lesions in senile dementia of the Alzheimer type: concordance between laboratories. A European multicenter study under the auspices of EURAGE. J Neurol Sci, 97, 295–323. El-Agnaf, O. M. & Irvine, G. B. (2000). Review: formation and properties of amyloid-like fibrils derived from alpha-synuclein and related proteins. J Struct Biol, 130, 300–9. Englund, E., Brun, A. & Alling, C. (1988). White matter changes in dementia of Alzheimer’s type. Biochemical and neuropathological correlates. Brain, 111, 1425–39. Erkinjuntti, T., Ketonen, L., Sulkava, R., Sipponen, J., Vuorialho, M. & Iivanainen, M. (1987). Do white matter changes on MRI and CT differentiate vascular dementia from Alzheimer’s disease? J Neurol Neurosurg Psychiatry, 50, 37–42. Erkinjuntti, T., Sulkava, R., Palo, J. & Ketonen, L. (1989). White matter low attenuation on CT in Alzheimer’s disease. Arch Gerontol Geriatr, 8, 95–104. Erkinjuntti, T., Lee, D. H., Gao, F. et al. (1993). Temporal lobe atrophy on magnetic resonance imaging in the diagnosis of early Alzheimer’s disease. Arch Neurol, 50, 305–10. Esch, F. S., Keim, P. S., Beattie, E. C. et al. (1990). Cleavage of amyloid beta peptide during constitutive processing of its precursor. Science, 248, 1122–4. Esiri, M. M. & Wilcock, G. K. (1984). The olfactory bulbs in Alzheimer’s disease. J Neurol Neurosurg Psychiatry, 47, 56–60. (1986). Cerebral amyloid angiopathy in dementia and old age. J Neurol Neurosurg Psychiatry, 49, 1221–6.

197

198


Esiri, M. M., Pearson, R. C. & Powell, T. P. (1986). The cortex of the primary auditory area in Alzheimer’s disease. Brain Res, 366, 385–7. Esiri, M. M., Matthews, F., Brayne, C. & Ince, P. G. (2001). Pathological correlates of late onset elementia in a multi-centre community-based population in England and Wales. Lancet, 357, 169–75. Evans, D. A., Funkenstein, H. H., Albert, M. S. et al. (1989). Prevalence of Alzheimer’s disease in a community population of older persons. Higher than previously reported. J Am Med Assoc, 262, 2551–6. Fabrigoule, C., Rouch, I., Taberly, A. et al. (1998). Cognitive process in preclinical phase of dementia. Brain, 121, 135–41. Farber, N. B., Newcomer, J. W. & Olney, J. W. (1998). The glutamate synapse in neuropsychiatric disorders. Focus on schizophrenia and Alzheimer’s disease. Prog Brain Res, 116, 421–37. Farlow, M., Murrell, J., Ghetti, B., Unverzagt, F., Zeldenrust, S. & Benson, M. (1994). Clinical characteristics in a kindred with early-onset Alzheimer’s disease and their linkage to a G–>T change at position 2149 of the amyloid precursor protein gene. Neurology, 44, 105–11. Farrar, L., Myers, R. & Connor, L. (1991). Segregation analysis reveals evidence for a major gene for Alzheimer disease. Am J Hum Genet, 48, 1026–33. Feany, M. B. & Dickson, D. W. (1996). Neurodegenerative disorders with extensive tau pathology: a comparative study and review. Ann Neurol, 40, 139–48. Fidani, L., Rooke, K., Chartier-Harlin, M. C. et al. (1992). Screening for mutations in the open reading frame and promoter of the beta-amyloid precursor protein gene in familial Alzheimer’s disease: identification of a further family with APP717 Val– >Ile. Hum Mol Genet, 1, 165–8. Fillit, H., Weinreb, H., Cholst, I. et al. (1986). Observations in a preliminary open trial of estradiol therapy for senile dementia–Alzheimer’s type. Psychoneuroendocrinology, 11, 337–45. Fischer, O. (1907). Miliare Nekrosen mit drusigen Wucherungen der Neurofibrillen, eine regelmässige Veränderung der Hirnrinde bei seniler Demenz. Monatsschrift fur ¨ Psychiatrie und Neurologie, 22, 361–72. Forstl, H. (1999). The Lewy body variant of Alzheimer’s disease: clinical, pathophysiological and conceptual issues. Eur Arch Psychiatry Clin Neurosci, 249, 64–7. Fratiglioni, L., Grut, M., Forsell, Y. et al. (1991). Prevalence of Alzheimer’s disease and other dementias in an elderly urban population: relationship with age, sex, and education. Neurology, 41, 1886–92. Friedland, R. P. (1993). Epidemiology, education, and the ecology of Alzheimer’s disease. Neurology, 43, 246–9. Galasko, D., Hansen, L. A., Katzman, R. et al. (1994). Clinicalneuropathological correlations in Alzheimer’s disease and related dementias. Arch Neurol, 51, 888–95. Garcia, M. L. & Cleveland, D. W. (2001). Going new places using an old MAP: tau, microtubules and human neurodegenerative disease. Curr Opin Cell Biol, 13, 41–8.

Gartner, U., Holzer, M. & Arendt, T. (1999). Elevated expression of p21ras is an early event in Alzheimer’s disease and precedes neurofibrillary degeneration. Neuroscience, 91, 1–5. Gearing, M., Mirra, S. S., Hedreen, J. C., Sumi, S. M., Hansen, L. A. & Heyman, A. (1995). The Consortium to Establish a Registry for Alzheimer’s Disease (CERAD). Part X. Neuropathology confirmation of the clinical diagnosis of Alzheimer’s disease. Neurology, 45, 461–6. Gentleman, S. & Roberts, G. (1991). Risk factors in Alzheimer’s disease. Br Med J, 304, 118–19. George, A. E., de Leon, M. J., Gentes, C. I. et al. (1986a). Leukoencephalopathy in normal and pathologic aging: 1. CT of brain lucencies. Am J Neuroradiol, 7, 561–6. George, A. E., Kalnin, A., Rosner, L., Goodgold, A. & Chase, N. (1986b). Leukoencephalopathy in normal and pathologic aging: 2. MRI of brain lucencies. Am J Neuroradiol, 7, 567–70. Gibbs, C. J., Jr., Joy, A., Heffner, R. et al. (1985). Clinical and pathological features and laboratory confirmation of Creutzfeldt–Jakob disease in a recipient of pituitary-derived human growth hormone. N Engl J Med, 313, 734–8. Gilbert, J. J. & Vinters, H. V. (1983). Cerebral amyloid angiopathy: incidence and complications in the ageing brain. I. Cerebral hemorrhage. Stroke, 14, 915–23. Gilleard, C. J., Kellett, J. M., Coles, J. A., Millard, P. H., Honavar, M. & Lantos, P. L. (1992). The St. George’s dementia bed investigation study: a comparison of clinical and pathological diagnosis. Acta Psychiatr Scand, 85, 264–9. Gillmore, J. D., Hawkins, P. N. & Pepys, M. B. (1997). Amyloidosis: a review of recent diagnostic and therapeutic developments. Br J Haematol, 99, 245–56. Glenner, G. G. & Wong, C. W. (1984). Alzheimer’s disease: initial report of the purification and characterization of a novel cerebrovascular amyloid protein. Biochem Biophys Res Commun, 120, 885–90. Goate, A., Chartier-Harlin, M. C., Mullan, M. et al. (1991). Segregation of a missense mutation in the amyloid precursor protein gene with familial Alzheimer’s disease. Nature, 349, 704–6. Goedert, M. (1993). Tau protein and the neurofibrillary pathology of Alzheimer’s disease. Trends Neurosci, 16, 460–5. Goedert, M. & Spillantini, M. G. (2001). Tau gene mutations and neurodegeneration. Biochem Soc Symp, 67, 59–71. Goedert, M., Spillantini, M. G., Jakes, R., Rutherford, D. & Crowther, R. A. (1989). Multiple isoforms of human microtubuleassociated protein tau: sequences and localization in neurofibrillary tangles of Alzheimer’s disease. Neuron, 3, 519–26. Goedert, M., Spillantini, M. G., Cairns, N. J. & Crowther, R. A. (1992). Tau proteins of Alzheimer paired helical filaments: abnormal phosphorylation of all six brain isoforms. Neuron, 8, 159–68. Golde, T. E., Estus, S., Younkin, L. H., Selkoe, D. J. & Younkin, S. G. (1992). Processing of the amyloid protein precursor to potentially amyloidogenic derivatives. Science, 255, 728–30. Gorelick, P. B., Freels, S., Harris, Y., Dollear, T., Billingsley, M. & Brown, N. (1994). Epidemiology of vascular and Alzheimer’s dementia among African Americans in Chicago, IL: baseline


frequency and comparison of risk factors. Neurology, 44, 1391– 6. Griffiths, R. A., Mortimer, T. F., Oppenheimer, D. R. & Spalding, J. M. (1982). Congophilic angiopathy of the brain: a clinical and pathological report on two siblings. J Neurol Neurosurg Psychiatry, 45, 396–408. Guo, Q., Sebastian, L., Sopher, B. L. et al. (1999). Increased vulnerability of hippocampal neurons from presenilin-1 mutant knock-in mice to amyloid beta-peptide toxicity: central roles of superoxide production and caspase activation. J Neurochem, 72, 1019–29. Haass, C., Koo, E. H., Mellon, A., Hung, A. Y. & Selkoe, D. J. (1992a). Targeting of cell-surface beta-amyloid precursor protein to lysosomes: alternative processing into amyloid-bearing fragments. Nature, 357, 500–3. Haass, C., Schlossmacher, M. G., Hung, A. Y. et al. (1992b). Amyloid beta-peptide is produced by cultured cells during normal metabolism. Nature, 359, 322–5. Hachinski, V. C., Iliff, L. D., Zilhka, E. et al. (1975). Cerebral blood flow in dementia. Arch Neurol, 32, 632–7. Hachinski, V. C., Potter, P. & Merskey, H. (1987). Leuko-araiosis. Arch Neurol, 44, 21–3. Hagnell, O., Ojesjo, L. & Rorsman, B. (1992). Incidence of dementia in the Lundby Study. Neuroepidemiology, 11, 61–6. Haltia, M., Viitanen, M., Sulkava, R. et al. (1994). Chromosome 14encoded Alzheimer’s disease: genetic and clinicopathological description. Ann Neurol, 36, 362–7. Hansen, L., Salmon, D., Galasko, D. et al. (1990). The Lewy body variant of Alzheimer’s disease: a clinical and pathologic entity. Neurology, 40, 1–8. Hardy, J., Mullan, M., Chartier-Harlan, M., Brown, J., Goate, A. & Rossor, M. (1991). Molecular classification of Alzheimer’s disease. The Lancet, 337, 1342–3. Hart, M. N., Merz, P., Bennett-Gray, J. et al. (1988). beta-amyloid protein of Alzheimer’s disease is found in cerebral and spinal cord vascular malformations. Am J Pathol, 132, 167–72. Hartmann, H., Eckert, A. & Muller, W. E. (1993). beta-Amyloid protein amplifies calcium signalling in central neurons from the adult mouse. Biochem Biophys Res Commun, 194, 1216–20. Hasegawa, M., Morishima-Kawashima, M., Takio, K., Suzuki, M., Titani, K. & Ihara, Y. (1992). Protein sequence and mass spectrometric analyses of tau in the Alzheimer’s disease brain. J Biol Chem, 267, 17047–54. Henderson, V. W. & Buckwalter, J. G. (1994). Cognitive deficits of men and women with Alzheimer’s disease. Neurology, 44, 90– 6. Henderson, V. W., Paganini-Hill, A., Emanuel, C. K., Dunn, M. E. & Buckwalter, J. G. (1994). Estrogen replacement therapy in older women. Comparisons between Alzheimer’s disease cases and nondemented control subjects. Arch Neurol, 51, 896–900. Hendrie, H. C., Hall, K. S., Hui, S. et al. (1995). Apolipoprotein E genotypes and Alzheimer’s disease in a community study of elderly African Americans. Ann Neurol, 37, 118–20. Hendriks, L., van Duijn, C. M., Cras, P. et al. (1992). Presenile dementia and cerebral haemorrhage linked to a mutation at codon

692 of the beta-amyloid precursor protein gene. Nat Genet, 1, 218–21. Heston, L. L. & Mastri, A. R. (1977). The genetics of Alzheimer’s disease: associations with hematologic malignancy and Down’s syndrome. Arch Gen Psychiatry, 34, 976–81. Heyman, A., Wilkinson, W. E., Stafford, J. A. et al. (1984). Alzheimer’s disease: a study of epidemiological aspects. Ann Neurol, 15, 335–41. Heyman, A., Fillenbaum, G., Prosnitz, B., Raiford, K., Burchett, B. & Clark, C. (1991). Estimated prevalence of dementia among elderly black and white community residents. Arch Neurol, 48, 594–8. Hill, L. R., Klauber, M. R., Salmon, D. P. et al. (1993). Functional status, education, and the diagnosis of dementia in the Shanghai survey. Neurology, 43, 138–45. Himmler, A. (1989). Structure of the bovine tau gene: alternatively spliced transcripts generate a protein family. Mol Cell Biol, 9, 1389–96. Hofman, A., Rocca, W. A., Brayne, C. et al. (1991). The prevalence of dementia in Europe: a collaborative study of 1980–1990 findings. Eurodem Prevalence Research Group. Int J Epidemiol, 20, 736–48. Hogervorst, E., Barnetson, L., Jobst, K. A., Nagy, Z., Combrinck, M. & Smith, A. D. (2000). Diagnosing dementia: interrater reliability assessment and accuracy of the NINCDS/ADRDA criteria versus CERAD histopathological criteria for Alzheimer’s disease. Dement Geriatr Cogn Disord, 11, 107–13. Holmes, C., Cairns, N., Lantos, P. & Mann, A. (1999). Validity of current clinical criteria for Alzheimer’s disease, vascular dementia and dementia with Lewy bodies. Br J Psychiatry, 174, 45–50. Homer, A. C., Honavar, M., Lantos, P. L., Hastie, I. R., Kellett, J. M. & Millard, P. H. (1988). Diagnosing dementia: do we get it right? Br Med J, 297, 894–6. Honjo, H., Ogino, Y., Naitoh, K. et al. (1989). In vivo effects by estrone sulfate on the central nervous system-senile dementia (Alzheimer’s type). J Steroid Biochem, 34, 521–5. Hope, T. & Fairburn, C. G. (1992). The Present Behavioural Examination (PBE): the development of an interview to measure current behavioural abnormalities. Psychol Med, 22, 223–30. Hu, J., Akama, K. T., Krafft, G. A., Chromy, B. A. & Van Eldik, L. J. (1998). Amyloid-beta peptide activates cultured astrocytes: morphological alterations, cytokine induction and nitric oxide release. Brain Res, 785, 195–206. Hubbard, B. M. & Anderson, J. M. (1981a). Age, senile dementia and ventricular enlargement. J Neurol Neurosurg Psychiatry, 44, 631–5. (1981b). A quantitative study of cerebral atrophy in old age and senile dementia. J Neurol Sci, 50, 135–45. Huff, F. J., Auerbach, J., Chakravarti, A. & Boller, F. (1988). Risk of dementia in relatives of patients with Alzheimer’s disease. Neurology, 38, 786–90. Hyman, B. T. & Tanzi, R. (1995). Molecular epidemiology of Alzheimer’s disease. N Engl J Med, 333, 1283–4. Ihara, Y. (1988). Massive somatodendritic sprouting of cortical neurons in Alzheimer’s disease. Brain Res, 459, 138–44.

199

200


Ingram, D. K. (2001). Vaccine development for Alzheimer’s disease: a shot of good news. Trends Neurosci, 24, 305–7. Iwatsubo, T., Odaka, A., Suzuki, N., Mizusawa, H., Nukina, N. & Ihara, Y. (1994). Visualization of A beta 42(43) and A beta 40 in senile plaques with end-specific A beta monoclonals: evidence that an initially deposited species is A beta 42(43). Neuron, 13, 45–53. Iwatsubo, T., Mann, D. M., Odaka, A., Suzuki, N. & Ihara, Y. (1995). Amyloid beta protein (A beta) deposition: A beta 42(43) precedes A beta 40 in Down syndrome. Ann Neurol, 37, 294–9. Janota, I., Mirsen, T. R., Hachinski, V. C., Lee, D. H. & Merskey, H. (1989). Neuropathologic correlates of leuko-araiosis. Arch Neurol, 46, 1124–8. Jarvik, L. F., Ruth, V. & Matsuyama, S. S. (1980). Organic brain syndrome and aging. A six-year follow-up of surviving twins. Arch Gen Psychiatry, 37, 280–6. Jellinger, K. (1976). Neuropathological aspects of dementias resulting from abnormal blood and cerebrospinal fluid dynamics. Acta Neurol Belg, 76, 83–102. Jellinger, K., Danielczyk, W., Fischer, P. & Gabriel, E. (1990). Clinicopathological analysis of dementia disorders in the elderly. J Neurol Sci, 95, 239–58. Jensson, O., Gudmundsson, G., Arnason, A. et al. (1987). Hereditary cystatin C (gamma-trace) amyloid angiopathy of the CNS causing cerebral hemorrhage. Acta Neurol Scand, 76, 102–14. Joachim, C., Morris, J. & Schoene, W. (1986). Neuropathological findings in six cases of dementia with severe focal atrophy. J. Neuropath. Exp. Neurol., 49, 275 (abstract). Joachim, C. L., Morris, J. H. & Selkoe, D. J. (1988). Clinically diagnosed Alzheimer’s disease: autopsy results in 150 cases. Ann Neurol, 24, 50–6. (1989). Diffuse senile plaques occur commonly in the cerebellum in Alzheimer’s disease. Am J Pathol, 135, 309–19. Jobst, K. A., Smith, A. D., Szatmari, M. et al. (1992). Detection in life of confirmed Alzheimer’s disease using a simple measurement of medial temporal lobe atrophy by computed tomography. Lancet, 340, 1179–83. Jobst, K. A., Hindley, N. J., King, E. & Smith, A. D. (1994a). The diagnosis of Alzheimer’s disease: a question of image? J Clin Psychiatry, 55 Suppl, 22–31. Jobst, K. A., Smith, A. D., Szatmari, M. et al. (1994b). Rapidly progressing atrophy of medial temporal lobe in Alzheimer’s disease. Lancet, 343, 829–30. Jorm, A. F., Korten, A. E. & Henderson, A. S. (1987). The prevalence of dementia: a quantitative integration of the literature. Acta Psychiatr Scand, 76, 465–79. Kalmijn, S., Mehta, K. M., Pols, H. A., Hofman, A., Drexhage, H. A. & Breteler, M. M. (2000). Subclinical hyperthyroidism and the risk of dementia. The Rotterdam study. Clin Endocrinol (Oxf ), 53, 733–7. Kang, J., Lemaire, H. G., Unterbeck, A. et al. (1987). The precursor of Alzheimer’s disease amyloid A4 protein resembles a cellsurface receptor. Nature, 325, 733–6. Karlinsky, H., Vaula, G., Haines, J. L. et al. (1992). Molecular and prospective phenotypic characterization of a pedigree with

familial Alzheimer’s disease and a missense mutation in codon 717 of the beta-amyloid precursor protein gene. Neurology, 42, 1445–53. Kato, T., Knopman, D. & Liu, H. (2001). Dissociation of regional activation in mild AD during visual encoding: a functional MRI study. Neurology, 57, 812–16. Katzman, R. (1993a). Clinical and epidemiological aspects of Alzheimer’s disease. Clin.Neurosci, 1, 165–70. (1993b). Education and the prevalence of dementia and Alzheimer’s disease. Neurology, 43, 13–20. Katzman, R., Terry, R., DeTeresa, R. et al. (1988). Clinical, pathological, and neurochemical changes in dementia: a subgroup with preserved mental status and numerous neocortical plaques. Ann Neurol, 23, 138–44. Katzman, R., Aronson, M., Fuld, P. et al. (1989). Development of dementing illnesses in an 80-year-old volunteer cohort. Ann Neurol, 25, 317–24. Kellett, J., Gilleard, C., Homer, A. & Millard, P. (1991). Diagnosis of Alzheimer’s disease. Br Med J, 302, 47. Kennedy, A. M., Newman, S., McCaddon, A. et al. (1993). Familial Alzheimer’s disease. A pedigree with a mis-sense mutation in the amyloid precursor protein gene (amyloid precursor protein 717 valine–>glycine). Brain, 116, 309–24. Khachaturian, Z. S. (1985). Diagnosis of Alzheimer’s disease. Arch Neurol, 42, 1097–105. Kido, D. K., Caine, E. D., LeMay, M., Ekholm, S., Booth, H. & Panzer, R. (1989). Temporal lobe atrophy in patients with Alzheimer disease: a CT study. Am J Neuroradiol, 10, 551–5. Kirazov, E., Kirazov, L., Bigl, V. & Schliebs, R. (2001). Ontogenetic changes in protein level of amyloid precursor protein (APP) in growth cones and synaptosomes from rat brain and prenatal expression pattern of APP mRNA isoforms in developing rat embryo. Int J Dev Neurosci, 19, 287–96. Klatka, L. A., Schiffer, R. B., Powers, J. M. & Kazee, A. M. (1996). Incorrect diagnosis of Alzheimer’s disease. A clinicopathologic study. Arch Neurol, 53, 35–42. Kokmen, E., Beard, C. M., Offord, K. P. & Kurland, L. T. (1989). Prevalence of medically diagnosed dementia in a defined United States population: Rochester, Minnesota, January 1, 1975. Neurology, 39, 773–6. Korczyn, A., Kahana, E. & Galper, Y. (1991). Epidemiology of dementia in Ashkelon, Israel. Neuroepidemiology, 10, 100. Kosaka, K. (1990). Diffuse Lewy body disease in Japan. J Neurol, 237, 197–204. Kosaka, K., Yoshimura, M., Ikeda, K. & Budka, H. (1984). Diffuse type of Lewy body disease: progressive dementia with abundant cortical Lewy bodies and senile changes of varying degree – a new disease? Clin Neuropathol, 3, 185–92. Kukull, W. A., Larson, E. B., Reifler, B. V., Lampe, T. H., Yerby, M. S. & Hughes, J. P. (1990). The validity of 3 clinical diagnostic criteria for Alzheimer’s disease. Neurology, 40, 1364–9. Kumar, A., Schapiro, M. B., Grady, C. L. et al. (1991). Anatomic, metabolic, neuropsychological, and molecular genetic studies of three pairs of identical twins discordant for dementia of the Alzheimer’s type. Arch Neurol, 48, 160–8.


Lambert, M. P., Viola, K. L., Chromy, B. A. et al. (2001). Vaccination with soluble Abeta oligomers generates toxicity-neutralizing antibodies. J Neurochem, 79, 595–605. Landfield, P. W., Thibault, O., Mazzanti, M. L., Porter, N. M. & Kerr, D. S. (1992). Mechanisms of neuronal death in brain ageing and Alzheimer’s disease: role of endocrine-mediated calcium dyshomeostasis. J Neurobiol, 23, 1247–60. Lee, V. M. & Trojanowski, J. Q. (1992). The disordered neuronal cytoskeleton in Alzheimer’s disease. Curr Opin Neurobiol, 2, 653–6. Lee, V. M., Balin, B. J., Otvos, L., Jr. & Trojanowski, J. Q. (1991). A68: a major subunit of paired helical filaments and derivatized forms of normal Tau. Science, 251, 675–8. Lennox, G., Lowe, J., Landon, M., Byrne, E. J., Mayer, R. J. & GodwinAusten, R. B. (1989). Diffuse Lewy body disease: correlative neuropathology using anti-ubiquitin immunocytochemistry. J Neurol Neurosurg Psychiatry, 52, 1236–47. Levine, D. N., Lee, J. M. & Fisher, C. M. (1993). The visual variant of Alzheimer’s disease: a clinicopathologic case study. Neurology, 43, 305–13. Levy, E., Carman, M. D., Fernandez-Madrid, I. J. et al. (1990). Mutation of the Alzheimer’s disease amyloid gene in hereditary cerebral hemorrhage, Dutch type. Science, 248, 1124–6. Levy-Lahad, E. & Bird, T. D. (1996). Genetic factors in Alzheimer’s disease: a review of recent advances. Ann Neurol, 40, 829–40. Levy-Lahad, E., Wasco, W., Poorkaj, P. et al. (1995a). Candidate gene for the chromosome 1 familial Alzheimer’s disease locus. Science, 269, 973–7. Levy-Lahad, E., Wijsman, E. M., Nemens, E. et al. (1995b). A familial Alzheimer’s disease locus on chromosome 1. Science, 269, 970– 3. Lim, A., Tsuang, D., Kukull, W. et al. (1999). Cliniconeuropathological correlation of Alzheimer’s disease in a community-based case series. J Am Geriatr Soc, 47, 564–9. Lindley, R. & Dennis, M. (1991). Diagnosis of Alzheimer’s disease. Br Med J, 302, 47. Lippa, C. F. (1999). Familial Alzheimer’s disease: genetic influences on the disease process (Review). Int J Mol Med, 4, 529– 36. Litchfield, S. & Nagy, Z. (2001). New temperature modification makes the Bielschowsky silver stain reproducible. Acta Neuropathol (Berl), 101, 17–21. Lopez, O. L., Swihart, A. A., Becker, J. T. et al. (1990). Reliability of NINCDS-ADRDA clinical criteria for the diagnosis of Alzheimer’s disease. Neurology, 40, 1517–22. Lopez, O. L., Becker, J. T., Klunk, W. et al. (2000). Research evaluation and diagnosis of probable Alzheimer’s disease over the last two decades: I. Neurology, 55, 1854–62. Lopez-Arrieta, J. M., Rodriguez, J. L. & Sanz, F. (2001). Efficacy and safety of Nicotine on Alzheimer’s disease patients (Cochrane Review). Cochrane Database Syst Rev, 2, Ma, J., Yee, A., Brewer, H. B., Jr., Das, S. & Potter, H. (1994). Amyloid-associated proteins alpha 1-antichymotrypsin and apolipoprotein E promote assembly of Alzheimer beta-protein into filaments. Nature, 372, 92–4.

Mackenzie, I. R., Hao, C. & Munoz, D. G. (1995). Role of microglia in senile plaque formation. Neurobiol Aging, 16, 797–804. McKee, A. C., Kosik, K. S. & Kowall, N. W. (1991). Neuritic pathology and dementia in Alzheimer’s disease. Ann Neurol, 30, 156–65. McKeith, I. G., Galasko, D., Kosaka, K. et al. (1996). Consensus guidelines for the clinical and pathologic diagnosis of dementia with Lewy bodies (DLB): report of the consortium on DLB international workshop. Neurology, 47, 1113–24. McLaurin, J., Yang, D., Yip, C. M. & Fraser, P. E. (2000). Review: modulating factors in amyloid-beta fibril formation. J Struct Biol, 130, 259–70. McMenemey, W. (1958). The dementias and progressive disease of the basal ganglia. In Neuropathology, ed. J. Greenfield, W. Blackwood, W. McMenemey, A. Meyer & R. Norman, pp. 475– 528. Edward Arnold. (1963). The dementias and progressive disease of the basal ganglia. In Greenfield’s Neuropathology, ed. W. Blackwood, W. McMenemey, A. Meyer, R. Norman & D. Russel, pp. 520–81. Edward Arnold. Mandybur, T. I. (1975). The incidence of cerebral amyloid angiopathy in Alzheimer’s disease. Neurology, 25, 120–6. Mandybur, T. I. & Gore, I. (1969). Amyloid in late postirradiation necrosis of brain. Neurology, 19, 983–92. Mann, D. M., Lincoln, J., Yates, P. O., Stamp, J. E. & Toper, S. (1980). Changes in the monoamine containing neurones of the human CNS in senile dementia. Br J Psychiatry, 136, 533–41. Manning, F. (1991). Diagnosis of Alzheimer’s disease. Br Med J, 302, 47–8. Martin, E. M., Wilson, R. S., Penn, R. D., Fox, J. H., Clasen, R. A. & Savoy, S. M. (1987). Cortical biopsy results in Alzheimer’s disease: correlation with cognitive deficits. Neurology, 37, 1201–4. Masliah, E., Terry, R. D., DeTeresa, R. M. & Hansen, L. A. (1989). Immunohistochemical quantification of the synapse-related protein synaptophysin in Alzheimer disease. Neurosci Lett, 103, 234–9. Masliah, E., Terry, R. D., Alford, M., DeTeresa, R. & Hansen, L. A. (1991). Cortical and subcortical patterns of synaptophysinlike immunoreactivity in Alzheimer’s disease. Am J Pathol, 138, 235–46. Matsuo, E. S., Shin, R. W., Billingsley, M. L. et al. (1994). Biopsyderived adult human brain tau is phosphorylated at many of the same sites as Alzheimer’s disease paired helical filament tau. Neuron, 13, 989–1002. Matthews, K. L., Chen, C. P., Esiri, M. M., Keene, J., Minger, S. L. & Francis, P. T. (2002). Noradrenergic changes, aggressive behavior, and cognition in patients with dementia. Biol Psychiatry, 51, 407–16. Mattson, M. P., Barger, S. W., Cheng, B., Lieberburg, I., SmithSwintosky, V. L. & Rydel, R. E. (1993). beta-Amyloid precursor protein metabolites and loss of neuronal Ca2+ homeostasis in Alzheimer’s disease. Trends Neurosci, 16, 409–14. Mattson, M. P., Zhu, H., Yu, J. & Kindy, M. S. (2000). Presenilin-1 mutation increases neuronal vulnerability to focal ischemia in vivo and to hypoxia and glucose deprivation in cell culture:

201

202


involvement of perturbed calcium homeostasis. J Neurosci, 20, 1358–64. Mayeux, R., Stern, Y., Ottman, R. et al. (1993). The apolipoprotein epsilon 4 allele in patients with Alzheimer’s disease. Ann Neurol, 34, 752–4. Mendez, M. F., Mastri, A. R., Sung, J. H. & Frey, W. H., II (1992). Clinically diagnosed Alzheimer disease: neuropathologic findings in 650 cases. Alzheimer Dis Assoc Disord, 6, 35–43. Menzel, H. J., Kladetzky, R. G. & Assmann, G. (1983). Apolipoprotein E polymorphism and coronary artery disease. Arteriosclerosis, 3, 310–15. Meyer, J. S., Rauch, G., Rauch, R. A. & Haque, A. (2000). Risk factors for cerebral hypoperfusion, mild cognitive impairment, and dementia. Neurobiol Aging, 21, 161–9. Minger, S. L., Esiri, M. M., McDonald, B., Keene, J., Carter, J., Hope, T. & Francis, P. T. (2000). Cholinergic deficits contribute to behavioral disturbance in patients with dementia. Neurology, 55, 1460–7. Mirra, S. S. (1997). Neuropathological assessment of Alzheimer’s disease: the experience of the Consortium to Establish a Registry for Alzheimer’s Disease. Int Psychogeriatr, 9, 263–8; discussion 9–72. Mirra, S. S. & Hyman, B. T. (2002). Ageing and dementia. In Greenfield’s Neuropathology, ed. D. I. Graham, & P. L. Lantos, pp. 195–272. Edward Arnold. Mirra, S. S., Heyman, A., McKeel, D. et al. (1991). The Consortium to Establish a Registry for Alzheimer’s Disease (CERAD). Part II. Standardization of the neuropathologic assessment of Alzheimer’s disease. Neurology, 41, 479–86. Mirra, S. S., Hart, M. N. & Terry, R. D. (1993). Making the diagnosis of Alzheimer’s disease. A primer for practicing pathologists. Arch Pathol Lab Med, 117, 132–44. Mirra, S. S., Gearing, M., McKeel, D. W., Jr. et al. (1994). Interlaboratory comparison of neuropathology assessments in Alzheimer’s disease: a study of the Consortium to Establish a Registry for Alzheimer’s Disease (CERAD). J Neuropathol Exp Neurol, 53, 303–15. Mirsen, T. R., Lee, D. H., Wong, C. J. et al. (1991). Clinical correlates of white-matter changes on magnetic resonance imaging scans of the brain. Arch Neurol, 48, 1015–21. Molsa, P. K., Marttila, R. J. & Rinne, U. K. (1982). Epidemiology of dementia in a Finnish population. Acta Neurol Scand, 65, 541– 52. Morris, J. C., Heyman, A., Mohs, R. C. et al. (1989). The Consortium to Establish a Registry for Alzheimer’s Disease (CERAD). Part I. Clinical and neuropsychological assessment of Alzheimer’s disease. Neurology, 39, 1159–65. Morris, J. C., McKeel, D. W., Jr., Storandt, M. et al. (1991). Very mild Alzheimer’s disease: informant-based clinical, psychometric, and pathologic distinction from normal aging. Neurology, 41, 469–78. Mortimer, J. A., van Duijn, C. M., Chandra, V. et al. (1991). Head trauma as a risk factor for Alzheimer’s disease: a collaborative re-analysis of case-control studies. EURODEM Risk Factors Research Group. Int J Epidemiol, 20, S28–35.

Mullen, M., Crawfird, F. & Axelman, K. (1992a). A new mutation in APP demonstrates that pathogenic mutations for probable Alzheimer’s disease frame the -amyloid sequence. Nature Genet, 1, 345–7. Mullen, M., Houlden, H. & Windelspecht, M. (1992b). A locus for familal early-onset Alzheimer’s disease on the long arm of chromosome 14, proximal to the a-1 antichymotrypsin gene. Nature Genet, 2, 340–2. Munch, G. & Robinson, S. R. (2002). Alzheimer’s vaccine: a cure as dangerous as the disease? J Neural Transm, 109, 537–9. Murphy, G. M., Jr., Forno, L. S., Higgins, L., Scardina, J. M., Eng, L. F. & Cordell, B. (1994). Development of a monoclonal antibody specific for the COOH-terminal of beta-amyloid 1–42 and its immunohistochemical reactivity in Alzheimer’s disease and related disorders. Am J Pathol, 144, 1082–8. Murrell, J., Farlow, M., Ghetti, B. & Benson, M. D. (1991). A mutation in the amyloid precursor protein associated with hereditary Alzheimer’s disease. Science, 254, 97–9. Nagy, Z. (2000). Cell cycle regulatory failure in neurones: causes and consequences. Neurobiol Aging, 21, 761–9. Nagy, Z., Esiri, M. M., Jobst, K. A. et al. (1995a). Influence of the apolipoprotein E genotype on amyloid deposition and neurofibrillary tangle formation in Alzheimer’s disease. Neuroscience, 69, 757–61. Nagy, Z., Esiri, M. M., Jobst, K. A. et al. (1995b). Relative roles of plaques and tangles in the dementia of Alzheimer’s disease: correlations using three sets of neuropathological criteria. Dementia, 6, 21–31. Nagy, Z., Jobst, K. A., Esiri, M. M. et al. (1996). Hippocampal pathology reflects memory deficit and brain imaging measurements in Alzheimer’s disease: clinicopathologic correlations using three sets of pathologic diagnostic criteria. Dementia, 7, 76– 81. Nagy, Z., Esiri, M. M., Jobst, K. A. et al. (1997). The effects of additional pathology on the cognitive deficit in Alzheimer disease. J Neuropathol Exp Neurol, 56, 165–70. Nagy, Z., Esiri, M. M., Hindley, N. J. et al. (1998a). Accuracy of clinical operational diagnostic criteria for Alzheimer’s disease in relation to different pathological diagnostic protocols. Dement Geriatr Cogn Disord, 9, 219–26. Nagy, Z., Esiri, M. M., Joachim, C. et al. (1998b). Comparison of pathological diagnostic criteria for Alzheimer disease. Alzheimer Dis Assoc Disord, 12, 182–9. Nagy, Z., Esiri, M. M. & Smith, A. D. (1998c). The cell division cycle and the pathophysiology of Alzheimer’s disease. Neuroscience, 87, 731–9. Nagy, Z., Yilmazer-Hanke, D. M., Braak, H., Braak, E., Schultz, C. & Hanke, J. (1998d). Assessment of the pathological stages of Alzheimer’s disease in thin paraffin sections: a comparative study. Dement Geriatr Cogn Disord, 9, 140–4. Nagy, Z., Hindley, N. J., Braak, H. et al. (1999a). Relationship between clinical and radiological diagnostic criteria for Alzheimer’s disease and the extent of neuropathology as reflected by ‘stages’: a prospective study. Dement Geriatr Cogn Disord, 10, 109–14.


Nagy, Z., Hindley, N. J., Braak, H. et al. (1999b). The progression of Alzheimer’s disease from limbic regions to the neocortex: clinical, radiological and pathological relationships. Dement Geriatr Cogn Disord, 10, 115–20. Namba, Y., Tomonaga, M., Kawasaki, H., Otomo, E. & Ikeda, K. (1991). Apolipoprotein E immunoreactivity in cerebral amyloid deposits and neurofibrillary tangles in Alzheimer’s disease and kuru plaque amyloid in Creutzfeldt-Jakob disease. Brain Res, 541, 163–6. Naruse, S., Igarashi, S., Kobayashi, H. et al. (1991). Mis-sense mutation Val—-Ile in exon 17 of amyloid precursor protein gene in Japanese familial Alzheimer’s disease. Lancet, 337, 978–9. Naruse, S., Thinakaran, G., Luo, J. J. et al. (1998). Effects of PS1 deficiency on membrane protein trafficking in neurons. Neuron, 21, 1213–21. Neary, D., Snowden, J. S., Mann, D. M. et al. (1986). Alzheimer’s disease: a correlative study. J Neurol Neurosurg Psychiatry, 49, 229–37. Neuropathology Group of the Medical Research Council Cognitive Function and Ageing Study (MRC CFAS) (2001). Pathological correlates of late-onset dementia in a multicentre, communitybased population in England and Wales. Lancet, 357, 169–75. Neve, R. L., Finch, E. A. & Dawes, L. R. (1988). Expression of the Alzheimer amyloid precursor gene transcripts in the human brain. Neuron, 1, 669–77. Noguchi, S., Murakami, K. & Yamada, N. (1993). Apolipoprotein E genotype and Alzheimer’s disease. Lancet, 342, 737. O’Connor, D. W., Pollitt, P. A. & Treasure, F. P. (1991). The influence of education and social class on the diagnosis of dementia in a community population. Psychol Med, 21, 219–24. Ogunniyi, A. O., Osuntokun, B. O., Lekwauwa, U. B. & Falope, Z. F. (1992). Rarity of dementia (by DSM-III-R) in an urban community in Nigeria. East Afr Med J, 69, 64–8. Oka, M., Katayama, S., Watanabe, C., Noda, K., Mao, J. J. & Nakamura, S. (1998). Argyrophilic structures stimulate glial reactions in neurofibrillary tangles and senile plaques. Neurol Res, 20, 121–6. Okazaki, H., Reagan, T. J. & Campbell, R. J. (1979). Clinicopathologic studies of primary cerebral amyloid angiopathy. Mayo Clin Proc, 54, 22–31. Osuntokun, B. O., Ogunniyi, A. O. & Lekwauwa, U. G. (1992). Alzheimer’s disease in Nigeria. Afr J Med Med Sci, 21, 71–7. Osuntokun, B. O., Ogunniyi, A., Akang, E. E. et al. (1994). Beta A4-amyloid in the brains of non-demented Nigerian Africans. Lancet, 343, 56. Osuntokun, B. O., Sahota, A., Ogunniyi, A. O. et al. (1995). Lack of an association between apolipoprotein E epsilon 4 and Alzheimer’s disease in elderly Nigerians. Ann Neurol, 38, 463–5. Palacino, J. J., Berechid, B. E., Alexander, P. et al. (2000). Regulation of amyloid precursor protein processing by presenilin 1 (PS1) and PS2 in PS1 knockout cells. J Biol Chem, 275, 215–22. Palmer, A. M., Stratmann, G. C., Procter, A. W. & Bowen, D. M. (1988). Possible neurotransmitter basis of behavioral changes in Alzheimer’s disease. Ann Neurol, 23, 616–20.

Panegyres, P. K. (2001). The functions of the amyloid precursor protein gene. Rev Neurosci, 12, 1–39. Pasquier, F., Bail, L., Lebert, F., Pruvo, J. P. & Petit, H. (1994). Determination of medial temporal lobe atrophy in early Alzheimer’s disease with computed tomography. Lancet, 343, 861–2. Passer, B. J., Pellegrini, L., Vito, P., Ganjei, J. K. & D’Adamio, L. (1999). Interaction of Alzheimer’s presenilin-1 and presenilin-2 with Bcl-X(L). A potential role in modulating the threshold of cell death. J Biol Chem, 274, 24007–13. Payami, H., Kaye, J., Heston, L. L., Bird, T. D. & Schellenberg, G. D. (1993). Apolipoprotein E genotype and Alzheimer’s disease. Lancet, 342, 738. Pearl, G. S. (1997). Diagnosis of Alzheimer’s disease in a community hospital-based brain bank program. South Med J, 90, 720–2. Pericak-Vance, M. A., Bebout, J. L., Gaskell, P. C., Jr. et al. (1991). Linkage studies in familial Alzheimer disease: evidence for chromosome 19 linkage. Am J Hum Genet, 48, 1034–50. Perry, E. K., Marshall, E., Kerwin, J. et al. (1990). Evidence of a monoaminergic-cholinergic imbalance related to visual hallucinations in Lewy body dementia. J Neurochem, 55, 1454–6. Perry, E. K., McKeith, I., Thompson, P. et al. (1991). Topography, extent, and clinical relevance of neurochemical deficits in dementia of Lewy body type, Parkinson’s disease, and Alzheimer’s disease. Ann NY Acad Sci, 640, 197–202. Perry, R. H., Irving, D., Blessed, G., Fairbairn, A. & Perry, E. K. (1990). Senile dementia of Lewy body type. A clinically and neuropathologically distinct form of Lewy body dementia in the elderly. J Neurol Sci, 95, 119–39. Podlisny, M. B., Lee, G. & Selkoe, D. J. (1987). Gene dosage of the amyloid beta precursor protein in Alzheimer’s disease. Science, 238, 669–71. Poirier, J. (1994). Apolipoprotein E in animal models of CNS injury and in Alzheimer’s disease. Trends Neurosci, 17, 525–30. Poirier, J., Minnich, A. & Davignon, J. (1995). Apolipoprotein E, synaptic plasticity and Alzheimer’s disease. Ann Med, 27, 663– 70. Polvikoski, T., Sulkava, R., Haltia, M. et al. (1995). Apolipoprotein E, dementia, and cortical deposition of beta-amyloid protein. N Engl J Med, 333, 1242–7. Prince, M., Cullen, M. & Mann, A. (1994). Risk factors for Alzheimer’s disease and dementia: a case-control study based on the MRC elderly hypertension trial. Neurology, 44, 97–104. Raghavan, R., Khin-Nu, C., Brown, A. G. et al. (1994). Gender differences in the phenotypic expression of Alzheimer’s disease in Down’s syndrome (trisomy 21). Neuroreport, 5, 1393–6. Rapoport, S. I., Pettigrew, K. D. & Schapiro, M. B. (1991). Discordance and concordance of dementia of the Alzheimer type (DAT) in monozygotic twins indicate heritable and sporadic forms of Alzheimer’s disease. Neurology, 41, 1549–53. Rizzo, G., Scifo, P., Gilardi, M. C. et al. (1997). Matching a computerized brain atlas to multimodal medical images. Neuroimage, 6, 59–69. Roberts, G. W., Allsop, D. & Bruton, C. (1990). The occult aftermath of boxing. J Neurol Neurosurg Psychiatry, 53, 373–8.

203

204


Roberts, G. W., Gentleman, S. M., Lynch, A., Murray, L., Landon, M. & Graham, D. I. (1994). Beta amyloid protein deposition in the brain after severe head injury: implications for the pathogenesis of Alzheimer’s disease. J Neurol Neurosurg Psychiatry, 57, 419–25. Rocca, W. A., Amaducci, L. A. & Schoenberg, B. S. (1986). Epidemiology of clinically diagnosed Alzheimer’s disease. Ann Neurol, 19, 415–24. Rocca, W. A., Bonaiuto, S., Lippi, A. et al. (1990). Prevalence of clinically diagnosed Alzheimer’s disease and other dementing disorders: a door-to-door survey in Appignano, Macerata Province, Italy. Neurology, 40, 626–31. Rogers, J. D., Brogan, D. & Mirra, S. S. (1985). The nucleus basalis of Meynert in neurological disease: a quantitative morphological study. Ann Neurol, 17, 163–70. Rohan de Silva, H. A. & Patel, A. J. (1997). Presenilins and early-onset familial Alzheimer’s disease. Neuroreport, 8, i–xii. Roher, A. E., Lowenson, J. D., Clarke, S. et al. (1993). beta-Amyloid(1–42) is a major component of cerebrovascular amyloid deposits: implications for the pathology of Alzheimer disease. Proc Natl Acad Sci USA, 90, 10836–40. Roses, A. D. (1994). Apolipoprotein E affects the rate of Alzheimer disease expression: beta-amyloid burden is a secondary consequence dependent on APOE genotype and duration of disease. J Neuropathol Exp Neurol, 53, 429–37. Roth, M., Tomlinson, B. E. & Blessed, G. (1966). Correlation between scores for dementia and counts of ‘senile plaques’ in cerebral grey matter of elderly subjects. Nature, 209, 109–10. Samuel, W., Terry, R. D., DeTeresa, R., Butters, N. & Masliah, E. (1994). Clinical correlates of cortical and nucleus basalis pathology in Alzheimer dementia. Arch Neurol, 51, 772–8. Sasaki, A., Yamaguchi, H., Ogawa, A., Sugihara, S. & Nakazato, Y. (1997). Microglial activation in early stages of amyloid beta protein deposition. Acta Neuropathol (Berl), 94, 316–22. Sato, N., Imaizumi, K., Manabe, T. et al. (2001). Increased production of beta-amyloid and vulnerability to endoplasmic reticulum stress by an aberrant spliced form of presenilin 2. J Biol Chem, 276, 2108–14. Saunders, A. M., Strittmatter, W. J., Schmechel, D. et al. (1993). Association of apolipoprotein E allele epsilon 4 with late-onset familial and sporadic Alzheimer’s disease. Neurology, 43, 1467– 72. Saykin, A. J., Flashman, L. A., Frutiger, S. A. et al. (1999). Neuroanatomic substrates of semantic memory impairment in Alzheimer’s disease: patterns of functional MRI activation. J Int Neuropsychol Soc, 5, 377–92. Scahill, R. I., Schott, J. M., Stevens, J. M., Rossor, M. N. & Fox, N. C. (2002). Mapping the evolution of regional atrophy in Alzheimer’s disease: unbiased analysis of fluid-registered serial MRI. Proc Natl Acad Sci USA, 99, 4703–7. Schellenberg, G. D., Deeb, S. S., Boehnke, M. et al. (1987). Association of an apolipoprotein CII allele with familial dementia of the Alzheimer type. J Neurogenet, 4, 97–108.

Schellenberg, G. D., Bird, T. D., Wijsman, E. M. et al. (1988). Absence of linkage of chromosome 21q21 markers to familial Alzheimer’s disease. Science, 241, 1507–10. Schellenberg, G. D., Bird, T. D., Wijsman, E. M. et al. (1992a). Genetic linkage evidence for a familial Alzheimer’s disease locus on chromosome 14. Science, 258, 668–71. Schellenberg, G. D., Boehnke, M., Wijsman, E. M., Moore, D. K., Martin, G. M. & Bird, T. D. (1992b). Genetic association and linkage analysis of the apolipoprotein CII locus and familial Alzheimer’s disease. Ann Neurol, 31, 223–7. Schenk, D., Barbour, R., Dunn, W. et al. (1999). Immunization with amyloid-beta attenuates Alzheimer-disease-like pathology in the PDAPP mouse. Nature, 400, 173–7. Schmechel, D. E., Saunders, A. M., Strittmatter, W. J. et al. (1993). Increased amyloid beta-peptide deposition in cerebral cortex as a consequence of apolipoprotein E genotype in late-onset Alzheimer disease. Proc Natl Acad Sci USA, 90, 9649–53. Schoenberg, B. S., Anderson, D. W. & Haerer, A. F. (1985). Severe dementia. Prevalence and clinical features in a biracial US population. Arch Neurol, 42, 740–3. Schoenberg, B. S., Kokmen, E. & Okazaki, H. (1987). Alzheimer’s disease and other dementing illnesses in a defined United States population: incidence rates and clinical features. Ann Neurol, 22, 724–9. Selkoe, D. J. (1994). Alzheimer’s disease: a central role for amyloid. J Neuropathol Exp Neurol, 53, 438–47. (2000). Toward a comprehensive theory for Alzheimer’s disease. Hypothesis: Alzheimer’s disease is caused by the cerebral accumulation and cytotoxicity of amyloid beta-protein. Ann NY Acad Sci, 924, 17–25. (2001). Alzheimer’s disease: genes, proteins, and therapy. Physiol Rev, 81, 741–66. Seubert, P., Vigo-Pelfrey, C., Esch, F. et al. (1992). Isolation and quantification of soluble Alzheimer’s beta-peptide from biological fluids. Nature, 359, 325–7. Sheng, J. G., Griffin, W. S., Royston, M. C. & Mrak, R. E. (1998). Distribution of interleukin-1-immunoreactive microglia in cerebral cortical layers: implications for neuritic plaque formation in Alzheimer’s disease. Neuropathol Appl Neurobiol, 24, 278–83. Sherrington, R., Rogaev, E. I., Liang, Y. et al. (1995). Cloning of a gene bearing missense mutations in early-onset familial Alzheimer’s disease. Nature, 375, 754–60. Shoji, M., Golde, T. E., Ghiso, J. et al. (1992). Production of the Alzheimer amyloid beta protein by normal proteolytic processing. Science, 258, 126–9. Skoog, I., Nilsson, L., Palmertz, B., Andreasson, L. A. & Svanborg, A. (1993). A population-based study of dementia in 85-year-olds. N Engl J Med, 328, 153–8. Smith, A. D. (2002). Imaging the progression of Alzheimer pathology through the brain. Proc Natl Acad Sci USA, 99, 4135–7. Smith, A. D., Johnston, C., Sim, E. et al. (1994). Protective effect of apo epsilon 2 in Alzheimer’s disease. Oxford Project to Investigate Memory and Ageing (OPTIMA). Lancet, 344, 473–4.


Smith, A. D., Jobst, K. A., Johnston, C., Joachim, C. & Nagy, Z. (1996). Apolipoprotein-E genotyping in diagnosis of Alzheimer’s disease. Lancet, 348, 483–4. Smith, C. & Anderton, B. (1994). The molecular pathology of Alzheimer’s disease: are we any closer to understanding the neurodegenerative process. Neuropath. Appl. Neurobiol, 20, 322–38. Smith, M. Z., Esiri, M. M., Barnetson, L., King, E. & Nagy, Z. (2001). Constructional apraxia in Alzheimer’s disease: association with occipital lobe pathology and accelerated cognitive decline. Dement Geriatr Cogn Disord, 12, 281–8. Smith, T. W., Anwer, U., DeGirolami, U. & Drachman, D. A. (1987). Vacuolar change in Alzheimer’s disease. Arch Neurol, 44, 1225–8. Snow, R. E. & Arnold, S. E. (1996). Psychosis in Neurodegenerative Disease. Semin Clin Neuropsychiatry, 1, 282–93. Snowden, D., Ostwald, S. & Kane, R. (1989). Education, survival and independence in elderly Catholic sisters, 1936–1988. Am J Epidemiol, 130, 999–1012. Sorbi, S., Nacmias, B., Forleo, P., Piacentini, S., Amaducci, L. & Provinciali, L. (1993). APP717 and Alzheimer’s disease in Italy. Nat Genet, 4, 10. Sparks, D. L., Hunsaker, J. C., 3rd, Scheff, S. W., Kryscio, R. J., Henson, J. L. & Markesbery, W. R. (1990). Cortical senile plaques in coronary artery disease, ageing and Alzheimer’s disease. Neurobiol Aging, 11, 601–7. St George-Hyslop, P. H., Tanzi, R. E., Polinsky, R. J. et al. (1987). The genetic defect causing familial Alzheimer’s disease maps on chromosome 21. Science, 235, 885–90. St George-Hyslop, P. H., Haines, J. L., Farrer, L. A. et al. (1990). Genetic linkage studies suggest that Alzheimer’s disease is not a single homogeneous disorder. FAD Collaborative Study Group. Nature, 347, 194–7. St George-Hyslop, P. H., Haines, J., Rogaev, E. et al. (1992). Genetic evidence for a novel familial Alzheimer’s disease locus on chromosome 14. Nat Genet, 2, 330–4. Steiner, H., Romig, H., Grim, M. G. et al. (1999). The biological and pathological function of the presenilin-1 Deltaexon 9 mutation is independent of its defect to undergo proteolytic processing. J Biol Chem, 274, 7615–18. Strijers, R. L., Scheltens, P., Jonkman, E. J., de Rijke, W., Hooijer, C. & Jonker, C. (1997). Diagnosing Alzheimer’s disease in community-dwelling elderly: a comparison of EEG and MRI. Dement Geriatr Cogn Disord, 8, 198–202. Strittmatter, W. J., Saunders, A. M., Schmechel, D. et al. (1993). Apolipoprotein E: high-avidity binding to beta-amyloid and increased frequency of type 4 allele in late-onset familial Alzheimer disease. Proc Natl Acad Sci USA, 90, 1977–81. Sugiyama, H., Hainfellner, J. A., Yoshimura, M. & Budka, H. (1994). Neocortical changes in Parkinson’s disease, revisited. Clin Neuropathol, 13, 55–9. Sulkava, R., Haltia, M., Paetau, A., Wikstrom, J. & Palo, J. (1983). Accuracy of clinical diagnosis in primary degenerative dementia: correlation with neuropathological findings. J Neurol Neurosurg Psychiatry, 46, 9–13.

Sulkava, R., Wikstrom, J., Aromaa, A. et al. (1985). Prevalence of severe dementia in Finland. Neurology, 35, 1025–9. Sun, J., Li, M., Han, J. & Gu, J. (2001). Sensitization of differentiated PC12 cells to apoptosis by presenilin-2 is mediated by p38. Biochem Biophys Res Commun, 287, 536–41. Swartz, R. H., Black, S. E. & St George-Hyslop, P. (1999). Apolipoprotein E and Alzheimer’s disease: a genetic, molecular and neuroimaging review. Can J Neurol Sci, 26, 77–88. Sweet, R. A., Hamilton, R. L., Lopez, O. L. et al. (2000). Psychotic symptoms in Alzheimer’s disease are not associated with more severe neuropathologic features. Int Psychogeriatr, 12, 547–58. Tabaton, M., Cammarata, S., Mancardi, G. et al. (1991). Ultrastructural localization of beta-amyloid, tau, and ubiquitin epitopes in extracellular neurofibrillary tangles. Proc Natl Acad Sci USA, 88, 2098–102. Tagliavini, F. & Pilleri, G. (1983). Neuronal counts in basal nucleus of Meynert in Alzheimer disease and in simple senile dementia. Lancet, 1, 469–70. Tanzi, R. E., Gusella, J. F., Watkins, P. C. et al. (1987). Amyloid beta protein gene: cDNA, mRNA distribution, and genetic linkage near the Alzheimer locus. Science, 235, 880–4. Terry, R. D., Hansen, L. A., DeTeresa, R., Davies, P., Tobias, H. & Katzman, R. (1987). Senile dementia of the Alzheimer type without neocortical neurofibrillary tangles. J Neuropathol Exp Neurol, 46, 262–8. Terry, R. D., Masliah, E., Salmon, D. P. et al. (1991). Physical basis of cognitive alterations in Alzheimer’s disease: synapse loss is the major correlate of cognitive impairment. Ann Neurol, 30, 572–80. Tierney, M. C., Fisher, R. H., Lewis, A. J. et al. (1988). The NINCDSADRDA Work Group criteria for the clinical diagnosis of probable Alzheimer’s disease: a clinicopathologic study of 57 cases. Neurology, 38, 359–64. Tolnay, M. & Probst, A. (1999). REVIEW: tau protein pathology in Alzheimer’s disease and related disorders. Neuropathol Appl Neurobiol, 25, 171–87. Tomlinson, B. E. (1989). The neuropathology of Alzheimer’s disease – issues in need of resolution (Second Dorothy S. Russell memorial lecture). Neuropathol Appl Neurobiol, 15, 491–512. Tomlinson, B. E., Blessed, G. & Roth, M. (1968). Observations on the brains of non-demented old people. J Neurol Sci, 7, 331–56. (1970). Observations on the brains of demented old people. J Neurol Sci, 11, 205–42. Tomonaga, M. (1981). Cerebral amyloid angiopathy in the elderly. J Am Geriatr Soc, 29, 151–7. Toran-Allerand, C. D., Miranda, R. C., Bentham, W. D. et al. (1992). Estrogen receptors colocalize with low-affinity nerve growth factor receptors in cholinergic neurons of the basal forebrain. Proc Natl Acad Sci USA, 89, 4668–72. Tozer, R. (1991). Diagnosis of Alzheimer’s disease. Br Med J, 302, 48. Trojanowski, J., Schmidt, M., Shin, R.-W., Bramblett, G., Rao, D. & Lee, V.-Y. (1993a). PHFt(A68): from pathological marker to potential mediator of neuronal dysfunction and neuronal degeneration in Alheimer’s disease. Clin Neurosci, 1, 184–91.

205

206


Trojanowski, J. Q., Schmidt, M. L., Shin, R. W., Bramblett, G. T., Rao, D. & Lee, V. M. (1993b). Altered tau and neurofilament proteins in neuro-degenerative diseases: diagnostic implications for Alzheimer’s disease and Lewy body dementias. Brain Pathol, 3, 45–54. Van Broeckhoven, C., Genthe, A. M., Vandenberghe, A. et al. (1987). Failure of familial Alzheimer’s disease to segregate with the A4amyloid gene in several European families. Nature, 329, 153–5. Van Broeckhoven, C., Backhovens, H., Cruts, M. et al. (1992). Mapping of a gene predisposing to early-onset Alzheimer’s disease to chromosome 14q24.3. Nat Genet, 2, 335–9. van Duijn, C. M., Clayton, D., Chandra, V. et al. (1991). Familial aggregation of Alzheimer’s disease and related disorders: a collaborative re-analysis of case-control studies. EURODEM Risk Factors Research Group. Int J Epidemiol, 20, S13–20. Varadarajan, S., Yatin, S., Aksenova, M. & Butterfield, D. A. (2000). Review: Alzheimer’s amyloid beta-peptide-associated free radical oxidative stress and neurotoxicity. J Struct Biol, 130, 184– 208. Victoroff, J., Ross, G. W., Benson, D. F., Verity, M. A. & Vinters, H. V. (1994). Posterior cortical atrophy. Neuropathologic correlations. Arch Neurol, 51, 269–74. Vincent, I., Zheng, J. H., Dickson, D. W., Kress, Y. & Davies, P. (1998). Mitotic phosphoepitopes precede paired helical filaments in Alzheimer’s disease. Neurobiol Aging, 19, 287–96. Vinters, H. V. (1987). Cerebral amyloid angiopathy. A critical review. Stroke, 18, 311–24. Vinters, H. V. & Gilbert, J. J. (1983). Cerebral amyloid angiopathy: incidence and complications in the ageing brain. II. The distribution of amyloid vascular changes. Stroke, 14, 924–8. Wade, J. P., Mirsen, T. R., Hachinski, V. C., Fisman, M., Lau, C. & Merskey, H. (1987). The clinical diagnosis of Alzheimer’s disease. Arch Neurol, 44, 24–9. Wernicke, T. F. & Reischies, F. M. (1994). Prevalence of dementia in old age: clinical diagnoses in subjects aged 95 years and older. Neurology, 44, 250–3. Whitehouse, P. J., Price, D. L., Clark, A. W., Coyle, J. T. & DeLong, M. R. (1981). Alzheimer disease: evidence for selective loss of cholinergic neurons in the nucleus basalis. Ann Neurol, 10, 122–6. Whitehouse, P. J., Price, D. L., Struble, R. G., Clark, A. W., Coyle, J. T. & Delon, M. R. (1982). Alzheimer’s disease and senile dementia: loss of neurons in the basal forebrain. Science, 215, 1237–9. Whitehouse, P. J., Hedreen, J., White, C., Clark, A. W. & Price, D. (1983). Neuronal loss in the basal forebrain cholinergic system is more marked in Alzheimer’s disease than in senile dementia of the Alzheimer type. Ann. Neurol., 13, 243–8. Wilcock, G. & Jacoby, R. (1991). In Psychiatry in the elderly, ed. R. Jacoby & C. Oppenheimer, pp. 586–605. Oxford: Oxford University Press.

Wilcock, G. K. & Esiri, M. M. (1982). Plaques, tangles and dementia. A quantitative study. J Neurol Sci, 56, 343–56. Wilcock, G. K., Esiri, M. M., Bowen, D. M. & Smith, C. C. (1982). Alzheimer’s disease. Correlation of cortical choline acetyltransferase activity with the severity of dementia and histological abnormalities. J Neurol Sci, 57, 407–17. Wilcock, G. K., Esiri, M. M., Bowen, D. M. & Hughes, A. O. (1988). The differential involvement of subcortical nuclei in senile dementia of Alzheimer’s type. J Neurol Neurosurg Psychiatry, 51, 842–9. Wolf, H., Ecke, G. M., Bettin, S., Dietrich, J. & Gertz, H. J. (2000). Do white matter changes contribute to the subsequent development of dementia in patients with mild cognitive impairment? A longitudinal study. Int J Geriatr Psychiatry, 15, 803–12. Wolfe, M. S., De Los Angeles, J., Miller, D. D., Xia, W. & Selkoe, D. J. (1999). Are presenilins intramembrane-cleaving proteases? Implications for the molecular mechanism of Alzheimer’s disease. Biochemistry, 38, 11223–30. Yamaguchi, H., Nakazato, Y., Shoji, M. et al. (1991). Secondary deposition of beta amyloid within extracellular neurofibrillary tangles in Alzheimer-type dementia. Am J Pathol, 138, 699– 705. Yamaguchi, H., Ishiguro, K., Sugihara, S. et al. (1994). Presence of apolipoprotein E on extracellular neurofibrillary tangles and on meningeal blood vessels precedes the Alzheimer betaamyloid deposition. Acta Neuropathol, 88, 413–9. Yang, Y., Geldmacher, D. S. & Herrup, K. (2001). DNA replication precedes neuronal cell death in Alzheimer’s disease. J Neurosci, 21, 2661–8. Yoshimura, M. (1983). Cortical changes in the parkinsonian brain: a contribution to the delineation of ‘diffuse Lewy body disease’. J Neurol, 229, 17–32. Yoshioka, K., Miki, T., Katsuya, T., Ogihara, T. & Sakaki, Y. (1991). The 717Val—-Ile substitution in amyloid precursor protein is associated with familial Alzheimer’s disease regardless of ethnic groups. Biochem Biophys Res Commun, 178, 1141–6. Younkin, S. G. (1995). Evidence that A beta 42 is the real culprit in Alzheimer’s disease. Ann Neurol, 37, 287–8. Zakzanis, K. K. (1998). Quantitative evidence for neuroanatomic and neuropsychological markers in dementia of the Alzheimer’s type. J Clin Exp Neuropsychol, 20, 259–69. Zhang, M. Y., Katzman, R., Salmon, D. et al. (1990). The prevalence of dementia and Alzheimer’s disease in Shanghai, China: impact of age, gender, and education. Ann Neurol, 27, 428–37. Zubenko, G. S., Moossy, J. & Kopp, U. (1990). Neurochemical correlates of major depression in primary dementia. Arch Neurol, 47, 209–14. Zweig, R. M., Ross, C. A., Hedreen, J. C. et al. (1988). The neuropathology of aminergic nuclei in Alzheimer’s disease. Ann Neurol, 24, 233–42.

10 Down’s syndrome and Alzheimer’s disease David M. A. Mann Clinical Neurosciences Research Group University of Manchester, UK

Introduction Down’s syndrome (DS) occurs in about 1/800 live births (Adams et al., 1981) and accounts for about 17% of the mentally handicapped population (Heller, 1969). An association between DS and dementia was first noted well over a century ago by Fraser & Mitchell (1876), who wrote ‘in not a few instances, however, death was attributed to nothing more than a general decay – a sort of precipitated senility’. Nevertheless, it was not until much later (Struwe, 1929; Jervis, 1948) that the linkage between this ‘senile decay’ and the occurrence within the brain of the pathological lesions of Alzheimer’s disease (AD), namely senile plaques (SP) and neurofibrillary tangles (NFT), was noted. More recently, a number of studies have shown there to be an excess of DS births among the relatives of AD patients, particularly early onset AD families (Heston et al., 1981; Heyman et al., 1983; Broe et al., 1990; Van Duijn et al., 1991). Conversely, there is an increased risk of AD among mothers of DS children (Yatham et al., 1988; Schupf et al., 1994). Such observations support the pathological observations of shared etiopathogenetic causes for AD and DS. Life expectancy for people with DS has progressively risen with nearly half individuals living to beyond 50 years of age (Dupont et al., 1986; Baird & Sadovnick, 1987; Holland & Moss 1997) and with this the problems of ‘precocious ageing’ and dementia in DS have gained prominence.

Pathological similarities between the brain in Down’s syndrome and Alzheimer’s disease The presence of SP or NFT, or both, within one or more brain regions in persons with DS at different times of life has

been the subject of numerous case reports and some more extensive surveys (e.g. Struwe, 1929; Bertrand & Koffas, 1946; Jervis, 1948; Solitaire & Lamarche, 1966; Neumann, 1967; Haberland, 1969; Olson & Shaw, 1969; Malamud, 1972; O’Hara, 1972; Burger & Vogel, 1973; Schochet, et al., 1973; Ellis et al., 1974; Reid & Maloney, 1974; Crapper et al., 1975; Rees, 1977; Murdoch & Adams, 1977; Wisniewski et al., 1979, 1985a,b; Ball & Nuttall, 1980; Ropper & Williams, 1980; Blumbergs et al., 1981; Pogacar & Rubio, 1982; Whalley, 1982; Yates et al., 1983; Sylvester, 1983; Ross et al., 1984; Mann et al., 1984; 1985a, 1986, 1987a, 1990a,b; Mann & Esiri, 1989; Belza & Urich, 1986; Motte & Williams, 1989; Giaccone et al., 1989; De La Monte & Hedley-White, 1990; Ferrer & Gullotta, 1990). These 39 studies encompass some 434 patients, aged from under 10 years to over 70 years. When classic silver (e.g. Bodian, Palmgren, Bielschowsky) staining methods were employed, the presence of (any) SP or NFT or both, was noted in 260 (59.9%) of these patients. When analysed by decade, typical SP (see later) together with, or without, NFT first appeared, infrequently, during the second decade of life and increased rapidly through the third and fourth decades such that nearly 100% prevalence was reported in patients aged 40–60 years and exactly 100% in those over 60 years of age. In summary, of those 211 patients over 40 years of age surveyed here, 208 (98.6%) showed SP and NFT whereas only 52 of 223 patients (23.3%) under 40 years showed either SP or NFT or both changes. It seems therefore that in DS there is a transitional period, usually between 20 and 40 years of age, during which the absence of SP and NFT in any part of the brain almost predictably changes into a widespread presence throughout the brain. So far, only a few persons over the age of 40 years with the DS phenotype have been claimed to show


207

208

D. M. A. Mann

no SP or NFT within their brains (see Murdoch & Adams, 1977; Whalley, 1982; Prasher et al., 1998). In two of these persons, one a 55-year-old woman, the karyotype was that of a chromosomal mosaic (Whalley, 1982) whereas in the other, a 78 year old woman, the karyotype was a partial trisomy (Prasher et al., 1998). In another individual, a 49year-old woman (Whalley, 1982) the karyotype was that of a full trisomy 21, though unusually the brain weighed 1520 g and the histological examinations were insufficiently extensive to definitely exclude the presence of SP and NFT anywhere in the brain. The fourth such potential patient was briefly referred to in a letter by Murdoch and Adams (1977) in which they reported on a 56 year-old man whose brain was said not to contain either SP or NFT. How extensive the pathological investigations had been in this case is not clear, nor were any details of karyotype presented. Therefore, from these limited data, it is still uncertain as to whether any exceptions from the usual association between DS and the presence of SP and NFT in later life do indeed exist, certainly as far as individuals with a full trisomy 21 are concerned. This present survey also shows that, while in most, if not all, patients over 50 years of age SP and NFT always occur together and in high numbers, in those patients under this age a more variable picture is seen. For example, within those studies (Struwe, 1929; Jervis, 1948; Olson & Shaw, 1969; Burger & Vogel, 1973; Schochet et al., 1973; Wisniewski et al., 1979, 1985a; Ball & Nuttall, 1080; Ropper & Williams, 1980; Yates et al., 1983; Sylvester, 1983; Ross et al., 1984, Mann et al., 1986; Mann & Esiri, 1989; Motte & Williams, 1989; Giaccone et al., 1989) in which patients aged between 6 and 50 years were represented and in which the presence of SP and NFT in different brain regions was investigated separately (giving a total of 94 brains), 43 patients showed neither SP nor NFT in any area examined, 41 showed SP and NFT together in all areas examined or in the hippocampus alone, but at least ten patients in five studies (see Struwe, 1929; Jervis, 1948; Burger & Vogel, 1973; Motte & Williams, 1989; Giaccone et al., 1989) seemed to show only SP alone in one or more regions. However, because in most instances, surveys were limited either to frontal cortex or hippocampus, or both, it is possible that some NFT might have occurred in other areas not investigated. In no patient, however, did NFT occur in the absence of SP. In all studies reported so far (Solitaire & Lamarche, 1966; Olson & Shaw, 1969; Burger & Vogel, 1973; Reid & Maloney, 1974; Crapper et al., 1975; Rees, 1977; Wisniewski et al., 1985b; Motte & Williams, 1989; Rafalowska et al., 1988), including our own (Mann et al., 1984, 1985a, 1986, 1987a, 1990b), the pattern of involvement of brain structures by SP and NFT, in persons with DS who live beyond 50 years

of age, seems to closely follow that typically seen in AD (for review, see Mann, 1985, 1988a). For example, in elderly and established cases of DS, the amygdala, hippocampus and association areas of frontal, temporal and parietal cortex, especially the outer laminae (Rafalowska et al., 1988), are all strongly favoured by SP formation, whereas the visual, motor and somatosensory cortex are less affected (Mann et al., 1986; Motte & Williams, 1989). Nerve cells in the olfactory nuclei and tracts are also affected by NFT (Mann et al., 1986; Mann et al., 1988a) and sometimes also by SP. Typical SP are not seen in the cerebellar cortex (see later) nor are NFT present in Purkinje cells though occasionally nerve cells of the dentate nucleus contain NFT (Mann et al, 1990a). The nucleus basalis, locus ceruleus and raphe are all severely affected by NFT (Mann et al. 1984, 1985b, 1986). Whether the density of SP and NFT within affected brain regions is also similar in middle-aged persons with DS to that in patients with AD is not clear. Ball and Nuttall (1980) estimated that NFT density in the hippocampus of two patients with DS fell within that range encountered in eight patients with AD; SP were not quantified. Ropper and Williams (1980) estimated SP and NFT densities in the hippocampus of 8 persons over 50 years of age with DS and, although no data was presented, these were said to be comparable with levels in ‘demented old people’. Work by us (Mann, 1988b) suggests differences might occur; for example, in 15 persons with DS over 50 years of age the mean SP density in the temporal cortex was less than that in ten patients of that age with AD whereas NFT densities were equivalent in both groups. However, in the hippocampus, in DS, both SP and NFT densities far exceeded those usually seen in AD.

The morphological and immunohistochemical appearance of SP and NFT in Down’s syndrome Plaques The similarities in appearance between SP and NFT of patients with DS and those typically seen in AD (see earlier and Wisniewski & Terry (1973) have been emphasized at both light (Struwe, 1929; Jervis, 1948; Solitaire & Lamarche, 1966; Olson & Shaw, 1969; Burger & Vogel, 1973) and electron (O’Hara, 1972; Schochet et al., 1973; Ellis et al., 1974) microscope levels. However, a high proportion of amorphous plaque cores dissimilar to those of AD, being larger and having the amyloid fibrils less compact and lacking, under Congo red birefringence, the well-defined typical polarization cross has been reported (Masters et al., 1985; Allsop et al., 1986). Although these observations were made on biochemically isolated plaque cores rather than on SP

Down’s syndrome and Alzheimer’s disease

in situ, we too have observed these amorphous SP to be more common in DS after 50 years of age, and especially so within the amygdala and entorhinal cortex (Mann et al., 1986). Assuming that, in DS, patients start to form and accumulate SP (and NFT) at 30–40 years of age, then the pathological picture seen at death may have been evolving in many patients for at least 20 years. In AD, the usual duration of illness is about 5 to 10 years, only occasionally exceeding 15 years. Differences in SP morphology between DS and AD may thus only represent variations in the ‘stages’ of their natural history consequent upon this longer pathological course. The proteinaceous, -pleated, congophilic material (amyloid) that comprises much of the ‘core’ of typical SP in AD (and DS), consists of a polypeptide of 39–42 amino acids and molecular weight, 4.2 kD (Masters et al., 1985) termed amyloid peptide (A). This is formed through an alternative cleavage of a precursor molecule, amyloid precursor protein, APP (Esch et al., 1990; Sisodia et al., 1990), involving a series of enzymes known as secretases. APP is initially cleaved by either - or -secretase. The former enzyme cuts the APP molecule between amino acids 16 and 17 of A sequence thereby precluding formation of A, whereas secretase cleavage mainly occurs immediately N-terminal to A sequence to leave a 100-amino acid carboxy-terminal fragment. Further variable carboxyl cleavage of this fragment by -secretase yields A peptides ranging from 39–43 amino acids in length, of which the most numerous are those with either 40 or 42 amino acids, known as A40 or A42 , respectively. Immunohistochemistry, using antibodies directed against A has revealed that, in the cortex in AD, besides the typical cored SP there are (often many more) ‘diffuse’ types of deposits (Ogomori et al., 1989; Ikeda et al., 1989a; Tagliavini et al.; 1988; Mann et al., 1990a). Similar deposits are widely seen in the cortex, hippocampus and amygdala in DS (Mann et al., 1989a, 1990a; Mann & Esiri, 1989; Allsop et al., 1989; Ikeda et al., 1989b; Giaccone et al., 1989; Rumble et al., 1989; Murphy et al., 1990; Snow et al., 1990; Spargo et al., 1990). By using antibodies that specifically recognize A40 or A42 , it is seen that these diffuse plaques in DS, as in AD, are comprised largely and usually exclusively of A42 (Fig. 10.1(a)) whereas A40 is largely present within the cored plaques (Fig. 10.1(b)) (Iwatsubo et al. 1995; Mann et al. 1995a,b; Lemere et al. 1996). There is also much N-terminal heterogeneity attached to A in both AD and DS, with A3(P E )−42 and A11(P E )−42 being prominent species (Saido et al., 1995, 1996; Iwatsubo et al., 1996; Kuo et al., 1997; Russo et al. 1997, 2001) and this may result from variable cleavages by -secretase involving amino acids at positions other than residue 1 of the A sequence.

In both AD and DS, diffuse plaques within the cerebral cortex are not associated with a neuritic component nor do they usually display (much) astrocytic reaction (Murphy et al., 1990; Mann et al., 1992a). Though they do however contain microglia (Mann et al., 1995b) with the numbers of these increasing in line with the amount of A40 (Fig. 10.1(c), (d )) (Mann et al., 1995b). Surrounding the amyloid core are various cellular and non-cellular (more A) elements. Unusual accumulation of glycoproteins, probably as oligosaccharides, are present in the plaque periphery (Szumanska et al., 1987; Mann et al., 1988b, 1989a) though the precise molecular nature of these, as well as the cellular elements within which they are contained, remain to be characterized. Heparan sulphate proteoglycan (HSPG) is also accumulated with the A deposits within SP, both in AD and DS (Snow et al., 1990), as are the apolipoproteins E and J (Kida et al., 1995). These, together with Amyloid P component, complement factors and -antichymotrypsin, may act as ‘chaperone’ proteins (Wisniewski & Frangione, 1992) which promote the -sheet structure and mediate fibrillogenesis. Recently, an as yet unidentified 100 KDa protein known as AMY has been identified by immunohistochemistry within plaques in both AD (Schmidt et al., 1997; Lemere et al., 1999) and DS (Lemere et al., 1999). AMY protein frequently, but not always, co-localizes with A immunostaining and is more often present within cored, neuritic plaques than diffuse deposits. In DS, it was clearly seen that A immunostaining preceded that of AMY with the many diffuse A42 deposits being negative for AMY (Lemere et al., 1999). The significance of this non-A amyloid associated protein within the context of plaque formation and evolution remains unclear. Since AMY protein accrues within plaques after A deposition, and within neuritic plaques preferentially, it may play an important role in their maturation, perhaps in relationship to the development of neuritic plaques or neurofibrillary changes. Importantly, A immunostaining has also shown that, in both AD and DS, non-cortical areas such as the cerebellum (Ogomori et al., 1989; Ikeda et al., 1989a; Joachim et al., 1989; Yamaguchi et al., 1989; Wisniewski et al., 1989; Mann et al., 1990a) and striatum (Ogomori et al., 1989; Suenaga et al., 1990) contain many similarly diffuse, A deposits, composed exclusively of A42 (Fig. 10.1(e)) (Mann & Iwatsubo 1996). These can also contain microglial cells but, in contrast to the cerebral cortical deposits are not associated with a neuritic element and astrocytes are only infrequently present (Mann et al., 1992a). Biochemically, Masters et al. (1985) report that A of plaque cores in DS, is identical to that of plaque cores in AD. Teller et al. (1996) showed that the soluble fraction of A extractable from the brains of individuals with

209

210

D. M. A. Mann

Fig. 10.1. (a) Diffuse plaques of the cerebral cortex contain predominantly amyloid protein (A) in the form of A42(43) (b) whereas cored plaques contain much A40 as well as A42(43) . (d ) Microglial cell activity is associated with cored A40 -containing plaques, rather than (c) diffuse A42(43) -containing plaques. (e) Diffuse plaques of the cerebellum are composed exclusively of A42(43) , as are the early deposits ( f ) appearing within the cerebral cortex in young persons with DS.


DS was composed mostly of species terminating at amino acid 42 (consistent with immunohistochemistry), though in later studies (Russo et al. 1997, 2001) these same workers showed considerable N-terminal heterogeneity, with full length A1−42 and truncated species A3(P E )−42 being most prominent. Indeed, the proportion of this latter peptide species increases with age, becoming the most dominant peptide species within plaques as the pathological process progresses (Russo et al., 1997). Similar findings were subsequently reported by Hosoda et al., 1998. These timeassociated modifications of the N-terminus of A peptide may render the amyloid deposits less susceptible to degradation by aminopeptidase and facilitate their transformation into cored deposits.

Tangles Immunohistochemistry indicates that the microtubule associated protein, tau, is most probably the major antigenic determinant of the paired helical filaments (PHF) of the NFT in AD (see earlier and Wood et al., 1986; Kosik et al., 1986; Ihara et al., 1986; Delacourte & Defossez, 1986). The presence of tau within PHF has been confirmed by direct protein analysis (Goedert et al., 1988; Wischik et al., 1988; Kondo et al., 1988) which in conjunction with other immunohistochemical studies (e.g. Kosik et al., 1988) implies that the whole of the tau molecule is incorporated into PHF via its microtubule binding region and exists in an abnormally phosphorylated state (Wood et al., 1986). Immunostaining (Perry et al., 1987; Lowe et al., 1988; Lennox et al., 1988) and direct protein analysis (Mori et al., 1987) have shown another protein, ubiquitin, also to form an important part of the NFT. Lectin histochemistry (Szumanska et al., 1987; Mann et al., 1988b; Sparkman et al., 1990) has revealed the NFT to contain, or at least to be associated with, certain saccharide sequences. These properties of the NFT in AD seem to hold in DS. Cross-reactivity between the NFT of AD and DS occurred with an antibody to neurofilament protein (Anderton et al., 1982). The NFT of DS is similarly immunoreactive to tau and ubiquitin as that of AD (Mann et al., 1989b; Murphy et al., 1990). Immunoblotting (Flament et al., 1990; Hanger et al., 1991) shows similar mobility profiles for tau proteins in elderly DS brains as in AD. However, in DS, NFT do not seem to interact with lectins (DMA Mann et al., unpublished data).

Other pathological similarities As in AD, a granulovacuolar degeneration of neurones in the hippocampus, particularly area CA1, is a feature of DS at middle age (Solitaire & Lamarche, 1966; Olson & Shaw,

1969; Burger & Vogel, 1973; Ellis et al., 1974; Ball & Nuttall, 1980) and Hirano bodies are likewise common in this part of the hippocampus (Burger & Vogel, 1973; Ellis et al., 1974). Patients with AD often show mild extrapyramidal signs late in the course of the illness associated with mild loss of cells from the substantia nigra; this is generally due to neurofibrillary degeneration, and nigral cells containing Lewy bodies are mostly absent. This is generally what is also experienced in elderly DS subjects (Gibb et al., 1989b; Raghavan et al., 1993; Hestnes et al., 1996; but see Bodhireddy et al., 1994 for description of a single case of DS with widespread Lewy body formation). Nonetheless, investigations using antibodies to -synuclein (a major protein component of Lewy bodies) have shown that Lewy bodies are commonplace within the amygdala and entorhinal cortex of patients with familial AD due to APP and presenilin mutations, but are rare in sporadic AD (Gibb et al., 1989a; Lippa et al., 1998). Likewise, Lewy bodies are present widely within these same brain regions (but not in the cerebral cortex) in many elderly DS people (Lippa et al., 1999). The reason as to why this particular part of the brain should be susceptible to Lewy body formation in these genetic forms of AD is unknown, nor is it clear whether they contribute to the clinical disability of AD in DS. Calcification of the walls of the larger arteries, and a deposition of calcified deposits (calcospherities) around capillaries, of the globus pallidus, is often seen in AD, particularly in late onset cases (Mann, 1988 c). Elderly persons with DS likewise show excessive calcification of this part of the basal ganglia (Wisniewski et al., 1982; Takashima & Becker, 1985; Mann, 1988 c). However, it is not clear whether this change is related to ageing alone or whether the additional burden of AD has a bearing on its frequent occurrence. Again, as in younger patients with AD uncomplicated by single or multiple lacunae or embolic infarcts, atherosclerosis of the major vessels at the base of the brain is a conspicuously negative finding in (elderly) patients with DS (Olson & Shaw, 1969; Burger & Vogel, 1973, Murdoch et al., 1977; Mann, 1988b). A is also present within the walls of arteries that in AD display ‘congophilic angiopathy’ (Glenner & Wong, 1984a; Joachim et al., 1988). A severe deposition of A protein within the walls of large meningeal arteries, especially those supplying the posterior hemispheres and cerebellum and within the walls of some intraparenchymal arteries, is also a feature of most middle aged patients with DS (Mann et al., 1990a). Glenner & Wong (1984b) have shown that A of DS blood vessels is biochemically identical to that in arterial walls in AD and immunohistochemistry shows this to be largely composed of A40 . Such a vascular amyloidosis is associated with a deposition of HSPG (Snow et al., 1990).

211

212

D. M. A. Mann

Other vascular changes, involving a deposition of glycoproteins within the endothelium or basement membrane and detectable by lectin histochemistry, occur after the age of 50 years in the small arteries of the cortex and hippocampus in DS (Mann et al., 1992b). Similar changes are present in AD, but in both instances seem unrelated to the deposition in vessels of A protein; their pathological significance remains uncertain though they may result or stem from a compromise in the blood–brain barrier that might have a bearing on the neuronal loss and atrophy taking place at that time of life.

Neuronal fallout and neurochemical changes in elderly persons with Down’s syndrome Because of developmental deficiencies, it is quite likely that individuals with DS do not start out life with the same complement of nerve cells as their non-DS counterparts in the general population. Hence, it cannot be assumed, prima facie, that ‘low values’ for cell number represent actual changes in cell complement in elderly DS subjects associated with the development of Alzheimer-type pathology. However, by comparing cell counts in elderly DS subjects with those from young DS individuals, rather than with young adults in the general population (Mann et al, 1987a, 1990b), it is apparent that an actual loss and atrophy of nerve cells in many brain regions does indeed occur. In elderly people with DS there is loss of pyramidal (Mann et al., 1985a) and non-pyramidal (Kobayashi et al., 1990) nerve cells from areas of the temporal cortex, hippocampus (Ball & Nuttall, 1980) and entorhinal cortex (Hyman & Mann, 1991). The nucleus basalis (Mann et al., 1985a; Mann et al., 1985b; Price et al., 1982; Casanova et al., 1985), locus ceruleus (Mann et al., 1985b; Marcyniuk et al., 1988; German et al., 1992), dorsal raphe (Mann et al., 1984, 1985b) and ventral tegmentum (Mann et al., 1987b; Gibb et al., 1989a,b) are also grossly depleted of nerve cells. Atrophy of surviving nerve cells of these types also takes place, as witnessed by reductions in nucleolar size (an index of ribosomal RNA synthesis and cellular protein synthetic activity) (Mann et al., 1984, 1985a, b, 1987a). This pattern of cell loss and atrophy is similar to that in AD, as compared to similarly aged persons in the general population (for reviews see Mann, 1985, 1988a). In AD, nerve cell atrophy and loss leads to associated reductions in neurochemical markers (i.e. transmitter levels, enzyme activities or receptor densities) in regions of brain containing the cell bodies or nerve terminals comprising such affected systems (see Mann & Yates, 1986 for review). In elderly individuals with DS, as

would be anticipated, low levels of choline acetyl transferase (ChAT) within the cerebral cortex and other brain regions have been reported (Yates et al., 1980, 1983; Godridge et al., 1987). Noradrenaline (Yates et al., 1981, 1983, Reynolds & Godridge, 1985; Godridge et al., 1987) and 5-hydroxytryptamine (Yates et al., 1986; Godridge et al., 1987) levels are reduced in cortex and other areas. Loss of glutamate (Reynolds & Warner, 1988) and GABA (Reynolds & Warner, 1988) in cerebral cortex has been reported, as have reductions in D-3 H aspartate binding (Simpson et al., 1989). Dopamine levels appear to be unaltered (Yates et al., 1983; Godridge et al., 1987). The amount of somatostatin also appears to be low in elderly DS brains (Pierotti et al., 1986).

Gross changes in the brain in Down’s syndrome The cerebral atrophy of AD is brought about by a shrinkage and loss of neurones within particular cortical and subcortical structures and by the loss of pathways connecting such areas. The weight of the brain in younger individuals with DS is usually low for age. For example, in the combined studies of Benda (1960), Solitaire & Lamarche (1967), Whalley (1982), Wisniewski et al. (1985a) and Mann and Esiri (1989), brain weight exceeded 1200 g in only 21 out of 106 patients, over 8 but under 50 years of age, and fell below 1000 g in 16 (15%) patients. Few mentally able persons of that age-range have a brain weight under 1200 g, and certainly none have a weight under 1000g. After the age of 50 years, however, the weight of the brain in DS falls when compared to younger DS individuals (see Wisniewski et al., 1985a). Hence, in the combined studies of Solitaire & Lamarche (1967), Whalley (1982), Wisniewski et al (1985a) and Mann (1988b), out of a total of 75 patients over 50 years of age, only 7 (9%) still had a brain weight of over 1200 g, but 39 (52%) now had a brain weight below 1000 g. Hence, similar reductions in brain weight (in percentage terms) occur in elderly, as compared to younger, DS individuals as in patients with AD, when related to non-demented persons in the general population (Mann et al., 1990b). Morphometric analysis (De La Monte & Hedley-White, 1990) shows the decrease in brain size in DS in later life is brought about, as in AD (De La Monte, 1989; Mann, 1991), by a loss of tissue, both grey and white, especially from posterior parts of the brain. Such changes undoubtedly relate to the onset and progression of Alzheimer-type pathology in such regions. Serial CT scanning (Schapiro et al., 1989b) shows that, while healthy young persons with DS have smaller brains than persons of that age within the general population (see


above), older DS subjects have a reduced brain size when compared to such younger DS individuals, and that this declines further with age and upon onset of dementia. Although their brains are smaller than usual, younger DS patients still have a normal (proportionately to size) cerebral regional glucose utilization (Schapiro et al., 1990) and blood flow (Risberg, 1980; Schapiro et al., 1988). In elderly patients, as in AD, both of these parameters are reduced, particularly in the temporal and posterior parietal cortex, in relationship to younger DS individuals (Schapiro et al., 1986, 1987) or to normal, non-Down’s persons of that age (Melamed et al., 1987). Hence, in general, differences in SP and NFT structure or chemistry between AD and DS seem slight; patterns of neuronal damage and loss of transmitters also appear similar. Any variations that may possibly occur might reflect differences in patient life history (e.g. community versus institutional life) or the differing time courses of evolution of pathology and not necessarily be of major aetiological or pathogenetic significance. Moreover, the changes of AD in elderly patients with DS are not due to mental handicap, per se. Malamud (1972) found that, while AD changes were present in all patients with DS over 40 years of age, these occurred in only 31 of 225 (14%) other (non-DS) mentally handicapped persons of that age; a prevalence rate probably similar to that of the elderly in the general population.

Temporal progression The predictability with which the pathological changes of AD develop in elderly persons with DS has provided a unique opportunity in humans to detect the earliest tissue changes of this destructive process. It therefore becomes possible to reconstruct a chronological course of pathological events by pooling cross-sectional data from young and old individuals with DS and to follow them in time to that end-point characteristically seen at autopsy in persons in the general population dying with AD itself. This kind of study cannot be performed on patients with AD itself, since brain tissues are only usually available at postmortem and then mostly from clinical and pathological ‘end-stage’ cases in whom the early changes of the disease will either no longer be present, or will not be easily identifiable. Moreover, it is not possible with any degree of certainty to differentiate those (non-demented) persons showing early pathological stages of AD from other patients also showing such minimal changes but who may not necessarily have developed the full-blown pathological picture of AD, and become demented, had they lived longer.

These kinds of studies in DS (Mann et al., 1989a, b, 1990a, b, 1992a; Mann & Esiri, 1989; Allsop et al., 1989; Ikeda et al., 1989b; Giaccone et al., 1989; Rumble et al., 1989; Murphy et al., 1990; Snow et al 1990; Wegiel et al., 1996; Kida et al., 1995; Eikelenboom & Veerhuis 1996; Lemere et al., 1996; Leverenz & Raskind 1998; Stoltzner et al., 2000; Head et al., 2001), using immunohistochemical probes to detect the presence of molecular and cellular elements such as A, tau, ubiquitin, PHF, oligosaccharides, proteoglycans, apolipoproteins, complement factors or glial cells, have shown that the sequence of changes within the cerebral cortex and hippocampus is initiated by deposition of A protein. This is as A42(43) , in the form of diffuse plaques (Figure 10.1f) but there is some N-terminal heterogeneity, even in these early deposits, with A3(P E )−42 and A11(P E )−42 being prominent species (Saido et al., 1995, 1996; Iwatsubo et al., 1996; Kuo et al., 1997; Russo et al., 1997, 2001). These latter species become increasingly present as the plaques evolve over time and racemization and isomerization may occur, particularly involving aspartate residues at positions 1 and 7 (Iwatsubo et al., 1996; Saido et al., 1996; Shapira et al., 1988; Fonseca et al., 1999; Azizeh et al., 2000). A deposition can commence in DS during the early teens (Fig. 10.1( f )) or sometimes even younger. Usually the cerebral cortex, particularly the parahippocampal, inferior and middle temporal gyri, is affected before the hippocampal formation. Soon afterwards activated microglial cells are present within A deposits and accumulations of glycoconjugates, apolipoproteins, complement factors and the proteoglycan HSPG, and other granular material detectable by anti-ubiquitin, appear as the A becomes fibrillar and the plaque begins to ‘mature’. Later, cored amyloid deposits, containing more activated microglia, much complement, apolipoproteins, ubiquitinated material and larger quantities of oligosaccharide and HSPG are seen. These cored deposits are reactive with anti-tau and anti-GFAP and contain filamentous structures (PHF), immunoreactive with antiubiquitin. At this stage, NFT are only occasionally present in the cerebral cortex, but are usually numerous in the hippocampus, especially in areas CAI and subiculum, entorhinal cortex and amygdala. After 50 years of age a pathological picture indistinguishable from that of AD is seen which involves a loss of neurones from cortical areas rich in NFT and in many patients (see later) a progressive deterioration in behaviour and personality. Deposition of A within the cerebellar cortex occurred about 5 years after that in the cerebral cortex, although such deposits never become associated with a neuritic change nor do NFT appear in Purkinje or other cerebellar cortical cells. Therefore it seems that the onset and progression of the pathological cascade of AD in persons with DS is triggered

213

214

D. M. A. Mann

by events that lead to a deposition of A protein within the cerebral cortex. Nonetheless, characteristic of the pathological process in DS is the prolonged prodromal period of as much as 25 years during which there is progressive formation of diffuse A deposits with minimal or no fibrillization of the A protein, no neuronal loss and little or no neurofibrillary changes. What impact, if any, these early diffuse plaques might have on brain function is not clear, and it remains to be defined as to whether they have any clinical repercussion. Functional impairment in later life seems to relate to the appearance and progress of neuritic plaques and neurofibrillary tangles, changes that undoubtedly lead to neuronal dysfunction and death and with that a loss of connectivity within the brain.

Genetic factors Although DS and AD therefore appear largely homogenous in pathological terms, the mechanisms driving the destructive process behind each seem different and diverse. In DS because of triplication of chromosome 21, there is overexpression and mismetabolism of APP, though this is perhaps only one of the (many) potential ways by which a common pathological end-point can be brought about. Other factors, some genetic, some possibly non-genetic, clearly exist and interplay within other persons in the general population to produce the same disease phenotype that we call ‘Alzheimer’s disease’. Most (around 95%) cases of DS are of trisomy 21 (Hook, 1981; Gilbert & Opitz, 1982) and as such will have three, rather than two, full copies of that chromosome. This abnormality arises from a failure of the chromosomes to separate correctly during the first or second stages of meiosis. However, some 4–6% of persons with phenotypical DS are ‘translocational’ (Hook, 1981; Gilbert & Opitz, 1982) with only a partial triplication of chromosome 21 (usually the most distal part of the long arm, a portion obligatory for expression of the DS phenotype). Here the extra chromosome 21 is joined to another chromosome, usually chromosome 14 or 22. In the remaining 1% of individuals a mosaic chromosomal abnormality is present in which only a proportion of the bodies cells carry the trisomy. Whether, and to what extent, these translocational or mosaical patients also show Alzheimer-type changes is not clear. In one study (Prasher et al., 1998) a 78-year-old woman with a partial trisomy karyotype (46,XX,rec(21)dup q, inv(21) (p12q22.1) bore all the physical features of DS but although her brain showed the typical developmental defects of DS there was no significant tau or A deposition. FISH analysis showed that the APP (and incidentally SOD-1) gene was present as

two copies whereas the genes within the DS ‘critical region’ (including S-100 gene) were triplicated. This important case shows that those genes dictating SP and NFT formation can be inherited independently (in normal amounts), and occur separately from those (in triplicate) which determine the DS phenotype. The case provides further support for the view that development of AD in DS is due to the presence of the additional copy of the APP gene. There have also been two reports of Alzheimer-type pathology being present in the brains of mosaic DS individuals (Rowe et al., 1989; Schapiro et al., 1989b) indicating that even a partial trisomy 21 can be sufficient to induce the pathological cascade. In concert with the triplication of chromosome 21, the APP gene seems to be overexpressed by some 4–5 times in trisomy 21-DS (Kang et al., 1987; Rumble et al., 1989; Neve et al., 1988; Oyama et al., 1994). It is therefore possible that imbalances in the handling of such an overproduction of APP lead to a progressive A deposition by ‘feeding’ the /-secretase cleavage pathway. The finding (Teller et al., 1996) of an increase in the ratio between A42 over A40 in DS brain, even from a very early age before plaques appear within the brain,would be consistent with this. In AD itself, the APP gene does not seem to exist in triplicate nor is APP gene over-expressed. However, it has been suggested (Geller & Potter, 1999) that mosaicism could be present in which a proportion of cells are trisomic and behave in an analogous way to those in DS. Some cases of AD can be inherited at an early age in an autosomally dominant fashion due to point mutations at codon 717 of the APP gene (Goate et al., 1991; Chartier-Harlin et al., 1991; Murrell et al., 1991; Naruse et al., 1991). Mutations at and around codon 717 of the APP molecule favour the processing of APP along the -secretase route elevating production (Scheuner et al., 1996) and deposition (Mann et al., 1996a) of A42(43) . Indeed transgenic mice overexpressing the APP gene, mutated as in these inherited forms of AD, show much A deposition though the other changes of AD (i.e. the neuritic plaques, NFT and nerve cell loss) are not present (Irizarry et al., 1997). In other families with inherited early onset AD mutations are in the presenilin (PS ) genes, PS1 and PS2 (see Nishimura et al., 1999 for review). As with the above APP mutations, PS1 and PS2 gene mutations affect -secretase activity (indeed PS-1 protein may be -secretase itself) to yield excessive deposition of A42(43) within the brain (Mann et al., 1996b, 1997b, 2001) – an effect also seen in transgenic mice habouring such mutations (Duff et al., 1996; Borchelt et al., 1996). Hence, a ‘mismetabolism’ of APP consequent upon genetic events involving the APP gene on chromosome 21 directly, or involving other genetic loci whose effect at protein level is to modulate APP processing, will result


in excess deposition of A in the brain both in AD and DS. Strong support for the view that triplication of the APP gene is indeed causal for the development of AD pathology in DS comes from correlative cytogenetic studies. Nonetheless, it has to be remembered that triplication of chromosome 21 may involve an overexpression of many other genes and these may have an impact upon the causation or progression of the pathological changes. Among these is the Cu/Zn-superoxide dismutase-1 (SOD-1) gene. A number of studies have shown SOD-1 to be elevated in a variety of cell types and organs, including brain, in DS in both young and elderly individuals in line with the extra gene copy (Brooksbank & Balazs, 1984; Sinet et al., 1975; Anneren & Epstein, 1987; Gulesserian et al., 2001). This could play a part in the generation of oxidative cell damage (see later). Again the translocational DS case described by Prasher et al., 1998, in which there was no AD-type pathology and the SOD-1 gene was not duplicated, illustrates the potential involvement of this gene in the pathogenetic cascade. Likewise, the gene for the neurotrophic factor S-100, produced by astrocytes, maps to chromosome 21 and both S-100 and its message are increased in young DS brain (Griffin et al., 1989; Mito & Becker, 1993), as are the numbers of S-100 immunoreactive astrocytes (Griffin et al., 1998). S-100 can induce expression of APP (Li et al., 1998) and may therefore exacerbate the elevation in APP expression due to the possession of the extra gene copy per se. However, in the translocational DS case described by Prasher et al., 1998, the S-100 gene was triplicated yet no significant AD pathology (including astrocytic changes) was present even though this individual lived to be 78 years old. Such findings would argue that the S-100 pathology seen in trisomy 21 DS is more likely to be reactive to the presence of A deposition than causal for this. Furthermore, the A producing enzyme -secretase (BACE2) is also encoded on chromosome 21. Elevated levels of BACE2 have been detected in the brain in DS (Acquati et al., 2000) and this may be responsible, in conjunction with the overexpression of the APP gene, for the high and earlier deposition of A in DS. Most AD is not, however, inherited in a ‘simple’ Mendelian fashion and represents a complex interaction between various genes and probably environmental factors. The strongest genetic factor associated to date with late onset sporadic and familial AD is the Apolipoprotein gene (APO E). This is a polymorphic gene occurring in three major allelic forms known as ε2, ε3, and ε4. In AD there is over representation of the ε4 allele (Corder et al., 1993), this increasing from a normal population level of around 14% to 30–50% depending on study and ethnic group. However,

most studies have shown that the APO E ε 4 allele frequency in DS does not differ from controls (Saunders et al., 1993; Hardy et al., 1994; Mann et al., 1995a; Martins et al., 1995; Van Gool et al., 1995; Hyman et al., 1995; Royston et al., 1994, Avramopoulous et al., 1996; Holder et al., 1996; Lambert et al., 1996; Del Bo et al., 1997; Rubinsztein et al. 1999). Indeed in some reports (Wisniewski et al., 1995; Cosgrave et al., 1996; Tyrrell et al., 1998) the ε4 allele frequency in elderly DS individuals is significantly lower than that in age matched controls, suggesting a premature death of bearers of ε4 allele in DS. As in the general population, DS individuals homozygous for the ε4 allele are rare. In 12 studies covering 882 subjects of all ages (Hardy et al., 1994; Mann et al., 1995a; Martins et al., 1995; Van Gool et al., 1995; Hyman et al., 1995; Royston et al., 1994; Avramopoulous et al., 1996; Holder et al., 1996; Lambert et al., 1996; Del Bo et al., 1997; Rubinsztein et al., 1999) 17 individuals (2%) (similar to control population frequencies) with this particular genotype were seen. However, the frequency of this fell to less than 1% when only those individuals dying after 50 years of age were considered. This suggests that subjects with the ε4 allele may actually have a lower life expectancy (Royston et al., 1994; Tyrrell et al., 1998; Cosgrave et al., 1996), especially those homozygous for the ε4 allele (Holder et al., 1996) with early death from cardiovascular disease or AD. Moreover, ε4 allele bearers are more likely to show a faster rate of cognitive decline during their first 4 decades of life (Del Bo et al., 1997) and are therefore more likely to become demented and at an earlier age than non-ε4 allele bearers (Royston et al., 1994; Martins et al., 1995; Schupf et al., 1996; Prasher et al., 1997; Rubinsztein et al., 1999). Again, this viewpoint is controversial (Van Gool et al., 1995; Lambert et al., 1996). Conversely, as in AD (Talbot et al., 1994), consensus opinion is that bearers of the ε2 allele in DS live longer, and are less likely to become demented (Royston et al., 1994; Tyrrell et al., 1998; Lambert et al., 1996; Rubinsztein et al., 1999). One of the biological effects of variations in APO E genotype in AD is its impact upon brain deposition of A, where there is a gene dosage effect on deposition of A40 with increasing number of ε4 alleles (Mann et al., 1997a; Gearing et al., 1996). However, in DS there seems to be no such relationship, with A40 and A42(43) levels both being similar whether ε4 allele is present or not (Mann et al., 1995a). In some studies late onset AD has been associated with increased frequencies of common polymorphisms in either the PS-1 gene (Wragg et al., 1996) or -1 antichymotrypsin (ACT ) gene (Kamboh et al., 1995) though this has not always been confirmed. The same polymorphisms in these two genes have been examined in DS (Tyrrell et al., 1999) but show no increased frequencies compared to

215

216

D. M. A. Mann

controls, neither were frequencies different in demented versus non-demented individuals. Involvement of such genetic variations in determining either the pathology of AD in DS, or whether dementia occurs or not, therefore seems unlikely.

Causative factors It is well known that the brains of children and young adults with DS are defective showing distinctive gross neuropathological changes as well as specific abnormalities of nerve cell number and neuronal connectivity; changes presumably responsible in some way for dictating the basic mental handicap. In appearance, the brain is ‘rounded’ and shows a foreshortening in the anterior-posterior dimension (Davidoff, 1928; Crome & Stern, 1972) with relative smallness of the frontal lobes (Crome & Stern, 1972; Wisniewski et al., 1985a). The cerebellum (Crome & Stern, 1972) and hippocampus (Sylvester, 1983; Wisniewski et al., 1985a) may also be small. The frontal and temporal gyri, especially the superior temporal gyrus, show incomplete eversion (Davidoff, 1928). There are structural abnormalities, with a low (for age) number of nerve cells in temporal (Ross et al., 1984; Mann et al., 1987a; Mann, 1988b) and other areas of cortex (Colon, 1972; Ross et al., 1984; Wisniewski et al., 1984; Wisniewski et al., 1986), hippocampus (Ball & Nuttall, 1980), subcortex and brainstem (Gandolfi et al., 1981; McGeer et al., 1985; Casanova et al., 1985; Mann et al., 1987a; Mann 1988b). Abnormalities in dendritic spines (Marin-Padilla, 1976; Suetsuga & Mehraein, 1980; Takashima et al., 1981; Becker et al., 1986; Ferrer & Gullotta, 1990), an arrested synaptogenesis (Wisniewski et al, 1984, 1986) and delayed post-natal myelination (Wisniewski & Schmidt-Sidor, 1986) have all been reported. All of these structural changes are likely to result from an abnormal modelling and wiring of the brain caused by a failure to properly integrate and coordinate the many growth and transcription factors that come into play at different times during the developmental and maturational periods of brain growth. Indeed, several transcription factors are encoded on chromosome 21, such as Ets-2. However, despite gene triplication, Ets-2 levels are not overexpressed in DS brain, but are in fact reduced (Greber-Platzer et al., 1999a). Ets-2 requires cooperation with other transcription factors like Fos and Jun for activation; JunD levels are low in DS brain (Labudova et al., 1998) whereas Fos levels are increased (Greber-Platzer et al., 1999b). Disordering of the complex relationships involving these, and perhaps other, transcription factors may therefore underpin (some of) the developmental abnormalities.

However, such structural changes, while of potential importance to the understanding of the pathophysiology conferring the basic mental handicap, offer no immediate clues as to why such persons develop AD in later life. Transgenic mice overexpressing S-100 gene initially show an accelerated development of dendrites though this soon gives way to cytoskeletal collapse and loss of dendrites; such mice experience learning difficulties (Whitaker-Azmitia et al., 1997). These changes may parallel the lack of dendritic arborization seen in DS brain (Takashima et al., 1994) though the high levels of S-100 seen in older DS subjects (Griffin et al., 1998) may reflect, at least partially, the increased astroglial activity in and around neuritic plaques (Sheng et al., 1997). Overexpression of SOD-1 in DS (Gulessarian et al., 2001) would be postulated to lead to increased formation of hydrogen peroxide, in the absence of compensatory detoxification by glutathione peroxidase and catalase. Indeed, glutathione and catalase levels do not seem to rise in DS to match SOD-1, and increased tissue hydrogen peroxide does occur (de Haan et al., 1997). Although there is some evidence of lipid peroxidation in DS (Ianello et al., 1999) this need not necessarily be due to increased SOD-1 and could plausibly arise through mitochondrial changes (Kim et al., 2000). Indeed, elevated expression of SOD-1 in older DS subjects (Gulessarian et al., 2001) might occur in response to oxidative stress imposed by the deposition of A (see later) or astrogliosis. Hence, the involvement of SOD-1 in either the developmental or degenerative phases of DS remains an open question. To date, therefore, it seems that the pathological cascade is triggered by deposition in the brain of A, this being consequent upon an excessive production and catabolism of APP. Whether A is itself neurotoxic and responsible, per se, for generating subsequent pathological events is still far from clear. Certainly there is a wealth of experimental data pointing towards roles for A in the generation of free radicals and oxidative stress, apoptosis and excitotoxicity. None of these are proven within the human brain and it still remains possible that deposition of A might simply represent a relatively innocuous tissue marker of a wider ranging process that carries in its wake other changes that cause the ‘malignant’ neurofibrillary alterations and eventual cell death. Indeed, it is clear that deposition of A does not always lead to neuritic changes and NFT formation in both AD and DS, particularly in areas such as the striatum (Suenaga et al., 1990) and cerebellum (Joachim et al., 1989; Mann et al., 1990a). Moreover, the early sites of A deposition (cerebral cortex, and particularly the temporal cortex) are not necessarily those in which tau accumulation leading to neurofibrillary degeneration take place (i.e. hippocampus). Hence, in AD and DS, PHF formation may


proceed in parallel to, but not necessarily as a direct consequence of A formation. A feature of maturing (into cored) diffuse plaques is the early appearance of microglial cells and complement factors (Mann et al., 1995b; Eikelenboom & Veerhuis 1996; Lemere et al., 1996; Stoltzner et al., 2000; Head et al., 2001). These features suggest that deposition of A induces a neuroinflammatory response that is initiated or exacerbated by the complement system. Indeed, clinical trials demonstrating that anti-inflammatory drugs slow the progression of dementia in AD (McGeer et al., 1996) and reduce the burden of A in transgenic mice (Lim et al., 2000) strengthen this viewpoint. Fibrillar forms of A avidly bind C1, the first component in the classical pathway (Jiang et al., 1994) which may trigger a chronic inflammatory cascade in which progressive cycles of complement deposition and microglial recruitment occur. Activated inflammatory cells, especially microglia (Sheng et al., 1997), may induce neuritic and neurofibrillary changes through release of further inflammatory mediators or free radicals and oxidative damage (Meda et al., 1995). Excessive production of S-100 by astrocytes may compound the problem by potentiating the formation and deposition of A, and promoting neuritic activity (Royston et al., 1999). Hence, the pathological process of AD in DS may be initially triggered by the early and excessive formation of A per se with its subsequent tissue deposition. This aspect of the cascade may, nonetheless, soon give way to a vicious cycle of neuroinflammation which promotes further A production, progressive glial cell activation and increasing neuritic and neurofibrillary change.

Dementia in Down’s syndrome The question as to whether all patients with DS who live beyond 40 years of age, and whose brains show SP and NFT in quantities sufficient to signal AD in the general population, do indeed also become demented has received much attention but is still unresolved. Many studies (Owens et al., 1971; Dalton et al., 1974; Wisniewski et al., 1978, 1985b, 1986; Lott & Lai, 1982; Miniszek, 1983; Thase et al., 1984; Dalton & Crapper, 1984, 1986; Hewitt et al., 1988; Schapiro et al., 1986; Fenner et al., 1987; Zigman et al 1987, 1996; Silverstein et al., 1988; Lai & Williams, 1989; Evenhuis, 1990; Burt et al., 1995; Holland et al., 1998; Devenny et al., 1996, 2000) show that many patients beyond this age display signs of mental deterioration or behavioural regression, though few(er) present an overt clinical deterioration that can be convincingly defined as dementia. Indeed, it is well recognized that certain individuals can live well into the sixth decade of life without

showing any evidence of behavioural or cognitive decline (Devenny et al., 2000). Arguments concerning the acquisition of relevant data that reflect the burden of an additional deficit upon a basic mental retardation, and the difficulties in extracting meaningful clinical data from retrospective records not specifically kept for the purpose of charting changes in cognitive function, are no doubt applicable and may go a long way towards explaining these apparent inconsistencies in many of the earlier studies. Furthermore, the changes in visual memory in later life, used to mark onset of dementia in several studies, could represent the effects of early deficiencies in the visual cortex (Wisniewski et al., 1984) aggravated by ageing instead of those of AD. However, such arguments cannot explain the continued findings of preserved elderly DS individuals in prospective studies of longitudinally assessed persons (Burt et al., 1995; Devenny et al., 1996, 2000). In order to explain this paradox, Wisniewski has argued (Wisniewski & Rabe 1986; Wisniewski et al., 1987) that a threshold effect might operate, dictating that a certain level of pathology must accrue before clinical dementia becomes apparent and that this threshold level might be higher in the DS brain. However, non-demented elderly DS persons were shown to have lost as many nerve cells as demented elderly DS persons, while still showing above ‘threshold’ numbers of SP and NFT (Mann et al., 1990b). However, like Wisniewski (Wisniewski & Rabe, 1986; Wisniewski et al., 1987) such non-demented patients had fewer neocortical SP and NFT than their demented counterparts, though the significance of this is unclear. Observations that many elderly persons with DS who bear APO E ε2 allele not only live longer, but are also less likely to become demented (Royston et al., 1994; Tyrrell et al., 1998; Lambert et al., 1996; Rubinsztein et al., 1999), may point towards crucial biological differences in their ability to maintain brain function, even in the face of massive pathology.

REFERENCES Acquati, F., Accarino, M., Nucci, C., Fumagalli, P., Jovine, L., Ottolenghi, S. & Tamarelli, R. (2000). The gene encoding DRAP (BACE2), a glycosylated transmembrane protein of the aspartic protease family, maps to the down critical region. FEBS Lett 468: 59–64. Adams, M. M., Erickson, J. D., Layde, P. M. & Oakley, G. P. (1981). Down’s syndrome. Recent trends in the United States. J Am Med Ass 246: 758–60. Allsop, D., Kidd, M., Landon, M. & Tomlinson, A. (1986). Isolated senile plaque cores in Alzheimer’s disease and Down’s syndrome show differences in morphology. J Neurol Neurosurg Psychiat 49: 886–92.

217

218

D. M. A. Mann

Allsop, D., Haga, S.-I., Haga, C., Ikeda, S.-I., Mann, D. M. A. & Ishii, T. (1989). Early senile plaques in Down’s syndrome brains show a close relationship with cell bodies of neurones. Neuropath Appl Neurobiol 15: 531–42. Anderton, B. H., Breinburg, D., Downes, M. J. et al. (1982). Monoclonal antibodies show that neurofibrillary tangles and neurofilaments share antigenic determinants. Nature 298: 87–96. Anneren, G. & Epstein, C. J. (1987). Lipid peroxidation and superoxide dismutase-1 and glutathione peroxidase activities in trisomy 16 fetal mice and human trisomy 21 fibroblasts. Pediatr Res 21: 88–92. Avramopoulos, D., Mikkelsen, M., Vassilopoulos, D., Grigoriadou, M. & Petersen, M. B. (1996). Apolipoprotein E allele distribution in parents of Down’s syndrome children. Lancet 347: 862–5. Azizeh, B. Y., Head, E., Ibrahim, M. A. et al. (2000). Molecular dating of senile plaques in the brains of individuals with Down syndrome and in aged dogs. Exp Neurol 163: 111–22. Baird, P. A. & Sadovnick, A. D. (1987). Life expectancy in Down’s syndrome. J Paediat 110: 849–54. Ball, M. J. & Nuttall, K. (1980). Neurofibrillary tangles and granulovacuolar degeneration and neurone loss in Down’s syndrome: quantitative comparison with Alzheimer’s dementia. Ann Neurol 7: 462–5. Becker, L. A., Armstrong, D. L. & Chan, F. (1986). Dendritic atrophy in children with Down’s syndrome. Ann Neurol 20: 520–7. Belza, M. G. & Urich, H. (1986). Cerebral amyloid angiopathy in Down’s syndrome. Clin Neuropathol 6: 257–60. Benda, C. E. (1960). The Child with Mongolism (Congenital Acromicria). Grune & Stratton, New York. Bertrand, I. & Koffas, D. (1946). Cad d’idiotie mongolienne adult avec nombreuses plaques seniles et concretions calcaires pallidales. Rev Neurol (Paris) 78: 338–45. Blumbergs, P., Beran, R. & Hicks, P. (1981). Myoclonus in Down’s syndrome. Association with Alzheimer’s disease. Arch Neurol 38: 453–4. Bodhireddy, S., Dickson, D. W., Mattiace, L. & Weidenheim, K. M. (1994). A case of Down’s syndrome with diffuse Lewy body disease and Alzheimer’s disease. Neurology 44: 159–61. Borchelt, D. R., Thinakaran, G., Eckman, C. B. et al. (1996). Familial Alzheimer’s disease-linked Presenilin 1 variants elevate A1– 42/1–40 ratio in vitro and in vivo. Neuron 17: 1005–13. Broe, G. A., Henderson, A. S., Creasey, H. et al. (1990). A case-control study of Alzheimer’s disease in Australia. Neurology 40: 1698– 707. Brooksbank, B. W. & Balazs, R. (1984). Superoxide dismutase, glutathione peroxidase and lipoperoxidation in Down’s syndrome fetal brain. Brain Res 318: 37–44. Burger, P. C. & Vogel, F. S. (1973). The development of the pathological changes of Alzheimer’s disease and senile dementia in patients with Down’s syndrome. Am J Path 73: 457–76. Burt, D. B., Loveland, K. A., Chen, Y.-W., Chuang, A., Lewis, K. R. & Cherry, L. (1995). Aging in adults with Down syndrome: Report from a longitudinal study. Am J Ment Retard 100: 262–70.

Casanova, M. F., Walker, L. C., Whitehouse, P. J. & Price, D. L. (1985). Abnormalities of the nucleus basalis in Down’s syndrome. Ann Neurol 18: 310–13. Chartier-Harlin, M. C., Crawford, F., Houlden, H. et al. (1991). Mutations at codon 717 of the amyloid precursor protein gene cause Alzheimer’s disease. Nature 353: 84–6. Colon, E. J. (1972). The structure of the cerebral cortex in Down’s syndrome. Neuropaediatrics 3: 362–76. Corder, E. H., Saunders, A. M., Strittmatter, W. J. et al. (1993). Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer’s disease in late onset families. Science 261: 921–3. Cosgrave, M., Tyrrell, J., Dreja, H., Hawi, Z., Lawlor, B. & Gill, M. (1996). Lower frequency of Apolipoprotein E4 allele in an ‘elderly’ Down’s syndrome population. Biol Psychiatry 40: 811–13. Crapper, D. R., Dalton, A. J., Skoptiz, M., Scott, J. W. & Hachinski, V. C. (1975). Alzheimer degeneration in Down’s syndrome. Arch Neurol 32: 618–23. Crome, L. & Stern, J. (1972). Pathology of Mental Retardation, 2nd edn., Williams & Wilkins, Baltimore. Dalton, A. J. & Crapper, D. R. (1984). Incidence of memory deterioration in ageing persons with Down’s syndrome. In J. M. Berg, (ed.). Perspectives and Progress in Mental Retardation, Vol. 2, University Park Press, Baltimore, pp. 55–62. (1986). Clinical expression of Alzheimer’s disease in Down’s syndrome. Psychiatry Clin N Am 4: 659–70. Dalton, A. J., Crapper, D. R. & Schlotterer, C. R. (1974). Alzheimer’s disease in Down’s syndrome: visual retention deficits. Cortex 10: 366–77. Davidoff, L. M. (1928). The brain in mongolian idiocy. Arch Neurol Psychiat 20: 1229–57. De Haan, J. B., Wolvetang, E. J., Ianello, R., Bladier, C., Kelner, M. J. & Kola, I. (1997). Reactive oxygen species and their contribution to pathology in Down syndrome. Adv Pharmacol 38: 379–402. Delacourte, A. & Defossez, A. (1986). Alzheimer’s disease tau proteins, the promoting factors of microtubule assembly, are major components of paired helical filaments. J Neurol Sci 76: 173–86. De la Monte, S. (1989). Quantitation of cerebral atrophy in preclinical and end-stage Alzheimer’s disease. Ann Neurol 25: 450–9. De la Monte, S. M. & Hedley-White, E. T. (1990). Small cerebral hemispheres in adults with Down’s syndrome. Contributions of developmental arrest and lesions of Alzheimer’s disease. J Neuropath Exp Neurol 49: 509–20. Del Bo R., Comi, G. P., Bresolin, N. et al. (1997). The apolipoprotein E ε4 allele causes a faster decline of cognitive performances in Down’s syndrome subjects. J Neurol Sci 145: 87–91. Devenny, D. A., Silverman, W. P., Hill, A. L., Jenkins, E., Sersen, E. A. & Wisniewski, K. E. (1996). Normal ageing in adults with Down’s syndrome: a longitudinal study. J Intellect Disability Res 40: 208–21. Devenny, D. A., Krinsky-McHale, S. J., Sersen, G. & Silverman, W. P. (2000). Sequence of cognitive decline in dementia in adults with Down’s syndrome. J Intellect Disability Res 44: 654–65.


Duff, K., Eckman, C. & Zehr, C. (1996). Increased amyloid-42(43) in brains of mice expressing mutant presenilin. Nature 383: 710–13. Dupont, A., Vaeth, M. & Videbech, P. (1986). Mortality and life expectancy of Down’s syndrome in Denmark. J Ment Defic Res 30: 111–20. Eikelenboom, P. & Veerhuis, R. (1996). The role of complement and activated microglia in the pathogenesis of Alzheimer’s disease. Neurobiol Ageing 17: 673–80. Ellis, W. G., McCulloch, J. R. & Corley, C. L. (1974). Presenile dementia in Down’s syndrome. Ultrastructural identity with Alzheimer’s disease. Neurology 24: 101–6. Esch, E. S., Kelm, P., Beattie, E. C. et al. (1990). Cleavage of amyloid peptide during constitutive processing of its precursor. Science 248: 1122–4. Evenhuis, H. M. (1990). The natural history of dementia in Down’s syndrome. Arch Neurol 47: 263–7. Fenner, M. E., Hewitt, K. E. & Torpy, D. M. (1987). Downs syndrome intellectual and behavioural functioning during adulthood. J Ment Defic Res 31: 241–9. Ferrer, I. & Gullotta, F. (1990). Down’s syndrome and Alzheimer’s disease: dendritic spine counts in the hippocampus. Acta Neuropathol 79: 680–5. Flament, S., Delacourte, A. & Mann, D. M. A. (1990). Phosphorylation of tau proteins: a major event during the process of neurofibrillary degeneration. Comparisons between Alzheimer’s disease and Down’s syndrome. Brain Res 516: 15–19. Fonseca, M. I., Head, E., Velasquez, P., Cotman, C. W. & Tenner, A. J. (1999). The presence of isoaspartic acid in amyloid plaques indicates plaque age. Exp Neurol 157: 277–88. Fraser, J. & Mitchell, A. (1876). Kalmuc idiocy: report of a case with autopsy with notes on 62 cases. J Ment Sci 22: 161–9. Gandolfi, A., Horoupian, D. S. & DeTeresa, R. M. (1981). Pathology of the auditory system in trisomies with morphometric and quantitative study of the ventral cochlear nucleus. J Neurol Sci 51: 43–50. Gearing, M., Mori, H. & Mirra, S. S. (1996). A peptide length and apolipoprotein E genotype in Alzheimer’s disease. Ann Neurol 39: 395–9. Geller, L. N. & Potter, H. (1999). Chromosome missegregation and trisomy 21 mosaicism in Alzheimer’s disease. Neurobiol Dis 6: 167–79. German, D. C., Manaye, K. F., White, C. L. et al. (1992). Disease specific patterns of locus caeruleus cell loss: Parkinson’s disease, Alzheimer’s disease and Down’s syndrome. Ann Neurol 32: 667–76. Giaccone, G., Tagliavini, F., Linoli, G. et al. (1989). Down patients: extracellular preamyloid deposits precede neuritic degeneration and senile plaques. Neurosci Lett 97: 232–8. Gibb, W. R. G., Mountjoy, C. Q., Mann, D. M. A. & Lees, A. J. (1989a). The substantia nigra and ventral tegmental area in Alzheimer’s disease and Down’s syndrome. J Neurol Neurosurg Psychiatry 52: 193–200.

(1989b). A pathological study of the association between Lewy body disease and Alzheimer disease. J Neurol Neurosurg Psychiatry 52: 701–8. Gilbert, E. F. & Opitz, J. M. (1982). Developmental and other pathological changes in syndromes caused by chromosomal abnormalities. Perspect Paediat Path 7: 1–63. Glenner, G. G., & Wong C. W. (1984a). Alzheimer’s disease: initial report of the purification and characterization of a novel cerebrovascular amyloid protein. Biochem Biophys Res Commun 120: 885–90. (1984b). Alzheimer’s disease and Down’s syndrome: sharing a unique cerebrovascular amyloid fibril. Biochem Biophys Res Commun 122: 1131–5. Goate, A., Chartier-Harlin, M. C., Mullan, M. et al. (1991). Segregation of a missense mutation in the amyloid precursor gene with familial Alzheimer’s disease. Nature 349: 704–6. Godridge, H., Reynolds, G. P., Czudek, C., Calcutt, N. A. & Benton, M. (1987). Alzheimer-like neurotransmitter deficits in adult Down’s syndrome brain tissue. J Neurol Neurosurg Psychiatry 50: 775–8. Goedert, M., Wischik, C., Crowther, R. A., Walker, J. E. & Klug, A. (1988). Cloning and sequencing of the cDNA encoding a core protein of the paired helical filament of Alzheimer’s disease: identification as the microtubule associated protein, tau. Proc Nat Acad Sci USA 85: 4051–5. Greber-Platzer, S., Balcz, B., Cairns, N. J. & Lubec, G. (1999a). C-fos expression in brain of patients with Down syndrome. J Neural Transm 57: 75–86. Greber-Platzer, S., Turhani-Schatzmann, D., Cairns, N. J., Balcz B. & Lubec, G. (1999b). The expression of the transcription factor ets-2 in the brains of patients with Down syndrome. Evidence against the overexpression-gene dosage hypothesis. J Neural Transm 57: 269–82. Griffin, W. S. T., Stanley, L. C., Ling, C. et al. (1989). Brain interleukin 1 and S100 immunoreactivity are elevated in Down’s syndrome and Alzheimer’s disease. Proc Natl Acad Sci USA 86: 7611–5. Griffin, W. S. T., Sheng, J. G., McKenzie, J. E. et al. (1998). Life long overexpression of S100 in Down’s syndrome: implications for Alzheimer pathogenesis. Neurobiol Ageing 19: 401–5. Gulesserian, T., Fountoulakis, M., Seidl, R., Hardmeier, R., Cairns, N. & Lubec, G. (2001). Superoxide dismutase SOD1, encoded on chromosome 21, but not SOD2 is overexpressed in the brains of patients with Down syndrome. J Invest Med 49: 41–46. Haberland, C. (1969). Alzheimer’s disease in Down’s syndrome: clinical and neuropathological observations. Acta Neurol Belg, 69: 369–80. Hanger, D. P., Brion, J.-P., Gallo, J.-M., Cairns, N. J., Luthert, P. J. & Anderton, B. H. (1991). Tau in Alzheimer’s disease and Down’s syndrome is insoluble and abnormally phosphorylated. Biochem J 275: 99–104. Hardy, J., Crook, R., Perry, R., Raghavan, R. & Roberts, G. (1994). Apo E genotype and Down’s syndrome. Lancet 343: 979–80. Head, E., Azizeh, B. Y., Lott, I. T., Tenner, A. J., Cotman, C. W. & Cribbs, D. H. (2001). Complement association with neurones and

219

220

D. M. A. Mann

-amyloid deposition in the brains of aged individuals with

Down syndrome. Neurobiol Dis 8: 252–65. Heller, J. H. (1969). Human chromosome abnormalities as related to physical and mental dysfunction. J Hered 60: 239–52. Hestnes, A., Daniel, S. E., Lees, A. J. & Brun, A. (1996). Down’s syndrome and Parkinson’s disease. J Neurol Neurosurg Psychiatry 53: 289. Heston, L. L., Mastri, A. R., Anderson, V. E. & White, J. (1981). Dementia of the Alzheimer type: Clinical genetics, natural history and associated conditions. Arch Gen Psychiatry 38: 1085– 90. Hewitt, K. E., Carter, G. & Jancar, J. (1985). Ageing in Down’s syndrome. Br J Psychiatry 147: 58–62. Heyman, A., Wilkinson, W. E., Hurwitz, B. J., Schmechel, D., Sigmon, A. H., Weinberg, T., Helms, M. J. & Swift, M. (1983). Alzheimer’s disease: genetic aspects and associated clinical disorders. Ann Neurol 14: 507–15. Holder, J. L., Habbak, R. A., Pearlson, G. D., Aylward, E. A., Pulsifer, M. & Warren, A. C. (1996). Reduced survival of apolipoprotein E4 homozygotes in Down’s syndrome. NeuroReport 7: 2455–6. Holland, A. J. & Moss, S. (1997). The mental health needs of older people with learning disabilities. In Jacoby R, Openheimer C eds, Psychiatry in the Elderly. Oxford University Press, Oxford, 494–506. Holland, A. J., Hon, J., Huppert, F. A., Stevens, F. & Watson, P. (1998). Population based study of the prevalence and presentation of dementia in adults with Down’s syndrome. Br J Psychiatry 172: 493–8. Hook, E. B. (1981). Down’s syndrome: its frequency in human populations and some factors pertinent to variations in rates. In F. F. de la Cruz, P. S. Gerald (eds) Trisomy 21 ( Down’s Syndrome): Research Perspectives, University Park Press, Baltimore USA, pp. 3–68. Hosoda, R., Saido, T. C., Otvos, L. et al. (1998). Quantification of modified amyloid peptides in Alzheimer’s disease and Down’s syndrome brains. J Neuropathol Exp Neurol 57: 1089–95. Hyman, B. T. & Mann, D. M. A. (1991). Alzheimer-type pathological changes in Down’s individuals of various ages. In K. Iqbal, D. R. C. McLachlan, B. Winblad & H. M. Wisniewski, (eds), Alzheimer’s disease: Basic Mechanisms, diagnosis and Therapeutic Strategies. Wiley, New York, pp. 105–13. Hyman, B. T., West, H. L., Rebeck, G. W., Lai, F. & Mann, D. M. A. (1995). Neuropathological changes in Down’s syndrome hippocampal formation: effect of age and apolipoprotein E genotype. Arch Neurol 52: 373–8. Ianello, R. C., Crack, P. J., De Haan, J. B. & Kola, I. (1999). Oxidative stress and neural dysfunction in Down syndrome. J Neural Transm 57: 256–67. Ihara, Y., Nukina, N., Miura, R. & Ogawara, M. (1986). Phosphorylated tau protein is integrated into paired helical filaments in Alzheimer’s disease. J Biochem (Jpn) 99: 1807–10. Ikeda, S.-I., Allsop, D. & Glenner, G. G. (1989a). The morphology and distribution of plaque and related deposits in the brains of Alzheimer’s disease and control cases: an

immunohistochemical study using amyloid protein antibody. Lab Invest 60: 113–22. Ikeda, S.-I., Yanagisawa, N., Allsop, D., Glenner, G. G. (1989b). Evidence of amyloid protein immunoreactive early plaque lesions in Down’s syndrome brains. Lab Invest 61: 133–7. Irizarry, M., Soriano, F., McNamara, M. et al. (1997). Abeta deposition is associated with neuropil changes but not with overt neuronal loss in the human amyloid precursor protein V717F (PDAPP) transgenic mouse. J Neurosci 17: 7053–9. Iwatsubo, T., Mann, D. M. A., Odaka, A., Suzuki, N. & Ihara, Y. (1995). Amyloid protein (A) deposition: A42(43) precedes A40 in Down syndrome. Ann Neurol 37: 294–9. Iwatsubo, T., Saido, T. C., Mann, D. M. A., Lee, V. M.-Y. & Trojanowski, J. Q. (1996). Full length amyloid--(1–42(43)) and amino-terminally modified and truncated amyloid-42(43) deposits in diffuse plaques. Am J Pathol 149: 1823–30. Jervis, G. A. (1948). Early senile dementia and mongoloid idiocy. Am J Psychiatry 105: 102–6. Jiang, H., Burdick, D., Glabe, C. G., Cotman, C. W. & Tenner, A. J. (1994). -amyloid activates complement by binding to a specific region of the collagen-like domain of C1qA chain. J Immunol 152: 5050–9. Joachim, C. L., Duffy, L. K., Morris, J. H. & Selkoe, D. J. (1988). Protein chemical and immunocytochemical studies of meningovascular amyloid protein in Alzheimer’s disease and normal ageing. Brain Res 474: 100–11. Joachim, C. L., Morris, J. H. & Selkoe, D. J. (1989). Diffuse amyloid plaques occur commonly in the cerebellum in Alzheimer’s disease. Am J Path 135: 309–19. Kamboh, M., Sanghera, D. Ferrell, R. & De Kosky, S. (1995). Apo E8 4-associated Alzheimer’s disease risk is modified by 1 antichymotrypsin polymorphism. Nat Genet 10: 486–8. Kang, J., Lemaire, H.-G., Unterbeck, A. et al. (1987). The precursor of Alzheimer’s disease amyloid A4 protein resembles a cell surface receptor. Nature 325: 733–6. Kida, E., Choi-Miura, N.-M. & Wisniewski, K. E. (1995). Deposition of apolipoproteins E and J in senile plaques is topographically determined in both Alzheimer’s disease and Down’s syndrome brain. Brain Res 685: 211–16. Kim, S. H., Vlkolinsky, R., Cairns, N. J. & Lubec, G. (2000). Decreased levels of complex III core protein I and complex V chain in brains from patients with Alzheimer’s disease and Down syndrome. Cell Molec Life Sci 57: 1810–16. Kobayashi, K., Emson, P. C., Mountjoy, C. Q., Thronton, S. N., Lawson, D. E. M. & Mann, D. M. A. (1990). Cerebral cortical calbindin D28k and parvalbumin neurones in Down’s syndrome. Neurosci Lett 113: 17–22. Kondo, J., Honda, T., Mori, H., Hamada, Y., Miura, R., Ogawara, M. & Ihara, Y. (1988). The carboxyl third of tau is tightly bound to paired helical filaments. Neuron 1: 827–34. Kosik, K. S., Joachim, C. L. & Selkoe, D. J. (1986). Microtubule associated protein tau is a major antigenic component of paired helical filaments in Alzheimer’s disease. Proc Nat Acad Sci USA, 83: 4044–8.


Kosik, K. S., Orecchio, L. D., Binder, L. I., Trojanowski, J. Q., Lee, V. M.-Y. & Lee, G. (1988). Epitopes that span the tau molecule are shared with PHF. Neuron 1: 817–25 Kuo, Y. M., Emmerling, M. R., Woods, A. S., Cotter, R. J. & Roher, A. E. (1997). Isolation, chemical characterization, and quantitation of A beta 3-pyroglutamyl peptide from neuritic plaques and vascular amyloid deposits. Biochem Biophys Res Commun 237: 188–91. Labudova, O., Krapfenbauer, K., Moenkmann, H., Rink, H., Kitzmueller, E., Cairns, N. & Lubec, G. (1998). Decreased transcription factor junD in brain of patients with Down syndrome. Neurosci Lett 252: 159–62. Lai, F. & Williams, R. A. (1989). Alzheimer’s disease in Down’s syndrome. Neurology 37: 332–9. Lambert, J.-C., Perez-Tur, J., Dupire, M.-J., Delacourte, A., Frigard, B. & Chartier-Harlin, M.-C. (1996). Analysis of APO E alleles impact in Down’s syndrome. Neurosci Lett 220: 57–60. Lemere, C. A., Blusztajn, J. K., Yamaguchi, H., Wisniewski, T., Saido, T. C. & Selkoe, D. J. (1996). Sequence of deposition of heterogenous amyloid beta peptides and APO E in Down syndrome: implications for initial events in amyloid plaque formation. Neurobiol Dis 3: 16–22. Lemere, C. A., Grenfell, J. & Selkoe, D. J. (1999). The AMY antigen co-occurs with A and follows its deposition in the amyloid plaques of Alzheimer’s disease and Down syndrome. Am J Pathol 155: 29–37. Lennox, G., Lowe, J. S., Morrell, K., Landon, M. & Mayer, R. J. (1988). Ubiquitin is a component of neurofibrillary tangles in a variety of neurodegenerative disorders. Neurosci Lett 94: 211–17. Leverenz, J. B. & Raskind, M. A. (1998). Early amyloid deposition in the medial temporal lobe of young Down syndrome patients: A regional quantitative analysis. Exp Neurol 150: 296–304. Li, Y. K., Wong, J. Z., Sheng, J. G., Jones, R. A., Mrak, R. E. & Griffin, W. S. T. (1998). S100 increases -amyloid precursor protein and its encoding mRNA in rat neuronal cultures. J Neurochem 71: 1421–8. Lim, G. P., Yang, F., Chu, T. et al. (2000). Ibuprofen suppresses plaque pathology and inflammation in a mouse model for Alzheimer’s disease. J Neurosci 20: 5709–14. Lippa, C. F., Fujiwara, H., Mann, D. M. A. et al. (1998). Lewy bodies contain altered alpha-synuclein in brains of many familial Alzheimer’s disease patients with mutations in presenilin and amyloid precursor protein genes. Am J Pathol 153: 1365–70. Lippa, C. F., Schmidt, M. L., Lee, V. M.-Y. & Trojanowski, J. Q. (1999). Antibodies to -synuclein detect Lewy bodies in many Down syndrome brains with Alzheimer’s disease. Ann Neurol 45: 353–7. Lott, I. T. & Lai, F. (1982). Dementia in Down’s syndrome; observations from a neurology clinic. Appl Res Ment Retard 3: 233–9. Lowe, J., Blanchard, A., Morrell, K. et al. (1988). Ubiquitin is a common factor in intermediate filament inclusion bodies of diverse type in man including those of Parkinson’s disease, Pick’s disease and Alzheimer’s disease, as well as Rosenthal fibres in cerebellar astrocytomas, cytoplasmic bodies in muscle and Mallory bodies in alcoholic liver disease. J Path 155: 9–15.

McGeer, E. G., Norman, M., Boyes, B., O’Kosky, J., Suzuki, J. & McGeer, P. L. (1985). Acetylcholine and aromatic amine systems in post mortem brain of an infant with Down’s syndrome. Exp Neurol 87: 557–60. McGeer, P. L., Schulzer, M. & McGeer, E. G. (1996). Arthritis and inflammatory agents as possible protective factors for Alzheimer’s disease: A review of 17 epidemiological studies. Neurology 47: 425. Malamud, N. (1972). Neuropathology of organic brain syndromes associated with ageing. In C. M. Gaitz (ed) Ageing and the Brain: Advances in Behavioural Biology Vol. 3, Plenum Press, New York, pp. 63–87. Mann, D. M. A. (1985). The neuropathology of Alzheimer’s disease: a review with pathogenetic, aetiological and therapeutic considerations. Mech Ageing Dev 31: 213–55. (1988a). Neuropathological and neurochemical aspects of Alzheimer’s disease. In L. L. Iversen, S. D. Iversen, S. Snyder (eds), Handbook of Psychopharmacology, vol 20. Plenum Press, New York, pp. 1–67. (1988b). Neuropathological association between Down’s syndrome and Alzheimer’s disease. Mech Ageing Dev 43: 99– 136. (1988c). Calcification of the basal ganglia in Down’s syndrome and Alzheimer’s disease. Acta Neuropathol 76: 595–8. (1991). The topographic distribution of brain atrophy in Alzheimer’s disease. Acta Neuropathol 83: 81–6. Mann, D. M. A. & Esiri, M. M. (1989). Regional acquisition of plaques and tangles in Down’s syndrome patients under 50 years of age. J Neurol Sci 89: 169–79. Mann, D. M. A. & Iwatsubo, T. (1996). Diffuse plaques in the cerebellum and corpus striatum in Down’s syndrome contain amyloid protein (A beta) only in the form of A beta 42(43). Neurodegeneration 5: 115–20. Mann, D. M. A. & Yates, P. O. (1986). Neurotransmitter deficits in Alzheimer’s disease and in other dementing disorders. Hum Neurobiol 5: 147–58. Mann, D. M. A., Yates, P. O. & Marcyniuk, B. (1984). Alzheimer’s presenile dementia, senile dementia of Alzheimer type and Down’s syndrome in middle age from an age-related continuum of pathological changes. Neuropath Appl Neurobiol 10: 185–207. (1985a). Some morphometric observations on the cerebral cortex and hippocampus in presenile Alzheimer’s disease, senile dementia of Alzheimer type and Down’s syndrome in middle age. J Neurol Sci 69: 139–59. Mann, D. M. A., Yates, P. O., Marcyniuk, B. & Ravindra, C. R. (1985b). Pathological evidence for neurotransmitter deficits in Down’s syndrome of middle age. J Ment Defic Res 29: 125–35. (1986). The topography of plaques and tangles in Down’s syndrome patients of different ages. Neuropath Appl Neurobiol 12: 447–57. (1987a). Loss of nerve cells from cortical and subcortical areas in Down’s syndrome patients at middle age: quantitative comparisons with younger Down’s patients and patients with Alzheimer’s disease. J Neurol Sci 80: 79–89.

221

222

D. M. A. Mann

(1987b). Dopaminergic neurotransmitter systems in Alzheimer’s disease and in Down’s syndrome at middle age. J Neurol Neurosurg Psychiatry 50: 341–4. Mann, D. M. A., Tucker, C. M. & Yates, P. O. (1988a). Alzheimer’s disease: an olfactory connection? Mech Ageing Dev 42: 1–15. Mann, D. M. A., Bonshek, R. E., Marcyniuk, B., Stoddard, R. W. & Torgerson, E. (1988b). Saccharides of senile plaques and neurofibrillary tangles in Alzheimer’s disease. Neurosci Lett 85: 277–82. Mann, D. M. A., Brown, A. M. T., Prinja, D. et al. (1989a). An analysis of the morphology of senile plaques in Down’s syndrome patients of different ages using immunocytochemical and lectin histochemical methods. Neuropath Appl Neurobiol 15: 317–29. Mann, D. M. A., Prinja, D., Davies CA et al. (1989b). Immunocytochemical profile of neurofibrillary tangles in Down’s syndrome patients of different ages. J Neurol Sci 92: 247–60. Mann, D. M. A., Jones, D., Prinja, D. & Purkiss, M. S. (1990a). The prevalence of amyloid (A4) protein deposits within the cerebral and cerebellar cortex in Alzheimer’s disease and Down’s syndrome. Acta Neuropathol 80: 318–27. Mann, D. M. A., Royston, M. C. & Ravindra, C. R. (1990b). Some morphometric observations on the brains of patients with Down’s syndrome: their relationship to age and dementia. J Neurol Sci 99: 153–64. Mann, D. M. A., Younis, N., Stoddard, R. W. & Jones, D. (1992a). The time course of pathological events concerned with plaque formation in Down’s syndrome with particular reference to the involvement of microglial cells. Neurodegeneration 1: 201–15. Mann, D. M. A., Purkiss, M. S., Bonshek, R. E., Jones, D., Brown, A. M. T. & Stoddard, R. W. (1992b). Lectin histochemistry of cerebral microvessels in ageing, Alzheimer’s disease and Down’s syndrome. Neurobiol Ageing, 13: 137–43. Mann, D. M. A., Pickering-Brown, S. M., Siddons, M. A. et al. (1995a). The extent of amyloid deposition in brain in patients with Down’s syndrome does not depend on the apolipoprotein E genotype. Neurosci Lett 196: 105–8. Mann, D. M. A., Iwatsubo, T., Fukumoto, H., Suzuki, N. & Odaka, A. (1995b). Microglial cells and amyloid ß protein (Aß) deposition; association with Aß40 containing plaques. Acta Neuropathol 90: 472–7. Mann, D. M. A., Iwatsubo, T., Ihara, Y. et al. (1996a). Predominant deposition of Amyloid-42(43) in plaques in cases of Alzheimer’s disease and hereditary cerebral hemorrhage associated with mutations in the Amyloid precursor protein gene. Am J Path 148: 1257–66. Mann, D. M. A., Iwatsubo, T., Cairns, N. J. et al. (1996b). Amyloid protein (A) deposition in chromosome 14-linked Alzheimer’s disease: predominance of A42(43) . Ann Neurol 40: 149–56. Mann, D. M. A., Iwatsubo, T., Pickering-Brown, S. M., Owen, F., Saido, T. C. & Perry, R. H. (1997a). Preferential deposition of amyloid protein (A) in the form A40 in Alzheimer’s disease is associated with a gene dosage effect of the apolipoprotein E E4 allele. Neurosci Lett 221: 81–4. Mann, D. M. A., Iwatsubo, T., Nochlin, D., Sumi, S. M., Levy-Lahad, E. & Bird, T. D. (1997b). Amyloid (Aß) protein deposition in

chromosome 1 linked Alzheimer’s disease – the Volga German kindreds. Ann Neurol 41: 52–7. Mann, D. M. A., Pickering-Brown, S. M., Takeuchi, A., Iwatsubo, T. and the familial Alzheimer’s disease study group (2001). Amyloid angiopathy and variability in amyloid (A) deposition is determined by mutation position in presenilin-1 linked Alzheimer’s disease. Am J Pathol 158: 1865–75. Marcyniuk, B., Mann, D. M. A. & Yates, P. O. (1988). Topography of nerve cell loss from the locus caeruleus in middle aged persons with Down’s syndrome. J Neurol Sci 83: 15–24. Marin-Padilla, M. (1976). Pyramidal cell abnormalities in the motor cortex of a child with Down’s syndrome. A Golgi study. J Comp Neurol 167: 63–82. Martins, R. N., Clarnette, R., Fisher, C., Broe, G. A., Brooks, W. S., Montgomery, P. & Gandy, S. E. (1995). Apo E genotypes in Australia: roles in early and late onset Alzheimer’s disease and Down’s syndrome. NeuroReport 6: 1513–16. Masters, C. L., Simms, G., Weinmann, N. A., Multhaup, G., McDonald, B. L. & Beyreuther, K. (1985). Amyloid plaque core protein in Alzheimer’s disease and Down’s syndrome. Proc Nat Acad Sci USA 82: 4245–9. Meda, L., Cassatella, M. A., Szendrei, G. I. et al. (1995). Activation of microglial cells by -amyloid protein and interferon-. Nature 374: 647–50. Melamed, E., Mildworf, B., Sharav, T., Benlenky, L. & Wertman, E. (1987). Regional cerebral blood flow in Down’s syndrome. Ann Neurol 22: 275–8. Miniszek, N. A. (1983). Development of Alzheimer’s disease in Down’s syndrome individuals. Am J Ment Defic 87: 377–85. Mito, T. & Becker & L. E. (1993). Developmental changes of S-100 protein and glial fibrillary acidic protein in the brain in Down syndrome. Exp Neurol 120: 170–6. Mori, H., Kondo, J. & Ihara, Y. (1987). Ubiquitin is a component of paired helical filament in Alzheimer’s disease. Science 235: 1641–4. Motte, J. & Williams, R. S. (1989). Age-related changes in the density and morphology of plaques and neurofibrillary tangles in Down’s syndrome brains. Acta Neuropathol 77: 535–46. Murdoch, J. C. & Adams, H. (1977). Reply to W. Hughes. Br Med J 2: 702. Murdoch, J. C., Rodger, J. C., Rao, S. S., Fletcher, C. D. & Donnigan, M. G. (1977). Down’s syndrome: an atheroma-free model. Br Med J 2: 226–8. Murphy, G. M., Eng, L. F., Ellis, W. G., Perry, G. Meissner, L. C. & Tinklenberg, J. R. (1990). Antigenic profile of plaques and neurofibrillary tangles in the amygdala in Down’s syndrome: a comparison with Alzheimer’s disease. Brian Res 537: 102–8. Murrell, J., Farlow, M., Ghetti, B. & Benson, M. D. (1991). A mutation in the amyloid precursor protein associated with hereditary Alzheimer’s disease. Science 254: 97–9. Naruse, S., Igarashi, S., Aoki, K. et al. (1991). Mis-sense mutation val→tile in exon 17 of amyloid precursor protein gene in Japanese familial Alzheimer’s disease. Lancet 337: 978–9. Neumann, N. A. (1967). Langdon Down syndrome and Alzheimer’s disease. J Neuropath Exp Neurol 26: 149–50.


Neve, R. L., Finch, C. E. & Dawes, L. R. (1988). Expression of the Alzheimer amyloid precursor gene transcripts in human brain. Neuron 1: 669–77. Nishimura, M., Yu, G. & St George-Hyslop, P. H. (1999). Biology of presenilins as causative molecules for Alzheimer disease. Clin Genet 55: 219–225. Ogomori, K., Kitamoto, T., Tateishi, J., Sato, Y., Suetsugu, M. & Abe, M. (1989). protein amyloid is widely distributed in the central nervous system of patients with Alzheimer’s disease. Am J Path 134: 243–51. O’Hara, P. T. (1972). Electronmicroscopical study of the brain in Down’s syndrome. Brain 95: 681–4. Olson, M. I. & Shaw, C. M. (1969). Presenile dementia and Alzheimer’s disease in mongolism. Brain 92: 147–56. Owens, D., Dawson, J. C. & Losin, S. (1971). Alzheimer’s disease in Down’s syndrome. Am J Ment Defic 75: 606–12. Oyama, F., Cairns, N. J., Shimada, H., Oyama, R., Titani, K. & Ihara, Y. (1994). Down’s syndrome: upregulation of beta-amyloid precursor protein and tau mRNAs and their defective coordination. J Neurochem 62: 1062–6. Perry, G., Friedman, R., Shaw, G. & Chau, V. (1987). Ubiquitin is detected in neurofibrillary tangles and senile plaque neurites of Alzheimer’s disease brains. Proc Nat Acad Sci, USA 84: 3033–6. Pierotti, A. R., Harmar, A. J., Simpson, J. & Yates, C. M. (1986). High molecular weight forms of somatostatin are reduced in Alzheimer’s disease and Down’s syndrome. Neurosci Lett 63: 141–6. Pogacar, S. & Rubio, A. (1982). Morphological features of Pick’s and atypical Alzheimer’s disease in Down’s syndrome. Acta Neuropathol 58: 249–54. Prasher, V. P., Chowdhury, T. A., Rowe, B. R. & Bain, S. C. (1997). Apo E genotype and Alzheimer’s disease in adults with Down syndrome: meta-analysis. Am J Ment Retard 102: 103–10. Prasher, V. P., Farrer, M. J., Kessling, A. M. et al. (1998). mapping of Alzheimer-type dementia in Down’s syndrome. Ann Neurol 43: 380–3. Price, D. L., Whitehouse, P. J., Struble, R. G. et al. (1982). Alzheimer’s disease and Down’s syndrome. Ann NY Acad Sci 396: 145–64. Rafalowska, J., Barcikowska, M., Wen, G. Y. & Wisniewski, H. M. (1988). Laminar distribution of neuritic plaques in normal ageing, Alzheimer’s disease and Down’s syndrome. Acta Neuropathol 77: 21–5. Raghavan, R., Khin-Nu, C. & Brown, A. (1993). Detection of Lewy bodies in trisomy 21. Can J Neurol Sci 20: 48–51. Rees, S. (1977). The incidence of ultrastructural abnormalities in the cortex of two retarded human brains (Down’s syndrome). Acta Neuropathol 37: 65–8. Reid, A. H. & Maloney, A. F. J. (1974). Giant cell arteritis and arteriolitis associated with amyloid angiopathy in an elderly mongol. Acta Neuropathol 27: 131–7. Reynolds, G. P. & Godridge, H. (1985). Alzheimer-like monoamine deficits in adults with Down’s syndrome. Lancet ii: 1368–9.

Reynolds, G. P. & Warner, C. E. J. (1988). Amino acid transmitter deficits in adult Down’s syndrome brain tissue. Neurosci Lett 94: 224–7. Risberg, J. (1980). Regional cerebral blood flow measurements by 133 Xe inhalation: methodology and application in neuropathology and psychiatry. Brain Lang 9: 9–34. Ropper, A. H. & Williams, R. S. (1980). Relationship between plaques and tangles and dementia in Down’s syndrome. Neurology 30: 739–44. Ross, M. H., Galaburda, A. M. & Kemper, T. L. (1984). Down’s syndrome: is there a decreased population of neurones? Neurology 34: 909–16. Rowe, I. F., Ridler, M. A. C. & Gibberd, F. B. (1989). Presenile dementia associated with mosaic trisomy 21 in a patient with a Down’s syndrome child. Lancet 2: 229. Royston, M. C., Mann, D., Pickering-Brown, S. et al. (1994). Apolipoprotein E ε2 allele promotes longevity and protects patients with Down’s syndrome from dementia. NeuroReport 5: 2583–5. Royston, M. C., McKenzie, J. E., Gentleman, S. M. et al. (1999). Overexpression of S100 in Down’s syndrome: correlation with patient age and -amyloid deposition. Neuropathol Appl Neurobiol 25: 387–93. Rubinsztein, D. C., Hon, J., Stevens, F. et al. (1999). APO E genotypes and risk of dementia in Down syndrome. Am J Med Genet 88: 344–7. Rumble, B., Retallack, R., Hilbich, C. et al. (1989). Amyloid (A4) protein and its precursor in Down’s syndrome and Alzheimer’s disease. New Engl J Med 320: 1446–52. Russo, C., Saido, T. C., DeBusk, L. M., Tabaton, M., Gambetti, P. & Teller, J. K. (1997). Heterogeneity of water-soluble amyloid beta-peptide in Alzheimer’s disease and Down’s syndrome brains. FEBS Lett 409: 411–16. Russo, C., Salis, S., Dolcini, V., Venezia, V., Song, X.-H., Teller, J. K. & Schettini, G. (2001). Identification of amino-terminally and phosphotyrosine-modified carboxy-terminal fragments of the amyloid precursor protein in Alzheimer’s disease and Down’s syndrome brain. Neurobiol Dis 8: 173–80. Saido, T. C., Iwatsubo, T., Mann, D. M., Shimada, H., Ihara, Y. & Kawashima, S. (1995). Dominant and differential deposition of distinct beta-amyloid peptide species, A beta N3 (pE), in senile plaques. Neuron 14: 457–66. Saido, T. C., Yamao-Harigaya, W., Iwatsubo, T. & Kawashima, S. (1996). Amino- and carboxyl-terminal heterogeneity of amyloid peptides deposited in human brain. Neurosci Lett 215: 173–6. Saunders, A. M., Schmader, K. & Breitner, J. (1993). Apolipoprotein E e4 allele distributions in late onset Alzheimer’s disease and in other amyloid forming diseases. Lancet 342: 710–11. Schapiro, M. B., Haxby, J. V., Grady, C. L. & Rapoport, S. I. (1986). Cerebral glucose utilization, quantitative tomography and cognitive function in adult Down’s syndrome. In C. J. Epstein (ed), Neurobiology of Down’s Syndrome, Raven Press, New York. pp. 89–108.

223

224

D. M. A. Mann

Schapiro, M. B., Haxby, J. V., Grady, C. L. et al. (1987). Decline in cerebral glucose utilization and cognitive function with ageing in Down’s syndrome. J Neurol Neurosurg Psychiatry 50: 766–74. Schapiro, M. B., Berman, K. F., Friedland, R. P., Weinberger, D. R. & Rapoport, S. I. (1988). Regional Blood flow is not reduced in young adult Down’s syndrome. Ann Neurol 24: 310. Schapiro, M. B., Kumar, A., White, B., Grady, C. L., Friedland, R. P. & Rapoport, S. I. (1989a). Alzheimer’s disease (AD) in mosaic/translocation Down’s syndrome (DS) without mental retardation. Neurology 39 Suppl. 1: 169. Schapiro, M. B., Luxemberg, J. S., Kaye, J. A., Haxby, J. V., Friedland, R. P. & Rapoport, S. I. (1989b). Serial quantitative CT analysis of brain morphometrics in adult Down’s syndrome at different ages. Neurology 39: 1349–53. Schapiro, M. B., Grady, C. L., Kumar, A. et al. (1990). Regional cerebral glucose metabolism is normal in young adults with Down’s syndrome. J Cereb Blood Flow Metab 10: 199–206. Scheuner, D., Eckman, C., Jensen, M. et al. (1996). Secreted amyloid -protein similar to that in the senile plaques of Alzheimer’s disease is increased in vivo by the presenilin 1 and 2 and APP mutations linked to familial Alzheimer’s disease. Nature Med 2: 864–9. Schmidt, M. L., Lee, V. M.-Y., Forman, M., Chiu, T.-S. & Trojanowski, J. Q. (1997). Monoclonal antibodies to a 100-kd protein reveal abundant A-negative plaques throughout gray matter of Alzheimer’s disease brains. Am J Pathol 151: 69–80. Schochet, S. S., Lampert, P. W. & McCormick, W. F. (1973). Neurofibrillary tangles in patients with Down’s syndrome: a light and electron microscope study. Acta Neuropathol 23: 342–6. Schupf, N., Kapell, D., Lee, J. H., Ottman, R. & Mayeux, R. (1994). Increased risk of Alzheimer’s disease in mothers of adults with Down’s syndrome. Lancet 344: 353–6. Schupf, N., Kapell, D. & Lee, J. H. (1996). Onset of dementia is associated with apolipoprotein E e4 in Down’s syndrome. Ann Neurol 40: 799–801. Shapira, R., Austin, G. E. & Mirra, S. S. (1988). Neuritic plaque amyloid in Alzheimer’s disease is highly racemized. J Neurochem 50: 69–74. Sheng, J. G., Mrak, R. E. & Griffin, W. S. T. (1997). Neuritic plaque evolution in Alzheimer’s disease is accompanied by transition of activated microglia from primed to enlarged to phagocytic forms. Acta Neuropathol 94: 1–5. Silverstein, A. B., Herbs, D. & Miller, T. J. (1988). Effects of age on the adaptive behaviour of institutionalized and noninstitutionalized individuals with Down’s syndrome. Am J Ment Retard 92: 455–60. Simpson, M. D., Slater, P., Cross, A. J. et al. (1989). Reduced D-[3 H] aspartate binding in Down’s syndrome brains. Brain Res 484: 273–8. Sinet, P. M., Michelson, A. M., Bazin, A., Lejeune, J. & Jerome, H. (1975). Increase in glutathione peroxidase activity in erythrocytes from trisomy 21 subjects. Biochem Biophys Res Commun 67: 910–5.

Sisodia, S. S., Koo, E. H., Beyreuther, K., Unterbeck, A. & Price, D. L. (1990). Evidence that amyloid protein in Alzheimer’s disease is not derived by normal processing. Science 248: 492–5. Snow, A. D., Mar, H., Nochlin, D. et al. (1990). Early accumulation of heparan sulphate in neurones and in the beta-amyloid protein containing lesions of Alzheimer’s disease and Down’s syndrome. Am J Path 137: 1253–70. Solitaire, G. B. & Lamarche, J. B. (1966). Alzheimer’s disease and senile dementia as seen in mongoloids: neuropathological observations. Am J Ment Defic 70: 840–8. (1967). Brain weight in the adult mongol. J Ment Defic Res 11: 79–84. Spargo, E., Luthert, P. J., Anderton, B. H., Bruce, M., Smith, D. & Lantos, P. L. (1990). Antibodies raised against different portions of A4 protein identify a subset of plaques in Down’s syndrome. Neurosci Lett 115: 345–50. Sparkman, D. R., Hill, S. J. & White, C. L. (1990). Paired helical filaments are not major binding sites for WGA and DBA agglutinins in neurofibrillary tangles of Alzheimer’s disease. Acta Neuropathol 79: 640–6. Stoltzner, S. E., Grenfell, T. J., Mori, C. et al. (2000). Temporal accrual of complement proteins in amyloid plaques in Down’s syndrome with Alzheimer’s disease. Am J Pathol 152: 489–99. Struwe, F. (1929). Histopathologische Untersuchungen uber Enstehung und Wesen der senilen plaques. Zeitsch Neurol Psychol 122: 291–307. Suenaga, T., Hirano, A., Llena, J. F., Yen, S.-H. & Dickson, D. W. (1990). Modified Bielschowsky staining and immunohistochemical studies on striatal plaques in Alzheimer’s disease. Acta Neuropathol 80: 280–6. Suetsuga, M. & Mehraein, P. (1980). Spine distribution along the apical dendrite of the pyramidal neurons in Down’s syndrome. A quantitative Golgi study. Acta Neuropathol 50: 207–10. Sylvester, P. E. (1983). The hippocampus in Down’s syndrome. J Ment Defic Res 27: 227–36. Szumanska, G., Vorbrodt, A. W., Mandybur, T. I. & Wisniewski, H. M. (1987). Lectin histochemistry of plaques and tangles in Alzheimer’s disease. Acta Neuropathol 73: 1–11. Tagliavini, F., Giaccone, G., Frangione, B. & Bugiani, O. (1988). Preamyloid deposits in the cerebral cortex of patients with Alzheimer’s disease and non-demented individuals. Neurosci Lett 93: 191–6. Takashima, S. & Becker, L. E. (1985). Basal ganglia calcification in Down’s syndrome. J Neurol, Neurosurg Psychiatry 48: 61–4. Takashima, S., Becker, L. E., Armstrong, D. L. & Chan, F. W. (1981). Abnormal neuronal development in the visual cortex of the human fetus and infant with Down’s syndrome. Brain Res 225: 1–21. Takashima, S., Iida, K., Mito, T. & Arima, M. (1994). Dendritic and histochemical development and ageing in patients with Down’s syndrome. J Intellect Disabil Res 38: 265–73. Talbot, C., Lendon, C., Craddock, N., Shears, S., Morris, J. & Goate, A. (1994). Protection against Alzheimer’s disease with APO E e2. Lancet 343: 1432–3.


Teller, J. K., Russo, C., DeBusk, L. M. et al. (1996). Presence of soluble amyloid beta-peptide precedes amyloid plaque formation in Down’s syndrome. Nature Med 2: 93–5. Thase, M. E., Tigner, R., Smeltzer, D. & Liss, L. (1984). Age-related neuropsychological deficits in Down’s syndrome. Biol Psychiatry 19: 571–85. Tyrrell, J., Cosgrave, M., Hawi, Z. et al. (1998). A protective effect of Apolipoprotein E e2 allele on dementia in Down’s syndrome. Biol Psychiatry 43: 397–400. Tyrrell, J., Cosgrave, M., McPherson, J. et al. (1999). Presenilin 1 and -1-antichymotrypsin polymorphisms in Down syndrome: No effect on the presence of dementia. Am J Med Genet 88: 616–20. Van Duijn, C. M., Clayton, D., Chandra, V. et al. (1991). Familial aggregation of Alzheimer’s disease and related disorders: A collaborative re-analysis of case-control studies – EURODEM Risk Factors Research Group. Int J Epidemiol 20 Suppl 2: S13–20. Van Gool, W. A., Evenhuis, H. M. & Van Duijn, C. M. (1995). A casecontrol study of apolipoprotein E genotypes in Alzheimer’s disease associated with Down’s syndrome. Ann Neurol 38: 225–30. Wegiel, J. & Wisniewski, H. M. (1990). The complex of microglial cells and amyloid star in 3-dimensional reconstruction. Acta Neuropath 81: 116–24. Wegiel, J., Wisniewski, H. M., Dziewiatkowski, J., Popovitch, E. R. & Tarnawski, M. (1996). Differential susceptibility to neurofibrillary pathology among patients with Down syndrome. Dementia 7: 135–41. Whalley, L. J. (1982). The dementia of Down’s syndrome and its relevance to aetiological studies of Alzheimer’s disease. Ann NY Acad Sci 396: 39–53. Whitaker-Azmitia, P. M., Wingate, M., Borella, A., Gerlai, R., Roder, J. & Azmitia, E. C. (1997). Transgenic mice overexpressing the neurotrophic factor S-100 show neuronal cytoskeletal and behavioural signs of altered ageing processes: implications for Alzheimer’s disease and Down’s syndrome. Brain Res 776: 51–60. Wischik, C., Novak, M., Thagersen, H. C. et al. (1988). Isolation of a fragment of tau derived from the core of the paired helical filament of Alzheimer’s disease. Proc Natl Acad Sci, USA, 85: 4506–10. Wisniewski, T. & Frangione, B. (1992). Apolipoprotein E: a pathological chaperone in patients with cerebral and systemic amyloid. Neurosci Lett 135: 235–8. Wisniewski, H. M. & Rabe, A. (1986). Discrepancy between Alzheimer-type neuropathology and dementia in persons with Down’s syndrome. Ann NY Acad Sci 477: 247–60. Wisniewski, K. E. & Schmidt-Sidor, B. (1986). Myelination in Down’s syndrome brains (pre- and post-natal maturation) and some clinical-pathological correlations. Ann Neurol 20: 429–30. Wisniewski, H. M. & Terry, R. D. (1973). Re-examination of the pathogenesis of the senile plaque. In H. M. Zimmerman (ed), Progress in Neuropathology vol 2. Grune and Stratton, New York, pp. 1–26.

Wisniewski, H. M., Rabe, A. & Wisniewski, K. E. (1987). Neuropathology and dementia in people with Down’s syndrome. Banbury Report 27: Molecular Neuropathology of Ageing, pp. 399–413. Wisniewski, H. M., Bancher, C., Barcikowska, M., Wen, G. Y. & Currie, J. (1989). Spectrum of morphological appearance of amyloid deposits in Alzheimer’s disease. Acta Neuropathol 78: 337–47. Wisniewski, K. E., Howe, J., Gwyn-Williams, D. & Wisniewski, H. M. (1978). Precocious ageing and dementia in patients with Down’s syndrome. Biol Psychiatry 13: 619–27. Wisniewski, K. E., Jervis, G. A., Moretz, R. C. & Wisniewski, H. M. (1979). Alzheimer neurofibrillary tangles in diseases other than senile and presenile dementia. Ann Neurol 7: 462–5. Wisniewski, K. E., Frenchy, J. H., Rosen, J. F., Kozlowski, P. B., Tenner, M. & Wisniewski, H. M. (1982). Basal ganglia calcification (BGC) in Down’s syndrome (DS) – another manifestation of premature ageing. Ann NY Acad Sci 396: 179–89. Wisniewski, K. E., Laure-Kamionowska, M. & Wisniewski, H. M. (1984). Evidence of arrest of neurogenesis and synaptogenesis in brains of patients with Down’s syndrome. New Engl J Med 311: 1187–8. Wisniewski, K. E., Wisniewski, H. M. & Wen, G. Y. (1985a). Occurrence of neuropathological changes and dementia of Alzheimer’s disease in Down’s syndrome. Ann Neurol 17: 278–82. Wisniewski, K. E., Dalton, A. J., Crapper-McLachlan, D. R., Wen, G. Y. & Wisniewski, H. M. (1985b). Alzheimer’s disease in Down’s syndrome. Clinicopathologic studies. Neurology 35: 957–61. Wisniewski, K. E., Laure-Kamionowska, M., Connell, F. & Wen, G. Y. (1986). Neuronal density and synaptogenesis in the postnatal stage of brain maturation in Down’s syndrome. In C. J. Epstein (ed), The Neurobiology of Down’s syndrome, Raven Press, New York, pp. 29–44. Wisniewski, T., Morelli, L., Wegiel, J., Levy, E., Wisniewski, H. M. & Frangione, B. (1995). The influence of apolipoprotein E isotypes on Alzheimer’s disease pathology in 40 cases of Down’s syndrome. Ann Neurol 37: 136–8. Wood, J. G, Mirra, S. S., Pollock, N. J. & Binder, L. I. (1986). Neurofibrillary tangles of Alzheimer’s disease share antigenic determinants with the axonal microtubule associated protein tau. Proc Nat Acad Sci, USA 83: 4040–3. Wragg, M., Hutton, M., Talbot, C. and the Alzheimer’s disease collaborative group (1996). Genetic association between intronic polymorphism in the presenilin-1 gene and late onset Alzheimer’s disease. Lancet 347: 509–12. Yamaguchi, H., Hirai, S., Morimatsu, M., Shoji, M. & Harigaya, Y. (1988). Diffuse type of senile plaque in the brains of Alzheimer type dementia. Acta Neuropath 76: 541–9. Yamaguchi, H., Hirai, S., Morimatsu, M., Shoji, M. & Nakazato, Y. (1989). Diffuse type of senile plaque in the cerebellum of Alzheimer type dementia as detected by -protein immunostaining. Acta Neuropath 77: 314–19. Yates, C. M., Simpson, J., Maloney, A. F. J., Gordon, A. & Reid, A. H. (1980). Alzheimer-like cholinergic deficiency in Down’s syndrome. Lancet ii: 979.

225

226

D. M. A. Mann

Yates, C. M., Ritchie, I. M., Simpson, J., Maloney, A. F. J. & Gordon, A. (1981). Noradrenaline in Alzheimer type dementia and Down’s syndrome. Lancet ii: 39–40. Yates, C. M., Simpson, A., Gordon, A. et al. (1983). Catecholamines and cholinergic enzymes in presenile and senile Alzheimertype dementia and Down’s syndrome. Brain Res 280: 119–26. Yates, C. M., Simpson, J. & Gordon, A. (1986). Regional brain 5hydroxytryptamine levels are reduced in senile Down’s syndrome as in Alzheimer’s disease. Neurosci Lett 65: 189–22.

Yatham, L. N., McHale, P. A. & Kinsella, A. (1988). Down’s syndrome and its association with Alzheimer’s disease. Acta Psychiatr Scand 77: 38–41. Zigman, W. B., Schupf, N., Lubin, R. A. & Silverman, W. P. (1987). Premature regression of adults with Down’s syndrome. Am J Ment Defic 92: 161–8. Zigman, W. B., Schupf, N., Sersen, E. & Silverman, W. P. (1996). Prevalence of dementia in adults with and without Down’s syndrome. Am J Ment Retard 100: 403–12.

11 Sporadic tauopathies: Pick’s disease, corticobasal degeneration, progressive supranuclear palsy and argyrophilic grain disease Dennis W. Dickson Department of Pathology, Mayo Clinic Jacksonville, FL, USA

INTRODUCTION Argyrophilic cytoplasmic neuronal and glial inclusions are characteristic of a number of sporadic neurodegenerative disorders, including Pick’s disease (PiD), corticobasal degeneration (CBD), progressive supranuclear palsy (PSP) and argyrophilic grain disease (ADG), that collectively are referred to as tauopathies (Lee et al., 2001). At the electron microscopic level inclusions in the tauopathies are non-membrane-bound filamentous aggregates composed of the microtubule-associated protein tau. The tau in the inclusions is abnormal in a number of respects, not the least of which is its fibrillar state. Tau is normally a soluble protein that is unstructured and resistant to heat and strong acids. It binds to microtubules and promotes polymerization and stabilization of microtubules. The functional attributes of tau are regulated by phosphorylation, with phosphorylated forms of tau having decreased ability to promote microtubule stability (Spillantini & Goedert, 1998; Delacourte & Buee, 1997). Fibrillar tau is abnormally phosphorylated with far more phosphate groups than normal tau. There are approximately 20 phosphorylation sites on tau and multiple kinases that are capable of phosphorylating tau. The degree of phosphorylation of the various sites contributes to great microheterogeneity in tau within the brain (Spillantini & Goedert, 1998). In human adult brain, alternative mRNA splicing of a single gene on chromosome 17 yields six major tau isoforms (Goedert et al., 1989a, 1995; Delacourte & Buee, 1997). Exons 2, 3 and 10 are alternatively spliced. Alternative splicing of exon 10 gives rise to tau isoforms with three (exon 10−) or four (exon 10+) conserved 30–32 amino acid repeats in the microtubule-binding domain, so-called three repeat

tau (3R tau) and four repeat tau (4R tau). The 3R and 4R tau isoforms have different properties in terms of microtubule binding, while alternative splicing of exons 2 and 3 at the amino-terminus (producing inserts of 29 or 58 amino acids) has unclear functional significance. Recent studies have shown that tau protein that accumulates within neurons and glia in tauopathies does so selectively (Delacourte & Buee 1997) (see Table 11.1). For example, both 3R and 4R tau isoforms are present in Alzheimer’s disease (AD) (Sergent et al., 1997), while 3R tau predominates in PiD (Delacourte et al., 1996) and 4R tau predominates in CBD, PSP and AGD (Buee Scherrer et al., 1996; Arai et al., 2001; Togo et al., 2002b). The mechanism of incorporation of selective isoforms in the various tauopathies is unknown. Evidence for increased expression of 3R or 4R tau is meagre (Chambers et al., 1999). The role of specific tau isoforms within variable neuronal populations has yet to be systematically explored. Nevertheless, the characteristic biochemical profile observed in the various disorders is the basis of a recent classification scheme for the non-Alzheimer degenerative dementias (Trojanowski & Dickson, 2001).

PICK’S DISEASE Pick’s disease (PiD) is a rare neurodegenerative disorder, estimated to be ten times less frequent than Alzheimer’s disease (Constantinidis, 1985). In the past PiD was a term used in a broad sense to describe the clinical presentations characterized by focal cortical degeneration regardless of the underlying histopathology, but in recent years with more refined molecular characterization of the disorder, it has come to have a much narrower definition (Dickson, 1998,


227

228

D. W. Dickson

Table 11.1. Biochemical and ultrastructural findings in sporadic tauopathies Feature

Pick’s disease

Corticobasal degeneration

Progressive supranuclear palsy

Argyrophilic grain disease

Insoluble tau EM of filaments

64 and 55 kDa 3R tau 14–16 nm straight filaments; 22–24 nm twisted ribbons

68 and 64 kDa; 4R tau 22–24 nm twisted ribbons

68 and 64 kDa; 4R tau 15–18 nm straight filaments

68 and 64 kDa; 4R tau 9–18 nm straight filaments

2001). In this discussion PiD is used to refer to a disorder with lobar atrophy, argyrophilic neuronal inclusions (Pick bodies), ballooned neurons and predominantly 3R tau in the pathologic lesions. In the past PiD included at least three distinct pathologic variants that Constantinidis (1985) termed Types A, B and C. Type A corresponds to the current use of the term PiD, while Type B corresponds to CBD and Type C to frontotemporal degeneration lacking distinctive histopathology (Table 11.2) (Chapter 12). The concept of an overarching ‘Pick complex’ has more recently been championed by Kertesz (Kertesz et al., 1994; Kertesz & Munoz, 2000), but is in some ways a throw back to the days when disparate entities were subsumed under a common name.

Clinical features PiD has a wide age of onset (40–80 years), but often begins before the age of 65 and is often considered a ‘presenile dementia’. The disease duration varies from 2 to 20 years. The sex incidence is nearly equal (Heston et al., 1987). While the clinical features of typical cases are sufficiently distinctive that a diagnosis can often be made with some degree of certainty (Mendez et al., 1993; Litvan et al., 1997),

there is overlap with other frontotemporal degenerations and with AD, particularly frontal variants of AD (Johnson et al., 1999). Virtually all cases have dementia and personality deterioration as early and prominent features. Depending upon the location of the brunt of the focal cortical atrophy, other clinical features may be present (Neary et al., 1993). Patients with marked medial temporal lobe pathology, especially affecting the amygdala, may have a ¨ Kluver–Bucy syndrome (Cummings & Duchen, 1981). Patients with more severe frontal lobe pathology usually have a frontal lobe syndrome (Neary et al., 1988). Depending upon whether the frontal atrophy predominates in orbital or convexity cortices, the personality deterioration may be characterized by disinhibition and antisocial behaviours or by apathy and moria (Constantinidis, 1985). Given the relative preservation of posterior cortical areas, patients may have preserved visuospatial skills. Language disorder leading to frank mutism is common, and some patients present with a progressive aphasia (Graff-Radford et al., 1990). Usually this is associated with dominant hemisphere, periSylvian atrophy. The parietal lobe is usually not affected severely, but cases with parietal involvement have been reported, and some of these cases have had a clinical syndrome of progressive apraxia suggestive of CBD (Lang et al., 1994).

Table 11.2. Historical Pick disease subtypes and modern terminology Historical term

Clinical

Distribution

Histopathology

Current term

Type A

Personality changes, amnestic syndrome, dementia Focal cortical signs, extrapyramidal signs, dementia

Frontal and temporal poles and limbic structures Superior frontal and parietal cortices; basal ganglia; substantia nigra

Pick bodies and ballooned neurons

PiD

Ballooned neurons

CBD

Neither Pick bodies nor ballooned neurons

FTLD (DLDH)

Type B

Type C

Dementia, personality changes

Sporadic tauopathies

Neuroimaging Structural findings with computed tomographic (CT) and magnetic resonance imaging (MRI) scans show circumscribed cerebral atrophy that may be asymmetric (Wechsler et al., 1982). There is often enlargement of the frontal or temporal horns and relatively normal occipital horns of the lateral ventricles. Cortical atrophy can be detected early in the clinical course. In some cases the caudate nucleus has atrophy. Functional imaging studies such as positron emission tomography (PET) or single photon emission tomography (SPECT) usually show bifrontal hypoperfusion deficits (Kamo et al., 1987; Salmon & Franck 1989; Neary et al., 1987).

Genetics Almost all cases of PiD are sporadic, but familial cases have been described (Heston et al., 1987; Morris et al., 2001). Some of these families have been shown subsequently to have mutations in the tau gene (Rosso et al., 2002; Pickering-Brown et al., 2000; Murrell et al., 1999) and are thus examples of frontotemporal dementia and Parkinsonism linked to chromosome 17 (FTDP-17), which is discussed in detail in Chapter 12. In some of these cases the pathological phenotype is close to that found in sporadic PiD, but there are usually histological features that differ, such as more prominent glial pathology or more pathology in subcortical regions than in typical PiD. Whether PiD is associated with increased frequency of apolipoprotein-E ε4 allele is unclear, since most studies have either been based upon clinical diagnoses, which clearly can include a variety of pathologic entities, or they have been based upon small autopsy series. The presence of the ε4 allele may influence the age of onset of PiD (Farrer et al., 1995; Minthon et al., 1997). There appears to be no association of PiD with the extended tau haplotype (Russ et al., 2001).

Pathological features Gross findings Brain atrophy is often marked in PiD, with total brain weight of as little as 750 grams. The atrophy is circumscribed, affecting anterior temporal and frontal lobes, the orbital frontal lobe and the medial temporal lobes; the posterior part of the superior temporal gyrus and the pre- and postcentral gyri and occipital lobes are usually spared (Fig. 11.1). Cortical atrophy may be asymmetrical. On sectioning, the cortical ribbon is thinner than usual and the grey–white junction is indistinct. The subcortical white matter may be retracted, grey and soft. The ventricles, especially the

Fig. 11.1. (a) Pick’s disease with sharply circumscribed atrophy of frontal lobe and in this case inferior parietal lobe; (b) midline view shows severe atrophy of the medial and superior temporal lobe; (c) coronal section at the level of the fornix shows severe atrophy of the amygdala and marked dilation of lateral ventricle; (d ) coronal section at the level of the anterior hippocampus shows severe medial temporal lobe atrophy with sparing of the superior temporal gyrus.

frontal and temporal horns of the lateral ventricles, are dilated. Degeneration in the corpus striatum, globus pallidus and substantia nigra may sometimes be detected. When there is basal ganglia degeneration, the striatum is more affected than lentiform nucleus. Despite presence of basal ganglia pathology, extrapyramidal features are described rarely in PiD (Kosaka et al., 1991). Microscopic findings In areas with severe pathology, the cortex has almost complete loss of large pyramidal neurons with status spongiosis and astrocytic gliosis. The cytoarchitectural features of the cortex become obscured. Neurons in the upper cortical layers contain round argyrophilic inclusions, Pick bodies, while ballooned neurons, also referred to as ‘swollen chromatolytic neurons’ (Clark et al., 1986) or ‘Pick cells’, are present in the middle and lower cortical layers (Fig. 11.2). Pick bodies are the most characteristic histopathological hallmark of PiD and have been shown to have high positive predictive value in blinded clinicopathological studies of various tauopathies (Litvan et al., 1996).

229

230

D. W. Dickson

Fig. 11.2. (a) Neocortex of PiD shows atrophy, gliosis and loss of normal architectural details and scattered swollen neurons in lower layers; (b) a microglial stain (HLA-DR) shows diffuse microgliosis in the neocortex in PiD.

Pick bodies are round, well-circumscribed argyrophilic cytoplasmic inclusions that are positive with Bodian and Bielschowsky silver stains, but usually negative with the Gallyas stain (Probst et al., 1996), which is a useful differential diagnostic feature (Fig. 11.3). Early immunohistochemical studies suggested that they were composed

of neurofilament protein, since they could be detected with antibodies that recognized phosphorylated neurofilament epitopes (Ulrich et al., 1987), but unlike phosphorylated neurofilament epitopes in axons, the phosphate epitopes in Pick bodies were resistant to phosphatases (Probst et al., 1983). Pick bodies were subsequently shown

Fig. 11.3. (a) The dentate fascia of hippocampus shows slight neuronal depopulation and subtle cytoplasmic inclusions (arrows); (b) Bielschowsky stain shows argyrophilic round Pick body inclusions in neurons of the dentate fascia (arrows); (c) Gallyas stain fails to stain Pick bodies (Arrows); (d ) tau immunostain shows intense staining of Pick bodies.


Fig. 11.4. (a) Three ballooned neurons with swollen eosinophilic cytoplasm within which are several granulovacuolar bodies; (b) ballooned neurons show intense immunoreactivity with phosphorylated neurofilament antibodies.

to have immunoreactivity with tau antibodies (Love et al., 1988; Probst et al., 1996) and neurofilament immunoreactivity in Pick bodies has been considered to probably be a cross-reactivity between neurofilament and tau. Extracellular Pick bodies lose tau immunoreactivity (Izumiyama et al., 1994). Ubiquitin is also present in Pick bodies (Love et al., 1988; Lowe et al., 1988), but ubiquitin immunostaining is not as robust or consistent as that with tau antibodies. Chromogranin A staining of Pick bodies has been described (Yasuhara et al., 1994), but the reactivity is not specific to Pick bodies. Pick bodies are negative for -synuclein, which differentiates Pick bodies from cortical Lewy bodies, but neurons that contain Pick bodies may have aberrant expression of synuclein in their cytoplasm. Markers of oxidative stress, including heme oxygenase (Castellani et al., 1995), advanced glycation end-products (Kimura et al., 1996) and 4-hydroxy-2-nonenal (Montine et al., 1997) have also been described in Pick bodies providing clues to the pathogenesis of this disorder. Ballooned neurons have weak argyrophilia with Bodian and modified Bielschowsky stains. They are best demonstrated with antibodies that recognize phosphorylated neurofilament epitopes (Dickson et al., 1986) (Fig. 11.4) or the low molecular weight stress protein, B-crystallin (Kato et al., 1992; Lowe et al., 1992). Unlike Pick bodies, the phosphorylated neurofilament epitopes in ballooned neurons are sensitive to enzymatic digestion by phosphatases, similar to neurofilament epitopes in axons (Dickson et al., 1986). Focal tau immunoreactivity is sometimes also detected in ballooned neurons, usually at the cell margins (Feany & Dickson, 1995). Pick bodies are most numerous in cingulate, insular, inferior parietal, inferior temporal, fusiform and lingular gyri (Yoshimura, 1989). They are less numerous in anterior

frontal and temporal lobes and occipital gyri. In some cases a few neurofibrillary tangles (NFT) may accompany these lesions, but the distribution of Pick bodies is different from the distribution of NFT (Hof et al., 1994). Neuronal loss in PiD affects large neurons more than small neurons, and neuronal loss is greater in frontal and temporal lobes than in the parietal lobe (Hansen et al., 1988). The pattern of neuronal loss in the temporal lobe in PiD is clearly distinct from that in AD, with greater loss of neurons in all cortical layers compared to the laminar neuronal loss in AD (Arnold et al., 1994). The hippocampus and amygdala are usually severely affected, with many Pick bodies in pyramidal neurons of Ammon’s horn and granular neurons of the dentate fascia (Fig. 11.3). The topographic distribution of Pick bodies in the hippocampus parallels that of NFT, granulovacuolar bodies and Hirano bodies in AD, except that the dentate fascia, which is relatively resistant to NFT, is particularly vulnerable to Pick bodies (Ball, 1979). In addition to Pick bodies, pyramidal neurons in Ammon’s horn are vulnerable to granulovacuolar degeneration, and Hirano bodies are also usually numerous. In the regions of the cerebrum where lobar atrophy is greatest, the subjacent white matter is attenuated and invariably displays loss of myelinated fibers. Accompanying the fibre loss is astrocytic gliosis and axonal degeneration. Although astrocytic gliosis is marked and microglia show signs of activation, there are relatively few lipid phagocytes. Glial tau-positive inclusions can be detected with immunocytochemical methods (Dickson et al., 1996; Feany et al., 1996). The basal nucleus of Meynert in PiD is relatively preserved (Mizukami & Kosaka, 1989; Tagliavini & Pilleri, 1983). When neuronal loss is present in the basal nucleus (Uhl

231

232

D. W. Dickson

Fig. 11.5. Subcortical Pick bodies: (a) basal ganglia has Pick bodies in the inner part of the putamen; (b) pigmented neurons in the substantia nigra have Pick bodies; (c) Pick bodies in the locus ceruleus are pleomorphic and multiple; (d ) Pick bodies are found in the neurons of the pontine base.

et al., 1983), the cell loss is usually neither extensive nor uniform in distribution. The latter has been attributed to a somatotopic retrograde degeneration secondary to focal cortical degeneration. Pick bodies can be found in the corpus striatum and the globus pallidus; when present, they are usually most numerous in the inner part of the putamen (Fig. 11.5). A few Pick bodies may also be detected in the hypothalamus and the periaqueductal gray matter. The substantia nigra is affected in many cases of PiD, with more severe neuronal loss in the rostral than caudal substantia nigra and more in the medial than lateral cell groups, again possibly reflecting a somatotopic retrograde degeneration since these areas provide dopaminergic innervation of the limbic system and the frontal lobes (Uchihara et al., 1990). Other cases show only minimal focal neuronal loss in the substantia nigra. A few Pick bodies are not uncommonly detected in the substantia nigra (Fig. 11.5). Neuronal loss has also been documented in the locus ceruleus (Arima & Akashi, 1990), and Pick bodies are often detected in the locus ceruleus (Forno et al., 1989) and less often in the neurons of the pontine base. Pick bodies in the locus ceruleus may be multiple or multilobulated, an appearance not found in other susceptible neuronal populations (Fig. 11.5). Other brainstem nuclei in which Pick bodies can be detected include the dorsal raphe, pontine base (Fig. 11.5) and the medullary reticular formation.

Immunocytochemistry for glial fibrillary acidic protein and S-100 protein demonstrates marked astrocytic gliosis in the cortex and white matter. The gliosis is often most marked in the upper cortex and at the grey–white matter junction. Tau-positive inclusions in glial cells are increasingly recognized in the tauopathies. While tau-positive inclusions can be detected in astrocytes and oligodendroglia in PiD, their numbers are variable, but they are usually more numerous in PSP, CBD and AGD (Feany et al., 1996). Microglia are increased in PiD and show morphological and phenotypic features of activation (Paulus et al., 1993). They can be readily detected with antibodies to class II major histocompatibility complex (MHC) antigens (HLADR) (Fig. 11.2). More recently class II immunoreactivity has been detected on neurons in PiD raising the possibility of immune-mediated injury to neurons (Hollister et al., 1997). Various complement components have been described in PiD using immunocytochemical methods (Singhrao et al., 1996). Complement activation may be one of the factors that contribute to widespread microglial activation in PiD.

Ultrastructural findings Pick bodies are composed of randomly arranged, straight filaments ranging from 14 to 16 nm in diameter and admixed with a variable number of twisted filaments with


Fig. 11.6. EM of Pick body. Straight filaments are randomly dispersed in the cytoplasm that is rich in vesicular structures.

a diameter ranging from 22 to 24 nm and a half period of 120–160 nm (Rewcastle & Ball, 1968; Wisniewski et al., 1972; Takauchi et al., 1984; Kato & Nakamura, 1990; Murayama et al., 1990) (Fig. 11.6). This contrasts with the twisted paired helical filaments of AD, which are 22 nm wide and have a half period of 60–80 nm. Ballooned neurons are composed of granulofilamentous material. Although occasionally filaments similar to those in Pick bodies are admixed (Murayama et al., 1990), ballooned neurons most often contain intermediate filaments (10 nm diameter) with fuzzy side arms, consistent with neurofilaments (Clark et al., 1986).

Biochemical findings Immunochemical studies have shown increased amounts of abnormal insoluble tau in PiD compared to normal brains, but many fold less than in AD (Wolozin & Davies,

1987). This correlates well with image analysis studies that show much fewer tau protein positive profiles in the neuropil in PiD compared to AD (Davis et al., 1992). In AD the predominant abnormal form of tau protein shows three bands on Western blots (55, 64 and 68 kDa), while in most cases of PiD the pattern is that of two bands (55 and 68 kDa) (Delacourte et al., 1996) (Fig. 11.7; summarized in Table 11.1). Using antibodies specific to alternatively spliced exons, it appears that abnormal tau protein in PiD lacks exon 10 (Sergeant et al., 1997) and is thus enriched in 3R tau. Tau in Pick bodies is also different from PHFtau in AD with respect to phosphorylation sites (e.g. serine 262 appears to be less phosphorylated in Pick’s than in AD) (Probst et al., 1996). The cholinergic innervation of the cortex is minimally affected (Sparks & Markesbery 1991; Wood et al., 1983) consistent with relative preservation of neurons in the basal

233

234

D. W. Dickson

Fig. 11.7. Western blot of insoluble tau in AD, PiD and CBD. Note three major bands and a minor higher molecular weight band in AD; note two bands in PiD and CBD, with the two major bands in PiD migrating at 64 and 55 kDa and the two in CBD migrating at 68 and 64 kDa.

nucleus of Meynert. Striatal dopamine is decreased in some cases of PiD, while decreases in gamma amino butyric acid are more variable (Kanazawa et al., 1988). Serotonin binding has been reported to be consistently decreased in all areas examined (Sparks & Markesbery, 1991). Synaptic loss has been demonstrated in PiD with quantitative immunocytochemical and immunoblotting assays (Weiler et al., 1990).

Corticobasal degeneration Clinical features CBD was first described as ‘corticodentatonigral degeneration with neuronal achromasia’ in a patient with an asymmetrical cortical syndrome (Rebeiz et al., 1968). Prevalence rates of CBD are difficult to estimate since clinical and pathological criteria for CBD have changed in recent years. The range of clinical presentations for CBD now include not only the classical progressive asymmetrical apraxia and rigidity syndrome (Gibb et al., 1989; Riley et al., 1990), but also other focal cortical syndromes, such as frontal lobe dementia and progressive aphasia (for review see Litvan et al., 2000). Increasingly, CBD is considered a focal cortical degenerative process with varied clinical phenotypes depending on the location of the dominant cortical pathology. The typical clinical phenotype corresponds to damage to the dorsal peri-Rolandic, superior frontal and superior parietal cortices, while cases with aphasia show pathology in the peri-Sylvian region. The clinical syndrome of progressive asymmetrical apraxia and rigidity can be due to a variety of pathological disorders (Boeve et al., 1999), as is true for frontal lobe dementia (Neary et al., 1993) and progressive aphasia (Kertesz et al., 1994). The initial signs of typical cases of CBD are unilateral or asymmetrical apraxia, rigidity and dystonia. This may be associated with myoclonic jerks, grasp reflex, cortical

sensory signs and the alien limb phenomenon. The affected hand may develop dystonic flexion contractures late in the disease. Cognitive impairment is common in CBD, and some cases come to autopsy with a primary diagnosis of frontal lobe degenerative dementia. Dementia in CBD is often characterized by personality change, disorder of conduct, impaired attention and distractibility. Frontal lobe signs, including grasp reflex, forced groping, utilization behaviour and inter-manual conflict, are common and may be unilateral or asymmetric. The alien limb phenomenon, which describes involuntary movement, such as elevation of the arm or leg into the air (Doody & Jankovic, 1992), is often emphasized in CBD, but it is neither specific to CBD nor found in all cases. Cortical sensory deficits due to parietal lobe involvement and characterized by graphesthesia and astereognosis are frequent in CBD.

Neuroimaging Magnetic resonance imaging may show asymmetrical atrophy of the superior parietal lobule variably extending into frontal regions, with less prominent atrophy elsewhere (Soliveri et al., 1999). Structural changes become more obvious as the disease progresses. Asymmetrical cortical atrophy can also be found in PiD, nonspecific frontotemporal dementia, primary progressive aphasia and some individuals with FTDP-17 and is hardly a specific diagnostic finding. MRI scans in CBD may show hyperintense signals in white matter in regions of brain atrophy and sometimes atrophy of the corpus callosum (Doi et al., 1999; Yamauchi et al., 1998). PET and SPECT scans show asymmetrical hypometabolism in superior frontal and parietal lobes (Blin et al., 1992; Nagasawa et al., 1996) and sometimes in caudate, putamen or thalamus. PET scans assessing dopamine metabolism [18 F]-DOPA may show reduction of striatal uptake (Sawle et al., 1991; Brooks, 2000). Striatal uptake is usually most severely impaired contralateral to the clinically most affected limbs.

Genetics Almost all reported cases of CBD have been sporadic, but there are rare familial reports of CBD (Brown et al., 1996). Whether these are actual cases of CBD or FTDP-17 remains to be determined by evaluation of these purported familial CBD cases for tau mutations. Some cases of FTDP-17 share clinical and pathologic features with CBD (Bugiani et al., 1999). In particular, familial multi-system tauopathy (Spillantini et al., 1997, 1998) and familial pallidoponto-nigral degeneration (Reed et al., 1998) have many


histological, ultrastructural and biochemical similarities to CBD. Most cases of FTDP-17 are caused by mutations in the tau gene in either the coding or non-coding regions (Hutton et al., 1998). A dinucleotide repeat polymorphism [with the number of repeats varying from 11 (a0) to 14 (a3)] in an intron in the tau gene (Conrad et al., 1997; Oliva et al., 1998; Hoenicka et al., 1999; Morris et al., 1999) as well as other tau gene polymorphisms (Higgins et al., 1999; Baker et al., 1999) that were originally reported to be associated with greater than chance frequency with PSP (see below) have more recently been shown to be increased in CBD as well (Di Maria et al., 2000; Houlden et al., 2001). Studies of the genetic structure of the tau gene have revealed essentially two major forms of the tau gene that have been termed H1 and H2 (Baker et al., 1999). The frequency of H1 is increased in CBD and many cases are H1/H1 homozygous.

Pathological features Gross findings The gross examination of the brain characteristically reveals subtle asymmetrical atrophy of cortical gyri, most marked in pre- and post-central regions (Fig. 11.8). The atrophy is rarely sharply circumscribed as in PiD. This dorsal frontoparietal atrophy merges with less severe atrophy in ventral frontal and posterior parietal regions. The temporal and occipital lobes are usually preserved. The brainstem does not have the consistent atrophy, but pigment loss is common in the substantia nigra. On cut sections the cortical ribbon is thinner than usual and the cerebral white matter in affected areas is often attenuated and may have a grey discolouration. The anterior corpus callosum may be thinned and the frontal horn of the lateral ventricle is usually dilated. The temporal horn is usually not dilated, as it is in PiD. The basal ganglia may have increased rust colour in the globus pallidus, but are otherwise grossly unremarkable. The midbrain sections invariably show decreased pigment in the substantia nigra, while the pigmentation in the locus ceruleus is usually better preserved. Microscopic findings Microscopic examination of atrophic regions of the frontoparietal cortex shows moderately severe neuronal loss and subcortical myelin pallor with gliosis. There is variable disruption of the cortical lamination with spongiosis, as well as astrocytic and microglial gliosis (Fig. 11.9). These findings are not dissimilar to those of non-specific focal cortical degenerations, but several histological features readily distinguish CBD from dementia lacking distinctive histopathology (Knopman et al., 1990). In affected cortex

Fig. 11.8. (a) Corticobasal degeneration has subtle atrophy of the superior frontal gyrus; (b) superior frontal gyrus atrophy is subtle on the medial view (arrow); note preservation of the medial temporal lobe; (c) coronal section at the level of the fornix shows normal amygdala and dilation of the frontal horn of the lateral ventricle; (d ) coronal section at the level of the anterior hippocampus shows focal superior frontal gyrus atrophy (arrow).

swollen and vacuolated, ballooned neurons are common in middle and lower cortical layers (Fig. 11.9). Ballooned neurons in CBD are similar in most respects to those seen in PiD. Ballooned neurons confined to the anterior cingulate, amygdala and claustrum are of less diagnostic significance, and often suggest the presence of AGD (see below). In CBD ballooned neurons are found in the superior frontal and parietal lobes, which are less affected in PiD. In subcortical regions the globus pallidus and putamen show neuronal loss with gliosis and occasional NFT. There may be granular foamy spheroids and increased iron pigment, as well in the pallidum and pars reticulata of the substantia nigra. The nucleus basalis of Meynert may have a few NFT, but the neuronal population is usually preserved. The subthalamic nucleus may show mild neuronal depletion and astrocytic gliosis, but severe neuronal loss is decidedly uncommon and should suggest another diagnosis. Thalamic nuclei, particularly the ventro-lateral and

235

236

D. W. Dickson

Fig. 11.9. (a) CBD cortex has microvacuolation in superficial cortical layers; (b) ballooned neurons are numerous in typical cases of CBD and are easily detected with H&E; (c) ballooned neurons are obvious with B-crystallin staining, which also shows swelling of proximal dendrites; (d ) Luxol fast blue stain of white matter shows staining pallor due to loss of myelinated fibres; (e) Gallyas silver stain shows thread-like processes in cerebral white matter and scattered coiled bodies (arrows) in oligodendrocytes.

-anterior nuclei, also may be affected. The substantia nigra has marked neuronal loss (Fig. 11.10). Mild neuronal depletion of the dentate nucleus is common, and grumose degeneration may be found, but it is far less common than in PSP. In addition to ballooned neurons, scattered neurons in atrophic cortical areas have tau immunoreactivity (Fig. 11.11). The tau-immunoreactive neuronal lesions are morphologically variable. In some neurons the immunoreactivity is diffuse and granular, consistent with so-called ‘pretangles’, while in other neurons it is densely packed into small inclusions somewhat reminiscent of Pick bodies or small NFT. In some neurons the filamentous inclusions are dispersed and skein-like. In contrast to NFT of AD, where lesions are readily detected with a host of diagnostic silver stains and thioflavin fluorescent microscopy, the neuronal lesions in CBD are not easily seen and often completely negative, especially with thioflavin-S. Other differences in

neurofibrillary pathology in CBD compared to AD include less ubiquitination and a restricted tau isoform composition. Neurofibrillary lesions in brainstem monoaminergic nuclei, such as the locus ceruleus and substantia nigra, sometimes resemble globose NFT, but more often are illdefined amorphous inclusions. In addition to fibrillary lesions in perikarya of neurons, the neuropil of CBD invariably contains numerous tauimmunoreactive thread-like cell processes (Wakabayashi et al., 1994; Mori et al., 1994; Feany & Dickson, 1995; Takahashi et al., 1996) (Fig. 11.12). In AD virtually all threads are neuronal in origin, but in CBD only a small fraction of thread-like structures are double labelled with neurofilament antibodies (Feany & Dickson, 1995), which indicates that many thread-like processes in CBD are probably glial derived. They are usually profuse in affected areas of grey and white matter. The numerous and widespread thread-like processes in both gray and white matter are an

Fig. 11.10. (a) Neuronal loss and gliosis in the substantia nigra with residual neurons having pale cytoplasmic inclusions; (b) round cytoplasmic inclusions are more obvious with tau immunostains; (c) astrocytic gliosis is the substantia nigra is marked; (d ) iron stains (Prussian blue) show increased iron pigment in the substantia nigra.

Fig. 11.11. (a) Astrocytic plaques in cortex are clusters of short stubby processes with Gallyas stain; (b) astrocytic plaques (arrow) are also positive for tau proteins; (c) double immunostaining for glial filament protein (brown) and tau (blue–grey) show astrocytes in the centre of astrocytic plaques (arrow); (d ) tau-immunoreactive neuronal inclusions are small round and sometime resemble small Pick bodies (arrow).

238

D. W. Dickson

Fig. 11.12. (a) Pre-tangles such as neurons in the basal nucleus of Meynert show diffuse to granular cytoplasmic tau immunoreactivity; (b) basal ganglia small neurons have tau inclusions, and there are also many tau-immunoreactive processes; (c) white matter tracts in the basal ganglia have numerous tau-immunoreactive processes; (d ) pontine nuclei also have many thread-like processes and a few pre-tangles.

important attribute of CBD and a useful feature in differentiating it from other tauopathies. In other disorders with which CBD can be confused, tau-related pathology is more often located in cell bodies (e.g. NFT and Pick bodies) and the proximal cell processes of neurons and glia. Astrocytic lesions are increasingly recognized in neurodegenerative tauopathies and this is the case for CBD, as well (for review see Chin & Goldman, 1996 and in Komori, 1999). The most characteristic tau-immunoreactive astrocytic lesion in CBD, particularly in the neocortex, is an annular cluster of short processes that may be highly suggestive of an Alzheimer type neuritic plaque (Uchihara et al., 1994; Feany & Dickson, 1995) (Fig. 11.11). The cell processes are shorter and stubbier than threads and have fuzzy or spiked rather than smooth profiles. These lesions, in contrast to Alzheimer plaques, do not contain amyloid based upon histochemical, fluorescent and immunocytochemical methods for detecting amyloid. Furthermore, the tau immunoreactivity is not within dystrophic processes, but rather processes of glial cells. Double immunostaining

for tau and glial fibrillary acid protein, vimentin, CD44 and more recently -B crystallin, which are all markers with varying degrees of specificity for reactive astrocytes, demonstrates that the tau-positive cell processes in these lesions are derived from astrocytes (Feany & Dickson 1995). These lesions have been referred to as ‘astrocytic plaques’ (Feany & Dickson, 1995). When the astrocytic plaque, with its distinct annular array of tau-immunoreactive processes, is detected with Gallyas silver stains or immunocytochemistry in the neocortex or neostriatum, it should suggest a diagnosis of CBD until proven otherwise. Somewhat similar lesions have been described in some cases of FTDP-17 (Goedert et al., 1999). The astrocytic plaque may be the most specific histopathologic lesion of CBD. In addition to cortical pathology, deep grey matter consistently has tau pathology in CBD (Fig. 11.12). In the basal ganglia, thread-like processes are often extensive in the pencil fibres of the striatum. There may also be a few astrocytic plaques in the striatum. Neuronal inclusions are also common in the striatum and globus pallidus.


Fig. 11.13. EM of neuron with tau filaments (immunogold labelled) in CBD. Inset shows the wide periodicity of the paired helical filaments (twisted ribbons) in CBD.

The internal capsule often has many thread-like processes, especially in the vicinity of the thalamic fasciculus. The subthalamic nucleus usually has neurons with tau inclusions, especially pre-tangles, and there may be a number of thread-like lesions in the nucleus. Many of the residual neurons in the substantia nigra contain ill-defined neurofibrillary inclusions, so-called ‘corticobasal bodies’ (Gibb et al., 1989). These inclusions have a homogeneous or faintly whorled or amorphous morphology with enmeshed granules of melanin pigment. Immunocytochemical studies have shown that corticobasal bodies are indistinguishable from NFT. The locus ceruleus and raphe nuclei have similar inclusions. The pontine base may have pre-tangles, as well as tau inclusions in glia and in cell processes. NFT are also detected in the tegmental grey matter.

Ultrastructural findings In CBD the filaments have a twisted ribbon appearance at the electron microscopic level, with a wider diameter (22–24 nm) and longer periodicity (160 nm) than in AD (Ksiezak-Reding et al., 1994) (Fig. 11.13). Similar filaments are found in some cases of FTDP-17.

Biochemical findings There are very few studies of postmortem neurochemical changes in CBD. In one of the first studies, Clark and coworkers showed that cortical choline acetyl transferase, a marker for cholinergic neurones, was not decreased (Clark et al., 1986), which demonstrated a significant difference

239

240

D. W. Dickson

between CBD and AD. Given the marked loss of pigmented neurons in the pars compacta of the substantia nigra, a population of neurons know to be the major dopaminergic innervation of the basal ganglia, it is not surprising that decreases in dopamine in the basal ganglia have been found in post-mortem studies of CBD (Marshall et al., 1997). Other neurotransmitter abnormalities have not been documented in CBD. It is unknown, for example, if the involvement of cortical neurons has any selectivity with respect to neurotransmitter. Most of the biochemical studies that have been conducted on CBD have focused on cytoskeletal proteins, especially tau protein (for review see Buee & Delacourte, 1999). Western blots of homogenates of detergent-soluble tau protein in CBD show an abnormal tau protein that migrates at a higher molecular weight than normal tau. In addition, the immunoblotting pattern is reduced to 2 major bands at about 64 and 68 kDa due to preferential accumulation of 4R tau in insoluble tau (Fig. 11.7).

Progressive supranuclear palsy Clinical features PSP is one of the major causes of levodopa-non-responsive Parkinsonism (Rajput et al., 1984). Given the fact that it is associated with additional clinical features not typical of Parkinson’s disease, it is considered one of the ‘Parkinsonism-plus’ syndromes. One of the earliest clinical features of PSP is unexplained falls, often backwards. Other characteristic clinical features include postural instability, vertical gaze paresis, nuchal and axial rigidity and dysarthria (Rajput et al., 1991; Golbe, 1993). Downwards gaze impairments are usually earlier than upwards, and impaired downwards gaze also carries more diagnostic significance, since upward gaze problems can be an ageassociated complaint. There is usually preservation of the oculocephalic reflexes (‘dolls’ eye’ test). Along with falls, gait disturbance is common in early stages of PSP. Gait problems are progressive and disabling, leading eventually to confinement to chair or bed within about 5 years (Golbe et al., 1988). Other clinical features commonly seen in PSP include dysarthria and dysphagia. Some degree of mental dysfunction, but not severe dementia, is common in PSP (Maher et al., 1985). Cognitive dysfunction in PSP is consistent with a subcortical or frontal lobe disorder. Attention is impaired and cognitive processing is slowed (Dubois et al., 1988). Frontal lobe signs may be prominent (Grafman et al., 1990).

PSP is an uncommon disorder, but considerably more common than either PiD or CBD. Previous estimates of the prevalence rate for PSP of about 1.5 per 100 000 (Golbe et al., 1988; Bower et al., 1999) may have been an underestimate based upon an active ascertainment survey of PSP in the United Kingdom (Nath et al., 2001), where crude and age-adjusted prevalences were determined to be 6.5 and 5.0 per 100 000, respectively. In comparison, the prevalence for Parkinson’s disease is estimated to be 100 to 150 per 100 000. New cases occur at a rate of about 3 to 4 per million population per year (Rajput et al., 1984).

Neuroimaging Structural imaging in PSP often shows midbrain atrophy, with narrowing of the anteroposterior midbrain diameter and dilation of the third ventricle (Golbe, 1993; GimenezRoldan et al., 1994; Cordato et al., 2000). There may also be increased signal intensity in the superior cerebellar peduncle (Oka et al., 2001). SPECT and PET scanning show bilateral frontal hypoactivity (D’Antona et al., 1985; Foster et al., 1988; Leenders et al., 1988). Striatal dopaminergic deficits are reflected in decreased [18 F]-DOPA uptake in the basal ganglia (Baron et al., 1986; Brooks et al., 1990). In addition to decreased dopamine there is also loss of dopamine receptors in PSP (Brooks et al., 1992). This reflects the fact that, in PSP, there is not only damage to dopaminergic neurons projecting from the substantia nigra to the striatum, but also to intrinsic neurons of the striatum.

Genetics Most PSP is sporadic, but some autosomal dominant cases of PSP have been described (Gazeley & Maguire, 1996; Rojo et al., 1999). Some cases of FTDP-17 have had a clinical phenotype that overlaps with PSP (Pastor et al., 2001; Stanford et al., 2000), but carefully screened cases with typical PSP, even with a family history, have failed to show tau mutations (Morris et al., 2002). As mentioned previously, a dinucleotide repeat polymorphism in an intron in the tau gene (a0) has been reported to be associated with PSP, with more than 95% of PSP cases having the a0 form of tau (Conrad et al., 1998). The increased frequency of this tau polymorphism in PSP has been confirmed in several studies among different ethnic groups (Oliva et al., 1998; Higgins et al., 1999; Hoenicka et al., 1999; Morris et al., 1999; Baker et al., 1999). Interestingly a0 is non-polymorphic in Japan, with almost all Japanese showing the a0 form of tau (Conrad et al., 1998). Additional studies of the genetic structure of the tau gene have revealed a number of polymorphisms included an extended tau haplotype (Baker et al., 1999), as


well as variants in the promoter region (Higgins et al., 1999). Most cases of PSP are H1 and a surprisingly large number of PSP patients are H1/H1 homozygous. The presence of an H1H1 genotype does not affect the pathology of PSP in any obvious manner (Liu et al., 2001).

Pathological findings Gross findings Gross examination of the brain may be unrevealing in PSP, but the most common pathological findings are a mild degree of frontal and midbrain atrophy (Fig. 11.14). The third ventricle and aqueduct of Sylvius may be dilated. The subthalamic nucleus may be noticeably smaller than expected (Fig. 11.15). The substantia nigra invariably shows some degree of depigmentation, with pigment loss often greater in ventrolateral regions than in dorsomedial regions of the substantia nigra. In contrast, the locus ceruleus usually has normal pigmentation. The superior cerebellar peduncle and the hilus of the cerebellar dentate nucleus may be attenuated and grey due to myelinated fibre loss (Fig. 11.16). As mentioned previously this latter finding can be detected on imaging studies. It is very common in PSP and is a useful gross feature for differentiating PSP from other disorders. Microscopic findings Microscopic findings include neuronal loss and fibrillary gliosis affecting mostly subcortical regions. The nuclei with the most marked and consistent pathology are the globus pallidus, subthalamic nucleus and substantia nigra. The cerebellar dentate nucleus and its outflow pathway are also consistently affected (Ishizawa et al., 2000). Other parts of the basal ganglia, diencephalon and brainstem are affected to a variable degree. In addition to the changes visible with routine histological methods, silver stains or immunostaining for tau reveal NFT as well as glial inclusions. The globus pallidus and pars reticularis of the substantia nigra may show in addition to neuronal loss, gliosis and NFT, iron pigment deposition and granular neuroaxonal spheroids (Fig. 11.17). The striatum and thalamus, especially ventral anterior and lateral nuclei, may have gliosis, but neuronal loss is usually subtle. The basal nucleus of Meynert usually has mild cell loss, but more noticeable neurofibrillary pathology. The brainstem regions that are affected include the superior colliculus, periaqueductal gray matter, oculomotor nuclei, locus ceruleus, pontine nuclei, pontine tegmentum, vestibular nuclei, medullary tegmentum and inferior olives. The cerebellar dentate nucleus is frequently affected and may have grumose degeneration (Cruz-Sanchez et al., 1992b; Ishizawa et al., 2000) (Fig. 11.18). The latter is a type

Fig. 11.14. (a) Progressive supranuclear palsy has subtle frontal atrophy; (b) frontal atrophy is subtle on the medial view, but note more marked tectal atrophy and dilation of the aqueduct of Sylvius (arrow); (c) coronal section at the level of the fornix shows normal amygdala and dilation of the frontal horn of the lateral ventricle; (d ) coronal section at the level of the anterior hippocampus shows dilation of the frontal horn of the lateral ventricle, but grossly unremarkable basal ganglia and thalamus.

Fig. 11.15. (a) Atrophy of the subthalamic nucleus (between arrows) is usually striking in PSP; (b) compared to normal subthalamic nucleus (between arrows).

of degeneration associated with clusters of degenerating presynaptic terminals around dentate neurons. The cerebellar cortex is generally preserved, but there may be mild

241

242

D. W. Dickson

Fig. 11.16. Sections of brainstem in PSP show atrophy of the midbrain with loss of pigment in the substantia nigra (black arrow), marked atrophy of the superior cerebellar peduncle (white arrow) and atrophy of the hilus of the dentate nucleus (arrowhead).

Purkinje and granular neuronal loss with scattered Purkinje cell dystrophic axons (‘torpedoes’). A more consistent finding in the cerebellum is presence of glial cells with fibrillary inclusions in the cerebellar white matter. Spinal cord involvement is common, where neuronal inclusions can be found in anterior horn and intermediolateral cells (Jellinger & Bancher, 1992). Neuronal loss in the substantia nigra is usually greatest in the ventrolateral tier, as in Parkinson’s disease, but often

with more extensive neuronal loss throughout all regions of the substantia nigra, as well (Jellinger, 1971). The neuronal loss in the substantia nigra is usually severe and affects not only the pigmented neurons, but also non-pigmented neurons (Oyanagi et al., 2001). Neurons in the substantia nigra have NFT in PSP, while they have more pretangles in CBD (Oyanagi et al., 2001). Neuronal loss in the corpus striatum is most marked in large neurons, with 30 to 40% loss reported (Oyanagi et al., 1988). Neuronal loss in the basal nucleus of Meynert ranges from 13 to 54% (Tagliavini et al., 1984). Cortical grey matter pathology is less pronounced than deep grey matter pathology, which differentiates PSP from CBD and PiD (Feany & Dickson, 1996), but lesions are increasingly recognized in the cortex, especially the periRolandic region (Hauw et al., 1990; Hof et al., 1992; Vermersch et al., 1994). Early studies emphasized the lack of neocortical pathology in PSP (Steele et al., 1964), but more recent studies using immunocytochemical and silver stains such as the Gallyas method show consistent involvement of the frontal cortex in PSP. Neocortical NFT and glial tangles are concentrated in the precentral gyrus (i.e. motor

Fig. 11.17. (a) Severe gliosis is present in the substantia nigra (the subthalamic nucleus usually has a similar appearance); (b) granular foamy axonal spheroids (arrows) and iron pigment are common in the substantia nigra; (c) residual neurons in the substantia nigra have globose shaped NFTs; (d) neurons in the pontine base also have pre-tangles and NFTs.


Fig. 11.18. (a) The neurons of the dentate nucleus have smudgy appearance due to grumose degeneration; (b) grumose degeneration is most obviously a degeneration of synaptic terminals around dentate neurons as seen with a synaptophysin stain; (c) severe atrophy of the superior cerebellar peduncle shows many lipid laden macrophages; (d ) macrophage reaction in the cerebellar peduncle is obvious with HLA-DR immunostaining (brown) and myelin degeneration is obvious with Luxol fast blue.

cortex). Recent studies suggest that cortical neurofibrillary pathology may be more widespread in cases with atypical clinical features, such as dementia (Bergeron et al., 1997; Bigio et al., 1999). The cerebral white matter beneath the motor cortex has glial pathology characterized by tau-positive inclusions in oligodendroglia. More widespread and severe cerebral white matter pathology is uncommon in PSP, and should raise suspicion of CBD. The limbic lobe is usually preserved in PSP. Neurofibrillary pathology in the hippocampus is variable and usually not greater than expected for age (Braak et al., 1992). When present, the distribution of NFT in the hippocampal formation is qualitatively similar to that seen in aging and AD. The exception to this rule is the frequent involvement of the dentate gyrus granule neurons in PSP (Hof et al., 1992) and preferential involvement of neurons in CA2 (Ishizawa et al., 2002).

NFT in PSP often have a rounded or globose appearance (Fig. 11.17); flame-shaped NFT are also detected. The Gallyas silver iodine method is sensitive to pathological structures in PSP and it demonstrates in addition to NFT, fibrillary inclusions in glial cells and cell processes. As in AD NFT in PSP contain abnormally phosphorylated tau protein. Immunocytochemistry for tau protein is a sensitive method for detecting NFT in PSP, but it also reveals a wide variety of inclusions in both neurons and glia, including pre-tangles, neuropil threads and tufts of abnormal fibers. Antibodies that are most sensitive for detecting pre-tangles and the host of other lesions that characterize the tauopathies are those that recognize phosphorylated epitopes or conformational changes in tau proteins (Jicha et al., 1999). Tufts of abnormal fibers, also known as tufted astrocytes, are most common in motor cortex and striatum (Fig. 11.19). They have been shown to be fibrillary lesions within

243

244

D. W. Dickson

Fig. 11.19. (a) Tau immunoreactive astrocytes (‘tufted astrocytes’) and present in the basal ganglia; (b) tufted astrocytes have tau immunoreactivity in cell bodies and proximal processes; (c) neurons in the basal nucleus of Meynert (and other cholinergic nuclei) have many NFTs; (d ) subthalamic nucleus has NFTs, coiled bodies and many thread-like processes.

astrocytes based upon double immunolabelling of tau and glial fibrillary acidic protein (Nishimura et al., 1992; Yamada et al., 1993; Iwatsubo et al., 1994). Tufted astrocytes account for much of the cortical pathology in PSP. The distribution of tufted astrocytes does not follow the distribution of neuronal loss, NFT and gliosis suggesting that astrocytic pathology in PSP is a degenerative process with uncertain functional significance (Togo & Dickson, 2002b). Tau immunohistochemistry also reveals tau-positive fibres, so-called ‘neuropil threads’ (Braak et al., 1986; Probst et al., 1988; Ikeda et al. 1994) and small round glial cells in the white matter. The latter tau-positive glia, are referred to as ‘coiled bodies’, (Braak & Braak, 1989) and have been shown to be oligodendroglial inclusions based upon double immunolabelling with tau and oligodendroglial markers (Iwatsubo et al., 1994; Yamada & McGeer, 1995; Nishimura et al., 1992). Thread-like processes in white matter are not as numerous in PSP as in CBD, but in both disorders have been shown to be not only within axons, but also glial processes. Immunoelectron microscopy in PSP

has shown these inclusions to be in the outer mesaxon of myelinated fibres (Arima et al., 1997). In grey matter taupositive threads are much less common in PSP than in AD (Davis et al., 1992) and CBD. The NFT in PSP, as in CBD, may be distinguished from NFT in AD by the paucity of ubiquitin immunoreactivity (Bancher et al., 1987; Cruz-Sanchez et al., 1992a). They also have far less fluorescence with thioflavin-S stains. Certain neurofilament epitopes may also distinguish NFT in PSP from those in AD (Schmidt et al., 1988). On the other hand, a study with a panel of anti-tau antibodies recognizing epitopes along the length of the tau molecule suggested that full length tau proteins are incorporated into NFT in PSP (Schmidt et al., 1996).

Ultrastructural findings NFT in PSP are composed of 15–18-nm diameter straight filaments (Tellez-Nagel & Wisniewski, 1973; Powell et al., 1974). The abnormal filaments in glial cells in PSP


Fig. 11.20. The NFT in PSP (a) are composed of bundles of straight filaments as well as granular material. They differ from the paired helical filaments of AD (b).

also contain straight filaments (Nishimura et al., 1992) (Fig. 11.20).

Biochemical findings The neuronal degeneration in the substantia nigra is associated with marked reduction in dopamine and tyrosine hydroxylase levels in the putamen and the caudate nucleus (Agid et al., 1987). While dopamine receptor cells are relatively preserved in Parkinson’s disease, the density of dopaminergic D2 receptors is reduced by almost half in the striatum in PSP (Pascual et al., 1992). Most cholinergic nuclei, including the basal forebrain, basal ganglia and pontine tegmental cholinergic cell groups (including the Edinger–Westphal nucleus and the pedunculopontine nucleus) show neuronal loss (Hirsch et al., 1987; Juncos et al., 1991). Consequently, there is widespread reduction in choline acetyl transferase in cortical and subcortical regions in PSP (Ruberg et al., 1985). Synaptic loss has been reported in the cortex in PSP and is more marked in cases with dementia (Bigio et al., 2001). Western blots of homogenates of detergent-soluble tau protein in PSP are similar to abnormal tau protein in CBD, with two major bands at about 64 and 68 kDa due to preferential accumulation of 4R tau in insoluble tau (Fig. 11.21).

disorder. It has been estimated to be present in 5–9% of autopsies for dementia (Braak & Braak, 1998; Tolnay et al., 1997; Martinez-Lage & Munoz, 1997; Togo et al. 2002a). The frequency of AGD increases with age. It is also common to find AGD in association with other disorders, particularly other tauopathies, including PiD, CBD and PSP (Martinez-Lage & Munoz, 1997; Braak & Braak, 1998; Togo et al., 2002b). AGD is found more often in PSP and CBD than in control cases, suggesting that it may be related to other 4R tauopathies (Togo et al., 2002a).

Clinical features The clinical features of AGD are not fully understood. Initial reports emphasized dementia (Braak et al., 1987), but well-characterized longitudinal studies of dementia in AGD have not been reported. More recently, personality change and emotional imbalance have been emphasized in AGD (Ikeda et al., 2000). Given that the brunt of the pathology in

Argyrophilic grain disease Argyrophilic grain disease (AGD), first reported by Braak and Braak (Braak & Braak, 1987), is an under-recognized

Fig. 11.21. Western blot of insoluble tau protein in PSP is characterized by two bands, in contrast to three bands for AD. This pattern is similar in CBD and AGD.

245

246

D. W. Dickson

Neuroimaging Ante-mortem imaging studies are limited in AGD. One report described only mild frontal atrophy (Seno et al., 2000).

Genetics

Fig. 11.22. (a) Argyrophilic grain disease is often grossly unremarkable; (b) coronal sections sometimes show atrophy of the hippocampus and entorhinal cortex (arrow).

AGD is in the medial temporal lobe, it is not surprising that some cases of AGD have had an amnestic disorder or even mild cognitive impairment (Ikeda et al., 2000; Parisi et al., 2000).

Tau haplotype analysis in AGD shows a similar trend as in PSP and CBD, with increased frequency of H1 compared with normal controls. In the only published study to address it, the frequency of the H1 haplotype in AGD was not different from PSP and CBD and showed a trend for higher frequency than in controls, although, the difference between AGD and non-tauopathy controls was not statistically significant, probably due to small sample size (Togo et al., 2002b). Whether AGD is associated with apolipoprotein E is controversial. When controlling for the degree of Alzheimer-type pathology, there seems to be no clear association between AGD and apolipoprotein E (Togo et al., 2002a).

Fig. 11.23. (a) Argyrophilic grains and glia (arrow) are tau positive; (b) glial cells in white matter also have tau immunoreactivity (arrow); (c) immunostaining with antibody specific to 3R tau fails to stain grains; (d ) immunostaining with antibody specific to 4R tau stains grains showing the AGD is a 4R tauopathy.


Table 11.3. Ballooned neurons and differential diagnosis Distribution of ballooned neurons PiD CBD

Fig. 11.24. (a), (b) Coiled bodies in white matter of limbic lobe are characteristic of AGD.

Pathological features Gross findings The gross findings are non-specific. The brain may show no obvious pathology that would differentiate it from an agematched control; however, in some cases there is medial temporal lobe atrophy (Fig. 11.22).

Microscopic findings The hallmark of AGD is the argyrophilic grain, which is a spindle- or comma-shaped argyrophilic structure found in the neuropil of entorhinal cortex, hippocampus and amygdala (Fig. 11.23). Argyrophilic grains are clearly demonstrated with the Gallyas silver stain and also with tau immunostains. They are indistinct with Bielschowsky and Bodian silver stains. Argyrophilic grains have been shown to be immunoreactive with a series of antibodies to hyperphosphorylated tau (Tolnay et al., 1997). In addition to grains in neuronal cell processes, AGD is usually accompanied by coiled bodies in the white matter underlying affected cortices (Fig. 11.24) and by ballooned neurons

AGD PSP

Association cortices (anterior frontotemporal) and limbic lobe Association cortices (superior frontoparietal) and motor cortex; Limbic lobe if concurrent AGD Limbic lobe (amygdala) Rare (usually due to concurrent AGD)

in the limbic lobe (Braak & Braak, 1989, 1987; Togo & Dickson, 2002a) (Fig. 11.25). Ballooned neurons in AGD are relatively restricted to the limbic lobe and most often found in the amygdala. If they are more widespread, it should suggest another diagnosis, such as PiD or CBD (see Table 11.3). On the other hand, if ballooned neurons are confined to limbic regions the diagnosis of AGD is the first one to consider (Togo & Dickson, 2002a). If ballooned neurons are found in PSP, for example, AGD can be found in most cases (Togo & Dickson, 2002a). The tau pathology of AGD is largely confined to medial temporal lobe structures; however, some pathology is found in the hypothalamus and diencephalon. As mentioned previously AGD is often found in concert with other pathological processes, most notable PSP and CBD. It can also be found with varying degrees of Alzheimer-type pathology. AGD becomes difficult to detect in advanced AD as the neuropil is overrun by a dense plexus of neuropil threads that makes it difficult to detect grains. A recent study has suggested that AGD should be considered when NFT are found preferentially in CA2 of the hippocampus (Ishizawa et al., 2002) (Fig. 11.26).

Fig. 11.25. (a) Ballooned neurons found in the limbic lobe in AGD show granular tau immunoreactive at the cytoplasmic border; (b) B-crystallin stains ballooned neurons in AGD.

247

Table 11.4. Differential diagnosis of the sporadic tauopathies Progressive supranuclear palsy

Argyrophilic grain disease

Premotor and frontal atrophy

Variable or none, temporal atrophy

Minimal or none None or sparse, limbic lobe (see AGD) Globose NFT, pretangles

None Frequent, limbic lobe Pretangles, grains

Numerous, grey and white matter cortical, striatal and brainstem Astrocytic plaques, coiled bodies

Variable, basal ganglia and diencephalon

Minimal, medial temporal lobe

Tufted astrocytes, coiled bodies

Minimal

Marked neuronal loss

Coiled bodies, tau-positive astrocytes Minimal

Variable Mild; pontine Pick bodies

Marked Mild; pontine threads and pretangles

None

Mild, Purkinje cell torpedoes

Marked Marked pathology, tegmental; pontine nuclei NFT Moderate, grumose degeneration in dentate

Feature

Pick’s disease

Corticobasal degeneration

Gross appearance

Sharply circumscribed lobar (frontotemporal) atrophy

Cortex Ballooned neurons

Status spongiosis Abundant

Circumscribed atrophy, parasagittal frontal and parietal Superficial spongiosis Abundant

Tau- or Gallyas positive neuronal lesions Threads and thread-like lesions Glial pathology

Pick body

Pretangles, skein-like, NFT-like, Pick body-like

Sparse to variable

Subthalamic nucleus Substantia nigra Pons & medulla

Cerebellum

Glial fibrillary tangles, Pick body-like inclusions in oligodendrocytes Minimal

Fig. 11.26. (a) Gallyas silver stain shows grains and sparse NFTs; (b) NFTs, including some extracellular NFTs, are present in CA2; (c) low power view of NFTs in CA2; (d) low power view of CA1 from same case; note the sparse NFTs in CA1 compared with CA2, a feature of AGD.

None None

None


Substantia nigra

Cerebellum (dentate)

Midbrain Substantia nigra

CA2 & amygdala

PSP (4R)

Pons and medulla

Thalamus Subthalamic nucleus

CBD (4R)

AGD (4R)

Basal ganglia

Motor cortex

Association cortices Limbic and frontotemporal Frontoparietal

Limbic lobe

PiD (3R)

CA1, dentate & amygdala

Disease

Table 11.5. Disease spectrum of the sporadic tauopathies

Note: The intensity of the shading reflects the severity of the neuronal and glial tau pathology in a given region. Note that in both PiD and CBD, the brunt of the pathology is in cortical areas, but that the distribution is different; limbic/frontotemporal in PiD vs. frontoparietal in CBD. AGD is a tauopathy limited to the temporal lobe. PSP is a tauopathy with the brunt of the pathology in brainstem and diencephalon.

This observation needs to be confirmed in a larger series of cases.

Ultrastructural findings Ultrastructurally, argyrophilic grains consist of aggregates of 9–18 nm diameter straight filaments (Braak & Braak,

1989; Tolnay et al., 1997) or smooth 25 nm diameter tubules (Ikeda et al., 1995) and are located mainly in dendrites of neurons (Ikeda et al., 1995; Tolnay et al., 1998). The ultrastructural features of coiled bodies and ballooned neurons are similar to those in PiD, PSP and CBD.

249

250

D. W. Dickson

Biochemical findings While it is known that argyrophilic grains contain hyperphosphorylated tau, the isoform composition of grains has only recently been shown to be similar to that in PSP and CBD. A recent study of a series of AGD cases with Western blots of sarkosyl-insoluble tau and immunohistochemistry with monoclonal antibodies specific to 3R and 4R tau demonstrated that pathological lesions in AGD were enriched in 4R tau (Togo et al., 2002b).

Summary The sporadic tauopathies have a number of overlapping clinical and pathologic features, but are sufficiently distinctive that they can be considered separate disorders. Table 11.4 summarizes the major pathological features of the sporadic tauopathies that can aid in their differential diagnosis. Table 11.5 shows schematically the different distribution of the pathology in these disorders.

Acknowledgements All electron micrographs were provided by Dr Wen-Lang Lin and Western blots were from Dr Shu-Hui Yen, both of Mayo Clinic. The study was supported by NIH grants AG 16574, AG 17216, AG 14449, AG 03949, NS 40256, The Mayo Foundation, The State of Florida Alzheimer Disease Initiative, and the Society for Progressive Supranuclear Palsy.

REFERENCES Agid, Y., Javoy-Agid, F., Ruberg, M. et al. (1987). Progressive supranuclear palsy: anatomoclinical and biochemical considerations. Adv Neurol, 45: 191–206. Arai, T., Ikeda, K., Akiyama, H. et al. (2001). Distinct isoforms of tau aggregated in neurons and glial cells in brains of patients with Pick’s disease, corticobasal degeneration and progressive supranuclear palsy. Acta Neuropathol, 101: 167–73. Arima, K. & Akashi, T. (1990). Involvement of the locus coeruleus in Pick’s disease with or without Pick body formation. Acta Neuropathol, 79: 629–33. Arima, K., Nakamura, M. Sunohara, N. et al. (1997). Ultrastructural characterization of the tau-immunoreactive tubules in the oligodendroglial perikarya and their inner loop processes in progressive supranuclear palsy. Acta Neuropathol, 93 558–66. Arnold, S. E., Hyman, B. T. & Van Hoesen, G. W. (1994). Neuropathological changes of the temporal pole in Alzheimer’s disease and Pick’s disease. Arch Neurol, 51: 145–50.

Baker, M., Litvan, I., Houlden, H. et al. (1999). Association of an extended haplotype in the tau gene with progressive supranuclear palsy. Hum Molec Gen, 8: 711–15. Ball, M. J. (1979). Topography of Pick inclusion bodies in hippocampi of demented patients. A quantitative study. J Neuropathol Exp Neurol, 38: 614–20. Bancher, C., Lassmann, H., Budka, H. et al. (1987). Neurofibrillary tangles in Alzheimer’s disease and progressive supranuclear palsy: antigenic similarities and differences. Acta Neuropathol, 74: 39–46. Baron, J. C., Maziere, B., Loch, C. et al. (1986). Loss of striatal [76 Br] bromospiperone binding sites demonstrated by positron tomography in progressive supranuclear palsy. J Cerebral Blood Flow Metab, 6: 131–6. Bergeron, C., Pollanen, M. S., Weyer, L. & Lang, A. E. (1997). Cortical degeneration in progressive supranuclear palsy. A comparison with cortical–basal ganglionic degeneration. J Neuropathol Exp Neurol, 56: 726–34. Bigio, E. H., Brown, D. F. & White, C. L. 3rd (1999). Progressive supranuclear palsy with dementia: cortical pathology. J Neuropathol Exp Neurol, 58: 359–64. Bigio, E. H., Vono, M. B. & Satumtira, S. (2001). Cortical synapse loss in progressive supranuclear palsy. J Neuropathol Exp Neurol, 60: 403–10. Blin, J., Vidailhet, M. -J., Pillon, B., Dubois, B., Feve, J. -R. & Agid, Y. (1992). Corticobasal degeneration: decreased and asymmetrical glucose consumption as studied with PET. Mov Disord, 7: 348–54. Boeve, B. F., Maraganore, D. M., Parisi, J. E. et al. (1999). Pathologic heterogeneity in clinically diagnosed corticobasal degeneration. Neurology, 53: 795–800. Bower, J. H., Maraganore, D. M., McDonnell, S. K. & Rocca, W. A. (1999). Incidence and distribution of parkinsonism in Olmsted County, Minnesota, 1976–1990. Neurology, 52: 1214–20. Braak, H. & Braak, E. (1987). Argyrophilic grains: characteristic pathology of cerebral cortex in cases of adult onset dementia without Alzheimer changes. Neurosci Lett, 76: 124–7. (1989). Cortical and subcortical argyrophilic grains characterize a disease associated with adult onset dementia. Neuropathol Appl Neurobiol, 15: 13–26. (1998). Argyrophilic grain disease: frequency of occurrence in different age categories and neuropathological diagnostic criteria. J Neural Transm, 105: 801–19. Braak, H., Braak, E., Grundke-Iqbal, I. & Iqbal, K. (1986). Occurrence of neuropil threads in the senile human brain and in Alzheimer’s disease: a third location of paired helical filaments outside of neurofibrillary tangles and neuritic plaques. Neurosci Lett, 65: 351–5. Braak, H., Jellinger, K., Braak, E. & Bohl, J. (1992). Allocortical neurofibrillary changes in progressive supranuclear palsy. Acta Neuropathol, 84: 478–83. Brooks, D. J. (2000). Functional imaging studies in corticobasal degeneration. In Litvan, I., Goetz, C. G., Lang, A. E. (eds). Corticobasal Degeneration and Related Disorders, Advances


in Neurology, Volume 82. Lippincott, Williams & Wilkins, Philadelphia, pp. 209–16. Brooks, D. J., Ibanez, V, Sawle, G. V. et al. (1990). Differing patterns of striatal 18 F-dopa uptake in Parkinson’s disease multiple system atrophy and progressive supranuclear palsy. Ann Neurol, 28: 547–55. Brooks, D. J., et al. (1992). Striatal D2 receptor status in patients with Parkinson’s disease, striatonigral degeneration, and progressive supranuclear palsy, measured with 11Craclopride and positron emission tomography. Ann Neurol, 31: 184–92. Brown, J., Lantos, P. L., Roques, P., Fidani, L. & Rossor, M. N. (1996). Familial dementia with swollen achromatic neurons and corticobasal inclusion bodies: a clinical and pathological study. J Neurol Sci, 135: 21–30. Buee, L., Delacourte, A. (1999). Comparative biochemistry of tau in progressive supranuclear palsy, corticobasal degeneration, FTDP-17 and Pick’s disease. Brain Pathol, 9: 681–93. Buee Scherrer, V., Hof, P. R., Buee, L. et al. (1996). Hyperphosphorylated tau proteins differentiate corticobasal degeneration and Pick’s disease. Acta Neuropathol, 91: 351–9. Bugiani, O., Murrell, J. R. & Giaccone, G. (1999). Frontotemporal dementia and corticobasal degeneration in a family with a P301S mutation in tau. J Neuropathol Exp Neurol, 58: 667–77. Castellani, R., Smith, M. A., Richey, P. L., Kalaria, R., Gambetti, P. & Perry, G. (1995). Evidence for oxidative stress in Pick disease and corticobasal degeneration. Brain Res, 696: 268–71. Chambers, C. B., Lee, J. M., Troncoso, J. C., Reich, S. & Muma, N. A. (1999). Overexpression of four-repeat tau mRNA isoforms in progressive supranuclear palsy but not in Alzheimer’s disease. Ann Neurol, 46: 325–32. Chin, S. S., Goldman, J. E. (1996). Glial inclusions in CNS degenerative diseases. J Neuropathol Exp Neurol, 55, 499–508. Clark, A. W., Manz, H. J., White, C. L. III, Lehmann, J., Miller, D., Coyle & J. T. (1986). Cortical degeneration with swollen chromatolytic neurons: its relationship to Pick’s disease. J Neuropathol Exp Neurol, 45: 268–74. Conrad, C., Andreadis, A., Trojanowski, J. Q. et al. (1997). Genetic evidence for the involvement of τ in progressive supranuclear palsy. Ann Neurol, 41: 277–81. Conrad, C., Amano, N., Andreadis, A. et al. (1998). Differences in a dinucleotide repeat polymorphism in the tau gene between Caucasian and Japanese populations: implication for progressive supranuclear palsy. Neurosci Lett, 250: 135–7. Constantinidis, J. (1985). Pick dementia. Anatomoclinical correlations and pathophysiological considerations. In Rose, F.C. (ed) Modern Approaches to the Dementias Part I: Etiology and Pathophysiology, Interdisciplinary Topics in Gerontology, Vol 19, Karger, Basel Switzerland, pp. 72–97. Cordato, N. J., Halliday, G. M., Harding, A. J., Hely, M. A. & Morris, J. G. L. (2000). Regional brain atrophy in progressive supranuclear palsy and Lewy body disease. Ann Neurol, 47: 718–28. Cruz-Sanchez, F.F., Rossi, M. L., Cardozo, A., Deacon, P. & Tolosa, E. (1992a). Clinical and pathological study of two patients

with progressive supranuclear palsy and Alzheimer’s changes. Antigenic determinants that distinguish cortical and subcortical neurofibrillary tangles. Neurosci Lett, 136: 43–6. Cruz-Sanchez, F. F., Rossi, M. L., Cardozo, A., Picardo, A. & Tolosa, E. (1992b). Immunohistological study of grumose degeneration of the dentate nucleus in progressive supranuclear palsy. J Neurol Sci, 110: 228–31. ¨ Cummings, J. L. & Duchen, L. W. (1981). Kluver–Bucy syndrome in Pick disease: clinical and pathologic correlations. Neurology, 31: 1415–22. D’Antona, R., Baron, J. C., Sanson, Y. et al. (1985). Subcortical dementia: frontal cortex hypometabolism detected by positron tomography in patients with progressive supranuclear palsy. Brain, 108: 785–99. Davis, D. G., Wang, H. Z. & Markesbery, W. R. (1992). Image analysis of neuropil threads in Alzheimer’s Pick’s diffuse Lewy body disease and in progressive supranuclear palsy. J Neuropathol Exp Neurol, 51: 594–600. Delacourte, A. & Buee, L. (1997). Normal and pathological Tau proteins as factors for microtubule assembly. Int Rev Cytol, 171: 167–224. Delacourte, A., Robitaille, Y., Sergeant, N. et al. (1996). Specific pathological tau protein variants characterize Pick’s disease. J Neuropathol Exp Neurol, 55: 159–68. Di Maria, E., Tabaton, M., Vigo, T. et al. (2000). Corticobasal degeneration shares a common genetic background with progressive supranuclear palsy. Ann Neurol, 47: 374–7. Dickson, D. W. (1998). Pick’s disease: A modern approach. Brain Pathol, 8: 339–54. Dickson, D. W. (2001). Neuropathology of Pick’s disease. Neurology, 56(Suppl 4): S16–20. Dickson, D. W., Yen, S. -H., Suzuki, K. I., Davies, P., Garcia, J. H. & Hirano, A. (1986). Ballooned neurons in select neurodegenerative diseases contain phosphorylated neurofilament epitopes. Acta Neuropathol, 71: 216–23. Dickson, D. W., Feany, M. B., Yen, S. H., Mattiace, L. A. & Davies, P. (1996). Cytoskeletal pathology in non-Alzheimer degenerative dementia: new lesions in diffuse Lewy body disease, Pick’s disease, and corticobasal degeneration. J Neural Transm Suppl, 47: 31–46. Doi, T., Iwasa, K., Makifuchi, T. & Takamori, M. (1999). White matter hyperintensities on MRI in a patient with corticobasal degeneration. Acta Neurol Scand, 99: 199–201. Doody, R. S. & Jankovic, J. (1992). The alien hand and related signs. J Neurol Neurosurg Psychiatry, 55: 806–10. Dubois, B., Pillon, B., Legault, F., Agid, Y. & Lhermitte, F. (1988). Slowing of cognitive processing in progressive supranuclear palsy: a comparison with Parkinson’s disease. Arch Neurol, 45: 1194–9. Farrer, L. A., Abraham, C. R., Volicer, L. et al. (1995). Allele epsilon 4 of apolipoprotein E shows a dose effect on age at onset of Pick disease. Exp Neurology, 136: 162–70. Feany, M. B. & Dickson, D. W. (1995). Widespread cytoskeletal pathology characterizes corticobasal degeneration. Am J Pathol, 146: 1388–96.

251

252

D. W. Dickson

(1996). Neurodegenerative disorders with extensive tau pathology: a comparative study and review. Ann Neurol, 40: 139–48. Feany, M. B., Mattiace, L. A. & Dickson, D. W. (1996). Neuropathologic overlap of progressive supranuclear palsy, Pick’s disease and corticobasal degeneration. J Neuropathol Exp Neurol, 55: 53–67. Forno, L. S., Eng, L. F. & Selkoe, D. J. (1989). Pick bodies in the locus ceruleus. Acta Neuropathol, 79: 10–17. Foster, N. L., Gilman, S., Berent, S., Morin, E. M., Brown, M. B. & Koeppe, R. A. (1988). Cerebral hypometabolism in progressive supranuclear palsy studied with positron emission tomography. Ann Neurol, 24: 399–406. Gazeley, S. & Maguire, J. A. (1996). Familial progressive supranuclear palsy. Clin Neuropathol, 15: 215–20. Gibb, W. R. G., Luthert, P. J. & Marsden, C. D. (1989). Corticobasal degeneration. Brain, 112: 1171–92. Gimenez-Roldan, S., Mateo, D., Benito, C., Grandas, F. & PerezGilabert, Y. (1994). Progressive supranuclear palsy and corticobasal ganglionic degeneration: differentiation by clinical features and neuroimaging techniques. J Neural Transm Suppl, 42: 79–90. Goedert, M., Spillantini, M. G., Jakes, R., Rutherford, D. & Crowther, R. A. (1989a). Multiple isoforms of human microtubuleassociated protein tau: sequences and localization in neurofibrillary tangles of Alzheimer’s disease. Neuron, 3: 519–26. Goedert, M., Spillantini, M. G., Potier, M. C., Ulrich, J. & Crowther, R. A. (1989b). Cloning and sequencing of the cDNA encoding an isoform of microtubule-associated protein tau containing four tandem repeats: differential expression of tau protein mRNAs in human brain. EMBO J, 8: 393–9. Goedert, M., Spillantini, M. G., Jakes, R. et al. (1995). Molecular dissection of the paired helical filament. Neurobiol Aging, 16: 325–34. Goedert, M., Spillantini, M. G., Crowther, R. A. et al. (1999). Tau gene mutation in familial progressive subcortical gliosis. Nature Med, 5: 454–7. Golbe, L. I. (1993). Progressive supranuclear palsy. In Stern, M. B., Koller, W. C. (eds) Parkinsonian Syndromes, Marcel Dekker, New York, pp. 227–47. Golbe, L. I., Davis, P. H., Schoenberg, B. S. & Duvoisin, R. C. (1988). Prevalence and natural history of progressive supranuclear palsy. Neurology, 38: 1031–4. Graff-Radford, N. R., Damasio, A. R., Hyman, B. T. et al. (1990). Progressive aphasia in a patient with Pick’s disease: a neuropsychological radiologic and anatomic study. Neurology, 40: 620–6. Grafman, J., Litvan, I., Gomez, C. & Chase, T. N. (1990). Frontal lobe function in progressive supranuclear palsy. Arch Neurol, 47: 553–8. Hansen, L. A., DeTeresa, R., Tobias, H., Alford, M. & Terry, R. D. (1988). Neocortical morphometry and cholinergic neurochemistry in Pick’s disease. Am J Pathol, 131: 507–18. Hauw, J. -J., Verny, M., Delaere, P., Cervera, P., He, Y. & Duyckaerts, C. (1990). Constant neurofibrillary changes in the neocortex in progressive supranuclear palsy: basic differ-

ences with Alzheimer’s disease and aging. Neurosci Lett, 119: 182–6. Heston, L. L., White, J. A. & Mastri, A. R. (1987). Pick’s disease: clinical genetics and natural history. Arch Gen Psychiatry, 44: 409–11. Higgins, J. J., Adler, R. L. & Loveless, J. M. (1999). Mutational analysis of the tau gene in progressive supranuclear palsy. Neurology, 53: 1421–4. Hirsch, E. C., Graybiel, A. M., Duyckaerts, C. & Javoy-Agid, F. (1987). Neuronal loss in the pedunculopontine tegmental nucleus in Parkinson disease and in progressive supranuclear palsy. Proc Natl Acad Sci USA, 84: 5976–80. Hoenicka, J., Perez, M., Perez-Tur, J. et al. (1999). The tau gene A0 allele and progressive supranuclear palsy. Neurology, 53: 1219–25. Hof, P. R., Delacourte, A. & Bouras, C. (1992). Distribution of cortical neurofibrillary tangles in progressive supranuclear palsy: a quantitative analysis of six cases. Acta Neuropathol, 84: 45–51. Hof, P. R., Bouras, C., Perl, D. P. & Morrison, J. H. (1994). Quantitative neuropathologic analysis of Pick’s disease cases: cortical distribution of Pick bodies and coexistence with Alzheimer’s disease. Acta Neuropathol, 87: 115–24. Hollister, R. D., Xia, M., McNamara, M. J. & Hyman, B. T. (1997). Neuronal expression of class II major histocompatibility complex (HLA-DR) in 2 cases of Pick disease. Arch Neurol, 54: 243–8. Houlden, H., Baker, M., Morris, H. R. et al. (2001). Corticobasal degeneration and progressive supranuclear palsy share a common tau haplotype. Neurology, 56: 1702–6. Hutton, M., Lendon, C. L., Rizzu, P. et al. (1998). Association of missense and 5’-splice-site mutations in tau with the inherited dementia (FTDP-17). Nature, 393: 702–5. Ikeda, K., Akiyama, H., Haga, C., Kondo, H., Arima, K. & Oda, T. (1994). Argyrophilic thread-like structure in corticobasal degeneration and supranuclear palsy. Neurosci Lett, 174: 157–9. Ikeda, K., Akiyama, H., Kondo, H. & Haga, C. (1995). A study of dementia with argyrophilic grains Possible cytoskeletal abnormality in dendrospinal portion of neurons and oligodendroglia. Acta Neuropathol, 89: 409–14. Ikeda, K., Akiyama, H., Arai, T., Matsushita, M., Tsuchiya, K. & Miyazaki, H. (2000). Clinical aspects of argyrophilic grain disease. Clin Neuropathol, 19: 278–84. Ishizawa, K., Lin, W. -L., Tiseo, P., Honer, W. G., Davies, P. & Dickson, D. W. (2000). A qualitative and quantitative study of grumose degeneration in progressive supranuclear palsy. J Neuropathol Exp Neurol, 59: 513–24. Ishizawa, T., Ko, L. -W., Cookson, N., Davies, P., Espinoza, M. & Dickson, D. W. (2002). Selective neurofibrillary degeneration of the hippocampal CA2 sector is associated with four-repeat tauopathies. J Neuropathol Exp Neurol, 61: 1040–7. Iwatsubo, T., Hasegawa, M. & Ihara, Y. (1994). Neuronal and glial tau-positive inclusions in diverse neurologic diseases share common phosphorylation characteristics. Acta Neuropathol, 88: 129–36. Izumiyama, Y., Ikeda, K. & Oyanagi, S. (1994). Extracellular or ghost Pick bodies and their lack of tau immunoreactivity: A


histological immunohistochemical and electron microscopic study. Acta Neuropathol, 87: 277–83. Jellinger, K. (1971). Progressive supranuclear palsy (subcortical argyrophilic dystrophy). Acta Neuropathol, 19: 347–52. Jellinger, K. A. & Bancher, C. (1992). Neuropathology. In Litvan I. & Agid Y. (eds) Progressive Supranuclear Palsy: Clinical and Research Approaches, Oxford University Press, New York, pp. 44–88. Jicha, G. A., Rockwood, J. M., Berenfeld, B., Hutton, M. & Davies, P. (1999). Altered conformation of recombinant frontotemporal dementia-17 mutant tau proteins. Neurosci Lett, 260: 153–6. Johnson, J. K., Head, E., Kim, R., Starr, A. & Cotman, C. W. (1999). Clinical and pathological evidence for a frontal variant of Alzheimer disease. Arch Neurol, 56: 1233–9. Juncos, J. L., Hirsch, E. C., Malessa, S., Duyckaerts, C., Hersh, L. B. & Agid, Y. (1991). Mesencephalic cholinergic nuclei in progressive supranuclear palsy. Neurology, 41: 25–30. Kamo, H., McGeer, P. L. & Harrop, R. et al. (1987). Positron emission tomography and histopathology in Pick’s disease. Neurology, 37: 439–45. Kanazawa, I., Kwak, S., Sasaki, H. et al. (1988). Studies on neurotransmitter markers of the basal ganglia in Pick’s disease with special reference to dopamine reduction. J Neurol Sci, 83: 63–74. Kato, S. & Nakamura, H. (1990). Presence of two different fibril subtypes in the Pick body: an immunoelectron microscopic study. Acta Neuropathol, 81: 125–9. Kato, S., Hirano, A., Umahara, T., Kato, M., Herz, F. & Ohama, E. (1992). Comparative immunohistochemical study on the expression of alpha B crystallin, ubiquitin, and stress-response protein 27 in ballooned neurons in various disorders. Neuropathol Appl Neurobiol, 18: 335–40. Kertesz, A. & Munoz, D. G. (2000). Diagnostic controversies: Is CBD part of the ‘Pick complex’? Adv Neurol, 82: 223–31. Kertesz, A., Hudson, L., MacKenzie, I. R. & Munoz, D. G. (1994). The pathology and nosology of primary progressive aphasia. Neurology, 44: 2065–72. Kimura, T., Ikeda, K., Takamatsu, J., Miyata, T., Sobue, G., Miyakawa, T. & Horiuchi, S. (1996). Identification of advanced glycation end products of the Maillard reaction in Pick’s disease. Neurosci Lett, 219: 95–8. Knopman, D. S., Mastri, A. R., Frey, F., Sung, J. H. & Rustan, T. (1990). Dementia lacking distinctive histologic features: A common non-Alzheimer degenerative dementia. Neurology, 40: 251–6. Komori, T. (1999). Tau-positive glial inclusions in progressive supranuclear palsy, corticobasal degeneration and Pick’s disease. Brain Pathol, 9: 663–79. Kosaka, K., Ikeda, K., Kobayashi, K. & Mehraein, P. (1991). Striatopallidonigral degeneration in Pick’s disease: a clinicopathological study of 41 cases. J Neurol, 238: 151–60. Ksiezak-Reding, H., Morgan, K., Mattiace, L. A. et al. (1994). Ultrastructure and biochemical composition of paired helical filaments in corticobasal degeneration. Am J Pathol, 145: 1496–508.

Lang, A. E., Bergeron, C., Pollanen, M. S. & Ashby, P. (1994). Parietal Pick’s disease mimicking cortical-basal ganglionic degeneration. Neurology, 44: 1436–40. Lee, V. M., Goedert, M. & Trojanowski, J. Q. (2001). Neurodegenerative tauopathies. Ann Rev Neurosci, 24: 1121–59. Leenders, K. L., Frackowiak, R. S. J. & Lees, A. J. (1988). Steele– Richardson–Olszewski syndrome: brain energy metabolism blood flow and fluorodopa uptake measured by positron emission tomography. Brain, 111: 615–30. Litvan, I., Hauw, J. J., Bartko, J. J. et al. (1996). Validity and reliability of the preliminary NINDS neuropathologic criteria for progressive supranuclear palsy and related disorders. J Neuropathol Exp Neurol, 55: 97–105. Litvan, I., Agid, Y., Sastry, N. et al. (1997). What are the obstacles for an accurate clinical diagnosis of Pick’s disease? A clinicopathologic study. Neurology, 49: 62–9. Litvan, I., Grimes, D. A. & Lang, A. E. (2000). Phenotypes and prognosis: clinicopathologic studies of corticobasal degeneration. In Litvan I., Goetz C. G, Lang, A. E. (eds). Corticobasal Degeneration and Related Disorders, Advances in Neurology, Volume 82. Lippincott, Williams & Wilkins, Philadelphia, pp. 183–96. Liu, W. -K., Le, T. V., Adamson, J. et al. (2001). Relationship of the extended tau haplotype to tau biochemistry and neuropathology in progressive supranuclear palsy. Ann Neurol, 50: 494–502. Love, S., Saitoh, T., Quijada, S., Cole, G. M. & Terry, R. D. (1988). Alz-50, ubiquitin and tau immunoreactivity of neurofibrillary tangles Pick bodies and Lewy bodies. J Neuropathol Exp Neurol, 47: 393–405. Lowe, J., Blanchard, A., Morrell, K. et al. (1988). Ubiquitin is a common factor in intermediate filament inclusion bodies of diverse type in man including those of Parkinson’s disease, Pick’s disease, and Alzheimer’s disease as well as Rosenthal fibres in cerebellar astrocytomas, cytoplasmic bodies in muscle, and Mallory bodies in alcoholic liver disease. J Pathol, 155: 9–15. Lowe, J., Errington, D. R. & Lennox, G. et al. (1992). Ballooned neurons in several neurodegenerative diseases and stroke contain alpha B crystallin. Neuropathol Appl Neurobiol, 18: 341–50. Maher, E. R., Smith, E. M. & Lees, A. J. (1985). Cognitive deficits in the Steele–Richardson–Olszewski syndrome (progressive supranuclear palsy). J Neurol Neurosurg Psychiatry, 48: 1234–9. Marshall, E. F., Perry, R. H., Perry, E. K. et al. (1997). Striatal dopaminergic loss without parkinsonism in a case of corticobasal degeneration. Acta Neurol Scand, 95: 287–92. Martinez-Lage, P. & Munoz, D. G. (1997). Prevalence and disease associations of argyrophilic grains of Braak. J Neuropathol Exp Neurol, 56: 157–64. Mendez, M. F., Selwood, A., Mastri, A. R. & Frey, W. H. II (1993). Pick’s disease versus Alzheimer’s disease: a comparison of clinical characteristics. Neurology, 43: 289–92. Minthon, L., Hesse, C., Sjogren, M., Englund, E., Gustafson, L. & Blennow, K. (1997). The apolipoprotein E epsilon4 allele frequency is normal in fronto-temporal dementia, but correlates with age at onset of disease. Neurosci Lett, 226: 65–7.

253

254

D. W. Dickson

Mizukami, K. & Kosaka, K. (1989). Neuropathological study on the nucleus basalis of Meynert in Pick’s disease. Acta Neuropathol, 78: 52–6. Montine, K. S., Kim, P. J., Olson, S. J., Markesbery, W. R. & Montine, T. J. (1997). 4-hydroxy-2-nonenal pyrrole adducts in human neurodegenerative disease. J Neuropathol Exp Neurol, 56: 866–71. Mori, H., Nishimura, M., Namba, Y. & Oda, M. (1994). Corticobasal degeneration: a disease with widespread appearance of abnormal tau and neurofibrillary tangles and its relation to progressive supranuclear palsy. Acta Neuropathol, 88: 113–21. Morris, H. R., Janssen, J. C. Bandmann, O. et al.(1999). The tau gene A0 polymorphism in progressive supranuclear palsy and related neurodegenerative diseases. J Neurol Neurosurg Psychiatry, 66: 665–7. Morris, H. R., Khan, M. N. & Janssen, J. C. (2001). The genetic and pathological classification of familial frontotemporal dementia. Arch Neurol, 58: 1813–16. Morris, H. R., Katzenschlager, R., Janssen, J. C. et al. (2002). Sequence analysis of tau in familial and sporadic progressive supranuclear palsy. J Neurol Neurosurg Psychiatry, 72: 388–90. Murayama, S., Mori, H., Ihara, Y. & Tomonaga, M. (1990). Immunocytochemical and ultrastructural studies of Pick’s disease. Ann Neurol, 27: 394–405. Murrell, J. R., Spillantini, M. G., Zolo, P. et al. (1999). Tau gene mutation G389R causes a tauopathy with abundant pick bodylike inclusions and axonal deposits. J Neuropathol Exp Neurol, 58:1207–26. Nagasawa, H., Tanji, H., Nomura, H. et al. (1996). PET study of cerebral glucose metabolism and fluorodopa uptake in patients with corticobasal degeneration. J Neurol Sci, 139: 210–17. Nath, U., Ben-Shlomo, Y., Thomson, R. G. et al. (2001). The prevalence of progressive supranuclear palsy (Steele–Richardson– Olszewski syndrome) in the UK. Brain, 124: 1438–49. Neary, D., Snowden, J. S. & Shields, R. A. (1987). Single photon emission tomography using 99mTc-HC-PAO in the investigation of dementia. J Neurol Neurosurg Psychiatry, 50: 1101–9. Neary, D., Snowden, J. S., Northern, B. & Goulding, P. J. (1988). Dementia of frontal lobe type. J Neurol Neurosurg Psychiatry, 51: 353–61. Neary, D., Snowden, J. S. & Mann, D. M. (1993). The clinical pathological correlates of lobar atrophy. Dementia, 4: 154–9. Nishimura, M., Namba, Y., Ideda, K. & Oda, M. (1992). Glial fibrillary tangles with straight tubules in the brains of patients with progressive supranuclear palsy. Neurosci Lett, 143: 35–8. Oka, M., Katayama, S., Imon, Y., Ohshita, T., Mimori, Y. & Nakamura, S. (2001). Abnormal signals on proton density-weighted MRI of the superior cerebellar peduncle in progressive supranuclear palsy. Acta Neurol Scand, 104: 1–5. Oliva, R., Tolosa, E., Ezquerra, M. et al. (1998). Significant changes in the tau A0 and A3 alleles in progressive supranuclear palsy and improved genotyping by silver detection. Arch. Neurol, 55: 1122–4. Oyanagi, K., Takahashi, H., Wakabayashi, K. & Ikuta, F. (1988). Selective decrease of large neurons in the neostriatum in progressive supranuclear palsy. Brain Res, 458: 218–22.

Oyanagi, K., Tsuchiya, K., Yamazaki, M. & Ikeda, K. (2001). Substantia nigra in progressive supranuclear palsy, corticobasal degeneration, and parkinsonism-dementia complex of Guam: specific pathological features. J Neuropathol Exp Neurol, 60: 393–402. Parisi, J. E., Dickson, D. W., Johnson, K. A. et al. (2000). Neuropathological substrate of mild cognitive impairment. Neurobiol Aging, 21 (Suppl 1): S198. Pascual, J., Berciano, J., Grijalba, B. et al. (1992). Dopamine D1 and D2 receptors in progressive supranuclear palsy: an autoradiographic study. Ann Neurol, 32: 703–7. Pastor, P., Pastor, E., Carnero, C. et al. (2001). Familial atypical progressive supranuclear palsy associated with homozygosity for the delN296 mutation in the tau gene. Ann Neurol, 49: 263–7. Paulus, W., Bancher, C. & Jellinger, K. (1993). Microglial reaction in Pick’s disease. Neurosci Lett, 161: 89–92. Pickering-Brown, S., Baker, M., Yen, S. H. et al. (2000). Pick’s disease is associated with mutations in the tau gene. Ann Neurol, 48: 859–67. Powell, H. C., London, G. W. & Lampert, P. W. (1974). Neurofibrillary tangles in progressive supranuclear palsy: electron microscopic observations. J Neuropathol Exp Neurol, 33: 98–106. Probst, A., Anderton, B. H., Ulrich, J., Kohler, R., Kahn, J. & Heitz, P. U. (1983). Pick’s disease: an immunocytochemical study of neuronal changes. Monoclonal antibodies show that Pick bodies share antigenic determinants with neurofibrillary tangles and neurofilaments. Acta Neuropathol, 60: 175–82. Probst, A., Langui, D., Lautenschlager, C., Ulrich, J., Brion, J. P. & Anderton, B. H. (1988). Progressive supranuclear palsy: extensive neuropil threads in addition to neurofibrillary tangles. Acta Neuropathol, 77: 61–8. Probst, A., Tolnay, M., Langui, D., Goedert, M. & Spillantini, M. (1996). Pick’s disease: hyperphosphorylated tau protein segregates to the somatoaxonal compartment. Acta Neuropathol, 92: 588–96. Rajput, A. H., Offord, K. P., Beard, C. M. & Kurland, L. T. (1991). Epidemiology of parkinsonism: incidence, classification, and mortality. Ann Neurol, 16: 278–82. Rebeiz, J. J., Kolodny, E. H. & Richardson, E. P., Jr (1968). Corticodentatonigral degeneration with neuronal achromasia. Arch Neurol, 18: 20–33. Reed, L. A., Schmidt, M. L., Wszolek, Z. K. et al. (1998). The neuropathology of a chromosome 17-linked autosomal dominant parkinsonism and dementia (‘pallido-ponto-nigral degeneration’). J Neuropathol Exp Neurol, 57: 588–601. Rewcastle, N. B. & Ball, M. J. (1968). Electron microscopic structure of the ‘inclusion bodies’ in Pick’s disease. Neurology, 18: 1205–13. Riley, D. E., Lang, A. E., Lewis, A. et al. (1990). Cortical-basal ganglionic degeneration. Neurology, 40: 1203–12. Rojo, A., Pernaute, R. S. & Fontan, A. (1999). Clinical genetics of familial progressive supranuclear palsy. Brain, 122: 1233–45. Rosso, S. M., van Herpen, E., Deelen, W. et al. (2002). A novel tau mutation, S320F, causes a tauopathy with inclusions similar to those in Pick’s disease. Ann Neurol, 51:373–6.


Ruberg, M., Javoy-Agid, F., Hirsch, E. et al. (1985). Dopaminergic and cholinergic lesions in progressive supranuclear palsy. Ann Neurol, 18: 523–9. Russ, C., Lovestone, S., Baker, M. et al. (2001). The extended haplotype of the microtubule associated protein tau gene is not associated with Pick’s disease. Neurosci Lett, 299: 156–8. Salmon, E. & Franck, G. (1989). Positron emission tomographic study in Alzheimer’s disease and Pick’s disease. Arch Gerontol Geriatr Suppl, 1: 241–7. Sawle, G. V., Brooks, D. J., Marsden, C. D. & Frackowiak, R. S. J. (1991). Corticobasal degeneration. A unique pattern of regional cortical oxygen hypometabolism and striatal fluorodopa uptake demonstrated by positron emission tomography. Brain, 114: 541–56. Schmidt, M. L., Lee, V. M.-Y., Hurtig, H. & Trojanowski, J. Q. (1988). Properties of antigenic determinants that distinguish neurofibrillary tangles in progressive supranuclear palsy and Alzheimer’s disease. Am J Pathol, 59: 460–6. Schmidt, M. L., Huang, R., Martin, J. A. et al. (1996). Neurofibrillary tangles in progressive supranuclear palsy contain the same tau epitopes identified in Alzheimer’s disease PHF-tau. J Neuropathol Exp Neurol, 55: 534–9. Seno, H., Kobayashi, S., Inagaki, T. et al. (2000). Parkinson’s disease associated with argyrophilic grains clinically resembling progressive supranuclear palsy: an autopsy case. J Neurol Sci, 178: 70–4. Sergeant, N., David, J. P., Lefranc, D., Vermersch, P., Wattez, A. & Delacourte, A. (1997). Different distribution of phosphorylated tau protein isoforms in Alzheimer’s and Pick’s diseases. FEBS Lett, 412: 578–82. Singhrao, S. K., Neal, J. W., Gasque, P., Morgan, B. P. & Newman, G. R. (1996). Role of complement in the aetiology of Pick’s disease? J Neuropathol Exp Neurol, 55: 578–93. Soliveri, P., Monza, D., Paridi, D. et al. (1999). Cognitive and magnetic resonance imaging aspects of corticobasal degeneration and progressive supranuclear palsy. Neurology, 53: 502–7. Sparks, D. L. & Markesbery, W. R. (1991). Altered serotonergic and cholinergic synaptic markers in Pick’s disease. Arch Neurol, 48: 796–9. Spillantini, M. G. & Goedert, M. (1998). Tau protein pathology in neurodegenerative diseases. Trends Neurosci, 21: 428–33. Spillantini, M. G., Goedert, M., Crowther, R. A., Murrell, J. R., Farlow, M. R. & Ghetti, B. (1997). Familial multiple system tauopathy with presenile dementia: a disease with abundant neuronal and glial tau filaments. Proc Natl Acad Sci (USA), 94: 4113–18. Spillantini, M. G., Murrell, J. R., Goedert, M., Farlow, M. R., Klug, A. & Ghetti, B. (1998). Mutation in the tau gene in familial multisystem tauopathy with presenile dementia. Proc Natl Acad Sci (USA), 95: 7737–41. Stanford, P. M., Halliday, G. M., Brooks, W. S. et al. (2000). Progressive supranuclear palsy pathology caused by a novel silent mutation in exon 10 of the tau gene: expansion of the disease phenotype caused by tau gene mutations. Brain, 123: 880–93.

Steele, J. C., Richardson, J. C. & Olszewski, J. (1964). Progressive supranuclear palsy: a heterogenous degeneration involving the brainstem, basal ganglia and cerebellum with vertical gaze and pseudobulbar palsy, nuchal dystonia, and dementia. Arch Neurol, 10: 333–9. Tagliavini, F. & Pilleri, G. (1983). Basal nucleus of Meynert: a neuropathological study in Alzheimer’s disease, simple senile dementia, Pick’s disease, and Huntington’s chorea. J Neurol Sci, 62: 243–60. Tagliavini, F., Pilleri, G., Bouras, C. & Constantinidis, J. (1984). The basal nucleus of Meynert in patients with progressive supranuclear palsy. Neurosci Lett 44: 37–42. Takahashi, T., Amano, N. & Hanihara, T. (1996). Corticobasal degeneration: widespread argentophilic threads and glia in addition to neurofibrillary tangles. Similarities of cytoskeletal abnormalities in corticobasal degeneration and progressive supranuclear palsy. J Neurol Sci, 138: 66–77. Takauchi, S., Hosomi, M., Marasigan, S., Sato, M., Hayashi, S. & Miyoshi, K. (1984). An ultrastructural study of Pick bodies. Acta Neuropathol, 64: 344–8. Tellez-Nagel, I., Wisniewski, H. M. (1973). Ultrastructure of neurofibrillary tangles in Steele-Richardson-Olszewski syndrome. Arch Neurol, 29: 324–7. Togo, T. & Dickson, D. W. (2002a). Tau accumulation in astrocytes in progressive supranuclear palsy is a degenerative rather than a reactive process. Acta Neuropath, 104: 53–56. Togo, T. & Dickson, D. W. (2002b). Ballooned neurons in progressive supranuclear palsy are usually due to concurrent argyrophilic grain disease. Acta Neuropathol, 104: 398–402. Togo, T., Cookson, N. & Dickson, D. W. (2002a). Argyrophilic grain disease: Neuropathology, frequency in a dementia brain bank and lack of relationship with apolipoprotein E. Brain Pathol, 12: 45–52. Togo, T., Sahara, N., Yen, S.-H. et al. (2002b). Argyrophilic grain disease is a sporadic 4-repeat tauopathy. J Neuropathol Exp Neurol, 61: 547–56. Tolnay, M., Spillantini, M. G., Goedert, M., Ulrich, J., Langui, D. & Probst, A. (1997). Argyrophilic grain disease: widespread hyperphosphorylation of tau protein in limbic neurons. Acta Neuropathol, 93: 477–84. Tolnay, M., Mistl, C., Ipsen, S. & Probst, A. (1998). Argyrophilic grains of Braak: occurrence in dendrites of neurons containing hyperphosphorylated tau protein. Neuropathol Appl Neurobiol, 24: 53–9. Trojanowski, J. Q. & Dickson, D. (2001). Update on the neuropathological diagnosis of frontotemporal dementias. J Neuropathol Exp Neurol, 60: 1123–6. Uchihara, T., Tsuchiya, K. & Kosaka, K. (1990). Selective loss of nigral neurons in Pick’s disease: a morphometric study. Acta Neuropathol, 81: 155–61. Uchihara, T., Mitani, K., Mori, H., Kondo, H., Yamada, M. & Ikeda, K. (1994). Abnormal cytoskeletal pathology peculiar to corticobasal degeneration is different from that of Alzheimer’s disease or progressive supranuclear palsy. Acta Neuropathol, 88: 379–83.

255

256

D. W. Dickson

Uhl, G. R., Hilt, D. C., Hedreen, J. C., Whitehouse, P. J. & Price, D. L. (1983). Pick’s disease (lobar sclerosis): depletion of neurons in the nucleus basalis of Meynert. Neurology, 33: 1470–3. Ulrich, J., Haugh, M., Anderton, B. H., Probst, A., Lautenschlager, C. & His, B. (1987). Alzheimer dementia and Pick’s disease: Neurofibrillary tangles and Pick bodies are associated with identical phosphorylated neurofilament epitopes. Acta Neuropathol, 73: 240–6. Vermersch, P., Robitaille, Y., Berneir, L., Wattez, A., Gauvreau, D. & Delacourte, A. (1994). Biochemical mapping of neurofibrillary degeneration in a case of progressive supranuclear palsy: evidence for general cortical involvement. Acta Neuropathol, 87: 572–7. Wakabayashi, K., Oyanagi, K., Makifuchi, T. et al. (1994). Corticobasal degeneration: etiopathological significance of the cytoskeletal alterations. Acta Neuropathol, 87: 545– 53. Wechsler, A. F., Verity, M. A., Rosenschein, S., Fried, I. & Scheibel, A. B. (1982). Pick’s disease: a clinical computed tomographic and histologic study with Golgi impregnation observations. Arch Neurol, 39: 287–90. Weiler, R., Lassmann, H., Fischer, P., Jellinger, K. & Winkler, H. (1990). A high ratio of chromogranin A to synaptin/ synaptophysin is a common feature of brains in Alzheimer and Pick disease. FEBS Lett, 263: 337–9.

Wisniewski, H. M., Coblentz, J. M. & Terry, R. D. (1972). Pick’s disease: a clinical and ultrastructural study. Arch Neurol, 26: 97–108. Wolozin, B. & Davies, P. (1987). Alzheimer-related neuronal protein A-68: specificity and distribution. Ann Neurol, 22: 521–6. Wood, P. L., Etienne, P., Lal, S. et al. (1983). A post-mortem comparison of the cortical cholinergic system in Alzheimer’s disease and Pick’s disease. J Neurol Sci, 62: 211–17. Yamada, T. & McGeer, P. L. (1995). Oligodendroglial microtubular masses: an abnormality observed in some human neurodegenerative diseases. Neurosci Lett, 120: 163–6. Yamada, T., Calne, D. B., Akiyama, H., McGeer, E. G. & McGeer, P. L. (1993). Further observations on tau-positive glia in the brains with progressive supranuclear palsy. Acta Neuropathol, 85: 308–15. Yamauchi, H., Fukuyama, H., Nagahama, Y. et al. (1998). Atrophy of the corpus callosum, cortical hypometabolism, and cognitive impairment in corticobasal degeneration. Arch Neurol, 55: 609–14. Yasuhara, O., Kawamata, T., Aimi, Y., McGeer, E. G. & McGeer, P. L. (1994). Expression of chromogranin A in lesions in the central nervous system from patients with neurological diseases. Neurosci Lett, 170: 13–16. Yoshimura, N. (1989). Topography of Pick body distribution in Pick’s disease: a contribution to understanding the relationship between Pick’s and Alzheimer’s diseases. Clin Neuropathol, 8: 1–6.

12 Hereditary tauopathies and idiopathic frontotemporal dementias Mark S. Forman, John Q. Trojanowski, and Virginia M.-Y. Lee Center for Neurodegenerative Disease Research, University of Pennsylvania, Philadelphia, PA, USA

List of abbreviations A, -amyloid; AD, Alzheimer’s disease; ALS, amyotrophic lateral sclerosis; CBD, corticobasal degeneration; CHO, Chinese hamster ovary; CNS, central nervous system; ESE, exon-splicing enhancer; ESS, exon-splicing silencer; FTD, frontotemporal dementia; FTDP-17, frontotemporal dementia with parkinsonism linked to chromosome 17; FTLD, frontotemporal lobar degeneration; IHC, immunohistochemistry; MAP, microtubule-associated protein; MND, motor neuron disease; MT, microtubule; NFT, neurofibrillary tangle; PD, Parkinson’s disease; PDC, parkinsonism-dementia complex; PHF, paired helical filament; PiD, Pick’s disease; PSG, progressive subcortical gliosis; PSP, progressive supranuclear palsy; SF, straight filament; TG, transgenic; WT, wild-type.

Introduction In 1892, Arnold Pick described a woman with lobar brain atrophy who presented clinically with presenile dementia and aphasia (Pick, 1892), and thus, the first description of what is now classified clinically as frontotemporal dementia (FTD) (Neary et al., 1998; McKhann et al., 2001). The clinical syndromes of FTD are associated with several neuropathological abnormalities (Lund and Manchester Groups, 1994; McKhann et al., 2001). A subset of these disorders is characterized by the intracellular accumulations of filamentous material composed of the microtubule-associated protein (MAP) tau. The term ‘tauopathies’ was coined to refer to this seemingly heterogeneous group of neurodegenerative disorders that

includes Pick’s disease (PiD), corticobasal degeneration (CBD), progressive supranuclear palsy (PSP), neurofibrillary tangle dementia, and argyrophilic grain disease (see ‘Sporadic tauopathies’, Chapter 11). In FTD patients, there is a family history of a similar dementing illness in approximately 38 to 50% of patients (Knopman et al., 1990b; Stevens et al., 1998; Chow et al., 1999). In 1998, pathogenic mutations in the MAP tau were identified in a heterogeneous group of families with FTD and autosomal dominant inheritance thus directly implicating tau in the pathogenesis of tauopathies (Poorkaj et al., 1998; Hutton et al., 1998; Spillantini et al., 1998c). This familial disorder was termed frontotemporal dementia with parkinsonism linked to chromosome 17 (FTDP-17) (Foster et al., 1997). Furthermore, the extensive tau pathology observed in amyotrophic lateral sclerosis (ALS)/parkinsonism–dementia complex (PDC) that afflicts the Chamorro population on the island of Guam suggests that environmental factors interact with genetics in the pathogenesis of tauopathies (Kurland and Mulder, 1955a,b; Hirano et al., 1961a; Plato et al., 1969). However, a second group of patients with FTD lacks filamentous tau inclusions. Several distinct patterns of pathology are observed in these individuals. The first pattern is characterized by the presence of ubiqitinated filamentous inclusions similar to that observed in motor neuron disease (MND) and is referred to as frontotemporal lobar degeneration with motor neuron disease-type inclusions (FTLDMND, also known as motor neuron disease-inclusion dementia) (Okamoto et al., 1991; Wightman et al., 1992; Jackson et al., 1996). Clinically, these patients often manifest MND although it is not present in all patients. A second group named progressive subcortical gliosis (PSG)


257

258

M. S. Forman, J. Q. Trojanowski and V. M.-Y. Lee

is characterized by cerebral atrophy, predominantly affecting the white matter of the frontal and temporal cortex associated with prominent astrocytosis (Neumann & Cohn, 1967). However, some of these patients may demonstrate abundant filamentous tau pathology similar to the tauopathies described above (Goedert et al., 1999b). A third group of FTD patients show non-specific pathology including degeneration of frontal and temporal lobes and lacking filamentous inclusions (Brun, 1987; Neary et al., 1988; Knopman et al., 1990a). Finally, Alzheimer’s disease (AD) and other neurodegenerative diseases such as dementia with Lewy bodies may manifest an FTD-like clinical picture. However, the specific type of pathological findings in all of these disorders does not correlate with the particular clinical manifestations. In the previous chapter, sporadic tauopathies were described including many that lead to the clinical phenotype of FTD. This chapter begins with a description of the neuropathology of disorders that present clinically with FTD in the absence of filamentous tau pathology. This will be followed by a discussion of familial disorders with FTD focusing on FTDP-17 including the effects of the various mutations identified in tau. Lastly, Guam ALS/PDC will be described, a unique tauopathy that implicates both environmental and genetic factors in the pathogenesis of the disease.

Clinical features of FTD FTD is the most common form of primary neurodegenerative disorder after AD accounting for approximately 20% of patients with presenile dementia. The disorder affects both sexes equally presenting between the ages of 35 and 75 with duration of illness ranging from 2 to 20 years. FTD patients present with 2 distinct clinical patterns characterized as either a gradual and progressive change in behaviour or a gradual and progressive language dysfunction (Lund & Manchester Groups, 1994; Neary et al., 1998; McKhann et al., 2001). The most common presentation of FTD is an early change in social and personal conduct with relative preservation of memory. These features are often associated with disinhibition resulting in impulsive, inappropriate, or compulsive behaviour, as well as lack of insight into his or her actions. Alternatively, FTD patients will present with a language disorder, usually classified as either a primary progressive aphasia or a semantic dementia that invariably progresses to mutism. Neurological signs are typically absent early in the disease; however, parkinsonism, particularly akinesia and rigidity, often develops with disease progression. A minority of

patients will also develop neurological signs of concurrent MND. Laboratory studies including routine electroencephalography are normal in the early stages of FTD. In contrast, structural imaging studies may show atrophy of the anterior temporal and frontal lobes that is occasionally asymmetrical. Functional imaging studies including single-photon emission computed tomography, positron emission tomography, and perfusion magnetic resonance imaging are more sensitive to such changes, often demonstrating decreased perfusion of frontal and temporal lobes. However, the clinical diagnosis of FTD remains one of exclusion, and other potential medical causes such as neuropsychiatric disorders, systemic conditions, and toxins must be ruled out.

Frontotemporal lobar degeneration with motor neuron disease-type inclusions In 1929, Meyer first noted an association between MND and dementia (Meyer, 1929). Since that time, this combination of clinical signs and symptoms has been increasingly recognized (Mitsuyama & Takamiya, 1979; Neary et al., 1990). Patients typically present with FTD while neurological signs of MND appear later. However, occasionally the dementia and physical symptoms present concurrently (Talbot, 1996; Mitsuyama, 2000; Nakano, 2000). In some instances FTD occurs in the absence of clinical signs of MND despite the presence of the characteristic pathological changes in the spinal cord (Jackson et al., 1996). Currently, FTLDMND, also referred to as ‘motor neuron disease-inclusion dementia’, is one of the more common neuropathological diagnoses in patients with FTD. However, the relationship between FTD patients with and without MND, as well as MND alone, remains unclear.

Neuropathology of FTLD-MND The gross pathological changes of FTD are relatively similar in FTD regardless of the underlying histopathology (Lund and Manchester Groups, 1994; Mann, 1998). In FTLDMND, there is variable atrophy predominantly affecting the frontal, anterior temporal, and anterior parietal lobes with relative sparing of the posterior third of the temporal lobe including the posterior superior temporal gyrus (Fig. 12.1). Upon coronal sectioning, there is mild to moderate dilatation of the lateral ventricles with thinning of the cortical ribbon in regions with prominent atrophy. There is often depigmentation of the substantia nigra, while the remainder of the brain stem, the hippocampus, and subcortical

Hereditary tauopathies

Fig. 12.1. Pathology in FTLD-MND and FTLD. (A) Brain from a patient with FTD with histopathology diagnostic of FTLD-MND. This lateral view of the right hemisphere shows moderate frontal and temporal lobe atrophy with relative sparing of the parietal and occipital lobes. (B) Neuron loss and spongiosis in upper cortical lamina of frontal cortex of patient with FTLD. Note the deeper cortical layers are relatively preserved. These pathologic changes are non-specific and are observed in most patients with FTD (hematoxylin and eosin; H&E). (C) FTLD-MND brain with with numerous ubiquitin positive cytoplasmic inclusions in granule cells of the fascia dentata. Although morphologically similar to Pick bodies, these inclusions are not immunoreactive with antibodies to the MAP tau. (D ) Spinal cord from FTLD-MND patient with prominent neurologic signs of MND. There is severe neuron loss in the anterior horn of the cervical spinal cord. One of the remaining motor neurons contains a pale eosinophilic inclusion (arrowhead). Inset shows high power magnification of this motor neuron (H&E). (E) High power magnification of ubiquitin positive inclusions in granular neurons of fascia dentata. (F ) Ubiquitin positive neurites and cytoplasmic inclusions in layer II of the frontal cortex from a patient with FTLD-MND. (G) Ubiquitin positive inclusions in motor neurons of anterior horn from a patient with FTLD-MND (C and E–G, IHC with anti-ubiquitin antibodies)

259

260


nuclei are grossly normal. In addition, the anterior roots of the spinal cord are often atrophic in patients with prominent MND (Neary et al., 1990). Histologically, there is superficial microvacuolation and gliosis with associated neuron loss in affected cortices predominantly in layers I to III with relative preservation of the deeper cortical layers (Fig. 12.1) (Morita et al., 1987; Neary et al., 1990). Occasionally, there is severe neuronal loss with transcortical gliosis and status spongiosis. In many patients, there is also loss of pigmented neurons in the substantia nigra, while the basal ganglia are variably affected. In contrast to other forms of FTD, there is loss of motor neurons in the spinal cord and brain stem with relative preservation of Clarke’s column and the intermediolateral nucleus, which are features similar to that observed in classic MND (Fig. 12.1). Residual motor neurons show prominent chromatolysis and/or pale eosinophilic inclusions. The changes in motor neurons are generally more prominent in those patients with clinically advanced MND (Jackson et al., 1996). By histochemical and immunohistochemical analysis, there are no senile plaques, amyloid angiopathy, tauimmunoreactive neurofibrillary lesions, or -synuclein pathology in FTLD-MND. In contrast, the defining histological feature is the presence of cytoplasmic neuronal inclusions in the granule cells of the dentate gyrus and the superficial layers of the neocortex, most prominently in the external granular layer (Fig. 12.1) (Okamoto et al., 1991; Wightman et al., 1992; Jackson et al., 1996). These inclusions are similar morphologically to Pick bodies, except they are immunoreactive with antibodies to ubiquitin while nonreactive with antibodies for tau and are not detected with conventional silver stains. Ubiquitin-immunoreactive dystrophic neurites are also detected in affected neocortices. Ultrastructurally, the inclusions are composed of filaments 10 to 15 nm in diameter (Kinoshita et al., 1997; Iseki et al., 1998). In addition, there are often ubiquitinated inclusions within motor neurons of the anterior horns of the spinal cord and brainstem that are identical to those observed in classic MND (Fig. 12.1) (Leigh et al., 1988; Lowe et al., 1988).

Relationship between FTLD-MND and classical MND While the relationship between FTLD-MND and classical MND remains uncertain, there are numerous similarities between the two disorders. In classic MND, there is often frontal lobe dysfunction in later stages of the illness despite the absence of cortical or hippocampal pathology (Montgomery & Erickson, 1987; Cooper et al., 1995). The ubiquitinated inclusions in the brain and spinal cord of FTLD-MND

patients are similar to those identified in the spinal cord of classic MND patients (Okamoto et al., 1991; Wightman et al., 1992; Jackson et al., 1996). Furthermore, several families with FTD-MND were described in which affected individuals clinically manifest FTD alone, MND alone, or both (Gunnarsson et al., 1991; Mann et al., 1993; Jackson et al., 1996). Thus, it is possible that FTLD-MND and classic MND represent different patterns of expression of a common disease process with an overlapping topographic distribution of pathology. The identification of mutations and/or polymorphisms in kindreds with FTLD-MND will provide insight into the pathogenesis of the disorder. However, it remains unclear whether FTLD-MND represents a primary disorder in ubiquitination or pathology of a distinct, as of yet unidentified protein, that is ubiquitinated during the process of aggregation.

Frontotemporal lobar degeneration In 1974, Constantinidis and colleagues subclassified PiD into several categories depending on the presence or absence of Pick bodies and/or ballooned neurons (Constantinidis et al., 1974). More recent neuropathological criteria for the classification of FTD require Pick bodies for the diagnosis of Pick’s disease (Lund and Manchester Groups, 1994; McKhann et al., 2001). A subset of these FTD patients (Constantinidis Group C) lacks filamentous inclusions within either neurons or glia. This subset, which accounts for up to 60% of patients with FTD, has received a variety of names including dementia of frontal lobe type, frontal lobe degeneration of non-Alzheimer type, and dementia lacking distinctive histopathology (Brun, 1987; Neary et al., 1988; Knopman et al., 1990a). A recent report from the ‘Work Group on FTD and PiD’ settled upon the name ‘frontal temporal lobar degeneration’ (FTLD) (McKhann et al., 2001).

Neuropathology of FTLD The macroscopic changes, including frontal and temporal lobe atrophy and variable depigmentation of the substantia nigra, are the same as that observed for FTLD-MND described above, except for the notable absence of atrophy of the anterior roots of the spinal cord (Fig. 12.1). In affected cortices, there is microvacuolation and neuron loss principally in layers II and III although occasionally, there is severe, transcortical neuron loss (Fig. 12.1) (Knopman, 1993; Mann et al., 1993; Brun, 1993; Giannakopoulos et al., 1995). There is associated astrocytic gliosis involving subpial regions, the junction between grey and white


matter, and the subcortical u-fibers, while deeper white matter is relatively uninvolved. There is also variable degeneration of the hippocampal formation, basal ganglia, and substantia nigra consisting of neuron loss and gliosis. However, these pathological features are non-specific, and the diagnosis of FTLD is one of exclusion. Specifically, by immunohistochemistry (IHC) there are no (or limited) filamentous inclusions identified with antibodies specific for amyloid, tau, -synuclein, or ubiquitin, and there is no protease-resistant prion protein (Lund and Manchester Groups, 1994; McKhann et al., 2001). However, Zhukareva et al. identified the first unique feature in FTLD patients (Zhukareva et al., 2001). In a subset of these patients, there is the selective loss of tau protein expression but with relative preservation of tau mRNA levels. This suggests that the level of tau protein is controlled post-transciptionally. Moreover, the loss of functional tau expression may disrupt axonal transport leading to neurodegeneration of affected neurons.

Progressive subcortical gliosis Neumann and Cohn introduced the term ‘progressive subcortical gliosis’ in 1967 to describe a series of 4 patients with FTD and neuropathology characterized by pronounced subcortical gliosis with relative preservation of the overlying cortex (Neumann & Cohn, 1967). Since the original publication, there have been several dozen autopsy-confirmed case reports of either sporadic or familial PSG (for review see Mann, 1998). The pathological features overlap with those of FTLD consisting of marked reactive gliosis especially involving the junction of the grey and white matter of the frontal and temporal lobes but without prominent neuron loss, myelin changes, or senile plaques. There is occasionally microvacuolation of the superficial cortical layers and variable neuron loss in the basal ganglia, thalamus, and substantia nigra. The original descriptions of PSG emphasized the absence of filamentous cellular inclusions (Lanska et al., 1994). More recently, a mutation in tau was identified in a large kindred with familial PSG (see FTDP-17 section below) (Petersen et al., 1995; Goedert et al., 1999b). IHC with antibodies specific for tau revealed abundant pathology in both neurons and glia. Thus, at this time it remains unclear whether PSG is a distinct disease entity or overlap between FTLD and FTDP-17.

Familial frontotemporal dementia In 1939, Sanders provided the first description of familial FTD in a Dutch kindred with behavioural and

language abnormalities including disinhibition, aggression, and an obsessive personality (Sanders et al., 1939; Schenk, 1959). However, localization of a genetic abnormality in FTD was not identified until 1994 when Wilhelmsen et al. demonstrated linkage to chromosome 17q21–22 in a large family with an autosomal dominant disorder named ‘disinhibition–dementia–parkinsonism– amyotrophy complex’ (Wilhelmsen et al., 1994). Subsequently, a number of related neurodegenerative disorders were linked to a similar locus on chromosome 17 and a consensus conference convened in Ann Arbor, Michigan, coined the term FTDP-17 (Wijker et al., 1996; Foster et al., 1997; Bird et al., 1997; Heutink et al., 1997; Murrell et al., 1997; Baker et al., 1997; Lendon et al., 1998). Since the tau gene also mapped to chromosome 17q21–22, it was an obvious candidate gene for the disease locus. In 1998, several groups identified pathogenic mutations in tau that segregated with affected individuals and not in control subjects (Table 12.1, Fig. 12.2) (Poorkaj et al., 1998; Hutton et al., 1998; Spillantini et al., 1998c). However, it is evident that tau mutations are not present in all kindreds with familial FTD. While there is a positive family history of a similar dementing illness in approximately 38 to 50% of patients with FTD (Knopman et al., 1990b; Stevens et al., 1998; Chow et al., 1999), the incidence of tau mutations in different studies is quite variable (Houlden et al., 1999; Rizzu et al., 1999; Fabre et al., 2001; Morris et al., 2001; Poorkaj et al., 2001a). For instance, in two studies of community-based practices, mutations were not identified in tau despite a positive family history in ∼38% of patients (Houlden et al., 1999; Fabre et al., 2001). In contrast, in studies of referral populations with FTD, the proportion of kindreds with tau mutations ranged from 3.6% to 17.8%, and if there was a positive family history of a similar dementing illness, the incidence increased to as much as 50% (range 9.4% to 50%) (Houlden et al., 1999; Rizzu et al., 1999; Morris et al., 2001; Poorkaj et al., 2001a). Moreover, if tau pathology was present at autopsy, the incidence of tau mutations increased further, from 33% to 100% (Morris et al., 2001; Poorkaj et al., 2001a). While additional mutations have not been identified, linkage analysis implicated several other genetic loci in specific kindreds with FTD. Chromosome 3 was linked to a large Dutch kindred with FTD and autosomal dominant inheritance (Brown, 1998; Ashworth et al., 1999). This pedigree shows anticipation suggestive of a trinucleotide repeat disorder. Hosler and colleagues demonstrated linkage to chromosome 9q21–q22 in a kindred with both ALS and FTD, but not ALS or FTD alone (Hosler et al., 2000). A second genetic locus was mapped to chromosome 9p13.3– p12 in four unrelated United States families (Kovach et al.,

261

262


Table 12.1. Tau mutations identified in FTDP-17 Mutation

Location

Exon 10 splicing

MT binding

Tau aggregationa

Reference

R5H R5L K257T I260V G272V E9+33 N279K 280K L284L N296N N296H N296b P301L P301S S305N S305S E10+3 E10+11 E10+12 E10+13 E10+14 E10+16 S320F V337M G342V K369I G389R R406W

Exon 1 Exon 1 E9, R1 E9, R1 E9, R1 I9 E10, IR1-2 E10, IR1-2 E10, IR1-2 E10, R2 E10, R2 E10, R2 E10, R2 E10, R2 E10, IR2-3 E10, IR2-3 I10 I10 I10 I10 I10 I10 E11 E12, IR3-4 E12, IR3-4 E12, IR3-4 E13 E13

No change No change No change ND No change ND Increasedc Decreasedd Increased Increased Increased No change No change No change Increased Increased Increased Increased Increased Increased Increased Increased No change No change Increased ND No change No change

Reduced Reduced Reduced ND Reduced NA Variable Reduced NA NA Decreased Decreased Reduced Reduced No effect NA NA NA NA NA NA NA Reduced Reduced ND Reduced Reduced Reduced

Increased ND Increased (3Rtau) ND Variable NA Variable Increased NA NA No change Increased Increased Increased ND NA NA NA NA NA NA NA ND Increased ND Reduced ND Increased

Hayashi et al. (2002) Schellenberg, personal communication Pickering-Brown et al. (2000) Rizzini et al. (2000) Hutton, personal communication Hutton et al. (1998) Rizzu et al. (1999) Clark et al. (1998) Rizzu et al. (1999) D’Souza et al. (1999) Spillantini et al. (2000) Iseki et al. (2001) Pastor et al. (2001) Hutton et al. (1998) Bugiani et al. (1999) Sperfield et al. (1999) Hasegawa et al. (1999) Iijima et al. (1999) Stanford et al. (2000) Spillantini et al. (1998a) Miyamoto et al. (2001) Yasuda et al. (2000a) Hutton et al. (1998) Hutton et al. (1998) Hutton et al. (1998) Rosso et al. (2002) Poorkaj et al. (1998) Lippa et al. (2000) Neumann et al. (2001) Murrell et al. (1999) Hutton et al. (1998)

E = exon; I = intron; R = MT binding repeat; IR = inter-repeat regions; ND = not determined; NA = not applicable. a Effects on tau fibril formation in vitro. b Homozygous mutation. c Increased indicates enhanced exon 10 utilization. d Decreased indicates reduced exon 10 utilization. e Variable indicates conflicting data in literature.

2001). These kindreds also show autosomal dominant inheritance with a unique constellation of clinical findings including FTD, inclusion body myopathy, and Paget’s disease of bone. Finally, two large kindreds with FTD showed linkage to chromosome 17q21–22, a locus which is similar to FTDP-17, but mutations in tau were not identified (Lendon et al., 1998; Rosso et al., 2001). Interestingly, these kindreds showed distinct pathology, one classified as FTLD-MND and the other as FTLD (Rosso et al., 2001; Zhukareva et al., 2001). Moreover, one of these kindreds that presented clinically with what has been termed ‘hereditary dysphasic disinhibition dementia’ showed reduced levels of tau protein, a biochemical feature of a FTLD (Zhukareva et al., 2001). Currently, efforts are focused on identifying the genes implicated in these kindreds as well as identify-

ing additional genetic factors involved in the pathogenesis of familial FTD.

Frontotemporal dementia with parkinsonism linked to chromosome 17 The initial identification of mutations in tau led to a flurry of research activity that attempted to identify additional FTDP-17 mutations. At least 28 distinct pathogenic mutations have been identified in tau in a large number of families with FTDP-17 (Table 12.1, Fig. 12.2). Sixteen missense mutations in coding regions of tau are known, including missense mutations in exon 1 [R5H (Hayashi et al., 2002) and R5L (G. Schellenberg, personal communication)], exon

G38 9R R40 6W

V337 G34 M 2V K369 I

N296 N/H, ∆2 9 6 P301 N S 3 0 5L / S N/S S320 F

N279 ∆2 8 0 K K L284 L

1

R5H

/L

K257 I260 T V G272 V


441

Exon 10 N279K ∆280K G ∆ AAG

L284L C

N296H N296N C C

P301S P301L UU

+3

S305N S305S A C

a u

AAUAAGAAGCUGGAUCUU--------------AAUAUCAAACACGUCCCGGGAGGCGGCAG ∆ AAG ∆296N

g c

U g

u

g

a g

a

a

c

u u

c

u

a c c

u +16

u

g

u

c

+14 +13 +12 +11

Fig. 12.2. Schematic representation of mutations in the tau gene identified in FTDP-17. The structure of the largest tau isoform is shown with known coding region mutations indicated above. The grey boxes near the amino terminus represent the alternatively spliced inserts encoded for by exons 2 and 3, while the black boxes represent each of the four MT binding repeats (not drawn to scale). The second MT binding repeat is encoded by exon 10. Part of the mRNA sequence encoding exon 10 and the intron following exon 10 is enlarged to visualize the 5 splice site as well as the mutations in both exon 10 and within the 5 splice site. Nucleotides that are part of intron 10 are shown in lower case.

9 [K257T (Pickering-Brown et al., 2000; Rizzini et al., 2000), I260V (M. Hutton, personal communication), and G272V (Hutton et al., 1998; Spillantini et al., 1998b)], exon 10 [N279K (Clark et al., 1998; Yasuda et al., 1999; Delisle et al., 1999; Arima et al., 2000), N296H (Iseki et al., 2001), P301L (Dumanchin et al., 1998; Hutton et al., 1998; Clark et al., 1998; Mirra et al., 1999; Houlden et al., 1999; Bird et al., 1999; Kodama et al., 2000; Tanaka et al., 2000), P301S (Bugiani et al., 1999; Sperfeld et al., 1999; Yasuda et al., 2000b), and S305N (Iijima et al., 1999)], exon 11 [S320F (Rosso et al., 2002)], exon 12 [V337M (Poorkaj et al., 1998), E342V (Lippa et al., 2000), and K369I (Neumann et al., 2001)], and exon 13 [G389R (Murrell et al., 1999; PickeringBrown et al., 2000) and R406W (Hutton et al., 1998; Van Swieten et al., 1999; Saito et al., 2002)]. Three silent mutations in exon 10 [L284L (D’Souza et al., 1999), N296N (Spillantini et al., 2000), and S305S (Stanford et al., 2000)] as well as two single amino acid deletions [K280 (Rizzu et al., 1999) and N296 (Pastor et al., 2001)] have also been identified; however, the N296 does not show autosomal dominant inheritance. In addition, seven different nucleotide substitutions were identified in the introns following exons 9 [+33 (Rizzu et al., 1999)] and 10 [+3 (Spillantini et al., 1998c; Tolnay et al., 2000), +11 (Miyamoto et al., 2001), +12 (Yasuda et al., 2000a), +13 (Hutton et al., 1998), +14 (Hutton et al., 1998; Clark et al., 1998), and +16 (Hutton et al., 1998;

Morris et al., 1999; Hulette et al., 1999; Goedert et al., 1999b)] that presumably play a role in the regulation of the alternative splicing of exon 10 which will be discussed below.

Biology of the tau protein Tau proteins are low-molecular-weight MAPs that are abundant in the central nervous system (CNS) where they are expressed predominantly in axons (Cleveland et al., 1977; Binder et al., 1985), and at low levels in astrocytes and oligodendrocytes (Shin et al., 1991; LoPresti et al., 1995). They are also expressed in axons of peripheral nervous system neurons (Couchie et al., 1992). Human tau proteins are encoded by a single copy gene on chromosome 17q21 composed of 16 exons with the CNS isoforms generated by the alternative mRNA splicing of 11 of these exons (Fig. 12.3) (Neve et al., 1986; Goedert et al., 1988; Andreadis et al., 1992). In adult human brain, alternative splicing of exons 2, 3, and 10 generates 6 tau isoforms ranging from 352 to 441 amino acids in length which differ by the presence of either 3 (3Rtau) or 4 (4Rtau) carboxy-terminal tandem repeat sequences of 31 or 32 amino acids each that are encoded by exons 9 to 12 (Goedert et al., 1989a,b). Additionally, alternative splicing of exons 2 and 3 leads to the absence (0N) or presence of inserted sequences of 29 (1N) or 58 (2N) amino acids in the amino-terminal third of the molecule. In the

263

264


Exons: 0

1

2 3

4

4a

5

6

7 8

9

10 11 12 13

14

Tau gene CNS Tau Isoforms 1 E3

E2 1

R1 R1

E2 1

R1

1

E2 1

E3 E2

1

R2 R2 R2 R1 R1 R1

R3 R3 R3 R3 R3 R3

R4 R4 R4 R4 R4 R4

441 412 383 410 381 352

4R/2N 4R/1N 4R/0N 3R/2N 3R/1N 3R/0N

Fig. 12.3. Schematic representation of the human tau gene and six human CNS tau isoforms generated by alternative splicing. The human tau gene contains 16 exons, including exon 0 that is part of the promoter. Exons 1, 4, 5, 7, 9, and 11 to 13 (blue boxes) are constitutively expressed. Alternative splicing of exons 2 (E2), 3 (E3) and 10 (yellow boxes) produces the six alternative tau isoforms. Exons 6 and 8 (stippled boxes) are not transcribed in the human CNS. Exon 4a (striped box), which is also not transcribed in the human CNS, is expressed in the PNS leading to the larger tau isoforms, termed ‘big tau’. The black bars depict the 18 amino acid MT binding repeats and are designated R1 to R4. The relative sizes of the exons and introns are not drawn to scale.

adult human brain, the ratio of 3Rtau to 4Rtau isoforms is approximately 1:1, while the 0N, 1N, and 2N tau isoforms comprise about 37%, 54%, and 9%, respectively, of total tau (Goedert & Jakes, 1990; Hong et al., 1998). Since its discovery over 25 years ago, a number of functions of tau have been characterized (for review see Buée et al., 2000; Lee et al., 2001). Most notably, tau binds to and stabilizes microtubules (MTs) and promotes MT polymerization (Weingarten et al., 1975; Cleveland et al., 1977). The MT binding domains of tau are localized to the carboxyterminal half of the molecule within the 4 MT binding motifs (Figs. 12.2 and 12.3). These motifs are composed of highly conserved 18-amino acid binding elements separated by less conserved 13 or 14 amino acid flexible interrepeat sequences (Himmler et al., 1989; Lee et al., 1989; Butner & Kirschner, 1991). The binding of tau to MTs is a complex process that is mediated by a flexible array of weak binding sites that are distributed throughout the MT binding domain delineated by these repeats and their interrepeat sequences (Lee et al., 1989; Butner & Kirschner, 1991). The function of tau as a MT binding protein is regulated by phosphorylation (Drechsel et al., 1992; Yoshida & Ihara, 1993; Bramblett et al., 1993; Biernat et al., 1993). There are 79 potential serine (Ser) and threonine (Thr) phosphate acceptor residues in the longest tau isoform, and phosphorylation at approximately 30 of these sites has been

reported in normal tau proteins (for review see Billingsley & Kincaid, 1997; Buée et al., 2000; Hong et al., 2000). Furthermore, at least 12 Ser/Thr protein kinases and 4 Ser/Thr protein phosphatases have been implicated in regulating the phosphorylation state and thus the function of tau. The phosphorylation sites are clustered in regions flanking the MT binding repeats, and increasing tau phosphorylation at multiple sites negatively regulates MT binding (Drechsel et al., 1992; Yoshida & Ihara, 1993; Bramblett et al., 1993; Biernat et al., 1993). However, in both sporadic and familial tauopathies including AD and FTDP-17, tau is hyperphosphorylated, and it is this ‘abnormal’ tau that is the principle component of the filamentous aggregates in neurons and glia that are the pathological hallmarks of these disorders (Hasegawa et al., 1996; Hoffmann et al., 1997; ZhengFischhofer et al., 1998).

Clinical features of FTDP-17 The term FTDP-17 was coined in 1997 to describe the common features found in the atypical kindreds that demonstrated linkage to chromosome 17 (Foster et al., 1997). These kindreds all exhibited autosomal dominant inheritance with age-dependent penetrance typically in the third through sixth decade of life. Duration of disease is approximately 10 years but with wide variability ranging from as


little as 3 years to as long as 30 years. The disease commonly presents insidiously with behavioural, language, and/or motor abnormalities. The behavioural abnormalities are typical of FTD, including alterations in personality, impaired social or occupation functioning, hyperorality, hyperphagia, and psychosis associated with disturbed executive function on neuropsychological examination (Lund and Manchester Groups, 1994; McKhann et al., 2001). In most patients, visuospatial function, orientation, and memory are preserved until late in the course of disease, features that are distinct from that of AD. Language disorders are typical of either an expressive or semantic aphasia, while motor abnormalities consist principally of l-dopaunresponsive parkinsonism including bradykinesia, rigidity, and postural instability. With reports of additional kindreds, it became apparent that the clinical presentation is highly variable, reflecting the specific pattern of neuronal loss in affected individuals (Table 12.2) (for review see Foster et al., 1997; Spillantini et al., 1998a; Reed et al., 2001). The clinical phenotype of patients with different mutations, as well as occasionally with the same mutation, is quite variable ranging from FTD to CBD to PSP to AD to a multisystem degeneration. However, some tau mutations cause a relatively similar phenotype. For instance, the N279K missense mutation typically causes a phenotype reminiscent of PSP with superimposed dementia (Reed et al., 1998; Yasuda et al., 1999; Delisle et al., 1999). Similarly, +16 mutation in the intron following exon 10 described in at least ten kindreds causes a very characteristic FTD with disinhibition and affective disturbances presenting in the fourth to sixth decade of life (Baker et al., 1997; Morris et al., 1999; Hulette et al., 1999; Goedert et al., 1999b; Janssen et al., 2002). In contrast, there are numerous clinical and pathologic descriptions of families with P301L mutations that demonstrate a highly variable clinical phenotype including PSP, CBD, and PiD (Spillantini et al., 1998b; Mirra et al., 1999; Nasreddine et al., 1999; Bird et al., 1999). Similarly, while several kindreds with the R406W mutation present with relatively late-onset, slowly progressive memory deficits similar to AD, other kindreds present with early-onset, rapidly progressive FTD (Reed et al., 1997; Van Swieten et al., 1999; Saito et al., 2002). Even more perplexing is a FTDP-17 family with the P301S mutation in which one individual presented clinically with FTD while his son presented clinically with CBD (Bugiani et al., 1999). The N296 mutation is unique it that it gives rise to atypical PSP in individuals homozygous for the mutation, but in heterozygous individuals this mutation is incompletely penetrant and associated with a phenotype similar to idiopathic Parkinson’s disease (Pastor et al., 2001). These reports, albeit anecdotal, suggest tremendous overlap between the various

tau-related disorders, and the clinical distinctions between them may be due to other genetic and/or epigenetic factors that modify the effects of the tau mutations.

Neuropathology of FTDP-17 Similar to the clinical phenotypes, the neuropathology of FTDP-17 is quite variable (Table 12.2) (Reed et al., 2001; Pickering-Brown et al., 2002). The most characteristic gross neuropathological feature is severe frontotemporal atrophy involving both cortex and the underlying white matter with relative sparing of the parietal and occipital lobes (Fig. 12.4) (Foster et al., 1997). The medial temporal lobe including the hippocampus, amygdala, and entorhinal cortex shows marked variability ranging from minimal involvement to marked atrophy. The basal ganglia often exhibit severe degeneration, while involvement of the thalamus and subthalamus is typically not prominent. The substantia nigra is characteristically depigmented, while atrophy of the remainder of the brainstem and cerebellum is variable, typically mirroring the clinical phenotype of the patient (Fig. 12.4). Microscopically, there is marked neuron loss in effected brain regions similar to that observed for sporadic tauopathies (Fig. 12.5). In affected cortices, the neuron loss is associated with rarefaction and spongiosis of the superficial cortex associated with gliosis of both the grey and underlying white matter (Fig. 12.5). Similarly, there is atrophy and neuron loss in the basal ganglia and substantia nigra (Fig. 12.5). Neuron loss in other brain regions is highly variable, paralleling both the clinical phenotype and pattern of gross atrophy. For instance, kindreds with the R406W mutation show profound neuron loss with severe gliosis in the medial temporal lobe including the amygdala, hippocampus, and entorhinal cortex with relative sparing of the basal ganglia and brainstem (Fig. 12.5, Table 12.2) (Reed et al., 1997; Van Swieten et al., 1999; Rosso et al., 2000; Saito et al., 2002). In contrast, the N279K mutation causes neuron loss in subcortical and brainstem nuclei with variable pathology in the medial temporal lobe structures (Fig. 12.5 and Table 12.2) (Kawai et al., 1993; Reed et al., 1998; Delisle et al., 1999; Arima et al., 2000). Despite the regional heterogeneity of the pathology of FTDP-17 kindreds, in all cases there is extensive neuronal or glial and neuronal fibrillary pathology composed of hyperphosphorylated tau protein, in the absence of -amyloid (A) deposits or other disease-specific brain lesions (Fig. 12.6) (for review see Spillantini et al., 1998a; Reed et al., 2001). These pathological inclusions are only variably identified with amyloid binding dyes such as Congo red and ThioflavinS as well as silver stains including

265

Table 12.2. Tau pathology in FTDP-17 Mutation

Regional distribution

Neurons

Glia

Biochemistry

Ultrastructure

R5H

FT, MT, SN

Pretangles (+/−)

4Rtau

Straight tubules, 15–20 nm

R5L

BG, Th, STN, BS, CB

Globose tangles

4Rtau

Straight filaments

K257T I260V G272V E9+33 N279K

Pick bodies Pretangles NA Pretangles Pick-body like NA Pretangles Balloon neurons

3Rtau > 4Rtau NA NA NA 4Rtau

280K L284L N296N

FT, MT NA FT, MT, Ca, Th, SN NA FT, MT, BG, Th, STN, BS, CB NA FT, MT, BG, SN FT, Hip, GP, BS

Oligodendrocytes Astrocytes (+/−) Astrocytes (tufted) Oligodendrocytes Negative NA Oligodendrocytes (+/−) NA Oligodendrocytes Astrocytes NA Glia Oligodendrocytes

NA NA NA

N296H N296 P301L

FT, MT, Ca, Th, SN NA FT, MT, BG, SN

NA Pretangles NFTs Pretangles Balloon neurons Corticobasal bodies Pretangles Balloon neurons (+/−) NA Pretangles Balloon neurons

Twisted ribbons NA NA NA Twisted ribbons Paired tubules NA NA NA

Astrocytes Oligodendrocytes NA Astrocytes Oligodendrocytes

4Rtau NA 4Rtau

P301S S305N S305S

FT, Ca, SN FT, MT PR, MT, GP, STN, SN

Pretangles Pretangles NFTs, ring-shaped Pretangles Balloon neurons

NA NA NA

E10+3

FT, SN, CB

Pretangles NFTs

E10+11

FT, BG, Th, BS, CB

Pretangles NFTs

E10+12 E10+13 E10+14 E10+16

FT, BG, Th, BS, CB FT, BG, SN FT, MT, BG, SN FT, BG, SN

S320F

Temp, Hip

Pretangles Pretangles NFTs (+/−) Pretangles NFTs Balloon neurons Pretangles NFTs (+/−) Pick bodies (+/−) Balloon neurons (+/−) Pick bodies Pretangles

Astrocytes Oligodendrocytes Oligodendrocytes Astrocytes, tufted Oligodendrocytes Oligodendrocytes Astrocytes (+/−) Astrocytes (tufted) Oligodendrocytes Glia Oligodendrocytes Oligodendrocytes Astrocytes Oligodendrocytes

V337M G342V

FT, MT, SN FT

K369I

FT, MT, Ca

NFTs Pretangles NFTs, Pretangles Pick bodies (+/−) Balloon neurons (+/−) Pick bodies Balloon neurons

G389R

FT, MT, Ca, SN

R406W

FT, MT

NA

Straight tubules, 15 nm NA Twisted ribbons Straight filaments Straight filaments, 10 nm Straight tubules, 15 nm Twisted filaments, 15 nm Straight filaments, 15 nm Straight filaments Twisted filaments NA

4Rtau 4Rtau > 3Rtau 4Rtau 4Rtau

Twisted ribbons, 5–23 nm Twisted filaments NA Twisted ribbons

Oligodendrocytes (+/−)

3Rtau = 4Rtau

NI Astrocytes

3Rtau = 4Rtau 4Rtau > 3Rtau (4R0N)

Straight filaments (major) Twisted filaments (minor) PHF-like Straight filaments PHF-like

3Rtau = 4Rtau

Twisted ribbons

Pretangles Pick bodies NFTs

Astrocytes (tufted) (+/−) Oligodendrocytes (+/−) NI

3Rtau = 4Rtau

NFTs Pretangles

Oligodendrocytes (+/−)

3Rtau = 4Rtau

Twisted filaments Straight filaments PHF-like Twisted ribbons Straight filaments

4Rtau

BG = basal ganglia; BS = brainstem; Ca = caudate; CB = cerebellum; FT = frontotemporal lobe; GP = globus pallidus; Hip = hippocampus; MT = medial temporal lobe (hippocampus, entorhinal cortex and amygdala); NFTs = neurofibrillary tangles; PHF = paired helical filaments; SN = substantia nigra; STN = subthalamic nucleus; NA = not available; NI = not identified; Temp = temporal lobe; Th = thalamus; +/− = scattered tau positive inclusions.


Fig. 12.4. Macroscopic pathology in FTDP-17. (A)–(C) FTDP-17 brain with intron 10, +16 mutation showing prominent frontal and temporal lobe atrophy with relative sparing of parietal and occipital lobes. Medial view (B) and coronal sections (C) show marked hydrocephalus of the lateral ventricles and moderate atrophy of the head of the caudate nucleus (arrow). (D ) Brainstem of patient with N279K mutation clinically classified as ‘pallido-ponto-nigral degeneration’. The mid-sagittal section shows atrophy of midbrain and pons with dilatation of the fourth ventricle (arrowhead).

Bodian, Bielschowsky, and Gallyas. However, the inclusions are most consistently identified by IHC for tau protein, particularly with antibodies specific for phosphorylationdependent epitopes that are characteristic of the insoluble inclusions. The filamentous aggregates exhibit remarkable heterogeneity in both neurons and glia. The neuronal tau pathology most often demonstrates a granular staining pattern, often referred to as ‘pretangles’ (Fig. 12.6) (Spillantini et al., 1998a; Reed et al., 2001). Classic neurofibrillary tangles (NFTs), similar to that observed in AD, are abundant in several of the FTDP-17 mutations (Fig. 12.6) (Sumi et al., 1992; Spillantini et al., 1996; Foster et al., 1997; Bird et al., 1997; Reed et al., 1997; Van Swieten et al., 1999; Saito et al., 2002). In addition, Pick bodies similar to those observed in PiD are abundant in several mutations including K257V (Pickering-Brown et al., 2000; Rizzini et al., 2000), G272V (Spillantini et al., 1998b), S320F (Rosso et al., 2002), K369I

(Neumann et al., 2001), and G389R (Murrell et al., 1999; Pickering-Brown et al., 2000). Ballooned neurons and corticobasal bodies similar to that originally described in CBD are also detected with many of the mutations (Foster et al., 1997; Spillantini et al., 2000; Reed et al., 2001). The tau pathology in astrocytes typically demonstrates a granular staining pattern (Fig. 12.6). However, occasional glial inclusions are observed that are reminiscent of the astrocytic plaques and tufted astrocytes described in CBD and PSP, respectively (Foster et al., 1997; Spillantini et al., 1998a; Komori, 1999; Reed et al., 2001). With many tau mutations, there is abundant tau pathology in oligodendrocytes, similar to the coiled bodies observed in many tauopathies (Fig. 12.6) (Foster et al., 1997; Spillantini et al., 1998a; Komori, 1999; Reed et al., 2001). Finally, tau-positive threads are abundant in either the grey matter or both the white and grey matter within the processes of both neurons and glia (Fig. 12.6) (Komori, 1999).

267

268


Ultrastructural analysis of the tau filaments in FTDP-17 also demonstrates marked heterogeneity in the tau pathology. In two mutations, R406W and V337M, paired helical filaments (PHFs) with a diameter of 8 to 20 nm and a periodicity of 80 nm as well as straight filaments (SFs) nearly identical to those present in AD are observed (Sumi et al., 1992; Spillantini et al., 1996; Reed et al., 1997). However, in the majority of mutations, the filaments demonstrate a range of morphologies including wide and narrow twisted ribbons, twisted filaments, paired tubules, and SFs that range in width from 8 to 25 nm typically with longer periodicities than that observed in the PHFs of AD (Spillantini et al., 1998a; Yen et al., 1999b; Reed et al., 2001). Perhaps the most unique feature of the pathology of FTDP-17 is that the pattern and nature of the tau inclusions fail to fit neatly into any of the known categories of sporadic tauopathies. However, the FTDP-17 mutations loosely fall into two general categories based on tau immunostaining (Table 12.2). In one group, the tau aggregates are primarily within neurons. By IHC, the tau pathology is localized predominantly in grey matter, although biochemically there is aggregation of all six tau isoforms (or predominantly 3Rtau with the K257T mutation) within both grey and white matter (Table 12.2, Figs. 12.7 and 12.8). In contrast, the second group shows extensive tau pathology within both glia and neurons in grey and white matter that is composed of 4Rtau only (Table 12.2, Figs. 12.7 and 12.8).

Effects of FTDP-17 mutations

Fig. 12.5. Microscopic pathology in FTDP-17. (A) Severe neuron loss with spongiosis of upper cortical lamina and prominent gliosis in frontal cortex of patient with P301L mutation (H&E). (B) Severe gliosis of entorhinal cortex of patient with R406W mutation. The gliosis is most prominent in the molecular layer (arrow) and at the junction of the grey and white matter (arrowheads) (IHC with anti-GFAP antibody). (C) Marked depletion of pigmented neurons in the substantia nigra of patient with N279K mutation. There is prominent gliosis with extracellular deposition of neuromelanin as well as several axonal spheroids (arrowheads) (H&E).

FTDP-17 mutations lead to tau dysfunction and presumably disease by several distinct mechanisms. Intronic and some exonic mutations affect the alternative splicing of exon 10 and consequently alter the relative proportions of 3Rtau and 4Rtau. The other exonic mutations impair the ability of tau to bind MTs and to promote MT assembly. Some of the mutations also promote the assembly of tau into filaments. Moreover, additional mechanisms may play a role in the case of some coding region mutations (for review see Lee et al., 2001). The intronic mutations clustered around the 5 splice site of exon 10, as well as several mutations within exon 10 (N279K, L284L, N296N, N296H, S305N, S305S, and G342V), increase the ratio of 4Rtau to 3Rtau by altering the splicing of this exon (Hutton et al., 1998; Yasuda et al., 1999; Delisle et al., 1999; Varani et al., 1999; Grover et al., 1999; D’Souza et al., 1999; Hasegawa et al., 1999; Stanford et al., 2000; D’Souza & Schellenberg, 2000; Gao et al., 2000; Spillantini et al., 2000; Yasuda et al., 2000a; Miyamoto et al., 2001; Grover et al., 2002). As a result of these mutations there is a relative increase in mRNA containing exon 10, reflecting increased utilization of the 5 splice site


Fig. 12.6. Tau pathology in FTDP-17. (a) Frontal cortex with numerous tau positive inclusions in patient with P301L mutation. (b) Numerous NFTs in the CA1 region of the hippocampus in patient with R406W mutation. (c) Granular tau inclusions (pretangles) in the frontal cortex of patient with P301L mutation. (d ), (e) NFTs in the hippocampus of patient with R406W mutation stained with anti-tau antibody (d ) or Thioflavin S (e). ( f ) Pick-body like inclusion in the fascia dentata of patient with R406W mutation. (g) Granular astrocytic inclusion in subcortical white matter of patient with intron 10, +16 mutation. (h) Coiled body in putamen of patient with N279K mutation. (i ) Thread pathology and coiled bodies in internal capsule of patient with N279K mutation. ( j ), (k). Thread pathology in white matter of frontal lobe in patients with intron 10, +16 ( j ) and R406W (k) mutations. Note the absence of white matter pathology detected by IHC with the R406W mutation in which there is aggregation of all 6 tau isoforms. In contrast, by Western blot analysis, there is abundant tau pathology in subcortical white matter (see Fig. 12.8). ((a)–(d ) and ( f )–(k), IHC with anti-tau antibodies.)

269

270


Soluble 4R/2N 3R/2N 4R/1N 3R/1N 4R/0N 3R/0N

4R/2N 3R/2N 4R/1N 3R/1N 4R/0N 3R/0N

Normal

AD ALS/PDC FTDP-17 Group A

PSP/CBD FTDP-17

PiD FTDP-17

Group B

Group C

—

—

FTDP-17 Group D

Insoluble Dephos.

—

+

72 kDa 68 kDa 64 kDa 60 kDa

+

+

—

+ 4R/2N 3R/2N 4R/1N 3R/1N 4R/0N 3R/0N

Fig. 12.7. Schematic representation of Western blot banding patterns of soluble and insoluble tau from different tauopathies. The cartoon depicts the typical banding pattern of soluble tau (top panels) and insoluble/filamentous tau (bottom panels) from the brains of patients with the diseases indicated following resolution with SDS–PAGE and immunoblotting with anti-tau antibodies. The FTDP-17 mutations show several different Western blot banding patterns of soluble and insoluble tau protein that are depicted as groups A to D. The soluble fraction from the brains of unaffected (normal) individuals, sporadic tauopathies, and FTDP-17 with mutations that do not affect tau splicing (Groups A,B, and C) show expression of all 6 tau isoforms. Insoluble tau from the brain of patients with AD, ALS/PDC and FTDP-17 group A (S320F, V337M, K369I, G389R, and R406W), resolve as three major proteins of 68-, 64- and 60-kDa; and a minor band of 72 kDa. When dephosphorylated, they resolve into six proteins that correspond to all six tau isoforms similar to the soluble fraction. In CBD, PSP, and FTDP-17 group B (R5H, P301L, and G342V), 2 prominent 68- and 64-kDa protein bands are detected (the 72 kDa minor band is variably detected) and align with 4Rtau following dephosphorylation indicating that there is selective aggregation of 4Rtau. Similarly, in PiD and FTDP-17 group C (K257T) the 64 and 60 kDa insoluble tau protein bands predominate and align with 3Rtau isoforms following dephosphorylation indicating selective aggregation of 3Rtau. In contrast, in FTDP-17 mutations that effect mRNA splicing (Group D: N279K, L284L, N296N, N296H, S305S, S305N, and intron 10 mutations), there is expression of predominantly 4Rtau throughout the entire brain which is reflected in the insoluble tau aggregates.

of exon 10 as demonstrated in exon trapping experiments. Biochemical analysis of insoluble tau extracted from autopsied FTDP-17 brain tissue of patients with these mutations reveals predominantly 4Rtau isoforms (Figs. 12.7 and 12.8) (Clark et al., 1998; Reed et al., 1998; Hong et al., 1998; Spillantini et al., 1998c; Hulette et al., 1999; Goedert et al., 1999b; Arima et al., 2000; Yasuda et al., 2000a; Iseki et al., 2001). Furthermore, 4Rtau protein levels are increased in both affected and unaffected regions of FTDP-17 brains (Hong et al., 1998; Spillantini et al., 1998c; Goedert et al., 1999b; Yasuda et al., 2000a). The regulation of splicing of exon 10 of tau is complex, and may involve multiple cis-acting regulatory elements that either enhance or inhibit the utilization of the 5 splice

site, many of which are affected by mutations identified in tau (Varani et al., 1999; D’Souza et al., 1999; Hasegawa et al., 1999; D’Souza & Schellenberg, 2000; Jiang et al., 2000; Gao et al., 2000; Grover et al., 2002). Splicing regulatory elements within exon 10 include an exon-splicing enhancer (ESE) and an exon-splicing silencer (ESS) (D’Souza et al., 1999; D’Souza & Schellenberg, 2000; Gao et al., 2000). The ESE consists of 3 domains, a potential SC35 binding element, a purine rich sequence, and an AC-rich sequence (D’Souza & Schellenberg, 2000). Immediately downstream of the ESE within exon 10 is the purine-rich ESS. The flanking exons of tau also affect exon 10 splicing (Gao et al., 2000). Exons 9 and 11 exert opposite effects; exon 9 promotes splicing of exon 10, while exon 11 suppresses it. Lastly, intronic


Fig. 12.8. Western blots of soluble and insoluble tau from FTDP-17 brains. Soluble (a) and insoluble (b), (c) tau fractions extracted from grey (G) and white (W) matter of patients with FTDP-17 were dephosphorylated with E. coli alkaline phosphatase, resolved by SDS-PAGE and immunoblotted with phosphorylation-independent tau antibodies. (a) Soluble fractions from the frontal cortex of patients with the R406W show all six tau isoforms. In contrast, in patients with the intron 10, +16 mutation, there is selective overexpression of 4Rtau isoforms. (b),(c) The insoluble tau protein from the indicated cortices are composed of either equimolar amounts of both 4Rtau and 3Rtau (R406W) or predominantly 4Rtau only (intron 10, +16). With both mutations the tau pathology is most abundant in the frontal and temporal lobes with relative sparing of the occipital lobe. In addition, there is abundant insoluble tau in the white matter of the brain with the R406W mutation that is not detected by IHC (see Fig. 12.6). Recombinant tau isoforms (rTau) are as indicated. Fr = frontal lobe; Te = temporal lobe; Pa = parietal lobe; Oc = occipital lobe.

sequences immediately downstream of exon 10 inhibit its splicing (Varani et al., 1999; D’Souza et al., 1999; Hasegawa et al., 1999; D’Souza & Schellenberg, 2000; Jiang et al., 2000; Gao et al., 2000; Grover et al., 2002). The inhibition may be secondary to the formation of a stem–loop structure that sequesters the 5 splice site from the splicing machinery including the U1- and U6-snRNP (Fig. 12.2) (Varani et al., 1999; Grover et al., 1999; Jiang et al., 2000). Interestingly, the relative proportions of 3Rtau and 4Rtau from other species correlates with the predicted stability of this stem–loop structure (Grover et al., 1999). However, another study concluded that the inhibitory effect of the intronic sequence is due to a linear sequence array that is independent of the stem–loop structure (D’Souza & Schellenberg, 2000). FTDP-17 mutations in tau may alter exon 10 splicing by affecting several of the regulatory elements described above. For example, the intronic mutations as well as the exonic mutations at codon 305 (S305N and S305S) may destabilize the inhibitory stem–loop structure (Fig. 12.2) (Hutton et al., 1998; Grover et al., 1999; D’Souza et al., 1999). The S305N and the +3 intronic mutations may also enhance exon 10 splicing by increasing the strength of the 5 splice site (GUgugagu to AUgugagu) (Senapathy et al., 1990). However, the finding that the S305S mutation that weakens the 5 splice site (GUgugagu to GCgugagu) also leads to predominantly 4Rtau argues against this effect of the mutation (Stanford et al., 2000). The N279K mutation may improve the function of the ESE by lengthening the purine-rich sequence within this regulatory element (TAAGAA to GAAGAA), thus enhancing exon 10 splicing (D’Souza & Schellenberg, 2000). Moreover, the thymidine nucleotide present in the wild-type (WT) sequence may function as an inhibitor of splicing (Tanaka et al., 1994). This hypothesis is supported by the observation that the K280 mutation, which deletes the three adjacent purine residues (AAG), reduces exon 10 splicing. The silent L284L mutation that enhances exon 10 splicing may do so by disrupting a potential exon splicing inhibitor (UUAG to UCAG) (Si et al., 1998; D’Souza et al., 1999). However, since mutation of this consensus sequence does not increase exon 10 splicing, a second possibility is that the mutation lengthens the AC-rich element within the ESE (Si et al., 1998; D’Souza & Schellenberg, 2000). Thus, the L284L mutation may affect either an enhancing or inhibitory splicing element. Lastly, the effect of the N296N and N296H mutations on splicing of exon 10 is due to disruption of the ESS or conversely the creation of a novel splice enhancer sequence (D’Souza & Schellenberg, 2000; Grover et al., 2002).

271

272


The mechanisms by which these changes in the ratio of 3Rtau to 4Rtau (3R/4Rtau) lead to neuronal and glial dysfunction and cell death remains unclear. However, 3Rtau and 4Rtau may bind to distinct sites on MTs (Goode & Feinstein, 1994), and it is possible that a specific ratio of tau isoforms is necessary for normal MT function (Goode et al., 1997). Thus, the altered ratio of 3R/4Rtau may directly affect MT function. In addition, overproduction of 4Rtau isoforms may lead to an excess of free tau in the cytoplasm that is prone to aggregate and polymerize into filaments over time. Another subset of the tau mutations has no effect on tau splicing but instead alters the ability of tau to interact with MTs. Specifically, missense mutations K257T, G272V, K280, N296, P301L, P301S, V337M, G389R, and R406W reduce the binding of tau to MTs and decrease its ability to promote MT stability and assembly in vitro (Hong et al., 1998; Hasegawa et al., 1998; Bugiani et al., 1999; Rizzu et al., 1999; Pickering-Brown et al., 2000; Rizzini et al., 2000; Barghorn et al., 2000; Grover et al., 2002). In contrast to those mutations that affect the splicing of tau, these mutations do not alter the expression pattern of 3Rtau and 4Rtau (Hong et al., 1998). In fact, the P301L mutation causes a moderate (25%) decrease in soluble 4Rtau due to the selective aggregation of mutant 4Rtau isoforms (Hong et al., 1998; Rizzu et al., 2000; Miyasaka et al., 2001). Biochemical analysis of insoluble tau extracted from brain tissue of patients with these mutations reveals a variety of patterns. Several mutations including S320F, V337M, K369I, G389R, and R406W are characterized by the aggregation of equal amounts of both 3Rtau and 4Rtau (Figs. 12.7 and 12.8) (Spillantini et al., 1996; Hong et al., 1998; Murrell et al., 1999; Neumann et al., 2001; Rosso et al., 2002). A second group (R5H, R5L, P301L, and G342V) leads to the selective aggregation of 4Rtau isoforms similar to that observed in PSP and CBD (Fig. 12.7) )(Clark et al., 1998; Spillantini et al., 1998b; Bugiani et al., 1999; Mirra et al., 1999; Nasreddine et al., 1999; Lippa et al., 2000; Hayashi et al., 2002). In contrast, the K257T mutation causes the selective aggregation of 3Rtau similar to PiD (Fig. 12.7) (Rizzini et al., 2000). This is intriguing because the neuropathology of both kindreds identified with the K257T mutation is characterized by the widespread deposition of Pick bodies comparable to those observed in PiD (Pickering-Brown et al., 2000; Rizzini et al., 2000). The effects of these FTDP-17 mutations on MT binding, assembly, and stability are not observed with the tau missense mutations that directly affect exon 10 splicing (Hong et al., 1998; D’Souza et al., 1999; Hasegawa et al., 1999). Similar effects on MT function are observed when mutant tau is expressed in a variety of cell lines including SHSY5Y neu-

roblastoma cells (Dayanandan et al., 1999), Chinese hamster ovary (CHO) cells (Matsumura et al., 1999; Dayanandan et al., 1999; Vogelsberg-Ragaglia et al., 2000), monkey kidney (COS) cells (Arawaka et al., 1999; Sahara et al., 2000), human embryonic kidney (HEK293) cells (Sahara et al., 2000; Nagiec et al., 2001), and SF9 insect cells (Frappier et al., 1999). Expression of a variety of tau missense mutations including G272V, P301L, V337M, and R406W in these cells caused reduced MT binding, MT instability, disorganized MT morphology, and defects in MT assembly to varying degrees. However, in 2 studies, many mutations had either no effect or only a modest effect on MT binding and/or function both in in vitro assays and in transfected cell lines (DeTure et al., 2000; Sahara et al., 2000). The discrepancies between these and other studies are most likely due to either differences in the levels of tau expression, the specific cells in which the FTDP-17 mutations were expressed, and/or the binding of tau to MTs. Nevertheless, even if the FTDP-17 mutations cause only a modest reduction in MT binding affinity, this effect could lead to large cumulative effects on neurons over the human life span. Furthermore, increased cytosolic concentrations of unbound mutant tau may facilitate aggregation of these abnormal proteins into filamentous inclusions. In support of this hypothesis, Yen and colleagues demonstrated that several mutations lead to decreased susceptibility to calpain I digestion, an enzyme involved in the normal degradation of tau, which might lead to an increase in the concentration of cytosolic tau (Yen et al., 1999a). The loss of function of tau could also cause a variety of toxic effects within affected cells. For example, the reduced capacity of these mutants to stabilize MTs might affect axonal transport in neurons. Alternatively, impairments in tau functions could lead to increased apoptotic vulnerability as suggested in a recent study in which two of the FTDP-17 mutations (N279K and V337M) showed increased cell death upon serum withdrawal in transfected neuroblastoma cell lines (Furukawa et al., 2000). A subset of missense tau mutations may cause FTDP17, at least in part, by promoting tau aggregation (Table 12.1). Several studies demonstrated that numerous mutations, including K257T, G272V, K280, N296, P301L, P301S, V337M, and R406W, promote heparin- or arachidonic acid-induced tau filament formation in vitro relative to WT tau (Arrasate et al., 1999; Nacharaju et al., 1999; Goedert et al., 1999a; Rizzini et al., 2000; Barghorn et al., 2000; Gamblin et al., 2000; Grover et al., 2002). The K280 and P301L promote in vitro filament formation more readily than other missense mutations and WT tau (von Bergen et al., 2001). This effect is most likely due to the enhancement of -structure around two hexapeptide motifs that are strong promoters of tau fibrillization


(von Bergen et al., 2000). Additionally, the P301L mutation causes the selective deposition of mutant tau in the pathological intracellular inclusions with corresponding depletion in the soluble fraction (Rizzu et al., 2000; Miyasaka et al., 2001). The aggregation of mutant tau in intact cells has also been demonstrated. Thus, CHO cells expressing tau with the K280 mutation, but not other mutations (V337M, P301L and R406W), formed insoluble aggregates (Vogelsberg-Ragaglia et al., 2000). It is unclear whether the filaments composed of mutant tau proteins are similar to those composed of WT tau. Indeed, Goedert et al., reported that, while a subset of the tau mutations stimulated filament formation, they were structurally identical to WT tau filaments as assessed by circular dichroism (Goedert et al., 1999a). However, Jicha et al., reported altered physical and structural properties of filaments composed of mutant tau as assessed by both circular dichroism and reverse phase high performance liquid chromatography (Jicha et al., 1999). The missense tau mutations may also affect tau function, and thus contribute to the pathogenesis of FTDP17, by altering the phosphorylation of tau. Numerous mutations decreased the binding affinity of tau for protein phosphatase 2A, a major phosphatase implicated in the regulation of the MT-binding activity of tau (Goedert et al., 2000). Goedert and colleagues suggest that this leads to both tau hyperphosphorylation and dissociation of tau from MTs therefore altering the normal balance between tau and MTs. However, expression of several mutations, particularly R406W, both in vitro and in transfected cell lines led to reduced levels of tau phosphorylation relative to WT tau, suggesting that increased tau phosphorylation is not required for pathogenesis of disease (Matsumura et al., 1999; Dayanandan et al., 1999; Sahara et al., 2000; Perez et al., 2000; Vogelsberg-Ragaglia et al., 2000; Connell et al., 2001). The biochemical and structural characteristics of the tau aggregates in FTDP-17 are somewhat predictable based on our understanding of the functions of tau proteins and tau splicing. However, the basis for the clinical phenotypes and topographical distributions of pathology of individuals with the various FTDP-17 mutations remains enigmatic. This finding suggests that the phenotypic differences may be due to other genetic and/or epigenetic factors that modify the effects of the primary mutation. The specific modifiers that mediate the ‘selective vulnerability’ of specific regions and/or cells associated with either a specific mutation or even of a specific individual remain unknown, but these are fields of active investigation and the generation of animal models of tau-mediated neurodegeneration may facilitate this research.

Animal models of FTDP17 (see also Chapter 24) Experimental and transgenic (TG) models of tauopathies will serve as informative systems for elucidating the role of abnormalities in tau in the onset and progression of disease as well as providing useful models for the development of novel therapies. Several models of tau pathology were produced by overexpressing human tau proteins in mice (for review see Götz, 2001). However, these mice were either asymptomatic or developed pathology that was localized to the spinal cord or lacked many of the features of tau-based disorders. In contrast, the introduction of the P301L mutation led to the development of TG mice that develop age- and gene dose-dependent accumulation of tau tangles in the brain and spinal cord with associated nerve cell loss and gliosis as well as behavioural abnormalities (Lewis et al., 2000; Götz et al., 2001a). Similar to human disease, the tau aggregates were composed of only mutant human tau further implicating the P301L change in promoting the selective aggregation of mutant tau. Gene expression profiling of brain tissue from these mice showed altered expression of genes contributing to the inhibition of apoptosis, inflammation, and intracellular transport (Ho et al., 2001). Interestingly, the co-expression or injection of A in mice expressing P301L mutant tau led to enhanced neurofibrillary degeneration with preferential involvement of the limbic system similar to that observed in AD (Lewis et al., 2001; Götz et al., 2001b). This observation suggests that there may be a mechanistic interaction between the A and tau pathologies that influence their distribution and abundance. Subsequently, TG mice were developed with different FTDP-17 mutations that also lead to neurodegeneration in either neurons (V337M) (Tanemura et al., 2001, 2002) or oligodendrocytes (G272V) (Götz et al., 2001c). Other systems were also developed to model aspects of human tauopathies. Overexpression of tau in lamprey reticulospinal neurons led to the formation of PHF-like tau inclusions with degeneration of a subset of neurons (Hall et al., 1997, 2000). In addition, over-expression of either WT or FTDP-17 mutant tau (R406W and V337M) in Drosophila melanogaster demonstrated key features of tauopathies including adult-onset progressive neurodegeneration with accumulation of abnormal tau (Wittmann et al., 2001). However, the neurodegeneration occurred in the absence of NFT formation. More recent studies demonstrated NFTlike pathology when tau was co-expressed with shaggy which is similar to glycogen synthase 3-kinase (Jackson et al., 2002). Nonetheless, all of these models show various features of tau-related disorders that will hopefully facilitate an understanding of the molecular mechanisms underlying tau neurotoxicity.

273

274


Amyotrophic lateral sclerosis/parkinsonism–dementia complex While FTDP-17 affects predominantly the Caucasian and Asian populations worldwide, ALS/PDC is a disease of specific ethnic groups, most commonly the Chamorros on the island of Guam. Guam is the largest of the Mariana islands, an archipelago composed of 15 islands located in the western Pacific Ocean. Guam has a population of approximately 154 000 and about 47% of these inhabitants are an indigenous native people referred to as Chamorros (United States Census, 2000). In the mid-1940s, Harry Zimmerman, a neuropathologist with the United States Navy, was asked to evaluate diseases of interest to the military and he identified a cluster of cases of ALS among the native Chamorro population in a report to his supervisors (Zimmerman, 1945). These initial observations led to a number of studies attempting to characterize two major Chamorro disorders that are now known as ALS and PDC in the scientific literature, (or ALS/PDC when they occur together), but they are referred to as lytico (paralysis) and bodig (lazy or slow), respectively, by the indigenous Chamorros on Guam (Koerner, 1952; Arnold et al., 1953; Kurland & Mulder, 1954, 1955b; Hirano et al., 1961a,b). The initial research efforts by US investigators identified ALS as highly prevalent among the native Chamorro population (Koerner, 1952; Arnold et al., 1953; Kurland & Mulder, 1954, 1955b). Subsequently, Hirano noted a second neurodegenerative disorder common among the Chamorros characterized by dementia and parkinsonism and coined the term PDC (Hirano et al., 1961a,b). These disorders were exceptionally common with a prevalence of each disorder of ∼120–140:100,000 among the indigenous Chamorro population (Kurland & Mulder, 1954; Lessel et al., 1962). Since Guam ALS/PDC is characterized neuropathologically by tau inclusions identical to the NFTs observed in AD, it was hoped that the study of the aetiopathogenesis of this disorder would help identify environmental and genetic factor(s) in the pathogenesis of AD and tauopathies (Hirano et al., 1968; Shankar et al., 1989; Buée-Scherrer et al., 1995; Mawal-Dewan et al., 1996).

Clinical features of ALS/PDC While Guam ALS and PDC were initially described as separate entities, there is extensive clinical and neuropathologic overlap between the two disorders. Guam ALS is clinically indistinguishable from ALS in the continental United States and elsewhere in the world, characterized by upper and lower motor neuron symptoms including progressive weakness, muscular atrophy with fasciculations, and

hyperreflexia (Kurland et al., 1956). The disease invariably progresses to flaccid paralysis with dysphagia and dysarthria leading to death usually within 3–4 years. In contrast, PDC of Guam is characterized clinically by progressive parkinsonism with associated cognitive decline (Hirano et al., 1961a,b; Murakami, 1999). Initial symptoms may be those of parkinsonism, dementia, or both. The parkinsonism typically consists of profound bradykinesia and rigidity with impaired postural reflexes that cause gait abnormalities and characteristic masked facies. The resting tremor typical of Parkinson’s disease (PD) is not a common feature. The dementia consists of prominent memory deficits, disorientation, and deterioration of intellectual function as well as variable personality and behavioural alterations. The disease invariably progresses with death in ∼4 to 6 years, although long-term survival of up to 15 years has been reported (Murakami, 1999). Patients with clinical features of both Guam ALS and PDC are common. Up to 38% of the Chamorro people initially diagnosed with PDC develop superimposed amyotrophy (Elizan et al., 1966; Rodgers-Johnson et al., 1986). Moreover, while it is uncommon for patients diagnosed with ALS to manifest parkinsonism, PET scans with 18 F-6-fluorodopa demonstrated reduced striatal uptake intermediate between Guamanian control and Guam PDC patients, which suggests the presence of subclinical pathology in the striatonigral system (Elizan et al., 1966; Rodgers-Johnson et al., 1986; Snow et al., 1990). Occasionally, affected Chamorros present with a pure dementing illness that is similar to AD in the absence of extrapyramidal symptoms, referred to as ‘Marianas dementia’ (Perl et al., 1994; Galasko et al., 2002). These patients are typically older than those with PDC since they have a mean age of onset of 73 years and women exceed men by almost 3 to 1. Although only a limited number of autopsies have been performed, the neuropathology is typical of either ALS/PDC or AD. Two additional foci of similar endemic neurodegenerative disorders were identified in the Kii peninsula of Japan and southwestern New Guinea. A high prevalence of ALS was identified in 2 villages in the Kii peninsula on the island of Honshu (Kimura et al., 1961; Shiraki & Yase, 1975). The neuropathology was similar to that described in Guam ALS. While initial reports of PDC were rare, more recently, several patients with PDC similar both clinically and pathologically to that observed in Guam have been described (Kuzuhara et al., 1996, 2001). In addition, Gajdusek described the increased prevalence of a disorder clinically resembling both ALS and PDC in the Auyu and Jakai people in southwestern New Guinea (Gajdusek, 1963; Gajdusek & Salazar, 1982; Rodgers-Johnson et al., 1986). However, to date there


Fig. 12.9. Microscopic pathology in ALS/PDC. (a)–(c) Spinal cord pathology in Chamorro with ALS. (a) Degeneration of lateral corticospinal tracts (arrowheads) (Kluver–Barrera stain). (b) Severe neuron loss in anterior horn. The single remaining motor neuron in the anterior horn shows chromatolytic change (arrowhead). (c) Chromatolytic change in motor neuron in anterior horn. (d )-(f ) Pathology in hippocampus of Guam PDC patient. (d ) Severe neuron loss in CA1 region of hippocampus. Note numerous ghost tangles (arrowheads). (e),(f ) Hirano bodies (arrowheads) and granulovacuolar degeneration (arrow) in large pyramidal neurons of CA1 region. ((b)–(f ), H&E).

is no neuropathology available from affected patients of this region.

Neuropathology of ALS/PDC The brains of affected individuals show generalized cortical atrophy, particularly in the frontal and temporal lobes, and hydrocephalus ex vacuo with brain weights often less than 1000 g. In Guam ALS, similar to sporadic ALS, there is degeneration of the lateral columns with atrophy of the ventral roots of the spinal cord. In Guam PDC, there is usually depigmentation of the substantia nigra and locus ceruleus. There is associated severe neuron loss in affected brain regions particularly the hippocampus, frontal, temporal, and entorhinal cortex, nucleus basalis of Meynert, tegmentum of the brainstem, and substantia nigra (Fig. 12.9) (Hirano et al., 1961b, 1966; Nakano & Hirano, 1983). The thalamus and basal ganglia show moderate neuron loss. Eosinophilic rod-like inclusions (Hirano bodies) identical to that observed in AD, composed of actin and

actin-binding proteins as well as medium weight neurofilaments, are abundant in the CA1 region of the hippocampus (Fig. 12.9) (Hirano, 1965; Hirano et al., 1968; Goldman, 1983; Schmidt et al., 1989). Similarly, granulovacuolar degeneration of Simchowicz is prominent in many of the neurons in Ammon’s horn (Fig. 12.9) (Hirano, 1965; Hirano et al., 1968). In patients with prominent parkinsonism, there is severe neuron loss and gliosis in the substantia nigra, particularly involving dopaminergic neurons (Hirano et al., 1961b; Goto et al., 1990). In contrast, in Guamanian ALS, there is profound loss of neurons in the anterior horn of the spinal cord, the hypoglossal nucleus, as well as Betz cells of the motor cortex with relative preservation of Clarke’s column, intermediolateral nucleus, and Onuf’s nucleus (Fig. 12.9) (Hirano, 1965; Hirano et al., 1966). The remaining anterior horn cells may be normal, shrunken, or chromatolytic with perikaryal collections of phosphorylated neurofilaments (Schmidt et al., 1987). In addition, similar to sporadic ALS, there are ubiquitinated filamentous inclusions (Fig. 12.10) and small eosinophilic, cytoplasmic inclusions

275

276


termed ‘Bunina bodies’ (Hirano, 1965; Matsumoto et al., 1990b). Furthermore, there is an associated degeneration of the lateral and uncrossed anterior corticospinal tracts secondary to loss of upper motor neurons (Fig. 12.9). The neuropathological hallmark of Guam ALS/PDC is the widespread formation of NFTs, especially in the hippocampal formation, neocortex, and basal forebrain (Fig. 12.10) (Malamud et al., 1961; Hirano et al., 1961b, 1966, 1968). In many cases, virtually every neuron in the CA1 region may show NFT formation, a finding that is associated with severe neuron loss and prominent extracellular NFT formation (Hof et al., 1994b; Schwab et al., 1998). Globose NFTs are frequently observed in the brainstem, in particular the substantia nigra, locus ceruleus, and periaquiductal grey matter (Hirano et al., 1961b). Occasional NFTs are also noted in both the anterior and posterior horns of the spinal cord (Fig. 12.10) (Hirano et al., 1966; Matsumoto et al., 1990a). By immunohistochemical and ultrastructural criteria, the NFTs of Guam ALS/PDC are indistinguishable from those found in AD (Fig. 12.10) )(Hirano et al., 1968; Shankar et al., 1989; Buée-Scherrer et al., 1995; Mawal-Dewan et al., 1996). Ultrastructurally, the NFTs contain both PHFs and SFs that are composed of the MAP tau. In addition, the NFTs react with antibodies to ubiquitin (Matsumoto et al., 1990b; Ikemoto et al., 1997), apolipoprotein E (Buée et al., 1996), and A peptide (Ito et al., 1991), as well as phosphorylated neurofilament proteins (Shankar et al., 1989). Biochemically, the NFTs in Guam ALS and PDC are composed of insoluble aggregates of hyperphosphorylated tau consisting of all six tau isoforms, similar to the NFTs in AD (Figs. 12.7 and 12.10) (Hof et al., 1994a; Buée-Scherrer et al., 1995; Mawal-Dewan et al., 1996; Schmidt et al., 2001). In addition, there are accumulations of -synuclein filaments in Lewy bodies in neurons of the amygdala in a large subset of patients with both disorders (Fig. 12.10) (Yamazaki et al., 2000; Forman et al., 2002). Lewy bodies are otherwise only rarely encountered in the substantia nigra or locus ceruleus of Guam ALS/PDC patients (Hirano et al., 1966; RodgersJohnson et al., 1986). There are differences between the tau pathology in these Guam disorders and AD. Hof and colleagues demonstrated that the NFTs are preferentially distributed in layers II and III of the neocortex, while in AD the pathology is more abundant in layers V and VI (Fig. 12.10) (Hof et al., 1991, 1994a,b). There is typically more extensive subcortical and brainstem pathology in Guam ALS/PDC than in AD (Hirano et al., 1966; Hirano & Llena, 1986). Furthermore, there is abundant glial pathology in Guam ALS/PDC, most prominently in the amygdala, inferior olives, and lateral funiculus of the spinal cord, including granular, hazy inclusions in as-

trocytes and coiled bodies in oligodendrocytes (Fig. 12.10) (Oyanagi et al., 1997). Finally, in contrast to AD, there is the relative absence of senile plaques and amyloid angiopathy with mild A deposition present in only about 20–25% of Guam ALS/PDC cases (Hirano et al., 1961b; Gentleman et al., 1991; Hof et al., 1994a; Schmidt et al., 1998).

Guam ALS and PDC: one disease or two The distribution of NFTs and neuron loss in the two major forms of Chamorro neurodegenerative disease varies with the burden of the neuropathology roughly reflecting the clinical manifestations of each disease )(Hirano et al., 1961a; Hirano & Llena, 1986; Hof et al., 1994a,b; Galasko et al., 2002). For example, the major neuropathology in ALS is in the caudal neuraxis, especially in the spinal cord where there is extensive neurodegeneration of spinal motor neurons in the anterior horn accompanied by accumulations of NFTs in surviving spinal cord neurons with lesser involvement of more rostral brain regions. In contrast, NFTs accumulate in rostral portions of the CNS in PDC, especially in hippocampus, other limbic regions, and in neocortex. However, the neuropathology of ALS and PDC of Guam shows significant overlap of these entities raising the possibility that they represent different ends of a spectrum of a single neurodegenerative disorder with a common pathogenesis (Hirano et al., 1961b). For instance, there is evidence of neuronal loss and depigmentation of the substantia nigra in 63% of Chamorros clinically diagnosed with ALS (Rodgers-Johnson et al., 1986). Similarly, over one-third of Guam PDC patients show pathological changes characteristic of ALS (Hirano et al., 1961b; Rodgers-Johnson et al., 1986). However, Oyanagi and colleagues argue that Guam ALS and PDC are distinct entities based on the widespread presence of NFTs in both affected and unaffected Chamorros (Oyanagi et al., 1994; Oyanagi & Wada, 1999). Neuropathological studies on the prevalence of neurofibrillary pathology in the brains of Chamorros without symptoms of ALS or PDC showed increased frequency with age, independent of clinical disease (Anderson et al., 1979; Chen, 1981; Perl et al., 1995). These studies demonstrated that the often widespread presence of NFTs is detected in Chamorros in their 20s and 30s and up to 100% in Chamorros over the age of 50. In light of this data, Oyanagi states that while Guam PDC is a distinct entity, NFT pathology in Guam ALS represents a background feature widely distributed in this population. However, it remains unclear whether the NFTs identified in asymptomatic individuals represent a preclinical stage of disease or background pathology in the Chamorro population.

Fig. 12.10. Tau pathology in ALS/PDC. (a) Abundant NFTs in CA1 region of hippocampus in Guam PDC patient. (b) NFTs in frontal cortex of Chamorro with PDC are preferentially distributed in layers II and III of the neocortex. (c)–(e) NFTs in the hippocampus (c), frontal cortex (d ), and anterior horn of spinal cord (e) of Chamorros with PDC stained with anti-tau antibody (d ),(e) or Thioflavin S (c). (f ) Granular, hazy tau inclusions in astrocytes of the amygdala of Guam PDC patient. (g) Ubiquitin positive inclusion in large motor neuron within the anterior horn in Chamorro with ALS. These inclusions are morphologically identical to those in sporadic ALS (IHC with anti-ubiquitin antibody). (h) Synuclein positive Lewy bodies and neurites in the amygdala of Guam PDC patient (IHC with anti-synuclein antibody). (a),(b),(d )–(f ), IHC with anti-tau antibody). (i) Insoluble fractions from AD and Guam ALS/PDC patients show similar patterns of tau isoform composition. Sarkosyl insoluble fractions from grey and white matter of different brain regions from a Caucasian patient with AD (left panel) and Guamanian ALS/PDC patient (right panel) were dephosphorylated with E. coli alkaline phosphatase, resolved by SDS-PAGE and immunoblotted with anti-tau antibodies. In both AD and ALS/PDC patients all six tau isoforms are detected. Recombinant tau isoforms are as indicated. Fr, frontal lobe; Te, temporal lobe; Pa, parietal lobe; Oc, occipital lobe; Ce, cerebellum; G, grey matter; W, white matter; rTau, recombinant tau.

278


Epidemiology In light of the neuropathologic similarities between Guam ALS/PDC and other tauopathies including AD, it was hoped that insights into the underlying etiologic factors of this perplexing enigma would provide important clues into mechanisms underlying the more common neurodegenerative disorders elsewhere in the world. Thus, the Chamorros appear to be a tractable population for elucidating genetic and/or environmental causes of disease because they are an island-bound, inbred population that has resided on Guam for centuries, although the population has fluctuated over time. The first epidemiological study of the prevalence of Guam ALS revealed a prevalence of ∼140 per 100 000 (incidence ∼60 per 100 000), vs. 1–6 per 100 000 in the Caucasian population (Kurland & Mulder, 1954). Males were affected approximately 1.5–2 times more frequently than females, with disease onset at 44 years of age. Similarly, the initial epidemiologic studies on Guam PDC indicated a prevalence of ∼120 per 100 000, with males affected 2.5 times more frequently and age at disease onset of 52 years (Lessel et al., 1962). Furthermore, there were villages where the new incidence rate of ALS approached 250 per 100 000 (Kurland & Mulder, 1954; Plato et al., 1969; Reed & Brody, 1975). In subsequent decades, from 1960–1983, the prevalence of ALS decreased sharply as did PDC, albeit to a lesser extent (Reed & Brody, 1975; Garruto et al., 1985; Rodgers-Johnson et al., 1986; Zhang et al., 1990). Moreover, the decreased incidence was associated with a delay in the age at onset of both disorders relative to patients affected by these diseases in the 1950s. For instance, the mean age of disease onset for Guam ALS shifted from 48 to 52 years while Guam PDC shifted from 53 to 59 years. More recent studies demonstrate the persistent occurrence of PDC among Chamorros on Guam, albeit about a decade later in their lifespan (mean age of onset is 68 years), while cases of ALS continue to diminish in frequency (McGeer et al., 1997; Wiederholt, 1999; Galasko et al., 2002). Thus, in the 1950s and 1960s, PDC was about as common as ALS in Chamorros. In contrast, the prevalence of PDC is about nine times that of ALS today (Galasko et al., 2002). Furthermore, the incidence of Mariana dementia appears to be increasing with a prevalence approaching that of Guam PDC (Galasko et al., 2002). These changes in the expression of Chamorro neurodegenerative diseases remain unexplained but are likely to reflect an interaction of environmental and genetic factors among the Chamorros living on Guam.

Aetiology/pathogenesis Numerous theories have been proposed to explain the high prevalence of ALS/PDC among the Chamorro population

(Kurland, 1988; Stone, 1993). Initially, ALS and PDC were thought to represent genetic disorders (Kurland & Mulder, 1954, 1955b). These early studies demonstrated that close relatives of affected subjects were at greater risk for disease than relatives of controls with up to 40% of patients having an affected relative (Koerner, 1952; Kurland & Mulder, 1954, 1955b). In addition, there were many families with multiple individuals in 2 or more generations including some pedigrees where four successive generations were affected (Koerner, 1952; Plato et al., 1967, 1969). However, subsequent prospective analyses did not support a genetic etiology (Reed et al., 1975; Plato et al., 1986). Furthermore, the identification of ALS and PDC among Filipino migrants to Guam implicated environmental factors in the pathogenesis of these disorders (Garruto et al., 1981). Infectious disease, including viruses or prions, have been proposed as possible etiological agents similar to that observed in post-encephalitic parkinsonism and Creutzfeld– Jakob disease, respectively (Gibbs, Jr. & Gajdusek, 1982; Hudson & Rice, 1990). However, to date, all attempts to recover an infectious agent or to transmit disease by intracerebral injection were unsuccessful. Cycads, the highly toxic seed of the false sago palm, were also proposed as a possible etiological agent by Whiting in 1964 (Whiting, 1964). Cycad, used as a food source in flour as well as a poultice in traditional medicine by the Chamorros of Guam, is highly hepatotoxic if ingested in its raw form. However, the seeds are detoxified by a ritualistic washing that lasts 7 days. The description of a degenerative locomotor disease in animals grazing on cycads fueled interest in the aetiological role of this plant (Conference Participants, 1964). Toxicological studies on the cycad seeds identified a potent alkylating agent, cycasin, that was both hepatotoxic and carcinogenic (Laqueur et al., 1963; Matsumoto & Strong, 1963). Yet, efforts to induce a disorder in the laboratory similar to ALS/PDC were unsuccessful (Yang et al., 1966). More recently, a metabolite of cycasin, methylazoxymethanol, was shown to damage neuronal DNA and modulate tau mRNA expression, indirectly implicating the alkylating agent in neuronal degeneration (Esclaire et al., 1999). Interest in an alternative cycad hypothesis arose in the late 1980s when Spencer and colleagues described the development of motor dysfunction, parkinsonian features, and behavioural abnormalities with associated degeneration of motor neurons in cerebral cortex and spinal cord in cynomolgus monkeys that were fed large doses of -N-methylaminol-alanine (BMAA) (Spencer et al., 1987). BMAA is an ‘unusual’ amino acid that is present in low quantities in cycad seeds. Unfortunately, the doses required to achieve toxicity were unrealistic, and BMAA is readily removed from the seed with washing (Duncan et al., 1990). However, Cox and


Sacks have yet again resurrected the cycad toxicity theory by arguing for ‘biomagnification’ (or concentration) of toxins by consumption of flying foxes, an indigenous animal of Guam. The flying foxes ingest large quantities of cycads and are eaten during ritualistic ceremonies by Chamorros (Cox & Sacks, 2002). However, data in support of this notion is lacking at this time. The possibility that neurotoxic trace elements might be related to ALS and PDC was first proposed by Yase in 1972 (Yase, 1972). Relatively high levels of aluminium and low levels of calcium and magnesium were found in samples of drinking water and soil in all three foci of ALS and PDC including Guam, the Kii peninsula of Japan and western New Guinea. Aluminium has been shown to induce neurofibrillary degeneration in animals exposed to aluminiumcontaining salts (Klatzo et al., 1965; Terry & Pena, 1965). Moreover, intraneuronal accumulation of aluminium was demonstrated in cortical and NFT-bearing neurons of patients with AD (Perl & Brody, 1980). In 1982, Perl and colleagues using electron probe microanalysis demonstrated intraneuronal accumulation of aluminium in hippocampal neurons in Chamorros of Guam both with and without neurological evidence of disease (Perl et al., 1982). Subsequent reports demonstrated the accumulation of not only aluminium, but also of calcium, silicon, and iron in NFT-bearing neurons (Garruto et al., 1984, 1986; Good & Perl, 1994). While the mechanism of aluminium accumulation is unknown, it was proposed that the deficiency of calcium and magnesium in the soil leads to increased uptake of alumnium as an alternative source of cations (Garruto et al., 1989). Alternatively, the accumulation of iron with aluminium in neurons causes oxidative stress leading to degeneration of vulnerable neurons (Olanow & Perl, 1994). Despite these efforts to identify environmental factor(s) that lead to ALS and PDC, definitive evidence of cause and effect is lacking. In contrast, the possibility of a role for genetics in susceptibility to Guam ALS and PDC is gaining popularity. Two recent long-term studies indicated that relatives of family members with ALS/PDC are at greater risk for disease than the Guamanian population, whereas relatives of controls have reduced disease risk (McGeer et al., 1997; Plato et al., 2002). Furthermore, segregation analysis rejects purely genetic or environmental hypothesis, but rather, is consistent with a two allele major locus hypothesis. This hypothesis suggests that a gene in combination with environmental factors may be involved in the etiology of ALS/PDC (Bailey-Wilson et al., 1993). In addition, a polymorphism in exon 9 of the gene CYP2D6, a genetic risk factor for PD and linked to slower metabolism of exogenous toxins, was higher in Chamorros and Filipinos than in Caucasian individuals (Chen et al., 1996). However, this

polymorphism had similar frequencies in both affected and healthy Chamorro individuals. The neuropathological similarities of ALS/PDC to AD led to the investigation of whether genetic factors that predispose to AD might also predispose to ALS/PDC. Sequencing of the amyloid precursor protein, the source of A in AD, failed to identify mutations (Harlin et al., 1993). In addition, the APOE-ε4 allele, a strong risk factor for AD, was uniformly low in both affected and healthy Chamorros (Waring et al., 1994; Buée et al., 1996; Chen et al., 1996; Galasko et al., 2002). The discovery of pathogenic tau mutations in FTDP-17 stimulated efforts to identify similar mutations in Chamorros. To date, no mutations in tau have been identified in Guam ALS/PDC (Perez-Tur et al., 1999; Poorkaj et al., 2001b). However, a polymorphism in the A0 allele (142 allele) associated with both PSP and CBD is in linkage disequilibrium in affected Chamorros implicating tau in the genetic susceptibility to both ALS and PDC (Poorkaj et al., 2001b). Thus, there are intriguing clues suggesting a powerful but unexplained mechanistic interaction between genes of the Chamorro population of Guam and the environment in the Mariana Islands to account for the shifting incidence and phenotypic manifestations of neurodegenerative diseases affecting this unique minority population.

Conclusions The accumulation of filamentous tau inclusions is a common feature of a wide variety of sporadic and familial neurodegenerative disorders that present clinically with FTD. These tauopathies are distinguished by the distinct topographical and cell type-specific distribution of inclusions. The biochemical and ultrastructural characteristics of the tau pathology also reveals a significant phenotypic overlap. The discovery of multiple mutations in tau that lead to the abnormal aggregation of tau and cause FTDP-17 demonstrates that tau dysfunction is sufficient to produce neurodegenerative disease. The mutations lead to specific alterations in the expression, function, and biochemistry of the tau protein. Furthermore, in light of the similarities between Guam ALS/PDC and other tauopathies, it is likely that elucidation of this perplexing enigma will provide important clues into mechanisms underlying the interaction between genetic and environmental factors in neurodegenerative disease. The identification of additional gene mutations or polymorphisms at distinct genetic loci in families with FTD that either cause or are risk factors for disease will provide additional insights into disease pathogenesis as well as in the development of novel strategies for treatment and prevention.

279

280


Acknowledgements V.M.-Y. Lee is the John H. Ware 3rd Professor of Alzheimer’s Disease Research at the University of Pennsylvania. This work was supported by grants from the National Institute of Aging of the National Institute of Health and the Alzheimer’s Association. We thank Dr Victoria Zhukareva who generously provided many of the Western blots. We also thank the many patients studied and their families for making the research reviewed here possible.

REFERENCES Anderson, F. H., Richardson, E. P., Jr., Okazaki, H. & Brody, J. A. (1979). Neurofibrillary degeneration on Guam: frequency in Chamorros and non-Chamorros with no known neurological disease. Brain, 102, 65–77. Andreadis, A., Brown, W. M. & Kosik, K. S. (1992). Structure and novel exons of the human tau gene. Biochemistry, 31, 10626–33. Arawaka, S., Usami, M., Sahara, N., Schellenberg, G. D., Lee, G. & Mori, H. (1999). The tau mutation (val337met) disrupts cytoskeletal networks of microtubules. Neuroreport, 10, 993–7. Arima, K., Kowalska, A., Hasegawa, M. et al. (2000). Two brothers with frontotemporal dementia and parkinsonism with an N279K mutation of the tau gene. Neurology, 54, 1787–95. Arnold, A., Edgren, D. C. & Palladino, V. S. (1953). Amyotrophic lateral sclerosis. Fifty cases observed on Guam. J. Nerv. Ment. Disord., 117, 135–9. Arrasate, M., Perez, M., Armas-Portela, R. & Avila, J. (1999). Polymerization of tau peptides into fibrillar structures. The effect of FTDP-17 mutations. FEBS Lett., 446, 199–202. Ashworth, A., Lloyd, S., Brown, J. et al. (1999). Molecular genetic characterization of frontotemporal dementia on chromosome 3. Dement. Geriatr. Cogn. Disord., 10 (Suppl. 1), 93–101. Bailey-Wilson, J. E., Plato, C. C., Elston, R. C. & Garruto, R. M. (1993). Potential role of an additive genetic component in the cause of amyotrophic lateral sclerosis and parkinsonism–dementia in the western Pacific. Am. J. Med. Genet., 45, 68–76. Baker, M., Kwok, J. B., Kucera, S. et al. (1997). Localization of frontotemporal dementia with parkinsonism in an Australian kindred to chromosome 17q21–22. Ann. Neurol., 42, 794–8. Barghorn, S., Zheng-Fischhofer, Q., Ackmann, M. et al. (2000). Structure, microtubule interactions, and paired helical filament aggregation by tau mutants of frontotemporal dementias. Biochemistry, 39, 11714–21. Biernat, J., Gustke, N., Drewes, G., Mandelkow, E. M. & Mandelkow, E. (1993). Phosphorylation of Ser262 strongly reduces binding of tau to microtubules: distinction between PHF-like immunoreactivity and microtubule binding. Neuron, 11, 153–63. Billingsley, M. L. & Kincaid, R. L. (1997). Regulated phosphorylation and dephosphorylation of tau protein: effects on microtubule interaction, intracellular trafficking and neurodegeneration. Biochem. J., 323, 577–91.

Binder, L. I., Frankfurter, A. & Rebhun, L. I. (1985). The distribution of tau in the mammalian central nervous system. J. Cell Biol., 101, 1371–8. Bird, T. D., Wijsman, E. M., Nochlin, D. et al. (1997). Chromosome 17 and hereditary dementia: linkage studies in three nonAlzheimer families and kindreds with late-onset FAD. Neurology, 48, 949–54. Bird, T. D., Nochlin, D., Poorkaj, P. et al. (1999). A clinical pathological comparison of three families with frontotemporal dementia and identical mutations in the tau gene (P301L). Brain, 122, 741–56. Bramblett, G. T., Goedert, M., Jakes, R., Merrick, S. E., Trojanowski, J. Q. & Lee, V. M.-Y. (1993). Abnormal tau phosphorylation at Ser396 in Alzheimer’s disease recapitulates development and contributes to reduced microtubule binding. Neuron, 10, 1089–99. Brown, J. (1998). Chromosome 3-linked frontotemporal dementia. Cell Mol. Life Sci., 54, 925–7. Brun, A. (1987). Frontal lobe degeneration of non-Alzheimer type. I. Neuropathology. Arch. Gerontol. Geriatr., 6, 193–208. (1993). Frontal lobe degeneration of non-Alzheimer type revisited. Dementia, 4, 126–31. Buée-Scherrer, V., Buée, L., Hof, P. R. et al. (1995). Neurofibrillary degeneration in amyotrophic lateral sclerosis/parkinsonism– dementia complex of Guam. Immunochemical characterization of tau proteins. Am. J Pathol, 146, 924–32. Buée, L., Perez-Tur, J., Leveugle, B. et al. (1996). Apolipoprotein E in Guamanian amyotrophic lateral sclerosis/parkinsonism– dementia complex: genotype analysis and relationships to neuropathological changes. Acta Neuropathol. (Berl), 91, 247–53. Buée, L., Bussière, T., Buée-Scherrer, V., Delacourte, A. & Hof, P. R. (2000). Tau protein isoforms, phosphorylation and role in neurodegenerative disorders. Brain Res. Rev., 33, 1–36. Bugiani, O., Murrell, J. R., Giaccone, G. et al. (1999). Frontotemporal dementia and corticobasal degeneration in a family with a P301S mutation in tau. J Neuropathol. Exp. Neurol., 58, 667–77. Butner, K. A. & Kirschner, M. W. (1991). Tau protein binds to microtubules through a flexible array of distributed weak sites. J Cell Biol., 115, 717–30. Chen, L. (1981). Neurofibrillary change on Guam. Arch. Neurol., 38, 16–18. Chen, X., Xia, Y., Gresham, L. S. et al. (1996). ApoE and CYP2D6 polymorphism with and without parkinsonism–dementia complex in the people of Chamorro, Guam. Neurology, 47, 779–84. Chow, T. W., Miller, B. L., Hayashi, V. N. & Geschwind, D. H. (1999). Inheritance of frontotemporal dementia. Arch. Neurol., 56, 817–22. Clark, L. N., Poorkaj, P., Wszolek, Z. et al. (1998). Pathogenic implications of mutations in the tau gene in pallido-ponto-nigral degeneration and related neurodegenerative disorders linked to chromosome 17. Proc. Natl. Acad. Sci. USA, 95, 13103–7. Cleveland, D. W., Hwo, S. Y. & Kirschner, M. W. (1977). Purification of tau, a microtubule-associated protein that induces assembly of microtubules from purified tubulin. J Mol. Biol., 116, 207–25.


Conference Participants (1964). Proceedings on the third conference on the identification of toxic elements of cycads. Fed. Proc., 23, 1335–88. Connell, J. W., Gibb, G. M., Betts, J. C. et al. (2001). Effects of FTDP17 mutations on the in vitro phosphorylation of tau by glycogen synthase kinase 3beta identified by mass spectrometry demonstrate certain mutations exert long-range conformational changes. FEBS Lett., 493, 40–4. Constantinidis, J., Richard, J. & Tissot, R. (1974). Pick’s disease. Histological and clinical correlations. Eur. Neurol., 11, 208–17. Cooper, P. N., Siddons, M. & Mann, D. M. (1995). Normal frontal cortex histology and immunohistochemistry in patients with motor neuron disease. J. Neurol. Neurosurg. Psychiatry, 59, 644. Couchie, D., Mavilia, C., Georgieff, I. S., Liem, R. K., Shelanski, M. L. & Nunez, J. (1992). Primary structure of high molecular weight tau present in the peripheral nervous system. Proc. Natl Acad. Sci. USA, 89, 4378–81. Cox, P. A. & Sacks, O. W. (2002). Cycad neurotoxins, consumption of flying foxes & ALS-PDC disease in Guam. Neurology, 58, 956–9. D’Souza, I. & Schellenberg, D. (2000). Determinants of 4 repeat tau expression: Coordination between enhancing and inhibitory splicing sequences for exon 10 inclusion. J Biol. Chem., 275, 17700–9. D’Souza, I., Poorkaj, P., Hong, M. et al. (1999). Missense and silent tau gene mutations cause frontotemporal dementia with parkinsonism-chromosome 17 type, by affecting multiple alternative RNA splicing regulatory elements. Proc. Natl Acad. Sci. USA, 96, 5598–603. Dayanandan, R., van Slegtenhorst, M., Mack, T. G. et al. (1999). Mutations in tau reduce its microtubule binding properties in intact cells and affect its phosphorylation. FEBS Lett., 446, 228–32. Delisle, M. B., Murrell, J. R., Richardson, R. et al. (1999). A mutation at codon 279 (N279K) in exon 10 of the Tau gene causes a tauopathy with dementia and supranuclear palsy. Acta Neuropathol. (Berl), 98, 62–77. DeTure, M., Ko, L. W., Yen, S. et al. (2000). Missense tau mutations identified in FTDP-17 have a small effect on tau-microtubule interactions. Brain Res., 853, 5–14. Drechsel, D. N., Hyman, A. A., Cobb, M. H. & Kirschner, M. W. (1992). Modulation of the dynamic instability of tubulin assembly by the microtubule-associated protein tau. Mol. Biol. Cell, 3, 1141–54. Dumanchin, C., Camuzat, A., Campion, D. et al. (1998). Segregation of a missense mutation in the microtubule-associated protein tau gene with familial frontotemporal dementia and parkinsonism. Hum. Mol. Genet., 7, 1825–9. Duncan, M. W., Steele, J. C., Kopin, I. J. & Markey, S. P. (1990). 2-Amino-3-(methylamino)-propanoic acid (BMAA) in cycad flour: an unlikely cause of amyotrophic lateral sclerosis and parkinsonism-dementia of Guam. Neurology, 40, 767–72. Elizan, T. S., Hirano, A., Abrams, B. M., Need, R. L., van Nuis, C. & Kurland, L. T. (1966). Amyotrophic lateral sclerosis and parkinsonism–dementia complex of Guam. Neurological evaluation. Arch. Neurol., 14, 356–68.

Esclaire, F., Kisby, G., Spencer, P., Milne, J., Lesort, M. & Hugon, J. (1999). The Guam cycad toxin methylazoxymethanol damages neuronal DNA and modulates tau mRNA expression and excitotoxicity. Exp. Neurol., 155, 11–21. Fabre, S. F., Forsell, C., Viitanen, M. et al. (2001). Clinic-based cases with frontotemporal dementia show increased cerebrospinal fluid tau and high apolipoprotein E epsilon4 frequency, but no tau gene mutations. Exp. Neurol., 168, 413–18. Forman, M. S., Schmidt, M. L., Kasturi, S., Perl, D. P., Lee, V. M.-Y. & Trojanowski, J. Q. (2002). Tau and alpha-synuclein pathology in amygdala of parkinsonism–dementia complex patients of Guam. Am. J. Pathol., 160, 1725–31. Foster, N. L., Wilhelmsen, K., Sima, A. A., Jones, M. Z., D’Amato, C. J. & Gilman, S. (1997). Frontotemporal dementia and parkinsonism linked to chromosome 17: a consensus conference. Ann. Neurol., 41, 706–15. Frappier, T., Liang, N. S., Brown, K. et al. (1999). Abnormal microtubule packing in processes of SF9 cells expressing the FTDP17 V337M tau mutation. FEBS Lett., 455, 262–6. Furukawa, K., D’Souza, I., Crudder, C. H. et al. (2000). Pro-apoptotic effects of tau mutations in chromosome 17 frontotemporal dementia and parkinsonism. Neuroreport, 11, 57–60. Gajdusek, D. C. (1963). Motor-neuron disease in natives of New Guinea. N. Engl. J Med., 268, 474–6. Gajdusek, D. C. & Salazar, A. M. (1982). Amyotrophic lateral sclerosis and parkinsonian syndromes in high incidence among the Auyu and Jakai people of West New Guinea. Neurology, 32, 107–26. Galasko, D., Salmon, D. P., Craig, U. K., Thal, L. J., Schellenberg, G. & Wiederholt, W. (2002). Clinical features and changing patterns of neurodegenerative disorders on Guam, 1997–2000. Neurology, 58, 90–7. Gamblin, T. C., King, M. E., Dawson, H. et al. (2000). In vitro polymerization of tau protein monitored by laser light scattering: method and application to the study of FTDP-17 mutants. Biochemistry, 39, 6136–44. Gao, Q. S., Memmott, J., Lafyatis, R., Stamm, S., Screaton, G. & Andreadis, A. (2000). Complex regulation of tau exon 10, whose missplicing causes frontotemporal dementia. J. Neurochem., 74, 490–500. Garruto, R. M., Gajdusek, D. C. & Chen, K. M. (1981). Amyotrophic lateral sclerosis and parkinsonism–dementia among Filipino migrants to Guam. Ann. Neurol., 10, 341–50. Garruto, R. M., Fukatsu, R., Yanagihara, R., Gajdusek, D. C., Hook, G. & Fiori, C. E. (1984). Imaging of calcium and aluminum in neurofibrillary tangle-bearing neurons in parkinsonism– dementia of Guam. Proc. Natl. Acad. Sci. USA, 81, 1875–79. Garruto, R. M., Yanagihara, R. & Gajdusek, D. C. (1985). Disappearance of high-incidence amyotrophic lateral sclerosis and parkinsonism–dementia on Guam. Neurology, 35, 193–8. Garruto, R. M., Swyt, C., Yanagihara, R., Fiori, C. E. & Gajdusek, D. C. (1986). Intraneuronal co-localization of silicon with calcium and aluminum in amyotrophic lateral sclerosis and parkinsonism with dementia of Guam. N. Engl. J. Med, 315, 711–12.

281

282


Garruto, R. M., Shankar, S. K., Yanagihara, R., Salazar, A. M., Amyx, H. L. & Gajdusek, D. C. (1989). Low-calcium, high-aluminum diet-induced motor neuron pathology in cynomolgus monkeys. Acta Neuropathol. (Berl), 78, 210–19. Gentleman, S. M., Perl, D., Allsop, D., Clinton, J., Royston, M. C. & Roberts, G. W. (1991). Beta (A4)-amyloid protein and parkinsonian dementia complex of Guam. Lancet, 337, 55–6. Giannakopoulos, P., Hof, P. R. & Bouras, C. (1995). Dementia lacking distinctive histopathology: clinicopathological evaluation of 32 cases. Acta Neuropathol. (Berl.), 89, 346–55. Gibbs, C. J., Jr. & Gajdusek, D. C. (1982). An update on long-term in vivo and in vitro studies designed to identify a virus as the cause of amyotrophic lateral sclerosis, parkinsonism dementia, and Parkinson’s disease. Adv. Neurol., 36, 343–53. Goedert, M. & Jakes, R. (1990). Expression of separate isoforms of human tau protein: correlation with the tau pattern in brain and effects on tubulin polymerization. EMBO J., 9, 4225–30. Goedert, M., Wischik, C. M., Crowther, R. A., Walker, J. E. & Klug, A. (1988). Cloning and sequencing of the cDNA encoding a core protein of the paired helical filament of Alzheimer disease: identification as the microtubule-associated protein tau. Proc. Natl Acad. Sci. USA, 85, 4051–5. Goedert, M., Spillantini, M. G., Jakes, R., Rutherford, D. & Crowther, R. A. (1989a). Multiple isoforms of human microtubuleassociated protein tau: sequences and localization in neurofibrillary tangles of Alzheimer’s disease. Neuron, 3, 519–26. Goedert, M., Spillantini, M. G., Potier, M. C., Ulrich, J. & Crowther, R. A. (1989b). Cloning and sequencing of the cDNA encoding an isoform of microtubule-associated protein tau containing four tandem repeats: differential expression of tau protein mRNAs in human brain. EMBO J., 8, 393–9. Goedert, M., Jakes, R. & Crowther, R. A. (1999a). Effects of frontotemporal dementia FTDP-17 mutations on heparin-induced assembly of tau filaments. FEBS Lett., 450, 306–11. Goedert, M., Spillantini, M. G., Crowther, R. A. et al. (1999b). Tau gene mutation in familial progressive subcortical gliosis. Nat. Med., 5, 454–7. Goedert, M., Satumtira, S., Jakes, R. et al. (2000). Reduced binding of protein phosphatase 2A to tau protein with frontotemporal dementia and parkinsonism linked to chromosome 17 mutations. J. Neurochem., 75, 2155–62. Goldman, J. E. (1983). The association of actin with Hirano bodies. J. Neuropathol. Exp. Neurol., 42, 146–52. Good, P. F. & Perl, D. P. (1994). A quantitative comparison of aluminum concentration in neurofibrillary tangles of Alzheimer’s disease and parkinsonism–dementia complex of Guam by laser microprobe mass analysis. Neurobiol. Aging, 15, S28. Goode, B. L. & Feinstein, S. C. (1994). Identification of a novel microtubule binding and assembly domain in the developmentally regulated inter-repeat region of tau. J. Cell Biol., 124, 769–82. Goode, B. L., Denis, P. E., Panda, D. et al. (1997). Functional interactions between the proline-rich and repeat regions of tau enhance microtubule binding and assembly. Mol. Biol. Cell, 8, 353–65. Goto, S., Hirano, A. & Matsumoto, S. (1990). Immunohistochemical study of the striatal efferents and nigral dopaminergic neu-

rons in parkinsonism–dementia complex on Guam in comparison with those in Parkinson’s and Alzheimer’s diseases. Ann. Neurol., 27, 520–7. Götz, J. (2001). Tau and transgenic animal models. Brain Res. Rev., 35, 266–86. Götz, J., Chen, F., Barmettler, R. & Nitsch, R. M. (2001a). Tau filament formation in transgenic mice expressing P301L tau. J. Biol. Chem., 276, 529–34. Götz, J., Chen, F., Van Dorpe, J. & Nitsch, R. M. (2001b). Formation of neurofibrillary tangles in P301L tau transgenic mice induced by Abeta 42 fibrils. Science, 293, 1491–5. Götz, J., Tolnay, M., Barmettler, R., Chen, F., Probst, A. & Nitsch, R. M. (2001c). Oligodendroglial tau filament formation in transgenic mice expressing G272V tau. Eur. J Neurosci., 13, 2131–40. Grover, A., Houlden, H., Baker, M. et al. (1999). 5 splice site mutations in tau associated with the inherited dementia FTDP-17 affect a stem–loop structure that regulates alternative splicing of exon 10. J. Biol. Chem., 274, 15134–43. Grover, A., DeTure, M., Yen, S. H. & Hutton, M. (2002). Effects on splicing and protein function of three mutations in codon N296 of tau in vitro. Neurosci. Lett., 323, 33–6. Gunnarsson, L. G., Dahlbom, K. & Strandman, E. (1991). Motor neuron disease and dementia reported among 13 members of a single family. Acta Neurol. Scand., 84, 429–33. Hall, G. F., Yao, J. & Lee, G. (1997). Human tau becomes phosphorylated and forms filamentous deposits when overexpressed in lamprey central neurons in situ. Proc. Natl Acad. Sci. USA, 94, 4733–8. Hall, G. F., Chu, B., Lee, G. & Yao, J. (2000). Human tau filaments induce microtubule and synapse loss in an in vivo model of neurofibrillary degenerative disease. J. Cell Sci., 113 (8), 1373– 87. Harlin, M. C., Crawford, F., Perl, D. P., Steele, J. & Hardy, J. (1993). Sequencing of exons 16 and 17 of the beta-amyloid precursor protein gene reveals the beta-amyloid sequence to be normal in cases of the parkinson dementia complex of Guam. J. Neural Transm. Park. Dis. Dement. Sect., 5, 63–5. Hasegawa, M., Jakes, R., Crowther, R. A., Lee, V. M.-Y., Ihara, Y. & Goedert, M. (1996). Characterization of mAb AP422, a novel phosphorylation-dependent monoclonal antibody against tau protein. FEBS Lett., 384, 25–30. Hasegawa, M., Smith, M. J. & Goedert, M. (1998). Tau proteins with FTDP-17 mutations have a reduced ability to promote microtubule assembly. FEBS Lett., 437, 207–10. Hasegawa, M., Smith, M. J., Iijima, M., Tabira, T. & Goedert, M. (1999). FTDP-17 mutations N279K and S305N in tau produce increased splicing of exon 10. FEBS Lett., 443, 93–6. Hayashi, S., Toyoshima, Y., Hasegawa, M. et al. (2002). Late-onset frontotemporal dementia with a novel exon 1 (Arg5His) tau gene mutation. Ann. Neurol., 51, 525–30. Heutink, P., Stevens, M., Rizzu, P. et al. (1997). Hereditary frontotemporal dementia is linked to chromosome 17q21–q22: a genetic and clinicopathological study of three Dutch families. Ann. Neurol., 41, 150–9. Himmler, A., Drechsel, D., Kirschner, M. W. & Martin, D. W., Jr. (1989). Tau consists of a set of proteins with repeated


C-terminal microtubule-binding domains and variable N-terminal domains. Mol. Cell Biol., 9, 1381–8. Hirano, A., (1965). Pathology of amyotrophic lateral sclerosis. In Slow, Latent and Temperate Virus Infections, ed. D. C. Gajdusek, C. J. Gibbs, and M. Alpers, US Government Printing Office, Washington, DC, pp. 23–37. Hirano, A. & Llena, J. F. (1986). Neuropathological features of parkinsonism–dementia complex on Guam: Reappraisal and comparative study with Alzheimer’s disease and Parkinson’s disease. Prog. Neuropathol., 6, 17–31. Hirano, A., Kurland, L. T., Krooth, R. S. & Lessell, S. (1961a). Parkinsonism–dementia complex, an endemic disease on the island of Guam. I. Clinical features. Brain, 84, 642–61. Hirano, A., Malamud, N. & Kurland, L. T. (1961b). Parkinsonism– dementia complex, an endemic disease on the island of Guam. II. Pathological features. Brain, 84, 662–79. Hirano, A., Malamud, N., Elizan, T. S. & Kurland, L. T. (1966). Amyotrophic lateral sclerosis and Parkinsonism–dementia complex on Guam. Further pathologic studies. Arch. Neurol., 15, 35–51. Hirano, A., Dembitzer, H. M., Kurland, L. T. & Zimmerman, H. M. (1968). The fine structure of some intraganglionic alterations. Neurofibrillary tangles, granulovacuolar bodies and ‘rod-like’ structures as seen in Guam amyotrophic lateral sclerosis and parkinsonism–dementia complex. J. Neuropathol. Exp. Neurol., 27, 167–82. Ho, L., Xiang, Z., Mukherjee, P. et al. (2001). Gene expression profiling of the tau mutant (P301L) transgenic mouse brain. Neurosci. Lett., 310, 1–4. Hof, P. R., Perl, D. P., Loerzel, A. J. & Morrison, J. H. (1991). Neurofibrillary tangle distribution in the cerebral cortex of parkinsonism–dementia cases from Guam: differences with Alzheimer’s disease. Brain Res., 564, 306–13. Hof, P. R., Nimchinsky, E. A., Buée-Scherrer, V. et al. (1994a). Amyotrophic lateral sclerosis/parkinsonism–dementia complex of Guam: quantitative neuropathology, immunohistochemical analysis of neuronal vulnerability, and comparison with related neurodegenerative disorders. Acta Neuropathol. (Berl.), 88, 397–404. Hof, P. R., Perl, D. P., Loerzel, A. J., Steele, J. C. & Morrison, J. H. (1994b). Amyotrophic lateral sclerosis and parkinsonism– dementia from Guam: differences in neurofibrillary tangle distribution and density in the hippocampal formation and neocortex. Brain Res., 650, 107–16. Hoffmann, R., Lee, V. M.-Y., Leight, S., Varga, I. & Otvos, L., Jr. (1997). Unique Alzheimer’s disease paired helical filament specific epitopes involve double phosphorylation at specific sites. Biochemistry, 36, 8114–24. Hong, M., Zhukareva, V., Vogelsberg-Ragaglia, V. et al. (1998). Mutation-specific functional impairments in distinct tau isoforms of hereditary FTDP-17. Science, 282, 1914–17. Hong, M., Trojanowski, J. Q. & Lee, V. M.-Y., (2000). Tau-based neurofibrillary lesions. In Neurodegenerative Dementias, ed. C. M. Clark and J. Q. Trojanowski, McGrawHill, New York, pp. 161–75. Hosler, B. A., Siddique, T., Sapp, P. C. et al. (2000). Linkage of familial amyotrophic lateral sclerosis with frontotemporal

dementia to chromosome 9q21–q22. J Am Med Assoc, 284, 1664–9. Houlden, H., Baker, M., Adamson, J. et al. (1999). Frequency of tau mutations in three series of non-Alzheimer’s degenerative dementia. Ann. Neurol., 46, 243–8. Hudson, A. J. & Rice, G. P. (1990). Similarities of Guamanian ALS/PD to post-encephalitic parkinsonism/ALS: possible viral cause. Can. J. Neurol. Sci., 17, 427–33. Hulette, C. M., Pericak-Vance, M. A., Roses, A. D. et al. (1999). Neuropathological features of frontotemporal dementia and parkinsonism linked to chromosome 17q21–22 (FTDP-17): Duke Family 1684. J. Neuropathol. Exp. Neurol., 58, 859–66. Hutton, M., Lendon, C. L., Rizzu, P. et al. (1998). Association of missense and 5 -splice-site mutations in tau with the inherited dementia FTDP-17. Nature, 393, 702–5. Iijima, M., Tabira, T., Poorkaj, P. et al. (1999). A distinct familial presenile dementia with a novel missense mutation in the tau gene. Neuroreport, 10, 497–501. Ikemoto, A., Hirano, A., Akiguchi, I. & Kimura, J. (1997). Comparative study of ubiquitin immunoreactivity of hippocampal granular cells in amyotrophic lateral sclerosis with dementia, Guamanian amyotrophic lateral sclerosis and Guamanian parkinsonism–dementia complex. Acta Neuropathol. (Berl.), 93, 265–70. Iseki, E., Li, F., Odawara, T., Hino, H. et al. (1998). Ubiquitinimmunohistochemical investigation of atypical Pick’s disease without Pick bodies. J. Neurol. Sci., 159, 194–201. Iseki, E., Matsumura, T., Marui, W. et al. (2001). Familial frontotemporal dementia and parkinsonism with a novel N296H mutation in exon 10 of the tau gene and a widespread tau accumulation in glial cells. Acta Neuropathol., 102, 285–92. Ito, H., Hirano, H., Yen, S. H. & Kato, S. (1991). Demonstration of beta amyloid protein-containing neurofibrillary tangles in parkinsonism–dementia complex on Guam. Neuropathol. Appl. Neurobiol., 17, 365–73. Jackson, J. R., Wiedau-Pazos, M., Sang, T.-K. et al. (2002). Human wild-type tau interacts with wingless pathway components and produces neurofibrillary pathology in Drosophila. Neuron, 34, 509–19. Jackson, M., Lennox, G. & Lowe, J. (1996). Motor neurone diseaseinclusion dementia. Neurodegeneration., 5, 339–50. Janssen, J. C., Warrington, E. K., Morris, H. R. et al. (2002). Clinical features of frontotemporal dementia due to the intronic tau 10(+16) mutation. Neurology, 58, 1161–8. Jiang, Z., Cote, J., Kwon, J. M., Goate, A. M. & Wu, J. Y. (2000). Aberrant splicing of tau pre-mRNA caused by intronic mutations associated with the inherited dementia frontotemporal dementia with parkinsonism linked to chromosome 17. Mol. Cell Biol., 20, 4036–48. Jicha, G. A., Rockwood, J. M., Berenfeld, B., Hutton, M. & Davies, P. (1999). Altered conformation of recombinant frontotemporal dementia-17 mutant tau proteins. Neurosci. Lett., 260, 153–6. Kawai, J., Sasahara, M., Hazama, F., Kuno, S., Komure, O., Nomura, S. & Yamaguchi, M. (1993). Pallidonigroluysian degeneration with iron deposition: a study of three autopsy cases. Acta Neuropathol. (Berl.), 86, 609–16.

283

284


Kimura, K., Yase, Y. & Higashi, Y. (1961). Epidemiological and geomedical studies on amyotrophic lateral sclerosis and allied diseases in Kii Peninsula (Japan) preliminary report. Proc. Japan Acad., 37, 417–21. Kinoshita, A., Tomimoto, H., Suenaga, T., Akiguchi, I. & Kimura, J. (1997). Ubiquitin-related cytoskeletal abnormality in frontotemporal dementia: immunohistochemical and immunoelectron microscope studies. Acta Neuropathol. (Berl.), 94, 67–72. Klatzo, I., Wisniewski, H. M. & Streicher, E. (1965). Experimental production of neurofibrillary pathology: I. Light microscopic observations. J Neuropathol. Exp Neurol., 24, 187–99. Knopman, D. S. (1993). Overview of dementia lacking distinctive histology: pathological designation of a progressive dementia. Dementia, 4, 132–6. Knopman, D. S., Mastri, A. R., Frey, W. H., Sung, J. H. & Rustan, T. (1990a). Dementia lacking distinctive histologic features: a common non-Alzheimer degenerative dementia. Neurology, 40, 251–6. Knopman, D. S., Mastri, A. R., Frey, W. H., Sung, J. H. & Rustan, T. (1990b). Dementia lacking distinctive histologic features: a common non-Alzheimer degenerative dementia. Neurology, 40, 251–6. Kodama, K., Okada, S., Iseki, E. et al. (2000). Familial frontotemporal dementia with a P301L tau mutation in Japan. J. Neurol. Sci., 176, 57–64. Koerner, D. R. (1952). Amyotrophic lateral sclerosis on Guam: a clinical study and review of the literature. Ann. Int. Med., 37, 1204–20. Komori, T. (1999). Tau-positive glial inclusions in progressive supranuclear palsy, corticobasal degeneration and Pick’s disease. Brain Pathol., 9, 663–79. Kovach, M. J., Waggoner, B., Leal, S. M. et al. (2001). Clinical delineation and localization to chromosome 9p13.3–p12 of a unique dominant disorder in four families: hereditary inclusion body myopathy, Paget’s disease of bone, and frontotemporal dementia. Mol. Genet. Metab., 74, 458–75. Kurland, L. T. (1988). Amyotrophic lateral sclerosis and Parkinson’s disease complex on Guam linked to an environmental neurotoxin. Trends Neurosci., 11, 51–4. Kurland, L. T. & Mulder, D. W. (1954). Epidemiologic investigations of amyotrophic lateral sclerosis. 1. Preliminary report on geographic distribution, with special reference to the Mariana Islands, including clinical and pathologic observations. Neurology, 4, 355–78. (1955a). Epidemiologic investigations of amyotrophic lateral sclerosis, 2: familial aggregations indicative of dominant inheritance. Neurology, 5, 249–68. (1955b). Epidemiologic investigations of amyotrophic lateral sclerosis, 2: familial aggregations indicative of dominant inheritance. Neurology, 5, 182–96. Kurland, L. T., Mulder, D. W. & Sayre, G. P. (1956). Amyotrophic lateral sclerosis of the Marianas islands. Arch. Neurol. Psych., 75, 435–41. Kuzuhara, S., Kokubo, Y. & Narita, Y. (1996). Neuropathology of a Kii ALS case from a family of ALS and parkinsonism–dementia in

Hobara in Kii Penninsula of Japan. J. Neuropathol. Exp Neurol., 55, 633. Kuzuhara, S., Kokubo, Y., Sasaki, R., Narita, Y., Yabana, T., Hasegawa, M. & Iwatsubo, T. (2001). Familial amyotrophic lateral sclerosis and parkinsonism–dementia complex of the Kii Peninsula of Japan: clinical and neuropathological study and tau analysis. Ann. Neurol., 49, 501–11. Lanska, D. J., Currier, R. D., Cohen, M. et al. (1994). Familial progressive subcortical gliosis. Neurology, 44, 1633–43. Laqueur, G. L., Mickelsen, O., Shitney, M. G. & Kurland, L. T. (1963). Carciniogenic properties of nuts from Cycas circinalis L. indigenous to Guam. J. Natl. Cancer Inst., 124, 919–59. Lee, G., Neve, R. L. & Kosik, K. S. (1989). The microtubule binding domain of tau protein. Neuron, 2, 1615–24. Lee, V. M.-Y., Goedert, M. & Trojanowski, J. Q. (2001). Neurodegenerative tauopathies. Ann. Rev. Neurosci., 24, 1121–59. Leigh, P. N., Anderton, B. H., Dodson, A., Gallo, J. M., Swash, M. & Power, D. M. (1988). Ubiquitin deposits in anterior horn cells in motor neurone disease. Neurosci. Lett., 93, 197–203. Lendon, C. L., Lynch, T., Norton, J. et al. (1998). Hereditary dysphasic disinhibition dementia: a frontotemporal dementia linked to 17q21–22. Neurology, 50, 1546–55. Lessel, S., Hirano, A., Torres, J. M. & Kurland, L. T. (1962). Parkinsonism–dementia complex, epidemiologic considerations in the Chamorros of the Marianas islands, including clinical and pathological observations. Arch Neurol., 7, 377–85. Lewis, J., McGowan, E., Rockwood, J. et al. (2000). Neurofibrillary tangles, amyotrophy and progressive motor disturbance in mice expressing mutant (P301L) tau protein. Nat. Genet., 25, 402–5. Lewis, J., Dickson, D. W., Lin, W. L. et al. (2001). Enhanced neurofibrillary degeneration in transgenic mice expressing mutant tau and APP. Science, 293, 1487–91. Lippa, C. F., Zhukareva, V., Kawarai, T. et al. (2000). Frontotemporal dementia with novel tau pathology and a Glu342Val tau mutation. Ann. Neurol., 48, 850–8. LoPresti, P., Szuchet, S., Papasozomenos, S. C., Zinkowski, R. P. & Binder, L. I. (1995). Functional implications for the microtubule-associated protein tau: localization in oligodendrocytes. Proc. Natl Acad. Sci. USA, 92, 10369–73. Lowe, J., Lennox, G., Jefferson, D. et al. (1988). A filamentous inclusion body within anterior horn neurones in motor neurone disease defined by immunocytochemical localization of ubiquitin. Neurosci. Lett., 94, 203–10. Lund and Manchester Groups (1994). Clinical and neuropathological criteria for frontotemporal dementia. J. Neurol. Neurosurg. Psychiatry, 57, 416–18. McGeer, P. L., Schwab, C., McGeer, E. G., Haddock, R. L. & Steele, J. C. (1997). Familial nature and continuing morbidity of the amyotrophic lateral sclerosis–parkinsonism dementia complex of Guam. Neurology, 49, 400–9. McKhann, G. M., Albert, M. S., Grossman, M., Miller, B., Dickson, D. & Trojanowski, J. Q. (2001). Clinical and pathological diagnosis of frontotemporal dementia: report of the Work Group on Frontotemporal Dementia and Pick’s Disease. Arch. Neurol., 58, 1803–9.


Malamud, N., Hirano, A. & Kurland, L. T. (1961). Pathoanatomic changes in amyotrophic lateral sclerosis of Guam. Neurology, 5, 401–14. Mann, D. M. (1998). Dementia of frontal type and dementias with subcortical gliosis. Brain Pathol., 8, 325–38. Mann, D. M., South, P. W., Snowden, J. S. & Neary, D. (1993). Dementia of frontal lobe type: neuropathology and immunohistochemistry. J. Neurol. Neurosurg. Psychiatry, 56, 605–14. Matsumoto, H. & Strong, F. M. (1963). The occurrence of methylazoxymethanol in Cycas circinalis L. Arch. Biochem. Biophys., 101, 299–310. Matsumoto, S., Hirano, A. & Goto, S. (1990a). Spinal cord neurofibrillary tangles of Guamanian amyotrophic lateral sclerosis and parkinsonism–dementia complex: an immunohistochemical study. Neurology, 40, 975–9. (1990b). Ubiquitin-immunoreactive filamentous inclusions in anterior horn cells of Guamanian and non-Guamanian amyotrophic lateral sclerosis. Acta Neuropathol. (Berl.), 80, 233–8. Matsumura, N., Yamazaki, T. & Ihara, Y. (1999). Stable expression in Chinese hamster ovary cells of mutated tau genes causing frontotemporal dementia and parkinsonism linked to chromosome 17 (FTDP-17). Am. J Pathol., 154, 1649–56. Mawal-Dewan, M., Schmidt, M. L., Balin, B., Perl, D. P., Lee, V. M.-Y. & Trojanowski, J. Q. (1996). Identification of phosphorylation sites in PHF-tau from patients with Guam amyotrophic lateral sclerosis/parkinsonism–dementia complex. J. Neuropathol. Exp. Neurol., 55, 1051–9. ¨ Meyer, A. (1929). Uber eine der amyotrophischen Lateralsklerose nahestende Erkrankung mit psychischen Storungen. Z. Ges. Neurol. Psychiat., 121, 107–28. Mirra, S. S., Murrell, J. R., Gearing, M. et al. (1999). Tau pathology in a family with dementia and a P301L mutation in tau. J. Neuropathol. Exp. Neurol., 58, 335–45. Mitsuyama, Y. (2000). Dementia with motor neuron disease. Neuropathology, 20 Suppl, S79–S81. Mitsuyama, Y. & Takamiya, S. (1979). Presenile dementia with motor neuron disease in Japan. A new entity? Arch. Neurol., 36, 592–3. Miyamoto, K., Kowalska, A., Hasegawa, M. et al. (2001). Familial frontotemporal dementia and parkinsonism with a novel mutation at an intron 10+11-splice site in the tau gene. Ann. Neurol., 50, 117–20. Miyasaka, T., Morishima-Kawashima, M., Ravid, R., Kamphorst, W., Nagashima, K. & Ihara, Y. (2001). Selective deposition of mutant tau in the FTDP-17 brain affected by the P301L mutation. J. Neuropathol. Exp. Neurol., 60, 872–84. Montgomery, G. K. & Erickson, L. M. (1987). Neuropsychological perspectives in amyotrophic lateral sclerosis. Neurol. Clin., 5, 61–81. Morita, K., Kaiya, H., Ikeda, T. & Namba, M. (1987). Presenile dementia combined with amyotrophy: a review of 34 Japanese cases. Arch. Gerontol. Geriatr., 6, 263–77. Morris, H. R., Perez-Tur, J., Janssen, J. C. et al. (1999). Mutation in the tau exon 10 splice site region in familial frontotemporal dementia. Ann. Neurol., 45, 270–71.

Morris, H. R., Khan, M. N., Janssen, J. C. et al. (2001). The genetic and pathological classification of familial frontotemporal dementia. Arch. Neurol., 58, 1813–16. Murakami, N. (1999). Parkinsonism–dementia complex on Guam. J. Neurol., 246 (Suppl 2), II16–II18. Murrell, J. R., Koller, D., Foroud, T. et al. (1997). Familial multiplesystem tauopathy with presenile dementia is localized to chromosome 17. Am. J. Hum. Genet., 61, 1131–8. Murrell, J. R., Spillantini, M. G., Zolo, P. et al. (1999). Tau gene mutation G389R causes a tauopathy with abundant pick body-like inclusions and axonal deposits. J. Neuropathol. Exp. Neurol., 58, 1207–26. Nacharaju, P., Lewis, J., Easson, C. et al. (1999). Accelerated filament formation from tau protein with specific FTDP-17 missense mutations. FEBS Lett., 447, 195–9. Nagiec, E. W., Sampson, K. E. & Abraham, I. (2001). Mutated tau binds less avidly to microtubules than wild type tau in living cells. J. Neurosci. Res., 63, 268–75. Nakano, I. (2000). Frontotemporal dementia with motor neuron disease (amyotrophic lateral sclerosis with dementia). Neuropathology, 20, 68–75. Nakano, I. & Hirano, A. (1983). Neuron loss in the nucleus basalis of Meynert in parkinsonism–dementia complex of Guam. Ann. Neurol., 13, 87–91. Nasreddine, Z. S., Loginov, M., Clark, L. N. et al. (1999). From genotype to phenotype: a clinical pathological, and biochemical investigation of frontotemporal dementia and parkinsonism (FTDP-17) caused by the P301L tau mutation. Ann. Neurol., 45, 704–15. Neary, D., Snowden, J. S., Northen, B. & Goulding, P. (1988). Dementia of frontal lobe type. J. Neurol. Neurosurg. Psychiatry, 51, 353–61. Neary, D., Snowden, J. S., Mann, D. M., Northen, B., Goulding, P. J. & Macdermott, N. (1990). Frontal lobe dementia and motor neuron disease. J. Neurol. Neurosurg. Psychiatry, 53, 23–32. Neary, D., Snowden, J. S., Gustafson, L. et al. (1998). Frontotemporal lobar degeneration: a consensus on clinical diagnostic criteria. Neurology, 51, 1546–54. Neumann, M., Schulz-Schaeffer, W., Crowther, R. A. et al. (2001). Pick’s disease associated with the novel Tau gene mutation K369I. Ann. Neurol., 50, 503–13. Neumann, M. A. & Cohn, R. (1967). Progressive subcortical gliosis, a rare form of presenile dementia. Brain, 90, 405– 18. Neve, R. L., Harris, P., Kosik, K. S., Kurnit, D. M. & Donlon, T. A. (1986). Identification of cDNA clones for the human microtubuleassociated protein tau and chromosomal localization of the genes for tau and microtubule-associated protein 2. Brain Res., 387, 271–80. Okamoto, K., Hirai, S., Yamazaki, T., Sun, X. Y. & Nakazato, Y. (1991). New ubiquitin-positive intraneuronal inclusions in the extramotor cortices in patients with amyotrophic lateral sclerosis. Neurosci. Lett., 129, 233–6. Olanow, C. W. & Perl, D. P. (1994). Free radicals and neurodegeneration. Trends Neurosci., 17, 193–4.

285

286


Oyanagi, K. & Wada, M. (1999). Neuropathology of parkinsonism– dementia complex and amyotrophic lateral sclerosis of Guam: an update. J. Neurol., 246 (Suppl 2), II19–II27. Oyanagi, K., Makifuchi, T. Othoh, T. et al. (1994). Amyotrophic lateral sclerosis of Guam: the nature of the neuropathological findings. Acta Neuropathol. (Berl.), 88, 405–12. Oyanagi, K., Makifuchi, T., Ohtoh, T., Chen, K. M., Gajdusek, D. C. & Chase, T. N. (1997). Distinct pathological features of the gallyasand tau-positive glia in the parkinsonism–dementia complex and amyotrophic lateral sclerosis. J. Neuropathol. Exp. Neurol., 56, 308–16. Pastor, P., Pastor, E., Carnero, C. et al. (2001). Familial atypical progressive supranuclear palsy associated with homozigosity for the delN296 mutation in the tau gene. Ann. Neurol., 49, 263–7. Perez-Tur, J., Buée, L., Morris, H. R. et al. (1999). Neurodegenerative diseases of Guam: analysis of tau. Neurology, 53, 411–13. Perez, M., Lim, F., Arrasate, M. & Avila, J. (2000). The FTDP-17-linked mutation R406W abolishes the interaction of phosphorylated tau with microtubules. J. Neurochem., 74, 2583–9. Perl, D. P. & Brody, A. R. (1980). Alzheimer’s disease: X-ray spectrometric evidence of aluminum accumulation in neurofibrillary tangle-bearing neurons. Science, 208, 297–9. Perl, D. P., Gajdusek, D. C., Garruto, R. M., Yanagihara, R. T. & Gibbs, C. J. (1982). Intraneuronal aluminum accumulation in amyotrophic lateral sclerosis and Parkinsonism–dementia of Guam. Science, 217, 1053–5. Perl, D. P., Hof, P. R., Steele, J. C., Purohit, D. P., Peterson, R. & Kurland, L. T. (1994). Neuropathologic studies of a pure dementing syndrome (Marianas dementia) among the inhabitants of Guam, a form of ALS/parkinsonism–dementia complex. Brain Pathol., 4, 529. Perl, D. P., Hof, P. R., Purohit, D. P., Loerzel, A. F. & Belli, D. (1995). Changes in the outbreak of ALS/parkinsonism–dementia complex: neuropathologic studies of asymptomatic Chamorros. J. Neuropathol. Exp. Neurol., 54, 416. Petersen, R. B., Tabaton, M., Chen, S. G. et al. (1995). Familial progressive subcortical gliosis: presence of prions and linkage to chromosome 17. Neurology, 45, 1062–7. ¨ Pick, A. (1892). Uber die Bejiehungen der senilen Hirnatrophie zur Aphasie. Prager Med. Wochenschr., 17, 165–7. Pickering-Brown, S., Baker, M., Yen, S. H. et al. (2000). Pick’s disease is associated with mutations in the tau gene. Ann. Neurol., 48, 859–67. Pickering-Brown, S. M., Richardson, A. M., Snowden, J. S. et al. (2002). Inherited frontotemporal dementia in nine British families associated with intronic mutations in the tau gene. Brain, 125, 732–51. Plato, C. C., Reed, D. M., Elizan, T. S. & Kurland, L. T. (1967). Amyotrophic lateral sclerosis-Parkinsonism–dementia complex of Guam. IV. Familial and genetic investigations. Am. J Hum. Genet., 19, 617–32. Plato, C. C., Cruz, M. T. & Kurland, L. T. (1969). Amyotrophic lateral sclerosis–Parkinsonism dementia complex of Guam: further genetic investigations. Am. J Hum. Genet., 21, 133–41. Plato, C. C., Garruto, R. M., Fox, K. M. & Gajdusek, D. C. (1986). Amyotrophic lateral sclerosis and parkinsonism–dementia on

Guam: a 25-year prospective case-control study. Am. J. Epidemiol., 124, 643–56. Plato, C. C., Galasko, D., Garruto, R. M. et al. (2002). ALS and PDC of Guam: forty-year follow-up. Neurology, 58, 765–73. Poorkaj, P., Bird, T. D., Wijsman, E. et al. (1998). Tau is a candidate gene for chromosome 17 frontotemporal dementia. Ann. Neurol., 43, 815–25. Poorkaj, P., Grossman, M., Steinbart, E. et al. (2001a). Frequency of tau gene mutations in familial and sporadic cases of nonAlzheimer dementia. Arch Neurol., 58, 383–7. Poorkaj, P., Tsuang, D., Wijsman, E. et al. (2001b). Tau as a susceptibility gene for amyotrophic lateral sclerosis–parkinsonism dementia complex of Guam. Arch. Neurol., 58, 1871–8. Reed, D. M. & Brody, J. A. (1975). Amyotrophic lateral sclerosis and parkinsonism–dementia on Guam, 1945–1972. I. Descriptive epidemiology. Am. J. Epidemiol., 101, 287–301. Reed, D. M., Torres, J. M. & Brody, J. A. (1975). Amyotrophic lateral sclerosis and parkinsonism–dementia on Guam, 1945–1972. II. Familial and genetic studies. Am. J. Epidemiol., 101, 302–10. Reed, L. A., Grabowski, T. J., Schmidt, M. L. et al. (1997). Autosomal dominant dementia with widespread neurofibrillary tangles. Ann. Neurol., 42, 564–72. Reed, L. A., Schmidt, M. L., Wszolek, Z. K. et al. (1998). The neuropathology of a chromosome 17-linked autosomal dominant parkinsonism and dementia (‘pallido-ponto-nigral degeneration’). J. Neuropathol. Exp. Neurol., 57, 588–601. Reed, L. A., Wszolek, Z. K. & Hutton, M. (2001). Phenotypic correlations in FTDP-17. Neurobiol. Aging, 22, 89–107. Rizzini, C., Goedert, M., Hodges, J. R. et al. (2000). Tau gene mutation K257T causes a tauopathy similar to Pick’s disease. J Neuropathol. Exp. Neurol., 59, 990–1001. Rizzu, P., Van Swieten, J. C., Joosse, M. et al. (1999). High prevalence of mutations in the microtubule-associated protein tau in a population study of frontotemporal dementia in the Netherlands. Am. J Hum. Genet., 64, 414–21. Rizzu, P., Joosse, M., Ravid, R. et al. (2000). Mutation-dependent aggregation of tau protein and its selective depletion from the soluble fraction in brain of P301L FTDP-17 patients. Hum. Mol. Genet., 9, 3075–82. Rodgers-Johnson, P., Garruto, R. M., Yanagihara, R., Chen, K. M., Gajdusek, D. C. & Gibbs, C. J., Jr. (1986). Amyotrophic lateral sclerosis and parkinsonism–dementia on Guam: a 30-year evaluation of clinical and neuropathologic trends. Neurology, 36, 7–13. Rosso, S. M., Kamphorst, W., Ravid, R. & Van Swieten, J. C. (2000). Coexistent tau and amyloid pathology in hereditary frontotemporal dementia with tau mutations. Ann. NY Acad. Sci., 920, 115–19. Rosso, S. M., Kamphorst, W., de Graaf, B. et al. (2001). Familial frontotemporal dementia with ubiquitin-positive inclusions is linked to chromosome 17q21–22. Brain, 124, 1948–57. Rosso, S. M., van Herpen, E., Deelen, W. et al. (2002). A novel tau mutation, S320F, causes a tauopathy with inclusions similar to those in Pick’s disease. Ann. Neurol., 51, 373–6. Sahara, N., Tomiyama, T. & Mori, H. (2000). Missense point mutations of tau to segregate with FTDP-17 exhibit site-specific


effects on microtubule structure in COS cells: A novel action of R406W mutation. J. Neurosci. Res., 60, 380–87. Saito, Y., Geyer, A., Sasaki, R. et al. (2002). Early-onset, rapidly progressive familial tauopathy with R406W mutation. Neurology, 58, 811–13. Sanders, J., Schenk, V. & van Veen, P. (1939). A family with Pick’s disease. Veerhandelingen de Koninklijke Nederlandse Akadamie van Wetenschappen. Schenk, V. (1959). Re-examination of a family with Pick’s disease. Ann. Hum. Genet., 23, 325–33. Schmidt, M. L., Carden, M. J., Lee, V. M.-Y. & Trojanowski, J. Q. (1987). Phosphate dependent and independent neurofilament epitopes in the axonal swellings of patients with motor neuron disease and controls. Lab. Invest., 56, 282–94. Schmidt, M. L., Lee, V. M.-Y. & Trojanowski, J. Q. (1989). Analysis of epitopes shared by Hirano bodies and neurofilament proteins in normal and Alzheimer’s disease hippocampus. Lab. Invest., 60, 513–22. Schmidt, M. L., Lee, V. M.-Y., Saido, T. et al. (1998). Amyloid plaques in Guam amyotrophic lateral sclerosis/parkinsonism– dementia complex contain species of A beta similar to those found in the amyloid plaques of Alzheimer’s disease and pathological aging. Acta Neuropathol. (Berl.), 95, 117–22. Schmidt, M. L., Zhukareva, V., Perl, D. P. et al. (2001). Spinal cord neurofibrillary pathology in Alzheimer disease and Guam Parkinsonism–dementia complex. J. Neuropathol. Exp. Neurol., 60, 1075–86. Schwab, C., Steele, J. C. & McGeer, P. L. (1998). Pyramidal neuron loss is matched by ghost tangle increase in Guam parkinsonism– dementia hippocampus. Acta Neuropathol. (Berl.), 96, 409–16. Senapathy, P., Shapiro, M. B. & Harris, N. L. (1990). Splice junctions, branch point sites, and exons: sequence statistics, identification, and applications to genome project. Methods Enzymol., 183, 252–78. Shankar, S. K., Yanagihara, R., Garruto, R. M., Grundke-Iqbal, I., Kosik, K. S. & Gajdusek, D. C. (1989). Immunocytochemical characterization of neurofibrillary tangles in amyotrophic lateral sclerosis and parkinsonism–dementia of Guam. Ann. Neurol., 25, 146–51. Shin, R. W., Iwaki, T., Kitamoto, T. & Tateishi, J. (1991). Hydrated autoclave pretreatment enhances tau immunoreactivity in formalin-fixed normal and Alzheimer’s disease brain tissues. Lab. Invest., 64, 693–702. Shiraki, H. & Yase, Y., (1975). Amyotrophic lateral sclerosis in Japan. In Handbook of Clinical Neurology, Vol. 22, System Disorders and Atrophies, ed. P. J. Vinken, G. W. Bruyn, H. L. Klawans & J. M. B. V. de Jong, Elsevier Scientific Publishing, New York, pp. 353–419. Si, Z. H., Rauch, D. & Stoltzfus, C. M. (1998). The exon splicing silencer in human immunodeficiency virus type 1 Tat exon 3 is bipartite and acts early in spliceosome assembly. Mol. Cell Biol., 18, 5404–13. Snow, B. J., Peppard, R. F., Guttman, M. et al. (1990). Positron emission tomographic scanning demonstrates a presynaptic dopaminergic lesion in Lytico-Bodig. The amyotrophic

lateral sclerosis/parkinsonism–dementia complex of Guam. Arch. Neurol., 47, 870–4. Spencer, P. S., Nunn, P. B., Hugon, J. et al. (1987). Guam amyotrophic lateral sclerosis–parkinsonism–dementia linked to a plant excitant neurotoxin. Science, 237, 517–22. Sperfeld, A. D., Collatz, M. B., Baier, H. et al. (1999). FTDP-17: an early-onset phenotype with parkinsonism and epileptic seizures caused by a novel mutation. Ann. Neurol., 46, 708– 15. Spillantini, M. G., Bird, T. D. & Ghetti, B. (1998a). Frontotemporal dementia and Parkinsonism linked to chromosome 17: a new group of tauopathies. Brain Pathol., 8, 387–402. Spillantini, M. G., Crowther, R. A., Kamphorst, W., Heutink, P. & Van Swieten, J. C. (1998b). Tau pathology in two Dutch families with mutations in the microtubule-binding region of tau. Am. J Pathol., 153, 1359–63. Spillantini, M. G., Crowther, R. A. & Goedert, M. (1996). Comparison of the neurofibrillary pathology in Alzheimer’s disease and familial presenile dementia with tangles. Acta Neuropathol. (Berl.), 92, 42–8. Spillantini, M. G., Murrell, J. R., Goedert, M., Farlow, M. R., Klug, A. & Ghetti, B. (1998c). Mutation in the tau gene in familial multiple system tauopathy with presenile dementia. Proc. Natl Acad. Sci. USA, 95, 7737–41. Spillantini, M. G., Yoshida, H., Rizzini, C. et al. (2000). A novel tau mutation (N296N) in familial dementia with swollen achromatic neurons and corticobasal inclusion bodies. Ann. Neurol., 48, 939–43. Stanford, P. M., Halliday, G. M., Brooks, W. S. et al. (2000). Progressive supranuclear palsy pathology caused by a novel silent mutation in exon 10 of the tau gene: expansion of the disease phenotype caused by tau gene mutations. Brain, 123, 880–93. Stevens, M., van Duijn, C. M., Kamphorst, W. et al. (1998). Familial aggregation in frontotemporal dementia. Neurology, 50, 1541–5. Stone, R. (1993). Guam: deadly disease dying out. Science, 261, 424–6. Sumi, S. M., Bird, T. D., Nochlin, D. & Raskind, M. A. (1992). Familial presenile dementia with psychosis associated with cortical neurofibrillary tangles and degeneration of the amygdala. Neurology, 42, 120–7. Talbot, P. R. (1996). Frontal lobe dementia and motor neuron disease. J. Neural Transm. Suppl, 47, 125–132. Tanaka, K., Watakabe, A. & Shimura, Y. (1994). Polypurine sequences within a downstream exon function as a splicing enhancer. Mol. Cell Biol., 14, 1347–54. Tanaka, R., Kobayashi, T., Motoi, Y., Anno, M., Mizuno, Y. & Mori, H. (2000). A case of frontotemporal dementia with tau P301L mutation in the Far East. J. Neurol., 247, 705–7. Tanemura, K., Akagi, T., Murayama, M. et al. (2001). Formation of filamentous tau aggregations in transgenic mice expressing V337M human tau. Neurobiol. Dis., 8, 1036–45. Tanemura, K., Murayama, M., Akagi, T. et al. (2002). Neurodegeneration with tau accumulation in a transgenic mouse expressing V337M human tau. J. Neurosci., 22, 133–41.

287

288


Terry, R. D. & Pena, C. (1965). Experimental production of neurofibrillary pathology: electron microscopy, phosphate histochemistry and electron probe analysis. J. Neuropathol. Exp Neurol., 24, 200–10. Tolnay, M., Grazia, S. M., Rizzini, C., Eccles, D., Lowe, J. & Ellison, D. (2000). A new case of frontotemporal dementia and parkinsonism resulting from an intron 10 +3-splice site mutation in the tau gene: clinical and pathological features. Neuropathol. Appl. Neurobiol., 26, 368–78. Van Swieten, J. C., Stevens, M., Rosso, S. M. et al. (1999). Phenotypic variation in hereditary frontotemporal dementia with tau mutations. Ann. Neurol., 46, 617–26. Varani, L., Hasegawa, M., Spillantini, M. G. et al. (1999). Structure of tau exon 10 splicing regulatory element RNA and destabilization by mutations of frontotemporal dementia and parkinsonism linked to chromosome 17. Proc. Natl Acad. Sci. USA, 96, 8229–34. Vogelsberg-Ragaglia, V., Bruce, J., Richter-Landsberg, C. et al. (2000). Distinct FTDP-17 missense mutations in tau produce tau aggregates and other pathological phenotypes in transfected CHO cells. Mol. Biol. Cell, 11, 4093–104. von Bergen, M., Friedhoff, P., Biernat, J., Heberle, J., Mandelkow, E. M. & Mandelkow, E. (2000). Assembly of tau protein into Alzheimer paired helical filaments depends on a local sequence motif ((306)VQIVYK(311)) forming beta structure. Proc. Natl Acad. Sci. USA, 97, 5129–34. von Bergen, M., Barghorn, S., Li, L. et al. (2001). Mutations of tau protein in frontotemporal dementia promote aggregation of paired helical filaments by enhancing local beta-structure. J. Biol. Chem., 276, 48165–74. Waring, S. C., O’Brien, P. C., Kurland, L. T. et al. (1994). Apolipoprotein E allele in Chamorros with amyotrophic lateral sclerosis/parkinsonism–dementia complex. Lancet, 343, 611. Weingarten, M. D., Lockwood, A. H., Hwo, S. Y. & Kirschner, M. W. (1975). A protein factor essential for microtubule assembly. Proc. Natl Acad. Sci. USA, 72, 1858–62. Whiting, M. (1964). Toxicity of cycads. Econ. Botany, 17, 64–71. Wiederholt, W. C. (1999). Neuroepidemiologic research initiatives on Guam: past and present. Neuroepidemiology, 18, 279–91. Wightman, G., Anderson, V. E., Martin, J. et al. (1992). Hippocampal and neocortical ubiquitin-immunoreactive inclusions in amyotrophic lateral sclerosis with dementia. Neurosci. Lett., 139, 269–74. Wijker, M., Wszolek, Z. K., Wolters, E. C. et al. (1996). Localization of the gene for rapidly progressive autosomal dominant parkinsonism and dementia with pallido-ponto-nigral degeneration to chromosome 17q21. Hum. Mol. Genet., 5, 151–4. Wilhelmsen, K. C., Lynch, T., Pavlou, E., Higgins, M. & Nygaard, T. G. (1994). Localization of disinhibition–dementia–

parkinsonism–amyotrophy complex to 17q21–22. Am. J Hum. Genet., 55, 1159–65. Wittmann, C. W., Wszolek, M. F., Shulman, J. M. et al. (2001). Tauopathy in Drosophila: neurodegeneration without neurofibrillary tangles. Science, 293, 711–14. Yamazaki, M., Arai, Y., Baba, M. et al. (2000). Alpha-synuclein inclusions in amygdala in the brains of patients with the parkinsonism–dementia complex of Guam. J. Neuropathol. Exp. Neurol., 59, 585–91. Yang, M. G., Mickelsen, O., Campbell, M. E., Laqueur, G. L. & Keresztesy, J. C. (1966). Cycad flour used by Guamanians: effects produced in rats by long-term feeding. J. Nutr., 90, 153–6. Yase, Y. (1972). The pathogenesis of amyotrophic lateral sclerosis. Lancet, 2, 292–6. Yasuda, M., Kawamata, T., Komure, O. et al. (1999). A mutation in the microtubule-associated protein tau in pallido-nigro-luysian degeneration. Neurology, 53, 864–8. Yasuda, M., Takamatsu, J., D’Souza, I. et al. (2000a). A novel mutation at position +12 in the intron following exon 10 of the tau gene in familial frontotemporal dementia (FTD-Kumamoto). Ann. Neurol., 47, 422–9. Yasuda, M., Yokoyama, K., Nakayasu, T. et al. (2000b). A Japanese patient with frontotemporal dementia and parkinsonism by a tau P301S mutation. Neurology, 55, 1224–7. Yen, S., Easson, C., Nacharaju, P., Hutton, M. & Yen, S. H. (1999a). FTDP-17 tau mutations decrease the susceptibility of tau to calpain I digestion. FEBS Lett., 461, 91–5. Yen, S. H., Hutton, M., DeTure, M., Ko, L. W. & Nacharaju, P. (1999b). Fibrillogenesis of tau: insights from tau missense mutations in FTDP-17. Brain Pathol., 9, 695–705. Yoshida, H. & Ihara, Y. (1993). Tau in paired helical filaments is functionally distinct from fetal tau: assembly incompetence of paired helical filament-tau. J. Neurochem., 61, 1183–6. Zhang, Z. X., Anderson, D. W., Lavine, L. & Mantel, N. (1990). Patterns of acquiring parkinsonism–dementia complex on Guam. 1944 through 1985. Arch. Neurol., 47, 1019–24. Zheng-Fischhofer, Q., Biernat, J., Mandelkow, E. M., Illenberger, S., Godemann, R. & Mandelkow, E. (1998). Sequential phosphorylation of Tau by glycogen synthase kinase-3beta and protein kinase A at Thr212 and Ser214 generates the Alzheimerspecific epitope of antibody AT100 and requires a pairedhelical-filament-like conformation. Eur. J. Biochem., 252, 542– 52. Zhukareva, V., Vogelsberg-Ragaglia, V., Van Deerlin, V. M. et al. (2001). Loss of brain tau defines novel sporadic and familial tauopathies with frontotemporal dementia. Ann. Neurol., 49, 165–75. Zimmerman, H. M. (1945). Monthly report to medical officer in command. US Naval Medical Research Unit, 2.

13 Vascular dementias James H. Morris1 , Hannu Kalimo2 and Matti Viitanen3 2

Introduction One of the most controversial and difficult areas in the pathology of dementing disorders is the role of cerebrovascular disease in dementia. Most of the main problems in assessing the significance and pathophysiology of cerebrovascular disease in the aetiology of dementia can be discussed under one of the following four questions: (i) How do we define cognitive impairment and dementia in the context of vascular disease? (ii) What is the frequency of dementia in association with cerebrovascular disease? (iii) What effects does cerebrovascular disease have on the nervous system and how does it produce cognitive impairment and dementia? (iv) What are the neuropathological criteria for vascular cognitive impairment and dementia?

Historical overview The history of dementing disorders, and specifically that of vascular dementia (VaD), has been profoundly and vividly covered by Roman (1999, 2002) and Mast et al. (1995). The most likely first description of post-apoplectic VaD was presented by Thomas Willis in the seventeenth century. On the basis of their clinicopathological studies at the turn of the nineteenth and twentieth century the two famous contemporary German neuropsychiatrists and pathologists, Alois Alzheimer and Otto Binswanger, distinguished VaD from what was, at the time very common, syphilitic dementia. The long-held fallacious view that the

1 Oxford Radcliffe Hospitals Oxford, UK Uppsala University Hospital, Sweden and Helsinki University Hospital, Finland 3 Karolinska Institutet Stockholm, Sweden

most frequent cause of senile dementia was arteriosclerosis (arteriosclerotic insanity) was due to Kraepelin’s summary of the above-mentioned studies (1910), in which the progressive narrowing and obliteration of large cerebral arteries was thought to lead to diminished cerebral blood flow (CBF) and consequently neuronal death and dementia. The problem with many of the studies until the 1970s was that in those days in most cases the nature of the cognitive decline was not extensively investigated. Thus, for example, in Corsellis’ (1962) paper the patients were classified only into ‘organic psychosis’. These studies, which seem to indicate a substantial element of vascular disease in the causation of dementia were performed on a largely uncharacterized population. The real breakthrough occurred in 1970 when Tomlinson and colleagues (1970) published their most influential article on the subject of VaD. In brief, they compared the brains of 50 demented old people with those of undemented controls and found that 18% had definite or probable ‘arteriosclerotic dementia’ and a similar fraction a combination of ‘arteriosclerotic dementia’ and Alzheimer’s disease (AD). Thus in approximately one-third of the patients studied, cerebrovascular disease was adjudged to play a major part in the causation of the dementia and in one-sixth it was the sole cause. Their observations of vascular disease were predominantly focused on the presence of infarctions, and they concluded that only when there was a large volume of infarction (usually more than 100 ml) was there a significant association with the presence of dementia. One of the interesting aspects of this study, which illustrates the continuing difficulties in this field, is that in their analysis of the infarcts


289

290

J. H. Morris, H. Kalimo and M. Viitanen

they did not consider ‘the possibility of ischaemic lesions affecting areas of brain likely to be particularly important in relation to producing features of dementia’. Completely to ignore the significance of location seems to conflict with the generally held view of the localization of function within the nervous system and the frequent observation that patients with damage to specific regions of the brain can have significant problems with cognitive function. The converse is also true; large infarcts with prominent focal signs can occur without producing a dementia. The findings of Tomlinson et al. (1970) underlined the fact that it was the focal destruction(s) of brain parenchyma, not gradual diffuse loss of neurons due to diminished circulation, that caused dementia. This view was further emphasized by the introduction of the term multiinfarct dementia (MID; Hachinski et al., 1974). However, it soon became clear that other cerebrovascular disturbances in addition to multiple infarcts can cause dementia, hence the term MID was replaced by the more comprehensive term VaD to cover a broad spectrum of vascular aetiologies that can lead to cognitive decline and later dementia (Roman et al., 1993). (A more correct term would actually be circulatory dementia, since it is the blood flow and exchange of substances between the blood and parenchyma and not only the vascular component that is decisive). More recently, it has been emphasized that VaD may be preventable and the cognitive decline due to cerebrovascular disease should be recognized as soon as possible to avoid unnecessary deterioration, i.e. already at the stage of mild vascular cognitive impairment (VCI, Bowler & Hachinski, 1995, Bowler, 2002). The exploitation of molecular genetics in research on familial dementias beginning in the 1980s disclosed several genetic defects causing familial VaDs (e.g. cerebral amyloid angiopathies and CADASIL, see below), clinical descriptions of which had been published long before the molecular genetic era (e.g. Worster-Draught et al. 1933; van Bogaert, 1955). Furthermore, analyses on these as such rare diseases has enormously contributed to understanding the pathogenesis of degenerative and vascular dementias in general.

Clinical criteria for vascular dementia Although this is not primarily a clinical treatise, some indication of the evolution of clinical criteria for VaD is appropriate. In spite of several attempts definite and reliable clinical (and pathological) criteria for VaD are still lacking as was again pointed out in two recent articles (Bowler, 2002; Korczyn, 2002).

Definition of dementia A significant part of the problem of determining the significance of vascular causes for the development of dementia lies in the definition that is adopted for the clinical syndrome of dementia. The presently used criteria tend to define dementia in terms of the archetypal patient with advanced AD and the genuinely global decline in higher mental function suffered by these patients as the quintessential ‘real dementia’; the gold standard against which other types of cognitive decline are tested. Applied rigorously, a requirement for truly generalized intellectual deterioration would exclude many cases of cognitive decline secondary to cerebrovascular disease. Whether a single stroke or focal cerebrovascular disease, can be said to cause a dementia largely depends on how strict the requirement is for a generalized deterioration in cognitive function. Such a strict requirement along the lines of that seen in AD would certainly exclude amnestic states such as those resulting from bilateral posterior cerebral infarctions that produce bilateral hippocampal damage, or anterior thalamic infarctions, which are the usual cause of the so-called ‘thalamic dementias’. Parenthetically, from a practical diagnostic standpoint it is the early stages where firm diagnostic criteria would be most valuable. By the time the patient is globally demented the diagnostic position is usually much clearer but dementia represents a late and usually irreversible stage of the cognitive deterioration (Bowler, 2002, see below). One factor in the debate over definition is the method used to assess cognitive function. There has been considerable difficulty in defining the most appropriate criteria for vascular dementia (Tatemichi et al., 1992; Chui et al., 1992; Roman et al., 1993). It may be pointed out that the degree to which the questions used in different dementia rating scales reflect different brain functions (and therefore different brain regions) will be biased to disease affecting those regions. As a straightforward example of the effect, as noted by Liu et al. (1992), many questions in dementia rating scales are concerned with language, and patients with language disorders will score particularly poorly in such scales. Other functions, such as, for example, right parietal may be relatively less emphasized and using such a rating scale, patients may score well even with severe right parietal functional deficits. When groups of patients are being surveyed for the presence of defects in higher mental function, the limitations of the way that the cognitive decline was defined and measured must be borne in mind. This caveat applies a fortiori in cases where cognitive decline is assessed without specific neuropsychologic assessment of cognitive function. It has been repeatedly pointed out that in AD

Vascular dementias

definition of dementia is heavily based on memory deficit, whereas in VCI/VaD memory loss is usually a relatively late symptom. Instead, subcortical vascular lesions appear to frequently cause executive dysfunction, whereas memory function is relatively well preserved (Kramer et al., 2002; Amberla et al., 2003). VaD patients with involvement of prefrontal cortical areas have been described, for example, as apathetic, irritable, impulsive, obsessive, disinhibited and repetitive (Korczyn, 2002). The situation is further complicated by the fact that the definitions of dementia markedly vary between the different sets of clinical criteria. This was clearly demonstrated by Erkinjuntti et al. (1997). In their study on the effects of different diagnostic criteria on the prevalence of dementia in a population of 1879 subjects aged > 65 years they showed that the proportion of subjects with dementia was 3.1% with ICD-10, 13.7% with DSM-IV, 17.3% with DSMIIIR and 29.1% with DSM-III. The difference is surprisingly great, since in all criteria diagnosis of dementia requires a cognitive decline of such an extent that it interferes with the social or occupational life, in other words a stage of severe cognitive impairment.

Presently used criteria for vascular dementia The oldest set of criteria, the Hachinski Ischaemic Score (HIS), was developed by Hachinski and his colleagues in 1974–1975 (later modified by Rosén et al., 1980). It is based on 13 clinical features, such as an abrupt onset, stepwise deterioration and the presence of focal symptoms and signs as indicators of a vascular etiology in patients with cognitive decline. Subsequent evaluation of the use of the ischaemic score has indicated that it attaches undue weight to the presence of focal signs. This had the effect of widening the clinical definition of VaD to the point where significant evidence of cerebrovascular disease in a patient with cognitive impairment will tend to classify the patient as having VaD. During the last decade these criteria have been considered inappropriate since they do not cover milder cases of vascular cognitive impairment (Bowler & Hachinski, 1995). Diagnostic and Statistical Manual of Mental Disorders DSM-III 1980 was the first set of clinical criteria that defined multi-infarct dementia and became widely used in clinical practice and epidemiological studies. These criteria were revised in 1987 (DSM-IIIR) and developed further in DSM-IV criteria in 1994. In DSM-criteria for dementia memory deficiency is required and one or more cognitive disturbances, which together cause social dysfunction. In DSM III and IIIR, diagnosis of multi-infarct dementia re-

quires a stepwise deterioration with patchy distribution of cognitive symptoms in the early phase of the disease, focal neurological signs and symptoms due to advanced cerebrovascular disease. In DSM-IV the more comprehensive diagnosis of VaD was adopted and defined. In International Classification of Disease 10 (ICD-10; WHO 1992) criteria for VaD the general criterion for dementia requires both verbal and non-verbal memory impairment for at least 6 months and decline in other cognitive abilities which interfere with everyday activities. To be VaD, it is required that there is ‘a history, examination or tests, of significant cerebrovascular disease, which may reasonably be judged to be aetiologically related to the dementia’. In the VaD criteria of Alzheimer’s Disease Diagnostic and Treatment Centers (ADDTC) of the State of California (Chui et al., 1992) dementia is defined as a ‘deterioration from a known or estimated prior level of intellectual function sufficient to interfere broadly with the conduct of the patient’s customary affairs of life’. They require evidence of two or more ischaemic strokes by history, neurological signs and/or neuroimaging. The criteria created by the National Institute of Neurological Disorders and Stroke and Asssociation Internationale pour la Recherche et l’Enseignement en Neurosciences (NINDS-AIREN; Roman et al., 1993) were meant for dementia research and therefore, have been considered too strict by many clinicians. Dementia is defined by cognitive decline from previously higher level of functioning and manifested by impairment of memory and of two or more cognitive domains, severe enough to interfere with activities of daily living not due to physical effects of stroke. Vascular aetiology is defined by focal signs in neurological examination and evidence of cerebrovascular disease by brain imaging, which must be related to cognitive impairment. Besides, criteria for probable and possible VaD were included in these criteria as well as a series of features that were considered to make the diagnosis of VaD uncertain or unlikely, including lack of evidence of vascular disease on neuroimaging.

Criteria for subcortical vascular dementia Because VaD is considerably heterogeneous as to the causes, pathogenetic mechanisms and structural changes in the brain, more specific (research) criteria for a subgroup of subcortical VaD were recently suggested (Erkinjuntti, 2002). Using such criteria a more homogeneous patient group with better predictable course of the VaD could

291

292


be distinguished, which would provide a better target for clinical trials (Inzitari et al., 2000). These criteria are a modification of the NINDS-AIREN criteria for probable VaD (see above). The cognitive syndrome includes the dysexecutive syndrome with some memory impairment with complex (‘executive’) activities of daily living. Evidence of relevant cerebrovascular disease (CVD) includes detailed brain imaging supported by some evidence of neurologic signs and/or history indicating cerebrovascular disease. The relationship between onset of dementia and CVD required in NINDS-AIREN criteria was omitted, because the onset of subcortical VaD is often insidious, while the strong relationship between the cognitive syndrome and brain imaging provides the evidence. Features supporting the diagnosis of subcortical VaD and those making that diagnosis uncertain or unlikely were also mentioned.

Criteria for vascular cognitive impairment Bowler and Hachinski (1995) introduced their recommendation of diagnosing imminent VaD at the earliest possible stage termed vascular cognitive impairment (VCI). Even though nearly a decade has elapsed since this logical requirement to clinicians was presented, ‘a suitable sensitive and discriminatory instrument for doing this does not yet exist’. It was suggested that the criteria should be very sensitive rather than specific and in designing them the methods developed for investigating cognitive decline in multiple sclerosis could be adaptable to VCI (Bowler, 2002).

Accuracy of the clinical diagnosis of vascular dementia One of the important reasons to try to develop clinical criteria to identify when vascular disease is significant in dementia is that vascular disease is treatable (Erkinjuntti, 1997). In a neuropathological study validating the HIS a similarly high level of sensitivity (89.0% at cut off of ≥7 score) and specificity (89.3%) was found (meta-analysis by Moroney et al., 1997). However, its strict application tends to overestimate the contribution of vascular disease to dementia largely drawn from the patients with mixed pathology, where cerebrovascular disease coexists with another, and often more significant, cause of dementia. Yet, the HIS was considered to be suitable for identifying most dementia patients with at least some cerebrovascular pathology and it has been recommended to be used in practice (Knopman et al., 2001). Inevitably the different clinical criteria for VaD give different proportions of VaD in a given population. For ex-

ample, Pohjasvaara et al. (2000) studied 107 patients who fulfilled the dementia criteria according to DSM-III. VaD criteria were met in 91.6% of dementia cases using DSM-IV, 86.9% using ADDTC, 36.4% using ICD-10 or DSM-III R and 32.7% using NINDS-AIREN criteria. Several studies have shown that the frequency of the diagnosis of VaD is highest if the modified HIS or DSM-IV criteria are used, intermediate with original HIS, ADDTC and ICD-10 and lowest using the NINDS-AIREN criteria. The main reasons for the differences are definition of dementia and how the certification of cerebrovascular disease is required. The sensitivities and specificities of the abovementioned criteria were analysed recently in detail by Gold et al. (2002) in a cohort of 89 autopsied patients with dementia from a geriatric and psychiatric hospital. The sensitivities for VaD according to DSM IV, ADDTC possible/probable, NINDS-AIREN possible/probable and ICD-10 criteria were 50%, 70/25%, 55/20% and 20%, respectively, and the specificities were 84%, 78/91%, 84/91% and 94%, respectively. It was concluded that the criteria are not interchangeable, ADDTC criteria for possible VaD being the most sensitive, whereas DSM IV and NINDS-AIREN criteria may be best in excluding mixed dementia. The high specificities of the probable ADDTC and NINDS-AIREN as well as ICD-10 criteria were achieved at the expense of low sensitivity, which was suggested to require revision of these criteria. Two fundamental realities make the development of clinical criteria for VaD that are both reasonably sensitive and reasonably specific extraordinarily difficult. The first is the well recognized heterogeneity of the pathological lesions in vascular dementia which embraces combinations of single infarcts in anatomically significant locations, multiple infarcts, incomplete ischaemic changes and widespread small vessel lesions and lacunes. The second is the challenge of mixed dementia, i.e. to be able to recognize when there are two, or possibly more, types of pathological processes contributing to the cognitive decline, most frequently vascular disease combined with Alzheimer changes (Barker et al., 2002). Studies of existing criteria (Erkinjuntti et al., 2000; Pohjasvaara et al., 2000; Gold et al., 2002), all demonstrate the limitations of the current criteria, particularly in respect of their tending to identify different subgroups of patients with vascular disease, so that, as criteria, they are not interchangeable. They are also limited, as noted by Roman (2001) by the fact that they do not recognise the frequent occurrence of the ‘mixed’ dementias of various types and, thus are not good at identifying when pure pathology is present (Holmes et al., 1999; Gold et al., 1997, 2002). One suggestion (Erkinjuntti et al., 2000) is that, for VaD, we would do better to accept

Vascular dementias

the pathological heterogeneity and concentrate on developing what are essentially pathology-specific criteria for the different types of vascular pathology that produce cognitive decline, at least for the purposes of clinical trials, in the particular paper cited, subcortical VaD. The opposite approach is recommended by Scheltens and Hijdra (1998), who suggest that improvement in our ability to identify vascular dementia is likely to come from the development of specific diagnostic tests for the degenerative diseases. However, while more specific tests for degenerative processes may become available, their availability will not solve the problem of identifying cases of mixed dementia. What is clear, however, is that in any effective set of criteria for vascular dementia cerebral imaging studies will be very important (Erkinjuntti, 1997; Roman, 2001).

Dementia following stroke Dementia and clinically certified stroke Another important and clinically relevant way of looking at vascular dementia is in terms of the frequency of the development of dementia in a population with a history of stroke. This type of study has tended to be purely clinical and thus to suffer from the major disadvantage of such studies which is the lack of pathological verification of the clinical diagnoses. This is particularly important in dementia related to vascular disease where small-scale studies have shown that, while the presence of vascular disease is reliably detected, there is poor discrimination between cognitive decline as a result of vascular disease alone and cognitive decline that is the result of the combination of vascular disease and a degenerative disease. Hence, in studies of this type, there is a tendency to overestimate the contribution of vascular disease to dementia. The authors of such studies are, of course, perfectly aware of the limitations of clinical methods in this respect and attempt to compensate for them (Henon et al., 1997). Tatemichi et al. (1992) studied 251 patients aged ≥ 60 years with ischaemic strokes. In this study the cognitive function of the patients was assessed 3 months after the stroke and dementia was found in just over 26% of the patients, compared to 3.2% in an age-matched control sample drawn from the community. This frequency of dementia in stroke patients is in general agreement with a number of other smaller studies. Tatemichi et al. (1994) went on to a further refinement of this incidence by distinguishing between dementia secondary to pre-existing degenerative disease (assumed to be AD), VaD (either directly related to the region of the

stroke or unrelated), and other causes of dementia. Based principally on a temporal association of the onset of the cognitive decline with the stroke, vascular disease was considered to be the cause in approximately two-thirds of the cases of dementia. Overall, they concluded that the incidence of dementia directly attributable to stroke was approximately 15% in the cohort surveyed. Of the group of patients with dementia associated with vascular disease they considered that, in approximately 40% of the cases, the location of the stroke(s) accounted for the dementia, while in the other 60% they considered that the location of the vascular disease did not explain the dementia syndrome (using ‘traditional concepts of functional neuroanatomy’). They also found that age and level of education were independent contributors to the incidence of dementia in the population surveyed. In a later study of dementia following stroke, Henon et al. (2001), reached the conclusion that in the almost 30% of patients who developed dementia in a 3-year follow-up, two-thirds were attributable to progression of vascular disease and only one-third to AD.

Can cerebrovascular disease produce a progressive dementing syndrome? From a practical standpoint it would be helpful to have some indication of whether and how often cerebrovascular disease alone can produce a progressive generalized dementing syndrome that clinically resembles that produced by a ‘degenerative’ disease such as AD (Tatemichi, 1990). In the study of Boller et al. (1989) on dementia patients with an obvious vascular aetiology excluded, only 4 out of 54 cases were considered to have multi-infarct dementia. In another study by Joachim et al. (1988) where patients were referred to an AD research group from a large variety of community sources with a clinical diagnosis of presumed AD, only 3 (2%) out of 150 cases were considered to be demented on the basis of cerebrovascular disease alone and in a further 11 cases strokes were thought to have contributed to the dementia. Pantoni et al. (1996) described three cases and Hulette et al. (1997) six cases of progressive cognitive decline without evidence of stepwise progression. The pathological findings in the patients of Pantoni et al. were almost entirely small vessel vascular disease, which is understandable since in that disease group the pathological changes (microangiopathy) are often widely and relatively diffusely distributed. Similarly, in CADASIL with definitely pure microangiopathy (see below), dementia may develop progressively without obvious stroke-like episodes (own unpublished observations). Another progressive dementing disease resembling a

293

294


Table 13.1. Year

Authors

Country

Number of patients

VaD N (%)

VaD+AD N (%)

All VaD N (%)

1970 1985 1985 1986 1986 1987 1987 1988 1989 1990 1994 1995 1995 1998 1999 2002 2002 2002

Tomlinson et al. Mölsä et al. Barclay et al. Esiri and Wilcock Ulrich et al. Alafuzoff et al. Wade et al. Joachim et al. Boller et al. Jellinger et al. Brun Ince et al. Markesbery Nolan et al. Seno et al. Jellinger Akatsu et al. Barker et al.

UK Finland USA UK Switzerland Sweden Canada USA USA Austria Sweden Norway USA USA Japan Austria Japan USA, Fla.

50 58 311 86 54 55 65 150 54 675 175 69 557 87 122 900 158 382

9 (18) 11 (19.0) 69 (22.2) 18 (20.9) 9/11 (16.7–20.4) 13 (23.6) 6 (9.2) 3 (2.0) 4 (7.4) 106 (15.7) 59 (33.7) 4 (5.8) 13 (2.3) 0 (0) 42 (34.4.) 85 (9.4) 34 (21.5) 12 (3.1)

9 (18) 6 (10.3) 43 (13.8) 11 (12.8) 6/10 (11.1–18.5) 15 (27.3)

18 (36) 17 (29.3) 112 (36.0) 29 (33.7) 15/21 (27.8–38.9) 28 (50.6)

3 (5.6) 53 (7.9) 63 (36)

7 (13.0) 159 (23.6) 122 (69.7)

32 (36.8)

32 (36.8)

43 (11.3)

55 (14.4)

degenerative dementia would be that with diffuse microangiopathy of subcortical leukoencephalopathy/Binswanger disease, a disputed syndrome (see below) that is considered to produce a diffuse loss of white matter which leads to progressive disconnection of the cerebral cortex from the deeper structures. However, the clinical picture in these two microangiopathies does not particularly resemble the dementia of AD (Caplan, 1995; Dichgans et al., 1998). These findings suggest that cerebrovascular disease may occasionally produce a dementia clinically resembling that of AD, and in these cases it is most often subcortical with diffuse lesions.

Epidemiological Studies Prevalence and incidence in Western countries A number of recent epidemiological studies of the incidence and prevalence of dementia in population and community based cohorts (von Strauss et al., 1999; Lobo et al., 2000; Fratiglioni et al., 2000; Andersen et al., 2000) have broadly similar findings and show that AD is much the commonest form of dementia (between 60 and 70% of all the cases of dementia) with vascular dementia a distant second (12–18%). However, in autopsy studies the frequency of VaD varies considerably (see Table 13.1.). In different Western countries the prevalence of VaD is fairly similar and it increases steadily but slowly with age. Notably, a similar rapid increase of prevalence of VaD around

80 years of age as in AD does not occur (Fig. 13.1). Men have higher prevalence at lower ages, 0.5% in male and 0.1% in female population in the age group of 65–69 years but females take over around 85 years of age and the prevalences reach 3.6% in male and 5.8% in female populations in the group of 90+ years of age (Lobo et al., 2000). Most studies have concerned the prevalence of VaD, but since it is affected by the rates of new and deceased cases (Dubois & Hébert, 2001) incidence, i.e. the occurrence of new patients, gives better understanding of the frequency of VaD. The incidence of VaD in Western populations has been analysed in two major studies. In the European collaborative study (Fratiglioni et al., 2000) the incidence of VaD increased with age, but much less steeply than that of AD (Fig. 13.2.). The incidence varied from 0.7 per 1000 person–years in the age group 65–69 years to 6.1 and 8.1 per 1000 person–years in the age groups 85–89 and 90+ years. In a comparative meta-analysis the age-standardized incidence in the Canadian population was found comparable to the rates above (0.94 in the age group of 65–69 years and 6.74 in the 85+ group; for all > 65 years of age the incidence was 2.52). The variation of the VaD prevalence and incidence in epidemiological studies can largely be explained by population selection, diagnostic procedure and criteria, and mortality. The sensitivity and specificity of the diagnostic criteria may be the most important explanation for the variation as discussed in the section on diagnostic criteria. Another reason for underestimation of the prevalence is the

Vascular dementias

% 30 25 20 VaD m VaD f AD m AD f

15 10 5 0 65-69

70-74

75-79

80-84

85-89

90+

age groups Fig. 13.1. Prevalence of vascular dementia (VaD) and Alzheimer’s disease (AD) in males (m) and females (f ) in the European collaborative study.

fact that the studies are most often made without imaging techniques and therefore the vascular changes in the brain are not discovered. On the other hand, white matter changes and brain infarctions increase with age and therefore the use of imaging techniques may overestimate the prevalence of VaD. In most studies screening instruments such as the Mini Mental State Examination have been used to detect individuals with possible dementia. These

screening instruments have relatively high sensitivity and specificity for AD, but not for VaD, and therefore they cause an underestimation of the prevalence of VaD. In many studies persons living at institutions are excluded and therefore the prevalence of dementia may be lower. Also, the number of persons 85 years and older is very low in several studies which may lead to underestimation of the prevalence. Pathological studies on selected populations indicate that

1000 person-years 70 60 50 VaD m VaD f AD m AD f

40 30 20 10 0 65-69

70-74

75-79

80-84

85-89

90+

age groups Fig. 13.2. Incidence of vascular dementia (VaD) and Alzheimer’s disease (AD) in males (m) and females (f ) in the European collaborative study.

295

296


the frequency of pure VaD is considerably lower than that suggested by epidemiological studies.

Higher incidence in some Asian countries In international epidemiological comparative studies the most striking observation is the way that especially Japan but also China and Korea, are outliers in terms of the relatively high frequency of vascular dementia (Kawano et al., 1990; Jorm, 1991; Jorm & Jolley, 1998; Fratiglioni et al., 1999; Suh & Shah, 2001) that has been found in both epidemiological (Ikeda et al., 2001) and pathological studies (Ueda et al., 1992; Goto et al., 1995; Seno et al., 1999). The ratio of AD to VaD among ethnic Japanese varied from 0.5 in Japan to 1.5 in Hawaii and to 2.0 in state of Washington, USA indicating interaction between genes and environment (Chui 2000). In Japan there is a trend of decreasing prevalence of VaD and increasing prevalence of AD. In the epidemiological studies conducted in China there are methodological issues regarding clinical criteria and patient selection that make the generalization of results unsure (Shen & Yu, 2001). A study from Beijing, China gave a ratio of AD to VaD in 1994 of 1.4 which indicates that vascular dementia is the second commonest cause for dementia in urban areas in China (Wang et al., 2000). In Korea the ratio is 2.8 (Lee et al., 2002) compared with the ratio of AD to VaD of 3.4 in Europe (Lobo et al., 2000).

phorylation (Munch et al., 1998). Elevated levels of cholesterol have been reported to be associated with the risk of dementia with stroke in elderly patients (Moroney et al., 1999); similarly they have been incriminated as a risk factor for AD by enhancing -secretase activity (Ehehalt et al., 2003). However, the role of cholesterol in stroke and VaD does not appear to be unequivocal: in one study statin therapy for 5 years showed significant reduction of strokes, whereas this effect was not sustained in another 3 years’ therapy. Furthermore, in neither of these studies did statins slow cognitive decline (Collins & Armitage, 2002). The negative lifestyle factors (smoking, inactivity, obesity) are supposed to be risk factors for VaD and are often associated with low education (Skoog, 1998; Hébert et al., 2000; Meyer et al., 2000). A high level of homocysteine has been presented as a newer risk factor for VaD, due to acceleration of atherosclerosis and inflammation in the vessel wall (Steinberg, 2002). Interestingly apolipoprotein E ε4 has been associated with increased risk for both cardiovascular diseases and AD, but in recent studies not for VaD (Travkov et al., 2002; Frank et al., 2002). Another difference from AD is the fact that in AD women are disproportionately affected, whereas in vascular dementia men have, especially at younger age, a higher prevalence (Andersen et al., 1999; Fratiglioni et al., 2000).

How does cerebrovascular disease produce dementia? Risk factors for vascular dementia The risk factors for stroke and other vascular diseases are supposed to increase the risk for VaD because presence of stroke is included in the definition of VaD. Hypertension is the most important modifiable risk factor for stroke but its role in VaD is tenuous. In the Canadian Study of Health and Aging (Hébert et al., 2000) hypertension was a risk factor for VaD only in females and in the study of Posner et al. (2002) hypertension was the risk factor for VaD in persons with cardiac disease or diabetes. Hypertension increases white matter changes but their association with cognitive decline is more complicated. Antihypertensive treatment has been shown to decrease the risk for stroke and Alzheimer’s disease but its effect on vascular cognitive decline is still unclear. Diabetes is reported to increase the risk for micro- and macroangiopathy in the brain and thereby the risk for stroke and cognitive decline, but the results are not consistent. Hyperglycemia also causes increased formation of advanced glycation end products (AGE) which are associated with increased -amyloid deposition and tau hyperphos-

Organization of cerebral vasculature Knowledge of the organization of the cerebral circulation, especially that of the smaller branches, is essential for understanding the pathogenetic mechanisms and distribution of the cerebrovascular lesions. (Pantoni & Garcia, 1997; Kalimo et al., 2002a). The three major arteries, anterior, middle and posterior cerebral arteries (ACA, MCA and PCA) branch in the superficial subarachnoid space to cover their supply territories, anastamosing with each other at the periphery of their territories in the ‘border zones’ also called ‘water shed zones’ (Fig. 13.3.). At this point the important distinction between focal (or regional) and global cerebral ischaemia may be emphasized: in focal ischaemia the brunt of the insult affects the centre of the supply zone of the occluded artery/arteriole, whereas if global hypoperfusion (e.g. in severe shock) ensues it is the border zones that first become ischaemic. Penetrating arteries arise at right angles from the main branches in the subarachnoid space. The short penetrators supply the cortex as well as the subcortical WM (U-fibres). The long penetrators enter the

Vascular dementias

(a)

Fig. 13.3. The supply territories of cerebral arteries. 1. Middle cerebral artery. 2. Border zone between anterior and middle cerebral artery. 3. Border zone between middle and posterior cerebral artery. 4. Border zone between the three cerebral arteries. (Reproduced with modification from Romanul 1970.)

WM as carrying vessels (rami medullares) of about 100–200 m in diameter, which remains fairly constant until the artery ends at approximately 20 to 50 mm from the surface. Short distributing arteries branch off from these various vessels and supply a cylindrical parenchymal unit without functional collateral circulation between neighbouring vessels (penetrating arteries are of terminal type). The deep penetrators extend to the periventricular WM and either meet there short ventriculofugal arteries originating from subependymal arteries with formation of a deep located border zone (de Reuck 1971, Fig. 13.4 (a) and (b)), or simply have their distal irrigation field periventricularly situated (Moody et al., 1990). All the deep grey nuclei have a similar pattern of blood supply, being fed by small arteries that are direct branches off large arteries, there being no progressive arterial arborization to progressively smaller diameters. This is the case both for the lenticulo-striate branches of the middle cerebral artery that supply the basal ganglia, and the branches from the initial segment of the posterior cerebral artery that enter through the posterior perforated substance to supply the thalamus. This pattern of blood supply is different from that of white matter and it also confers on these nuclei a peculiar vulnerability to vascular disease. Where the white matter is vulnerable because the long penetrating arteries are at the distal end of a very long arterial arborization, the deep grey nuclei are vulnerable because their blood supply lacks the protection of an arborization. Being direct branches of large arteries, the arterial supply of the deep grey nuclei is immediately exposed to the effects of hypertension, wherefore all three regions supplied in this

(b)

Fig. 13.4. The two systems of penetrating arteries are: (a) pial (dashed lines) and deep (solid lines) penetrating branches (b) Pial arteries give rise to short (1), medium (2) and long (3) penetrating branches to cerebral cortex as well as the longest branches (4) to the white matter. (Reproduced from Romanul 1970.)

manner, the basal ganglia, thalamus and pons (supplied by branches from the basilar artery) are sites of predilection for hypertensive haemorrhages and small vessel disease.

297

298


Pathogenetic mechanisms of cerebrovascular lesions that cause VCI or VaD The NINDS-AIREN International workshop entitled Vascular Dementia: diagnostic criteria for research studies (Roman et al., 1993) suggested six general ways that cerebrovascular disease can produce dementia: (i) Multi-infarct dementia, (ii) Strategic single infarct dementia, (iii) Small vessel disease, (iv) Hypoperfusion, (v) Haemorrhagic dementia, and (vi) Other mechanisms which include combinations of the five preceding causes and other as yet unknown mechanisms. The first three categories are the most important and they represent focal hypoxia–ischaemia caused by different conditions in the blood vessels (such as athero/arteriosclerosis, embolic or thrombotic occlusions, specific angiopathies or inflammatory diseases). In the fourth category, hypoperfusion, dementia is caused by an episode of global cerebral ischaemia following, e.g. cardiac arrest or hypotension. The fifth category is dementia resulting from chronic subdural haematoma, subarachnoid haemorrhage or intracerebral haematoma. This formulation needs to be recognized as a considerable simplification and these categories are themselves not exclusive, it being perfectly possible to have both large cystic infarcts and small vessel disease in the same patient (Vinters et al., 2000). Nevertheless it fulfils a need for a framework into which to fit a description of the range of vascular disease that is encountered in patients with cognitive decline. Hypoxia–ischaemia and infarction In the above-mentioned three main categories the cerebrovascular disturbance is focal hypoxic–ischaemic, i.e. there is insufficient supply of oxygen and substrates to the brain to meet the energy requirements. Hypoxia alone without concomitant ischaemia (reduced blood flow) is, on the basis of experimental studies, less deleterious, because the removal of harmful metabolic waste products (especially acidic constituents) is maintained. On the other hand, in clinical situations, hypoxia is rapidly followed by reduction in CBF, i.e. ischaemia also ensues. Threshold values of cerebral ischaemia In the grey matter it has been shown that there are definite threshold values for the CBF needed for maintenance of functional and/or living neurons. These were first determined experimentally (Astrup et al., 1981; Heiss, 1992) and then verified to also apply to humans (Baron 1999, 2001). The CBF values in the grey matter are from 50 up to over 100 ml/min per 100 g of tissue depending on the intensity of neuronal activity. When the CBF is reduced to about 20–25 ml/min/100 g (40–50% of the base flow value)

the neuronal electrical activity begins to decrease with appearance of K+ /Ca2+ transients and cytotoxic oedema begins to develop, i.e. tissue is at risk of ischaemic injury. At about 25–30% the threshold of electrical failure is reached and the neural transmission ceases though the neurons still survive (penumbra zone). Further decrease to about 8–12 ml/min/100 g (15–20%) of basal flow value results in membrane failure in neurons with uncontrolled ion fluxes and irreversible nerve cell death (necrosis) unless rapid restoration of the CBF occurs (Baron, 2001). The corresponding threshold values for the WM, where neuronal processes, oligodendrocytes and myelin are the predominating (and vulnerable) components, are not known, but PET studies have shown that normally the CBF in the WM is about 40% of that in the grey matter, i.e. 20 to 40 ml/min/100 g (Law et al., 2000). Ischaemic tissue damage and infarction Different diseases in the vessel wall (atherosclerosis, vasculitis, ‘lipohyalinosis’, etc) or within the vessel (embolus, local thrombus) may cause circulatory impedance and ischaemia of such degree that tissue destruction (necrosis) of variable severity ensues. When all cellular components become necrotized, a complete infarct has occurred (Fig. 13.5), whereas in incomplete infarction only a selected, most/more vulnerable population of cells has become necrotized. ‘Lacunar infarcts’ or ‘microinfarcts’ differ from ‘common’ infarcts only by the size of the infarct, meaning that blood flow in a small artery or arteriole has been compromised, but it does not indicate a specific cause for the ischaemia and consequent tissue necrosis. The presently used subtyping of infarcts due to focal cerebral ischaemia with their relative frequencies in the Stroke Data bank of NINDS (Sacco, 1994) are given in Table 13.2. As the figures imply infarcts due to large vessel occlusions are more often caused by emboli (5.4 + 19.3 = 24.7%) than by local thrombosis in atheromatous arteries (8.9%). Embolic Table 13.2. Relative frequencies of different subtypes of symptomatic infarcts (all causing strokes, not only causing VCI/VaD)1 Atherothrombotic, large artery thrombosis Atherothrombotic, artery-to-artery embolism Cardioembolic Lacunar Undetermined cause

8.9 % 5.4 % 19.3% 26.6 % 39.9 %

1 In Stroke Data bank of NINDS. Adapted from Sacco (1994).

Vascular dementias

(a)

(b)

Fig. 13.5. Multi-infarct dementia: two adjacent brain slices from a patient with multi-infarct dementia showing the variety and severity of tissue destruction commonly seen in these patients. (a) In the more anterior slice there is infarction in the territories of the inferior division of both middle cerebral arteries, part of the superior division of the right middle cerebral artery and in the left caudate and putamen involving the intervening internal capsule. This damage has resulted in significant enlargement of the right lateral ventricle and third ventricle and gross enlargement of the left lateral ventricle. (b) In the posterior slice, there are still infarctions in both inferior divisions of the middle cerebral arteries, the left putamen and an additional extensive area of damage in the white matter of the left centrum semiovale. There are also at least three identifiable lacunes in the thalami.

infarcts often become haemorrhagic when the embolus is broken down and recirculation into ischaemically injured vessels occurs, whereas the majority of large vessel infarcts in VCI/VaD seem to be of the pale, anaemic type. Knowing the pattern of embolisation it is likely that in VCI/VaD embolic aetiology is even more common than in strokes in

general, since multiple infarcts are much more common than strategic single infarcts as the cause of cognitive decline (Esiri et al., 1997; Esiri, 2000). The size of these single or multiple infarcts vary from large hemispheric to small infarcts, the diameter of which should exceed 20 mm, since this value is given as the upper limit for the size of lacunar infarcts (Lammie, 2002). Small vessel disease The character of the vascular and parenchymal changes in small vessel disease has been considered to differ from those in larger vessel disease to such an extent that this entity is usually described separately. Small vessel disease includes two historic patterns of vascular disease, the e´ tat lacunaire described by Pierre Marie in 1901 and socalled Binswanger’s disease (subcortical (arteriosclerotic) leukoencephalopathy), widespread diffuse disease affecting mainly the white matter, first described in 1894. The combination of a history of hypertension and the almost invariable presence of some degree of lacunar disease together with white matter lesions is strong evidence for the importance of hypertension in the pathophysiology of small vessel disease. There are however, many other causes, both hereditary (see below) and sporadic diseases such as diabetes, antiphospholipid antibody syndrome, some forms of amyloid angiopathy and severe hyperlipidaemia (Caplan, 1995). Lacunes Lacunes are small cavitary lesions found within the brain parenchyma which range in size between 1 and 20 mm. Lacunes have been classified into three types based on their aetiology (Poirier and Derouesne 1984, 1998, Lammie, 2002). The most common and clinically most significant type I lacunes are small infarcts, type Ia being the complete (focal pan-necrosis, Fig. 13.5.(b) 13.6. and 13.11.) and type Ib the incomplete (necrosis of selectively vulnerable components only) variant. Type II lacunes are due to small haemorrhages or small haemorrhagic infarcts, and type III are just dilated perivascular spaces. The small size of the lacunar infarcts indicates that the diseased vessel must be small, the lenticulo-striate arteries below 100–800 m and the white matter penetrating arteries 100 to 200 m in diameter (Lammie, 2002). Consequently, they are most typically found in the deep grey nuclei, internal capsule and the basis pontis as well as in the deep hemispheric white matter. The modern imaging methods have made the identification of lacunar infarcts relatively easy, which is also reflected in their becoming the most common type of infarct (26.6%, Table 13.2.). When lacunes are numerous, indicating the presence of widespread severe small vessel disease,

299

300


Fig. 13.6. Lacunar state: in this patients who suffered from a stepwise cognitive decline there are at least seven lacunes of varying size in this plane of section of the thalamus and posterior putamen. Other planes of section in this case showed additional lacunes with very little focal vascular disease elsewhere in the brain.

the condition has been called the lacunar state or, more evocatively for Anglo-Saxons, l’état lacunaire (Figs. 13.6 and 13.11; Marie, 1901). Lacunar infarcts are frequent in patients with a history of hypertension and in his classic studies Fisher (1969) described marked changes in the vascular wall, descriptively called ‘lipohyalinosis’. However, lacunar infarcts have most certainly multiple aetiologies. At present fibrinoid necrosis (which later develops into ‘lipohyalinosis’, i.e smooth muscle cells are replaced by hyaline fibrosis and occasional lipid-aden macrophages) is regarded as the main destructive lesion in smaller arterioles 40 to 300 m in diameter. Small atherosclerotic plaques have been considered responsible for stenosis or occlusion of 200 to 800 m diameter vessels. In these cases the final occlusion has been suggested to be caused by local thrombosis or post-stenotic hypoperfusion (Lammie, 2002). Furthermore microemboli have been strongly advocated by some researchers, e.g. Millikan and Futrell (1990) who in general considered lacunar infarcts just small in size but otherwise similar, also etiologically, as their larger counterparts. In addition, small arterioles may be affected by other specific hereditary diseases with very interesting aetiologies and pathogeneses (see below). White matter lesions The discussion of the white matter lesions is complicated by the somewhat confusing terminology. Leukoaraiosis

was introduced by Hachinski (1987) as a descriptive term for the relatively diffuse white matter imaging change, which is a frequent finding in hypertensive and aged people (see below). This introduction of leukoaraiosis ‘unleashed a veritable diagnostic epidemic of the controversial Binswanger’s disease’ (BD) (Hachinski, 1991). It is uncertain what the disease was that Otto Binswanger’s patients at the end of the nineteenth century suffered from and to which an eponym with his name was later given. Therefore this ill-defined eponym, in the view of some authors, ‘lacks medical significance or relevance’ and should not be used (Hachinski, 1991). Instead, a more descriptive term subcortical leukoencephalopathy was recommended (Pantoni & Garcia, 1995). Caplan (1995) was not as negative, though he did not regard BD as a distinct entity. According to him BD is a heterogeneous group of small vessel diseases due to diverse causes and pathophysiologies (see above), hypertension being the most common cause of this condition which he called chronic microvascular leukoencephalopathy (Caplan, 1995; Loeb, 2000). Thickening of the walls of small arteries and arterioles together with haemodynamic and rheological factors cause chronic ischaemia, which consequently leads to white matter lesions (WML), i.e. leukoaraiosis in imaging and loss of oligodendrocytes and myelin (incomplete infarction) in pathology. If more severe the ischaemia leads to multiple lacunar infarcts (lacunar state). Positron emission tomography (PET) scanning of patients with symptomatic vascular disease and with identified infarcts, lacunes or leukoaraiosis has demonstrated a widespread reduction in perfusion compared to non-symptomatic patients (except in the cerebellum) and, using cobalt-55, ongoing and progressive ischaemic injury in the white matter (De Reuck, 1996; De Reuck et al., 1998, 2001). The cause of the thickening of the walls of the penetrating arteries has remained open. In addition to general hypertension, Okeda (1973) suggested that autoregulatory activity in the short penetrating arteries supplying the cortex could result in a relative ‘shunting’ of flow to the arteries supplying the deep white matter which results in inappropriate thickening of these arteries and an impaired microcirculation. Hypoperfusion (global cerebral ischaemia) Global cerebral ischaemia (GCI) is a less common cause of VaD. It has also received less attention than different forms of focal ischaemia, all the more, since GCI often causes incomplete infarctions, which are more difficult to detect both by imaging and in neuropathological examination. GCI commonly occurs in severe hypotension of different causes, for example in association with cardiac

Vascular dementias

Table 13.3. Causes of non-traumatic intracerebral haemorrhages Cause

%

Hypertension Cerebral amyloid angiopathy Anticoagulant therapy Tumours Illicit and licit drugs Aneurysms and arteriovenous malformations Other

50 12 10 8 6 5 9

Adapted from Feldman (1994).

arrhythmias, myocardial infarction, cardiac arrest, congestive heart disease, any major surgery associated with general anaesthesia, or hypovolemic shock. Those areas of tissue damage relevant to VCI/VaD are the arterial border zones, i.e. in the most distant regions supplied by branches of ACA, MCA and PCA, both in cerebral cortex and in the deep white matter (Figs. 13.3 and 13.4), and hippocampus supplied by small branches of the PCA or anterior choroidal artery. In cerebral cortex layers III, V and VI are the most vulnerable and in hippocampus CA1 area and subiculum (Auer & Sutherland, 2002). Furthermore, different disorders of the vessel walls, such as atherosclerotic stenoses in larger arteries and loss of adaptability in small vessel diseases (see above) most likely predispose to hypoperfusion, since the normal compensatory mechanisms are compromised by these diseases. Haemorrhagic dementia Haemorrhages are relatively rare causes of dementia. The frequencies of different causes of intracerebral haemorrhages are given in Table 13.3. Despite the presently available effective treatment for hypertension, the cause of about half of the intracerebral haemorrhages (ICH) is still hypertension. They are most commonly deep and relatively large ICHs located in the basal ganglia and thalamus. Most often they cause motor deficits or if they break into the ventricular system they are almost invariably fatal. ICHs in more superficial locations, lobar ICHs, are usually non-hypertensive with specific aetiologies. Cerebral amyloid angiopathy (CAA) is responsible for about 12% of all non-traumatic ICHs (Kalimo et al., 2002a). CAA is described in detail in Chapter 14. Vasculitides, both infectious and immunologically mediated, are another cause of non-traumatic ICHs, which may lead to VaD. Many of the ICHs caused by use of illicit drugs are associated with inflammatory cells and necrosis

in the arterial wall, but the drug-induced acute rise in the blood pressure may as such be a sufficient noxious factor to cause an arterial breach. Toxic components in the abused drugs may also contribute to neuronal damage. Among immunologically mediated vasculitides in which ICHs occur are polyarteritis nodosa and systemic lupus erythematosus (Kalimo et al., 2002a; Jennekens & Kater, 2002). In addition to having high mortality subarachnoid haemorrhage (SAH) causes disabling cognitive dysfunction and poor health-related quality of life in up to 50% of the survivors (Kreiter et al., 2002; Mayer et al., 2002; Hadjivassiliou et al., 2001). The cognitive impairment was fairly general in character affecting most cognitive domains, but psychomotor speed, language function and verbal memory seemed to be most severely deteriorated. The general deleterious effects of SAH, appeared to be aggravated by the incidence of single or multiple small infarcts in the vascular territory of the aneurysm, which were more common in patients operated on by clipping than in those treated by coiling (Hadjivassiliou et al., 2001). Patients with chronic subdural haematomas (CSH) have frequently marked cognitive decline. In a recent Japanese study 69.2% of the 26 patients operated for CSH were considered demented on the basis of their performance on Mini Mental State Examination and the Hasegawa Dementia Scale-Revised (Ishikawa et al., 2002). The cause of impaired cognitive performance is considered to be related to the increased intracranial pressure due to the space occupying haematoma. Fortunately, CSH does not readily cause marked irreversible brain damage and surgical removal often results in good recovery (Ishikawa et al., 2002), if CSH is diagnosed early enough.

Significance of the volume the cerebrovascular lesions Imaging studies The modern imaging technology has provided a method to visualize tissue changes in the CNS much earlier than by pathology. At the same time it should be possible to quantify the volume of tissue changes (and assess their localization, see below) in the images at selected section thickness more easily than in post-mortem brain slices, although the histopathological substrate of the imaging findings is not always certain. In many studies the quantification is based only on semiquantitative rating scores, but there are also available semi-automated techniques in which volumes of lesions can be measured on the basis of their isointensity with the indicated reference intensity. In aged persons without neurological disease the significance of extent or severity (principally volume) of WML for

301

302


cognitive function has been analysed in large correlative studies (Boone et al., 1992; Longstreth et al., 1996; de Groot et al., 2000, 2002). These have shown that the extent of WML is associated with cognitive decline, but there seems to be a threshold which must be surpassed before impairment of cognitive functions appear/become detectable (Pantoni et al., 1999; Desmond, 2002). Such a threshold appears to vary depending on the topographic distribution of WML (see below, site of vascular lesions). In clinically diagnosed cases of VaD the extent of WML has been mainly related to impaired executive performance, but not to more global cognitive dysfunction (Mungas et al., 2001, Cohen et al., 2002). In CADASIL, a hereditary form of VaD with early and characteristic WML, the gene defect makes the development of the dementia inescapable. This allows monitoring of the development of dementia in comparison with increase in WML (see below). Although several MRI studies on patients with clinically diagnosed VaD have been published in recent years, only a few studies with neuropathological follow-up are available. The difficulty of assigning a useful interpretation to the presence of leukoaraiosis has been confirmed by Clarke et al. (2000) who, in a clinico-pathological study, noted its presence in all histological varieties of dementia, its occurrence being most closely related age and cardiovascular risk factors. Quantitative imaging studies have addressed especially VCI of subcortical ischemic vascular disease (SIVD) (Mungas et al., 2001; Du et al., 2002), and have given findings somewhat discrepant with pathological analyses. The finding that subcortical lacunes (with an attempt to exclude perivascular vacuoles) were not good predictors of performance in neuropsychological tests contrasts with the results of Esiri et al. (1997) according to which cribriform change and microinfarction were more common in demented patients with vascular disease. On the other hand, in MRI studies it is mainly the volume of the remaining tissue that has usually been quantified, not that of infarcted parenchyma. Such quantitative studies have shown that cognitive impairment is associated with reduced cortical grey matter and hippocampal volumes, not only in AD but also in VaD (Mungas et al., 2001). This was considered not to reflect concomitant AD pathology in VaD patients, since AD pathology was not present in three of their neuropathologically analysed patients (Fein et al., 2000; Mungas et al., 2001). In accordance, in another study the whole brain volume in VaD was strongly associated with the level of global cognitive function (Cohen et al., 2002). Autopsy studies In their pioneering study Tomlinson et al. (1970) reported that a volume of 100 ml or greater most often resulted in

dementia and volumes from 50 to 100 ml did so less consistently. They also had some patients whose dementia appeared to be caused by a lesser volume of lost tissue, and in most VaD patients in later studies considerably smaller volumes have been incriminated as a cause of impaired cognition. The estimation of the volume in neuropathological examination requires a lot of work and relatively few researchers have taken the trouble to perform such studies. Erkinjuntti et al. (1988) analysed the brains of 27 clinically diagnosed VaD patients and in 23 patients found multiple cerebral infarcts with mean number 5.8 (range 2 to 9) and with mean volume of 39 ml with range from 1 to 229 ml. Thus, surprisingly small volumes of tissue loss were considered pathogenic, though the authors conluded that larger infarcts are of major importance in causing VaD, subcortical lesions being contributory. Later Erkinjuntti et al. (1997) repeated the study in 22 of the above 23 patients with mean number of focal infarcts 5.7 (range 2 to 9) and their mean volume 40.7 ml (range 0.9 to 229.1 ml). In a similar study by del Ser et al. (1990) the mean number of vascular lesions in the carotid territory was 4.9 (SD 3.4). The exact volume values were not given but the total relative volume of vascular lesions in 28 demented patients examined was 33.2 (SD 33) 100 parts/1000 (∼3.32%) which was significantly (over 2fold) greater than in 12 non-demented persons (11.8, SD 12). The significance of the amount of tissue loss was also demonstrated by Esiri et al. (1997): the volume of infarcted tissue was greater (2.48% vs. 1.93%) in demented than in undemented patients with vascular disease as were the frequencies of cribriform change and microinfarction. On the other hand, Esiri et al. (1997) reported that single infarcts were more common in undemented than in demented patients. Such results indicate that microvascular disease, not macroscopic infarction can be the chief substrate of VaD, which agrees with several other studies, which have shown that patients with even relatively large single infarcts do not necessarily become cognitively impaired (for further details, see the paragraph on dementia following stroke, above).

Significance of the site of the cerebrovascular lesions It has been emphasized that the pathological lesions in vascular dementia are heterogeneous, i.e. the same patient may have both large cystic infarcts and small vessel disease (Rockwood et al., 1999; Vinters et al., 2000). The commonest neuropathological findings are multiple small vessel lesions affecting both grey and white matter (Pantoni et al., 1996; Esiri et al., 1997; Esiri, 2000; Vinters et al., 2000; Jellinger, 2001).

Vascular dementias

Infarctions Multi-infarct dementia This is the classical concept where the combined effect of a number of different infarcts has a cumulative effect on cognitive function sufficient to produce a dementia. The common expression of this type of cerebrovascular disease is illustrated in Fig. 13.5., but occasionally the multiple infarcts can be very large numbers of microinfarcts (Fig. 13.7.). There are many variations on this particular theme, but most cases will be found to have bilateral infarcts affecting parts of the middle cerebral artery territories of both hemispheres, often with additional damage in the basal ganglia and thalamus. Strategic single infarct dementia The striking feature of individual cerebral infarctions is how few of them cause mental dysfunction resembling a dementia. Major left hemisphere infarcts with significant hemiparesis and language disorder may not produce gross cognitive decline. A caveat here is that it is very difficult to evaluate the cognitive function of patients with severe aphasia. However, there are a number of areas in the brain where a single infarct can produce a restricted but significant impairment in cognitive function (Tatemichi et al., 1995). As has been discussed, if the dementia associated with advanced neurodegenerative disease is used as the necessary standard, many of the reported cases are not sensu stricti examples of dementia because the cognitive decline associated with these strategic single infarcts cannot be said to be generalized. As listed by NINDS-AIREN, the major examples of vascular territories infarction of which is associated with changes in cognitive function are: Angular gyrus infarcts Cognitive function abnormalities are a fluent aphasia, alexia with agraphia, memory disturbance, spatial disorientation and constructional disturbances. Bilateral posterior cerebral artery infarcts (Fig 13.8.) This will produce bitemporal and occipital infarctions. The bilateral hippocampal infarction produces an amnestic state. This may be accompanied by psychomotor agitation, confusion, and visual impairment and hallucinations. Very occasionally, a rather similar amnestic state can occur when bilateral convexity subdural haematomas produce chronic distortion and compression of the temporal lobes that results in temporal lobe atrophy. This situation would probably be included in the category of ‘haemorrhagic dementia’ in the NINDS-AIREN classification (see below).

Fig. 13.7. Multi-infarct dementia: the pattern and severity of granular atrophy of the cortex seen in patients who have large numbers of microemboli. In this brain it is conspicuous that the most severe damage is seen in the parasagittal boundary zone regions.

Fig. 13.8. Infarction of the hippocampus and part of the parahippocampal gyrus as a result of occlusion of the posterior cerebral artery. Bilateral occlusion of the posterior cerebral artery can result in bilateral hippocampal infarction with severe effects on memory.

Vinters et al. (2000) recorded remote hippocampal injury in 11 of 20 patients with ischemic VaD, though given the presence of this lesion in 16% of old non-demented patients and in 26% of those with dementia of any aetiology (Dickson et al., 1994) the true significance of its presence in VaD is hard to evaluate. Anterior cerebral artery infarcts Associated with abulia, transcortical motor aphasia, memory impairment and dyspraxia.

303

304


These examples almost certainly do not cover the full spectrum of focal lesions that can be associated with some degree of cognitive impairment.

Fig. 13.9. Cognitive decline as a result of strategic specific infarctions: the macroscopic appearance of the thalamus in a patient with mesencephalic artery syndrome that has produced bilateral thalamic damage with cognitive impairment. Some indication of the degree of tissue loss from the thalamus is provided by the degree of dilation of the third ventricle.

Inferior genu of the capsula interna infarcts Tatemichi et al. (1992) described six patients with a focal infarct in the inferior genu of the capsula interna. In five with left-sided lesions neuropsychological testing revealed severe verbal memory loss and additional cognitive deficits in four of them. In the sixth patient the right-sided infarct caused transient impairment of visuospatial memory. Pantoni et al. (2001) also described cognitive impairment, in addition to abulia, after capsular genu infarction in two patients, the cognitive effects of which they hypothesised to be a combined consequence of these and other subclinical lesions, revealing the difficulty in ascribing cognitive decline to a strategic single infarct. Right middle cerebral artery infarcts May be associated with confusional episodes and psychosis. Parietal lobe infarcts Cognitive and behavioural abnormalities in addition to the spatial perceptive disorder. Thalamic infarctions (Fig 13.9.) Particularly of the anterior and intralaminar nuclei. Infarctions in this region produce predominantly a profound memory disturbance similar to that seen in Korsakoff’s psychosis and what is termed an amnestic dementia. Basal forebrain lesions Amnesia and behavioural changes.

Small vessel disease Lacunar infarcts Lacunar infarcts tend to be either asymptomatic or, if located in a clinically eloquent area of the brain such as the internal capsule or the basis pontis, produce sharply defined focal neurological symptoms and signs. Clinically, a large number of individual neurological syndromes have been described in association with single lacunes (Fisher, 1982). Examples include pure sensory strokes, pure motor hemiparesis or monopareses. Equally important is that dementia is hardly ever caused by individual lacunes, although a single example, memory loss, as a result of bilateral infarction of the pillars of the fornix has been described. However, when present in large numbers lacunar infarcts are associated with cognitive decline (del Ser et al., 1990), clearly manifested also in CADASIL for which multiple lacunar infarcts are characteristic. On the other hand, the small vessel disease often causes, at the same time, more diffuse lesions in the white matter (see below) which produce a diffuse loss of white matter, leading to progressive disconnection of the cerebral cortex from the deeper structures. White matter lesions WML have been demonstrated to also have a selective distribution, which most likely reflects the pathogenesis of these lesions (Pantoni & Garcia, 1997; de Groot et al., 2000; O’Sullivan et al., 2002). WML have been reported to appear first in the periventricular region, to extend later at more severe stages of the disease more widely into the centrum semiovale. On the other hand, the subcortical WM seems to be relatively better spared. The topography – periventricular–subcortical gradient – of the WML has been suggested to reflect the vascularization pattern, the arteries to the periventricular regions being the longest and most vulnerable for degenerative changes (Fig. 13.4, for details see below). The association of WML with mainly decline in executive functions and psychomotor speed has been ascribed to anatomic factors: the periventricular region as well as the centrum semiovale contains long association fibres connecting cortex with subcortical nuclei as well as with other distant cortical regions, whereas the subcortical short looped U-fibres mainly connect neighbouring gyri. It must also be emphasized that diffuse WML are not usually complete infarcts, but most probably incomplete infarcts damaging the most vulnerable cells in WM, oligodendrocytes, and hence myelin, whereas axons

Vascular dementias

may have retained their integrity though with impaired function.

Hereditary vascular dementia Hereditary liability to ischaemic stroke or intracerebral haemorrhages There are a number of well-established, most often rare conditions that confer a hereditary liability to recurrent cerebral infarcts or haemorrhages and are therefore potential causes of vascular dementia (Hademenos et al., 2001). Those causing ischaemic strokes include conditions such as disorders of different coagulation factors, certain connective tissue diseases, metabolic diseases such as homocysteinuria (Bertsch et al., 2001), varieties of the dyslipoproteinaemias and Fabry’s disease (Mendez et al., 1997), mitochondrial disorder MELAS (Tanahashi et al. 2000) and several inherited small vessel diseases (see below). Among hereditary disorders associated with haemorrhagic strokes there are different amyloid angiopathies (see Chapter 14), vascular malformations, some connective tissue disorders and polycystic kidney disease. Presenting all these entities is beyond the scope of this chapter and the reader is referred to a recent review by Hademenos et al. (2001).

Hereditary small vessel diseases, including CADASIL Several hereditary diseases affect small vessels of the brain (Dichgans, 2003), among them at least the following may lead to dementia: cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL), hereditary endotheliopathy with retinopathy, nephropathy and stroke (HERNS; Ophoff et al., 2001) and cerebral autosomal recessive arteriopathy with subcortical infarcts and leukoencephalopathy (CARASIL; Yanagawa et al., 2002). Among them the disease that has received the greatest attention is CADASIL, which is presented here in greater detail.

History and background Hereditary forms of neurodegenerative diseases have been of utmost importance in clarifying pathogeneses of dementing disorders. The first hereditary form of vascular dementia with an identified gene defect has been a multiinfarct disease with a lengthy but at the same time descriptive name cerebral autosomal dominant arteriopathy with

subcortical infarcts and leukoencephalopathy (CADASIL). Most probably the first CADASIL family was described by van Bogaert (1955) as a familial form of Binswanger’s disease, though the honour has been traditionally ascribed to Sourander and W˚alinder (1977), whose family actually most likely suffers from another hereditary vascular dementia (unpublished observations). Tournier-Lasserve et al. in the early 1990s mapped the defective gene to chromosome 19 and at the same time gave CADASIL its present name (Tournier-Lasserve et al., 1993). In 1996 the defective gene was identified to be Notch3 at locus chr19p13 (Joutel et al., 1996) and the following year the stereotypic character of the defect was described (Joutel et al., 1997). Interestingly, Notch3 was found to have a close relationship also with Alzheimer’s disease (AD), since upon ligand binding Notch3 appears to be proteolytically cleaved by the same -secretase/presenilin1 as -amyloid precursor protein (APP) (De Strooper et al., 1999; for details see below), mutations of presenilin1 being the major cause of early onset familial AD (Chapter 9). The number of CADASIL patients increased rapidly, when molecular genetic identifcation of the gene defects and diagnostic ultrastructural findings in diseased arteries made definite diagnosis possible, but CADASIL appears to be still underdiagnosed and better awareness of this entity is desirable. Clinical picture Epidemiology CADASIL occurs worldwide and in many different ethnic groups. The greatest number of families have so far been identified among European Caucasians, while in the USA and Canada the number of reported cases has been surprisingly low considering the size of their populations and the high standard of neurology in these countries. The clinical picture of CADASIL has been described in detail in members from French families (Chabriat et al., 1995a) and German/Austrian families (Dichgans et al., 1998). Exact prevalence numbers have not been reported, but the estimation of prevalence in Finland, with a relatively high frequency and fairly good general awareness of this entity, is about 4 per 100 000. The worldwide number of CADASIL families is estimated to exceed 500 making CADASIL much more common than hereditary AD. Onset and duration The four cardinal symptoms of CADASIL are (i) migraine with aura, (ii) ischaemic attacks (transient or strokes), (iii) psychiatric symptoms and (iv) cognitive decline and dementia. The age of onset varies greatly. Migraine attacks may begin even before the age of 10 years, but most often

305

306


drag on, the oldest living patient having reached the age of 95 years (S. Tuisku, personal communication). Cognitive decline and dementia Cognitive decline can be detected in the prestroke phase of the disease and it becomes clinically manifest between 40 and 70 years of age. CADASIL affects predominantly frontal lobe functions and, thus, the patients have impaired executive and organizing functions, general mental slowing, poor concentration and slowing of motor functions, whereas memory is affected only at later stages. The end stage is a subcortical vascular dementia (Taillia et al., 1998; Amberla et al., 2003) present in about 80% of the CADASIL patients aged over 65 years (Dichgans et al., 1998).

Fig. 13.10. In a 54-year old moderately demented CADASIL patient with R133C mutation the hyperintensities in T2 -weighted MRI are confluent reflecting severe involvement. (Courtesy of Timo Kurki, Turku University Hospital.)

they appear at 26 to 38 years of age (Chabriat et al., 1995a,b; Dichgans et al., 1998; Tuominen et al., 2001). However, because migraine is so common as an independent disease, the age of onset is usually given on the basis of the first ever stroke. On the other hand, it may be difficult to distinguish between a transient ischaemic attack (TIA) and migrainous aura, which may be very severe and long-lasting in CADASIL. The age at which the first ischaemic attack strikes is also highly variable. The peak of ischaemic attacks is around 40–50 years of age with the youngest age at the first stroke being 28 years and the latest age in the 70s. The reported mean ages of onset varies between 37 and 45 years (Sonninen & Savontaus, 1987; Chabriat et al., 1995a,b; Dichgans et al., 1998). The variation in onset is wide even within the same family and even between monozygotic twins (Kalimo et al., 2002c), which indicates that the variation does not depend on the type of mutation and that other factors must also modify the course of CADASIL. In general, CADASIL shows slowly progressive exacerbations occurring during recurrent strokes. Death usually ensues within 10 to 20 years from the onset, but a few patients

Imaging Hyperintensities are seen in T2w and FLAIR MRI in cerebral white matter, especially in anterior temporal lobes (Fig. 13.10). The first changes are detectable already in the early asymptomatic stage and, on the other hand, periventricular hyperintensities are so common (present in 96% of patients) that their absence in practice excludes CADASIL (Chabriat et al., 1998). The findings are compatible with the imaging diagnosis of leukoaraiosis and they are deceptively similar to multiple sclerosis. The white matter lesions in CADASIL generally begin as small periventricular punctiform hyperintensities, progress through a nodular stage in parallel with cognitive decline to confluent lesions (Chabriat et al., 1998) in which the total lesion volume correlates significantly with disability (Dichgans et al., 1999). Infarction superimposed on white matter lesions causes marked aggravation of the cognitive decline in CADASIL as exemplified in a pair of cousins (Desmond, 2002) and in a pair of monozygotic twins (Kalimo et al., 2002c) with WML of similar severity in each pair. Diffusion tensor MRI of the white matter reveals marked increase in water diffusivity, and the diffusion can occur more freely in any direction meaning loss of anisotropy (Chabriat et al., 1999). This is considered to indicate a microstructural change with enlargement of the extracellular space due to vasogenic oedema, possibly associated with myelin and axonal damage. Thus, this imaging finding corresponds well with the histopathological alterations in CADASIL. Furthermore increased diffusivity is detectable also in the normal appearing white matter outside the T2w MRI lesions indicating an early microstructural abnormality. In symptomatic patients, who have sustained strokes (in 10–15% of cases subclinical), T1 -weighted MRI and computer-assisted tomography (CT) disclose a variable number of small infarcts, which are most commonly

Vascular dementias

located in the white matter and deep grey matter (basal ganglia), whereas the cerebral cortex remains virtually intact (Figs. 13.11 and 13.12(a)). T2 -weighted gradient echo MRI reveals small hyperintensities corresponding to microbleeds, i.e. perivascular clusters of siderophages indicating focal extravasation of red blood cells, mainly in cerebral cortex or thalamus (Lesnik Oberstein et al., 2001; Dichgans et al., 2002). Microbleeds do not appear significant for the pathogenesis of CADASIL, since they occur outside the T2w MRI hyperintensities and also in other small vessel diseases of CNS, such as hypertension and amyloid angiopathy. Reduced cerebral blood flow (CBF) has been demonstrated in CADASIL with several functional imaging methods, e.g. with single photon emission computed tomography (SPECT) (Mellies et al., 1998), positron emission tomography (PET) (Chabriat et al., 1995a) and MRI bolus tracking method (Chabriat et al., 2000). It is already present in the white matter in presymptomatic subjects, whereas glucose consumption does not decrease until a later stage in parallel with tissue loss and development of dementia (Chabriat et al., 1995a, b; Tuominen et al., 2001). Furthermore the arteriopathy is reflected in the prolongation of arteriovenous cerebral transit time as well as decreased response to the vasodilatory drug acetazolamide (Chabriat et al., 2000). Genetics and Notch signalling The humanNotch3 gene has 33 exons and encodes Notch3 receptor protein of 2321 amino acids with a single transmembrane domain (Joutel et al., 1996). The structure of Notch3 is presented in Fig. 13.13. Notch3 gene/protein is one of the four members (in mammals) of the Notch family. Notch receptor molecules are the name-giving elements of the Notch signalling pathway, which is a widely used signalling pathway during animal development. Highly homologous Notch molecules exist in organisms ranging from nematodes to man (Artavanis-Tsakonas et al., 1999; Beatus & Lendahl, 1998). The Notch signalling pathway appears to transduce signals between neighbouring cells in immediate contact, since the ligands (Delta and Serrate type) binding to Notch receptors are considered to also be strictly cell bound. Notch receptor molecules (most likely also Notch3) undergo three regulated proteolytic cleavages (Fig. 13.13, S1–S3). The primary protein product is first constitutively cleaved by a furin-like convertase (S1). The cleavage products form a dimer, which is inserted in the plasma membrane. The next cleave (S2) is performed upon ligand binding by TNF-converting enzyme (TACE), also called metalloprotease ADAM-17, 12 amino acids outside of the plasma membrane – a similar process to APP cleavage by

-secretase (Kimberly et al., 2001). This cleavage releases

the extracellular domain to the extracellular space (this may be of central importance in the pathogenesis of CADASIL, see below). The third cleavage (S3) is performed within the plasma membrane by -secretase-like activity which appears to be the same as in the cleavage of A in AD and which is considered to be a complex of presenilin1 and nicastrin (Esler et al., 2002). It releases the Notch3 intracellular domain (NICD), which then translocates to the nucleus and binds a DNA binding protein called CSL (also called RBPJk) to regulate transcription of genes (Artavanis-Tsakonas et al., 1999; Beatus & Lendahl, 1998). This represents a novel signalling paradigm called regulated intramembrane proteolysis (RIP), which appears to be an important phenomenon generating proteins for nuclear signalling. During development the Notch pathway regulates tissue differentiation (Artavanis-Tsakonas et al., 1999; Beatus & Lendahl, 1998). However, the exact role of Notch3 in the development of vascular smooth muscle cells (VSMC) is unknown, as is the exact function of Notch3 in adult animals, in which Notch3 is expressed exclusively in VSMCs (Joutel et al., 2000). Recent data have provided evidence that Notch3 may be essential for preventing apoptosis of VSMCs (Wang et al., 2002). More than 95% of CADASIL cases to date are due to missense point mutations in the extracellular domain of

Fig. 13.11. Numerous, mainly lacunar infarcts are seen in the basal ganglia (asterisk, corresponding lacunar state) and in the white matter (arrowheads) in the coronal slice of the brain from a 62-year-old woman with CADASIL.

307

(c)

(a)

(d)

* *

(b)

(e)

Fig. 13.12. Microscopic changes in the brain parenchyma. (a). Cortex is well preserved, whereas the underlying white matter is severely damaged with cystic infarct (asterisk). (b) The wall of a small deep penetrating artery is noticeably thickened and fibrotic with accumulation of basophilic granular material. (c) In PAS staining the granular material accumulated in the arterial wall is intensely red. (d) The number of smooth muscle cells in the arterial wall are markedly reduced due to their severe degeneration. (e) Immunostaining reveals massive deposition of type I collagen. a and b: H&E, c: PAS, d. Anti--smooth muscle actin and haematoxylin counterstain, e: anti-type I collagen and haematoxylin counterstain.

Vascular dementias

Fig. 13.13. Schematic presentation of Notch3 structure and signaling (below) and, for comparison, A-PP structure and signalling, (above). The extracellular N-terminal part of the Notch3 molecule contains 34 (truncated in the picture) epidermal growth factor (EGF) type repeats followed by three notch/lin-12 repeats. On the cytoplasmic side there are six ankyrin repeats (30). Subsequent to translation the primary protein product is constitutively cleaved by furin like convertase (S1) and the subunits form a homodimer, which is inserted in the plasma membrane (PM). Upon ligand (Serrate and Delta) binding the extracellular (EC) part (ectodomain) is released at S2 by TNFalpha-converting enzyme (TACE) and the intracellular (IC) domain (NICD) is cleaved by the same -secretase as is A in Alzheimer’s disease, and translocated to the nucleus, where it regulates transcription bound to a DNA binding protein CSL (RBP-Jk). (Reproduced from Kalimo et al., 2002b.) NM:nuclear membrane.

Notch3. Over 90 different pathogenic point mutations have been identified. There is a marked clustering of mutations at the 5’ end of the Notch3 gene: in about 70% of patients the mutation is located within exons 3 and 4, which encode the first five EGF repeats (Joutel et al., 1997) (Fig. 13.13). All the point mutations result in amino acid substitutions, either a replacement of a cysteine with another amino acid or vice versa. Thus, the mutated EGF repeat contains an uneven number, either five or seven cysteines instead of the normal six cysteines. In addition, four different small deletions have been described, which cause a loss of either one or three cysteine residues and thereby also result in an uneven number of cysteines in the deleted EGF repeat (Kalimo et al., 2002b). The pattern of inheritance is dominant with 100% penetrance. Recently, a CADASIL patient homozygous for the relatively common R133C substitution was described (Tuominen et al., 2001). Although the homozygous male patient had early onset and severe clinical and imaging findings, the homozygosity was not considered to definitely aggravate symptoms, suggesting that CADASIL follows the classical definition of a dominant disease, according to which the heterozygous and homozygous patients are clin-

ically indistinguishable. While sporadic cases of CADASIL may occur these are likely to be rare. Several families with typical CADASIL features have failed to reveal any abnormalities in Notch3 suggesting that other inherited CADASIL-like disorders exist. Pathology Biopsy findings Although the symptoms of CADASIL are almost exclusively neurological, vascular changes are present in mediumsized and small arteries of almost all organs. This makes intra vitam diagnosis on the basis of the pathognomonic structural changes relatively easy, since a diagnostic biopsy can be taken from skin, and common neuropathological biopsies from muscle or peripheral nerve can also be used (Ruchoux et al., 1994; Schröder et al., 1995). In electron microscopic examination pathognomonic granular osmiophilic material (GOM; Fig. 13.14) can virtually always be detected between degenerating vascular smooth muscle cells (VSMCs) in the walls of deeply located dermal arterioles (Ruchoux et al., 1994). Recently a light microscopic method was introduced: the pathological accumulation of Notch3 extracellular domains in the vessel walls can be

309

310


Fig. 13.14. The same CADASIL patient as in Fig. 13.10. In electron microscopy smooth muscle cells in the wall of a small dermal artery are irregular in shape and between them within thickened basal lamina or in indentations of smooth muscle cells there are deposits of characteristic granular osmiophilic material (GOM, arrowheads).

demonstrated immunohistochemically (Fig. 13.15(a) and (b)) (Joutel et al., 2000, 2001). Both accumulation of GOM and degeneration of VSMCs appear to begin early being detectable in skin biopsies of CADASIL subjects before 20 years of age (Tuominen et al., 2001; Brulin et al., 2002). The pathognomonic GOM is located either in indentations of degenerating arterial VSMCs or free between these cells, often within the thickened basal lamina (Fig. 13.14; Ruchoux & Maurage, 1997; Kalimo et al., 2002b). There is no filamentous component in GOM. The composition of GOM has not yet been definitely identified, even though Notch3 protein was suspected to be one component. The pathogenetic significance of the small caveolae common in VSMCs close to GOM deposits is open. Post-mortem brain findings In accordance with the imaging, multiple small (lacunar) infarcts are found in the white matter or deep grey matter (Fig. 13.11), and also in the brain stem. In accordance with the CBF observations, the cerebral cortex is relatively

preserved (Fig. 13.12(a)). Unlike in -amyloid angiopathy (Kalaria, 2001; Kalimo et al., 2002b) intracerebral haemorrhages are uncommon, having occurred most often in patients treated with anticoagulants or antiaggregants, or subjected to arteriography. The walls of small and medium-sized leptomeningeal and penetrating arteries are markedly thickened. The degenerating tunica media contains characteristic granular material, which is basophilic in H&E staining, PAS-positive (Fig. 13.12(b) and (c)) and immunopositive for Notch3 ectodomains (Fig. 13.15(b)). Destruction of the smooth muscle cells can be demonstrated by immunostaining for smooth muscle -actin (Fig. 13.12(d)). The thickening of the arterial walls appears to be mainly due to accumulation of extracellular matrix proteins (Kalaria, 1996, 2001), including laminin and various types of collagen (Fig. 13.12(e)). The presence of the granular material (Fig. 13.12(b) and (c)) distinguishes CADASIL from the fibrotic vasculopathy occurring for example in arterial hypertension and Binswanger’s disease (Kalaria, 2001). The arteries in CADASIL

Vascular dementias

(a)

(b)

Fig. 13.15. Notch3 ectodomain is accumulated as granular material within arterial walls in a skin biopsy (a) from a CADASIL patient and (b) in the white matter of the brain of a deceased CADASIL patient. Sections stained with antibody against extracellular domain of Notch3 (kind gift from A. Joutel) with haematoxylin counterstain.

are completely negative for amyloid. Electron microscopy confirms the destruction of smooth muscle cells in the arterial walls and also discloses in greater detail the accumulation of pathognomonic GOM (cf. Fig. 13.14., see above).

Pathogenesis It has been suggested that the alteration from an even to uneven number of cysteines affects the formation of sulphur bridges and thereby the three-dimensional structure of the extracellular part of the Notch3 receptor molecule. Consequently, processing or trafficking of Notch3 molecules, binding ligands, disposal of the cleaved extracellular domain and other molecular interactions may be affected (Joutel et al., 1997; Kalaria, 2001). These alterations presumably lead to defective signalling in the VSMCs and cause their degeneration. Since CADASIL is an autosomally dom-

inantly inherited disease, both alternatives, haploinsufficiency (i.e. one wild-type allele of Notch3 not being sufficient to main normal cellular function) and toxic gain of function, are plausible. One explanation is that the pathological deposits of mutated Notch3 ectodomains on the surface of VSMCs ‘sops up ligand without transmitting signal’ (Spinner, 2000). On the other hand, it has been suggested that impaired Notch3 signalling could lead to apoptosis of VSMCs by causing reduced production of FLIP, an inhibitor of Fas ligand (Wang et al., 2002). How the arteriopathy causes CBF disturbance of such severity that infarction ensues is not known. The progressive thickening of arterial walls obviously reduces the arterial lumina and their compliance and consequently CBF as discussed above in the imaging section. The small cerebral arteries do become severely obliterated by the fibrotic process, but thrombosis of affected arteries seems to also occur, since fibrin degradation products can be detected in the plasma of CADASIL patients with recent infarcts (Ilveskero et al., unpublished observation). The tissue destruction leads to cognitive decline, which finally develops into dementia of subcortical vascular type. The predominant localization of infarcts to the white matter and deep grey matter can be explained by the fact that those areas of the brain are supplied by relatively long penetrating arteries of the end artery type without efficient collaterals. Furthermore, penetrating arteries become tortuous with age. In the cerebral cortex the density of arteries is greater and they are shorter. The walls of the cortical arteries are also significantly less thickened than in the deeper parenchyma (Miao et al., in preparation).

Diagnosis and differential diagnosis in CADASIL Diagnosis The first affected member in a CADASIL family most likely seeks for medical help either for a transient ischaemic attack or stroke at an exceptionally young age or for severe migraine. If examined by MRI, the characteristic, though not specifically diagnostic white matter changes should alert the radiologist. The appearance of similar symptoms in relatives supports the possibility of CADASIL. Positive MRI findings necessitate electron microscopic or immunohistochemical examination of skin biopsy, which are relatively easy to perform. Presence of GOM or immunopositivity for Notch3 ectodomains in the arterial walls can be considered definitely diagnostic of CADASIL, because GOM has not been found in any other disease (Ruchoux & Maurage, 1998). In many biopsies of suspected but not genetically verified patients various types of granular debris can be found between degenerative SMCs, but such

311

312


debris is different and can be distinguished from true GOM. If properly performed, skin biopsy is almost always diagnostic, but GOM deposits may be too focal or the biopsy not deep enough, in which cases repeating the biopsy may solve the problem. The identification of GOM and morphological diagnosis of CADASIL have been long made electron microscopically. Recently, the development of a monoclonal antibody to the extracellular domain of Notch3, which accumulates in the arterial wall, has provided a technically easier immunohistochemical method (Fig. 13.15) for morphological diagnosis is with high sensitivity (96%) and specificity (100%) (Joutel et al., 2001). Genetic screening for Notch3 mutations still provides the most definitive diagnosis. The large number of different mutations (over 90) makes the search tedious and expensive if the mutations are not found in exons 3 and 4 where 60% of the mutations appear and can be readily screened. Differential diagnosis Since all CADASIL cases diagnosed have so far been familial the presence of CADASIL in affected relatives of appropriate age would be implicated. However, a similar familial disease in which migraine is associated with strokelike symptoms is familial hemiplegic migraine (FHM). The defective gene CACNL1A4 in one type of FHM is located close to Notch3 in Chr 19p13, but the FHM patients usually recover from attacks. Stroke may also associate with ‘independent’ migraine, and migraine has been traditionally included among risk factors for stroke. Recurrent stroke of other etiologies as well as other vascular dementias must also be taken into consideration. Recurrent embolizations may cause multiple lacunar infarcts and they should be excluded by adequate examinations. White matter infarcts are a feature of Binswanger’s disease and a hereditary form of Binswanger’s disease has been described, although this disorder could be a CADASIL-like disease of unknown etiology (Gutiérrez-Molina et al., 1994). Hypertension is commonly associated with Binswanger’s disease, whereas CADASIL patients are usually normotensive (Chabriat et al., 1998; Kalaria et al., 2000). Yet, risk factors of cerebrovascular disease are not uncommon in CADASIL. Chabriat et al. (1998) identified at least one risk factor for stroke in 44% of their patients. Finally, strokes in young persons may be caused by mitochondrial encephalopathy, especially MELAS, but in these diseases infarcts are usually cortical and located in occipital lobes (Allard et al., 1988; Majamaa et al., 1998). Interestingly, a novel pathogenic mutation 5650G > A in the tRNAAla gene in mtDNA was reported in an R133C CADASIL patient (Majamaa et al., 1998) and in patients with migraine, mi-

tochondrial halpotype U is a risk factor for occipital stroke (Majamaa et al., 1998).

Other diseases associated with vascular dementia Hypertension and atherosclerosis The direct effects of hypertension are of necessity diffuse in that, when it is present, all vessels are exposed to hypertension. However, although all vessels are exposed to hypertension, the effects are not the same. This is made most obvious by considering the effects of hypertension on the short penetrating arteries that supply the cerebral cortex and the long penetrating arteries that pass through the cortex to supply the white matter. While the arterioles in the white matter can become markedly thickened and hyalinised and the white matter diffusely attenuated, the short penetrating arteries and the overlying cortex rarely show any changes. As has been described small vessel changes are also prominent in the deep grey nuclei of the basal ganglia and thalamus. By contrast to the diffuse and widespread effects of hypertension, atherosclerosis is almost always focal and exerts its influence locally either through stenosis, thrombosis or embolism. Probably the commonest site of vascular thrombosis secondary to atherosclerosis is the internal carotid artery at the bifurcation, and it is good practice always to examine the carotid bifurcations in patients who come to post-mortem examination. Thrombosis of the intracranial arteries is distinctly rarer, with the basilar artery being the most commonly affected.

Embolic disease Although atheroemboli are very important and probably the commonest type of embolus encountered in vascular dementia, the existence of cognitive decline secondary to strategic single infarcts means that other sources of emboli can certainly be associated with cognitive decline and that cognitive decline does not invariably occur in the context of widespread cerebrovascular disease.

Vasculitides Vasculitis occurs in a number of clinical forms which have varying effects on the nervous system (Moore & Cupps, 1983). The principal clinically defined varieties of vasculitis are the polyarteritis nodosa group, hypersensitivity angiitis, Wegener’s granulomatosis, primary angiitis of the nervous

Vascular dementias

system, giant cell arteritis, and Bechet’s disease (Nishino et al., 1993; Ferro, 1998; Kalimo et al., 2002c). These different entities affect the nervous system to different degrees and in different ways, but problems with higher nervous system function and cognitive decline are one of the ways that vasculitis affecting the nervous system can become manifest. With the possible exception of hypersensitivity angiitis, it is probably true to say that any of the major groups of arteritis mentioned above can present with generalised encephalopathy, memory loss and behavioural changes that could fall within the category of dementia. It is certainly reasonable to add that such general features are frequently accompanied by other focal symptoms and signs such as cranial neuropathies and focal cerebral defects that suggest a multifocal process. Since they are also systemic diseases, effects on other organs systems are usually present and may dominate the clinical picture. Further to distinguish these conditions from atherosclerotic vascular disease, the timescale of the process is almost invariably compressed and spread over weeks to months rather than the months to years that elapse for atherosclerotic disease. Primary angiitis of the central nervous system The most challenging situation is perhaps that which occurs in primary angiitis of the central nervous system (PACNS; previously called isolated or granulomatous angiitis) when systemic effects are not present (Calabrese, 1995; Rollnik et al., 2001). In this condition there is focal inflammation and necrosis in the walls of leptomeningeal and parenchymal arteries. The inflammation is typically lymphoplasmacytic with macrophages and, in most cases, also giant cells and formation of granulomas; hence the alternative name of granulomatous angiitis of the CNS. The condition may present with confusion and intellectual deterioration although this is nearly always accompanied by headache. Although sometimes absent early in the course, focal neurological signs usually occur at some time. In most cases granulomatous angiitis is subacutely progressive over weeks to months but has been recorded to ‘grumble’ on for years. The pathologist becomes involved in these cases because they usually produce considerable diagnostic uncertainty and even when angiography demonstrates segmental narrowing and beading of affecting arteries there is sometimes a need to confirm this pathologically. There have also been cases where angiography has been normal but biopsy has shown the presence of arteritis in the smaller vessels not visualized by radiographical techniques (Moore, 1989). In most cases only leptomeningeal and parenchymal biopsy provide the definite diagnosis of cere-

bral angiitis (Harrison, 1976; Jellinger, 1977; Zuber et al., 1999). Giant cell (temporal) arteritis This arteritis affects arteries in the distribution of the carotid and, less frequently, vertebral arteries (Caselli & Hunder, 1997). Visual involvement is the most common manifestation and CNS abnormalities are relatively uncommon. Bilateral carotid involvement has been reported to produce global mental deterioration (Howard et al., 1984; Inafuku et al., 1998). Buerger’s disease Another condition within the spectrum of vasculitis that can present with dementia is the cerebral form of thromboangiitis obliterans (Buerger’s disease) (Lie, 1988). This condition, as is widely recognized, is almost invariably associated with a history of cigarette smoking, but the precise nature of the etiologic connection between the two remains elusive. Although rare in Europe and North America, thromboangiitis obliterans is common in other parts of the world, notably Japan, India and Israel. In most cases, the disease is generalized and thus occasionally affects the brain, but Lindenberg and Spatz (1939) described a cerebral form of the disease that they divided into types I and II which tended to affect large basal arteries (type I) or distal branches (type II) (Larner et al., 1999). Both these variants were associated with the presence of vascular dementia, although it was reported to be more common in the type II disease. In a report of what has been called Spatz– Lindenberg disease (SLD) Zhan et al. (1993) attempted to compare synaptophysin loss in SLD with that in Alzheimer’s disease and showed similar degrees of loss, although the significance and interpretation of this finding was not clear.

Collagen vascular disease Systemic lupus erythematosis (SLE) SLE is the collagen vascular disease with the most frequent involvement of the central nervous system. Although neurological disease is rarely a presenting sign, there is a high frequency of involvement at some time during the course of the disease, estimates varying between 10 and 75% of patients in different series (Lim et al., 1988; Bluestein, 1987), with neurological symptoms being more frequent in the later and more severe stages (Johnson & Richardson, 1968). SLE has a range of effects on the nervous system, some of which, such as the cranial neuropathies, cerebrovascular accidents and chorea reflect focal disease. Given the role of single strategic infarction in vascular

313

314


dementia even focal disease can produce cognitive decline although this is not common (Mitsias & Levine 1994). Diffuse effects such as confusional states, and acute delirium, delusions and other more generalized disturbances of cerebral function are quite common and dementia has also been described. Although vasculitis has always been regarded as one of the major pathologic mechanisms in SLE, the extent of the identifiable vasculitic changes in the brain has always been much less than would be expected from the severity of the symptoms and signs (Johnson & Richardson, 1968). Occasional examples of a conventional acute vasculitis, with an acute inflammatory reaction within the arterial wall, sometimes with a giant cell reaction, and necrosis of the vessel wall, are seen in SLE, but this is quite a rare finding. The more frequent observation is a ‘vasculopathy’ with hyaline thickening of the wall of the blood vessel, fibrinoid degeneration and endothelial proliferation (Ellis & Verity, 1979; Devinsky et al., 1988; Hanley et al., 1992; Scolding & Joseph, 2002) (Fig. 13.16). The aetiology of this vascular lesion is not clear but Ellison et al. (1993) have shown the presence of platelet-derived material within the thickened wall of these abnormal vessels. The disparity of the relative paucity of pathological findings in patients with rather prominent, and often fluctuating, clinical findings has precipitated a search for other pathophysiologic mechanisms by which the function of the central nervous system could be compromised (Bluestein, 1987; Mitchell et al., 1994). One process that has been heavily canvassed as a probable mechanism underlying the ephemeral neuropsychiatric symp-

toms is an immune-complex mediated ‘vasculitis’ associated with complement activation. This process could be very short lasting and produce tissue ischaemia short of infarction. The vascular endothelial damage attendant on ischaemia has been shown to be associated with platelet deposition in small vessels (Jafar et al., 1989). The presence of antiphospholipid antibodies has also been associated with cerebral infarction in SLE (Asherson & Lubbe, 1988; Asherson et al., 1989), and this could be a further mechanism by which tissue ischaemia could develop in the brain. Another more diffuse mechanism that has been suggested to have a role in the production of the neuropsychiatric symptoms of SLE is antineuronal antibodies, which are hypothesized to gain access to the cerebral parenchyma as a consequence of breakdown of the blood–brain barrier following the immune mediated vasculitis. Hanson et al. (1992) reported the presence of antibodies to a 50 kD neuronal membrane protein in 19 of 20 patients with lupus and CNS involvement but in only 8 of 23 where there was no CNS involvement and 1 of 34 patients with other neurological diseases. It is quite plausible to imagine that access to the brain by an antibody to a widely expressed neuronal protein, particularly if the protein was involved in synaptic transmission, could produce widespread and variable effects on CNS function. Nevertheless, as emphasized by Mitchell et al. (1994) the pathogenesis of cerebral lupus is likely to be multifactorial. Sneddon’s syndrome Sneddon’s syndrome is the idiopathic coupling of widespread livedo reticularis with a liability to ischaemic stroke (Francis & Piette, 2000) with features in common with both SLE and antiphospholipid syndrome. In addition to migraine and seizures it has also been associated with a dementia of both multi-infarct type (Antoine et al., 1994) and cognitive decline without stepwise progression (Wright & Kokmen, 1999; Adair et al., 2001). Accounts of neuropathological findings in Sneddon’s syndrome are inconsistent with findings of a lupus like vasculopathy (Geschwind et al., 1995), possible association with a meningeal granulomatous process (Boortz-Marx et al., 1995) and non-specific changes (Devuyst et al., 1996).

Illicit drug related diseases Fig. 13.16. Systemic lupus erythematosus (SLE). The vascular changes in the vasculopathy of SLE are concentrated in the intima, with, as in this illustration, relative sparing of the media. In arteritides such as polyarteritis nodosa, the principal focus of the inflammatory infiltrate and tissue damage is in the muscular media of the artery. (H&E x 125)

Cerebral vasculitis can either be stimulated or mimicked by both cocaine and amphetamines. Cocaine use is associated with an increased frequency of transient ischaemic attacks, cerebral infarcts and haemorrhages which may be subarachnoid or intraparenchymal (Levine et al., 1990). The strokes and intraparenchymal haemorrhages, while

Vascular dementias

usually producing focal neurological symptoms may, if they affect appropriate regions of the brain, also cause cognitive defects. A prominent example would be thalamomesencephalic infarcts (Rowley et al., 1989) that, depending on the precise thalamic structures involved (Graff-Redford et al., 1985) may have a wide variety of neuropsychological and cognitive problems. Although the vast majority of lesions associated with cocaine abuse are probably manifestations of the acute vasospastic effects of cocaine, there are a few reports where cocaine seems to be associated with a histologically demonstrated vasculitis (Krendal et al., 1990; Morrow & McQuillen, 1993). A vasculitis with vessel wall necrosis has also been described in association with amphetamine abuse (Citron et al., 1970).

Tumour-related vascular disease Lymphomatoid granulomatosis This condition was first described by Liebow et al. (1972) and is a lymphoreticular proliferative disorder with an angiocentric and angiodestructive polymorphic cellular infiltrate. In a significant minority of cases the condition evolves into or is associated with a non-Hodgkin’s lymphoma. Although the lung is usually the primary site of disease the CNS is involved in between a fifth and a quarter of the cases and neurological symptoms may be the mode of presentation (Katzenstein et al., 1979). In a very small number of cases the disease is clinically restricted to the nervous system (Schmidt et al., 1984; Kerr et al., 1987.) although there may be pathological evidence of disease in the lungs. Within the brain the lesions may be multiple and show mass effect (Simon et al., 1981). Histologically, there is often necrosis and haemorrhage with parenchymal involvement by a markedly angiocentric and angiodestructive mixed inflammatory infiltrate composed of lymphoctes (often with marked nuclear pleomorphism and irregularity), histiocytes, plasma cells and plasmacytoid lymphocytes together with other bizarre mononuclear cells (Anders et al., 1989; Ironside et al., 1984). The vasculopathic nature of the condition means that its nervous system presentation very much resembles that of a cerebral arteritis with a wide spectrum of neurological abnormalities, including cognitive decline. Angiotropic (intravascular) large cell lymphoma This is a rare and generally fatal disease characterised by growth of malignant B-lymphocytes within the lumens of small vessels that affects predominantly the skin and central nervous system. This condition has had a number of different names including neoplastic angioendotheliomatosis – a portmanteau term that reflects an earlier confusion

over the nature of the neoplastic cells (Sheibani et al., 1986). In more modern publications it is more correctly called angiotropic large cell lymphoma or malignant intravascular lymphomatosis. How these patients present depends on the location of the disease, but if the principal burden of disease is in the cerebral hemispheres, it can take the form of a subacutely progressive dementia that may mimic a vasculitis (Reinglass et al., 1977; Petito et al., 1978; LeWitt et al., 1983; Drlicek et al., 1991; Treves et al., 1995; Beristain and Azzarelli, 2002). As might be suspected from the nature of the underlying condition, survival tends to be measured in months rather than years.

Miscellaneous diseases As a pendant to this list of unusual vascular causes of cognitive decline and dementia, there are, as might be anticipated, individual case reports of, probably, unique aetiologies for vascular dementia including watershed infarction (Hashiguchi et al., 2000), venous hypertensive encephalopathy secondary to dural arterio-venous fistulas (Hurst et al., 1998) and a familial syndrome associated with occipital calcifications, haemorrhagic strokes and dementia (Inglesias et al., 2000). A final rare cause of vascular dementia (Gil-Nagel et al., 1995; Kariya et al., 2000; Kageyama et al., 2000) and also encountered by one of us, is illustrated in Fig. 13.17. The dementia in this case was the result of a very large number of cavernous angiomas. The multiple and recurrent haemorrhages from the angiomas produced, along with focal symptoms and signs, widespread cerebral damage and a progressive cognitive decline the cause of which, in the pre-MRI era, was obscure until post-mortem examination.

Pathology of vascular dementia Multi infarct and single strategic infarct dementia The pathology of infarcts causing dementia is the same as any brain infarcts. Since the dementia has usually developed over a longer time period most lesions are old, but in multi-infarct dementia infarcts of various ages can be detected. In both larger and smaller complete infarctions the tissue destruction occurs similarly and it follows the common sequence of events described in cerebral infarctions (Kalimo et al., 2002a). The necrotic tissue is invaded within a day by neutrophil leukocytes, during the next days followed by macrophages, the length of whose presence depends on the size of the infarct, i.e. the amount of cell debris to be removed and it may be many months. Depending on

315

316


macrophages and cholesterol crystals. In complicated unstable plaques the fibrous cap is usually thinner and inflammatory cells accumulate in the plaques, especially in those which are ulcerated. There may be inflammatory changes indicating either an immune or infectious aetiology (Fieschi et al., 1998; see above, ‘Vasculitides’) or there may be accumulation of specific foreign substances, e.g. amyloid (see Chapter 14).

Small vessel disease

Fig. 13.17. An unusual cause of vascular dementia: this patient, who developed a slowly progressive dementia, with other focal neurological signs, had the syndrome of multiple cavernous angiomas.

Fig. 13.18. Microscopic examination of the same case as in Fig. 13.6. showed, in addition to distinct type 1a lacunes, scattered areas of non-cavitated focal ischaemic damage (lacune type 1b). The neuropil around the focal lesion is well preserved. (H&E × 50)

the volume of the infarcted tissue a cavity of corresponding size will gradually develop. In haemorrhagic infarcts a variable number of siderophages may be found in the lining of the cystic cavity. Astrocytic reaction, often of gemistocytic type, around the forming cavity begins within a week and remains for ever. Understandably emboli having occluded the larger vessels are rarely found, but other changes in the vessel walls may reveal the cause of the infarct. Atherosclerosis has the characteristic appearance of plaques covered by a fibrous cap under which within the necrotic core there are lipid

Small vessel disease is usually a diffuse vascular disease: its effects may be most obvious in one particular region, but small vessels throughout the brain are commonly affected and the pathological changes are manifested as a continuum of progressively more severe damage. It also needs to be emphasized here that the two different pathological expressions of small vessel disease, small vessel occlusions that produce lacunes, and more diffuse damage, often coexist. The brunt of the damage is borne by the WM and the deep grey nuclei, whereas the cerebral cortex despite its sensitivity to ischaemia, shows little if any damage in diffuse small vessel disease when examined with the normal neuropathological techniques. Lacunar infarcts The work needed to identify the occluded vessel in lacunar infarcts is seldom done (Fisher, 1991), but microscopic analysis of small arteries may reveal above mentioned fibrinoid necrosis/lipohyalinosis, significant arteriosclerosis with fibrosis of the vessel walls or diagnostic deposits of foreign substances (e.g. granular material of CADASIL, see above). Since the lesions produced by the small vessel disease very rarely cause a patient’s death, most lesions are old or subacute. The tissue lesions in incomplete lacunar infarcts (type 1b) are characterized by loss of selectively vulnerable cellular elements, falling short of pan-necrosis (Fig. 13.18.). In deep grey matter these lacunes appear as perivascular rarefactions with loss of neurons, some oligodendroglia and patchy astrogliosis (Lammie, 2002). Complete lacunar infarcts are better circumscribed, small cavities containing one or a few blood vessels and scattered macrophages and in the surrounding tissue there is reactive astrogliosis of variable degree (Fig. 13.12(a). and 13.19.). Pathology of diffuse small vessel lesions The major morphological changes in diffuse small vessel disease can be described under four headings: (a) arterial wall changes. (b) expansion of the perivascular spaces, (c) perivascular parenchymal rarefaction and gliosis. (d)

Vascular dementias

Fig. 13.19. There is a small lacunar infarct (lacune type 1a) on the left and expansion of the perivascular space on the right The arterial walls are thickened and fibrotic. Van Gieson staining.

diffuse WM rarefaction (subcortical leukoencephalopathy/Binswanger change). In general, these changes can be thought of as sequential and additive, so that, for example, the arterial wall changes can be seen in the absence of any of the other morphologic features, but if there is perivascular parenchymal rarefaction and gliosis, there will also be expansion of the Perivascular spaces and arterial wall changes.

(a) Blood vessel wall changes In diffuse small vessel disease, as already indicated, the blood vessels of interest are the smaller arteries and arterioles. Large vessel cerebral atherosclerosis appears to have no specific relationship to diffuse small vessel disease of the WM. The small vessels particularly affected are the long penetrating arteries that supply the hemispheric WM and the penetrating arteries supplying the deep grey nuclei (basal ganglia and thalamus) and their branches. In the WM the

characteristic change is a hyaline thickening of their walls with loss of the normal smooth muscle cell population of the vessel wall (Figs. 13.19. and 13.20.). The thickening can be very marked, constituting 75% of the total diameter of the arteriole. It can also be seen in the absence of any other change and in patients with a history of hypertension but no evidence of cognitive decline. ´ (b) Expansion of the perivascular spaces (L’Etat Criblé) The expansion of the perivascular space reflects both changes in the arteries and tissue damage with consequent reduction in parenchymal volume. It can be seen in the WM, but it is most conspicuous in the deep grey nuclei. reflecting the fact that the arteries in the deep grey nuclei are larger (type III lacunes, see above) with more conspicuous perivascular space than in the WM. The putative pathophysiological mechanism of expansion of the perivascular spaces is the spiralling of the artery that occurs in hypertension, and that is why the artery is most often seen to one side of the enlarged perivascular space (Fig. 13.20).

317

318


Fig. 13.20. In mild small vessel disease the perivascular spaces in the white matter are asymmetrically expanded around the vessels.

(c) Perivascular parenchymal rarefaction and gliosis This change consists of a loosening, vacuolation and gliosis of the neuropil surrounding the small arteries and arterioles both white and grey matter. Neurones do not appear to be particularly susceptible to this process and may still be present in affected regions. There are considerable variations in severity in different brain regions the globus pallidus usually showing the most severe changes, while the thalamus shows less severe changes and the caudate/putamen is the least dramatically affected. This difference probably reflects the different structure of the neuropil in the various nuclei. (d) Diffuse white matter rarefaction (subcortical leukoencephalopathy/Binswanger change) As suggested above, the balance of opinion seems now to be swinging towards the view that neither the clinical nor the neuropathological criteria are sufficiently secure or distinctive to regard Binswanger’s disease as anything other than a particular expression of small vessel hypertensive cerebrovascular disease, for which the descriptive terms subcortical (arteriosclerotic) leukoencephalopathy or chronic microvascular leukoencephalopathy have been recommended (Olszewski, 1962, Pantoni & Garcia, 1995; Caplan, 1995; Loeb, 2000). The external appearance of the brain in cases of subcortical leukoencephalopathy is not usually at all impressive. The brain weight tends to be well maintained, and there is no marked gyral atrophy. There is also no particular tendency for large vessel atherosclerosis. Fisher (1989) reported that large vessel atherosclerosis was absent or virtually absent in 22% of the cases. On brain slices, the most marked feature is usually dilatation of the lateral and third

ventricles which were moderately or greatly enlarged in 81% of cases (Fig. 13.21(a)). Whole mount sections stained for myelin show marked fairly symmetric pallor (Fig. 13.21(a)). The pallor is variable in degree, most severe in the deeper regions of the WM in frontal and parietal planes and periventricularly, often with perivascular accentuation and sometimes with small foci of tissue lysis. More peripherally, the pallor becomes less intense and more focal. In regions of more compacted myelin, the internal capsule tends to be spared, and the corpus callosum, although reduced in thickness shows less severe pallor than the periventicular WM. The WM of the cerebellar hemispheres is relatively spared. Microscopic examination of the affected WM shows widespread but variable, in some places almost total, pruning of the parenchymal elements of the white matter (Fig. 13.21(b)). The loss of myelin is usually greater than the axonal pruning, described by Fisher (1989) as ‘diffuse incomplete demyelination’, There is a reduction in the number of oligodendrocytes and an accompanying reactive astrocytosis. Surprisingly, in view of the degree of tissue loss, macrophages are often not very conspicuous, though in the regions of recent tissue destruction, foamy macrophages are present and within the more diffusely damaged areas, scattered macrophages may also be found in the perivascular spaces. The subcortical U-fibres tend to be spared. The blood vessel changes are very much those previously described above with marked hyaline thickening of the long penetrating arteries, but vascular occlusion is notably rare. Recent studies of the medullary blood vessels in ‘Binswanger’s disease’ (Zhang et al., 1994; Zhang & Olsson, 1997; Tanoi et al., 2000; Lin et al., 2000) have demonstrated markedly increased collagen of several types in the media and adventitia together with segmental loss of the smooth muscle cells and disruption and reduplication of the internal elastic lamina. There was also evidence of smooth muscle cell proliferation in the terminal arterioles. Despite these major changes in the media and adventitia, the endothelial cell layer of the blood vessels was surprisingly well preserved.

Neuropathological criteria of vascular dementia and assessment of brains As noted in the opening paragraphs of this chapter, one of the big three questions in dementia associated with cerebrovascular disease is ‘what are the neuropathological criteria for the diagnosis of vascular dementia?’ At present, although the matter has been extensively discussed, there are no accepted neuropathological criteria on which to

Vascular dementias

(a)

(b)

Fig. 13.21. (a) Subcortical leukoencephalopathy (Binswanger’s disease). In the whole mount section there is severe attenuation of stainable myelin in the hemispheric white matter of fronto-parietal lobe. The overlying cortex is well preserved and there are no focal infarcts. The severe attenuation of the corpus callosum and dilation of the lateral ventricle are noteworthy and they reflect the axonal damage that has occurred. Damage of this severity could be expected to produce significant disconnection between the deep grey nuclei and the cortex. (b) Myelin stain of the white matter shows only scattered preserved myelinated axons with most marked damage around the blood vessel in the centre of the picture. Luxol fast blue staining. × 50.

make this diagnosis. The inability of the neuropathologic community to produce diagnostic criteria is not solely the result of insouciance since the problems of arriving at a generally agreed set of criteria are not inconsiderable. So far, correlative clinico-pathological studies are few and in addition the number of patients examined is still relatively low (Gold et al., 2002). Proposed neuropathologic assessment scheme For the neuropathologist, the challenge is to find a reasonably concise yet meaningful way to describe the distribution and severity of cerebrovascular disease that is useful for correlative clinico-pathological studies of dementia. The essential components of any workable assessment scheme for cerebrovascular disease are that it should be: (i) Based on clear morphological criteria. (ii) Simple to perform (iii) Repeatable. At a practical level it requires:

(i) Description of the vascular morbid anatomy: particularly the presence, distribution and severity of atherosclerosis, fibrosis of the vessels, deposition of specific substances such as amyloid or the basophilic granular substance of CADASIL, as well as of other vascular processes. (ii) Description of the focal lesions (infarcts and lacunes) in terms of the vascular territory(ies), number, and volume of tissue (including the major anatomical structures involved). (iii) Description of the diffuse non-occlusive small vessel disease. Both the description of the vascular morbid anatomy and the focal lesions can be accomplished by gross examination of the brain, while description of the diffuse non-occlusive small vessel disease requires microscopic examination of selected regions of the brain. CERAD, although principally concerned with the problem of AD, have produced a protocol that is perfectly

319

320


adequate to describe the vascular morbid anatomy and focal lesions that are detectable on gross examination. Encouragingly, in the latest revision of the CERAD guide the Consortium is beginning to approach methods of describing small vessel disease by asking for a record of WM pallor and Binswanger change on microscopic examination of specific WM regions. What is proposed here is an extension of these recommendations. In diffuse non-occlusive cerebrovascular disease the problem is to define the locations within the brain that are most consistently affected and which therefore need to be sampled, the morphologic features to be recorded and a severity rating scale for the features selected. Selection of blocks for microscopic examination The most generally affected regions of white and grey matter are: White matter Centrum semiovale – at level of striatum Parietal lobe – at level of pulvinar Occipital lobe – at occipital pole of lateral ventricle. Grey matter Basal ganglia Thalamus Pons Basis pontis In the assessment of dementia associated with cerebrovascular disease, blocks from these regions should always be submitted together with any other relevant regions. A general scheme of block selection for the brain in dementia was outlined in Chapter 3. Morphological features The major morphological features of diffuse non-occlusive small vessel disease have already been described on pp. 317–18. They differ somewhat in grey and white matter. In grey matter, in terms of progressive severity they are: Arteriolar sclerosis (Vascular thickening), Expansion of the perivascular spaces Perivascular gliosis and rarefaction. While in white matter the changes are: Arteriolar sclerosis Expansion of the perivascular spaces Diffuse myelin pallor and gliosis (Binswanger change) Severity rating scale As has been discussed, a semiquantitive scale of severity is required for each of the morphological features that we

have suggested as possibly significant. Although imperfect, probably the most effective way of defining the various grades of severity in a semiquantitive scheme such as this is to provide oneself with representative illustrations of each grade of severity. A possible assessment pro forma for recording the aspects of cerebrovascular disease that have been mentioned is provided on page 328–329. It is the nature of these protocols that they teeter unsteadily along a fine line between over- and under-specification. Upon which side they are considered to err is entirely governed by the level of interest of the user; what is absurd oversimplification to one person is gross information overload to another. We make no advance apologies for our suggestions, but if they seem over-prescriptive it is because specific pathological information in this area is hard to come by. Laudable as these proposals may be, they do not address the final component of a diagnostic protocol which must be verification of the diagnostic significance of the pathological changes that are recorded. The pathological changes that we have suggested for recording seem to us to include at least some that are likely to be diagnostically significant and some that are not, but this remains to be demonstrated. Pressed for our own opinion, we think it likely that in brains without concomitant Alzheimer change and where the bulk of the disease is of the diffuse type, changes restricted to the arteriolar sclerosis and perivascular space expansion are not likely to be diagnostically significant, however severe. At the other end of the spectrum, patients with severe white matter pallor and gliosis are very likely to be symptomatic. A major uncertainty is the significance of perivascular gliosis and parenchymal attenuation in the deep grey nuclei, and particularly the thalamus, which can be very prominent in some patients.

REFERENCES Adair, J. C., Digre, K. B., Swanda, R. M. et al. (2001). Sneddon’s syndrome: a cause of cognitive decline in young adults. Neuropsychiatry Neuropsychol Behav Neurol, 14, 197–204. Akatsu, H., Takahashi, M., Matsukawa, N. et al. (2002). Subtype analysis of neuropathologically diagnosed patients in a Japanese geriatric hospital. J Neurol Sci, 196, 63–9. Alafuzoff, I., Iqbal, K., Friden, H., Adolfsson, R. & Winblad, B. (1987). Histopathological criteria for progressive dementia disorders: clinical-pathological correlation and classification by multivariate data analysis. Acta Neuropathol (Berl), 74, 209–25. Allard, J. C., Tilak, S. & Carter, A. P. (1988). CT and MR of MELAS syndrome. Am J Neuroradiol, 9, 1234–8. Amberla, K., Wäljas, M., Tuominen, S. et al. (2002). Insidious cognitive decline in CADASIL. Submitted.

Vascular dementias

Anders, K. H., Harrison, L., Change, B. S. et al. (1989). Lymphomatoid granulomatosis and malignant lymphoma of the central nervous system in the acquired immunodeficiency syndrome. Hum Pathol, 20, 326–34. Andersen, K., Launer, L. J., Dewey, M. E. et al. (1999). Gender differences in the incidence of AD and vascular dementia: The EURODEM Studies. EURODEM Incidence Research Group. Neurology, 53, 1992–7. Andersen, K., Lolk, A., Nielsen, H. & Kragh-Sorensen, P. (2000). [Prevalence and incidence of dementia in Denmark. The Odense study] Ugeskr Laeger, 162, 4386–90. Antoine, J. C., Michel, D., Garnier, P. et al. (1994). Sneddon syndrome: 9 cases. Rev Neurol (Paris), 150, 435–43. Artavanis-Tsakonas, S., Rand, M. D. & Lake, R. J. (1999). Notch signaling: cell fate control and signal integration in development. Science, 284, 770–6. Asherson, R. A. & Lubbe, W. F. (1988). Cerebral and valve lesions in SLE: association with antiphospholipid antibodies. J Rheumatol, 15, 539–43. Asherson, R. A., Khamashta, M. A., Gil, A. et al. (1989). Cerebrovascular disease and antiphospholipid antibodies in systemic lupus erythematosus, lupus-like disease and the primary phospholipid syndrome. Am J Med, 86, 391–9. Astrup, J., Siesjö. B. K., Symon, L. (1981). Thresholds in cerebral ischemia – the ischemic penumbra. Stroke, 12, 723–5. Auer, R. N. & Sutherland, G. R. (2002). Hypoxia and related conditions. In: Greenfield’s Neuropathology, ed. D. I. Graham and P. L. Lantos, 7th edn, pp. 233–280, London: Arnold. Barclay, L., Zemcov, A., Blass, J. P. et al. (1985). Survival in Alzheimer’s disease and vascular dementia. Neurology, 35, 834–40. Barker, W. W., Luis, C. A., Kashuba, A. et al. (2002). Relative frequencies of Alzheimer disease, lewy body, vascular and frontotemporal dementia, and hippocampal sclerosis in the state of Florida brain bank. Alzheimer Dis Assoc Disord, 16, 203–12. Baron, J. C. (1999). Mapping the ischaemic penumbra with PET: implications for acute stroke treatment. Cerebrovasc Dis, 9, 193–201. (2001). Perfusion thresholds in human cerebral ischemia: historical perspective and therapeutic implications. Cerebrovasc Dis, 11, Suppl 1, 2–8. Beatus, P., Lendahl, U. (1998). Notch and neurogenesis. J Neurosci Res, 54, 125–36. Beristain, X. & Azzarelli, B. (2002). The neurological masquerade of intravascular lymphomatosis. Arch Neurol, 59, 439–43. Bertsch, T., Mielke, O., Holy, S. et al. (2001). Homocysteine in cerebrovascular disease: an independent risk factor for subcortical vascular encephalopathy. Clin Chem Lab Med, 39, 721–4. Bluestein, H. G. (1987). Neuropsychiatric manifestations of systemic lupus erythematosus. N Engl J Med, 317, 309–11. Boller, F., Lopez, O. L. & Moossy, J. (1989). Diagnosis of dementia: clinicopathologic correlations. Neurology, 38, 76–9. Boone, K. B., Miller, B. L., Lesser, I. M. et al. (1992). Neuropsychological correlates of white-matter lesions in healthy elderly subjects. A threshold effect. Arch Neurol, 49, 549–54.

Boortz-Marx, R. L., Clark, H. B., Taylor, S. et al. (1995). Sneddon’s syndrome with granulomatous leptomeningeal infiltration. Stroke, 26, 492–5. Bowler, J. V. (2002). The concept of vascular cognitive impairment. J Neurol Sci, 203–204, 11–15. Bowler, J. V. & Hachinski, V. (1995). Vascular cognitive impairment: a new approach to vascular dementia. Baillieres Clin Neurol, 4, 357–76. Brulin, P., Godfrain, C., Leteurtre, E. & Ruchoux, M. M. (2002). Morphometric analysis of ultrastrcutural vascular changes in CADASIL: analysis of 50 skin biopsy specimens and pathogenic implications. Acta Neuropathol (Berl), 104, 241–8. Brun, A. (1994). Pathology and pathophysiology of cerebrovascular dementia: pure subgroups of obstructive and hypoperfusive etiology. Dementia, 5, 145–7. Calabrese, L. H. (1995). Vasculitis of the central nervous system. Rheum Dis Clin North Am, 21, 1059–76. Caplan, L. R. (1995). Binswanger’s disease–revisited. Neurology, 45, 626–33. Caselli, R. J. & Hunder, G. G. (1997). Giant cell (temporal) arteritis. Neurol Clin, 15, 893–902. Chabriat, H., Vahedi, K., Iba-Zizen, M. et al. (1995a). Clinical spectrum of CADASIL: a study of 7 families. Lancet, 346, 934–9. Chabriat, H., Bousser, M.-G. & Pappata, S. (1995b). Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy: a positron emission tomography study in two affected family members. Stroke, 26, 1729–30. Chabriat, H., Levy, C., Taillia, H. et al. (1998). Patterns of MRI lesions in CADASIL. Neurology, 51, 452–7. Chabriat, H., Pappata, S., Poupon, C. et al. (1999). Clinical severity in CADASIL related to ultrastructural damage in white matter: in vivo study with diffusion tensor MRI. Stroke, 30, 2637–43. Chabriat, H., Pappata, S., Ostergaard, L. et al. (2000). Cerebral hemodynamics in CADASIL before and after acetazolamide challenge assessed with MRI bolus tracking. Stroke, 31, 1904–12. Chui, H. C. (2000). Vascular dementia, a new beginning: shifting focus from clinical phenotype to ischemic brain injury. Neurol, Clin, 18, 951–77. Chui, H. C., Victoroff, J. I., Margolin, D. et al. (1992). Criteria for the diagnosis of ischaemic vascular dementia proposed by the State of California Alzheimer’s disease diagnostic and treatment centers, Neurology, 42, 473–80. Citron, B. P., Halpern, M., McCarron, M. et al. (1970). Necrotising angiitis associated with drug abuse. N Engl J Med, 283, 1003–11. Clarke, R., Joachim, C., Esiri, M. et al. (2000). Leukoaraiosis at presentation and disease progression during follow-up in histologically confirmed cases of dementia. Ann N Y Acad Sci, 90, 497–500. Cohen, R. A., Paul, R. H., Ott, B. R. et al. (2002). The relationship of subcortical MRI hyperintensities and brain volume to cognitive function in vascular dementia. J Int Neuropsychol Soc, 8, 743–52. Collins, R., Armintage, J. (2002). High risk elderly patients PROSPER from cholesterol lowering therapy. Lancet, 360, 1618–19.

321

322


Corsellis, J. A. N. (1962). Mental Illness and the Aging Brain. The Distribution of Pathological Change in a Mental Hospital Population. London: Oxford University Press. Cupps, T. R., Moore, P. M. & Fauci, A. S. (1983). Isolated angiitis of the central nervous system. Prospective diagnostic and therapeutic experience. Am J Med, 74, 97–105. del Ser, T., Bermejo, F., Portera, A., Arredondo, J. M., Bouras, C. & Constantinidis, J. (1990). Vascular dementia. A clinicopathological study. J Neurol Sci, 96, 1–17. De Groot, J. C., de Leeuw, F. E., Oudkerk, M. et al. (2000). Cerebral white matter lesions and cognitive function: the Rotterdam Scan Study. Ann Neurol, 47, 145–51. De Groot, J. C., de Leeuw, F. E., Oudkerk, M. et al. (2002). Periventricular cerebral white matter lesions predict rate of cognitive decline. Ann Neurol, 52, 335–41. De Reuck, J. (1971). The human periventricular arterial blood supply and the anatomy of cerebral infarctions. Eur Neurol, 5, 321–34. De Reuck, J. (1996). Evidence for chronic ischaemia in the pathogenesis of vascular dementia: from NeuroPATH to neuroPET. Acta Neurol Belg, 96, 228–31. De Reuck, J., Decoo, D., Marcheau, M. et al. (1998). Positron emission tomography in vascular dementia. J Neurol Sci, 154, 55–61. De Reuck, J., Santens, P., Strijckmans, K. et al. (2001). Cobalt55 positron emission tomography in vascular dementia: significance of white matter changes. J Neurol Sci, 193, 1–6. Desmond, D. W. (2002). Cognition and white matter lesions. Cerebrovasc Dis, 13, Suppl 2, 53–7. De Strooper, B., Annaert, W., Cupers, P. et al. (1999). A presenilin-1dependent gamma-secretase-like protease mediates release of Notch intracellular domain. Nature, 398, 518–22. Devinsky, O., Petito, C. K. & Alonso, D. R. (1988). Clinical and neuropathological findings in systemic lupus erythematosis: the role of vasculitis, heart emboli and thrombotic thrombocytopenic purpura. Ann Neurol, 23, 380–4. Devuyst, G., Sindic, C., Laterre, E. C. & Brucher, J. M. (1996). Neuropathological findings of a Sneddon’s syndrome presenting with dementia not preceded by clinical cerebrovascular events. Stroke, 26, 492–5. Dichgans, M. (2003). A new cause of hereditary small vessel disease: angiopathy of retina and brain. Neurology, 60, 8–9. Dichgans, M., Mayer, M., Uttner, D. P. et al. (1998). The phenotypic spectrum of CADASIL: Clinical findings in 102 cases. Ann Neurol, 44, 731–9. Dichgans, M., Filippi, M., Bruning, R. et al. (1999). Quantitative MRI in CADASIL: correlation with disability and cognitive performance. Neurology, 52, 1361–7. Dichgans, M., Holtmannspotter, M., Herzog, J., Peters, N., Bergmann, M. & Yousry, T. A. (2002). Cerebral microbleeds in CADASIL: a gradient-echo magnetic resonance imaging and autopsy study. Stroke, 33, 67–71. Dickson, D. W., Davies, P., Bevona, C. et al. (1994). Hippocampal sclerosis: a common pathological feature of dementia in very old (greater than or equal to 80 years of age) humans. Acta Neuropathol, 88, 212–21.

Drlicek, M., Grisold, W., Liszka, U. et al. (1991). Angiotropic lymphoma (malignant angioendotheliomatosis) presenting with rapidly progressive dementia. Acta Neuropathol (Berl), 82, 533–55. Du, A. T., Schuff, N., Laakso, M. P. et al. (2002). Effects of subcortical ischemic vascular dementia and AD on entorhinal cortex and hippocampus. Neurology, 58, 1635–41. Dubois, M. F. & Hébert, R. (2001). The incidence of vascular dementia in Canada: a comparison with Europe and East Asia. Neuroepidemiology, 20, 179–87. Ehehalt, R., Keller, P., Haass, C., Thiele, C. & Simons, K. (2003). Amyloidogenic processing of the Alzheimer beta-amyloid precursor protein depends on lipid rafts. J Cell Biol, 160, 113–23. Ellis, S. G. & Verity, M. A. (1979). Central nervous system involvement in systemic lupus erythematosis: a review of neuropathological findings in 57 cases 1955–1977. Semin Arthritis Rheum, 8, 212–21. Ellison, D., Gatter, K., Heryet, A. & Esiri, M. (1993). Intramural platelet deposition in cerebral vasculopathy of systemic lupus erythematosus. J Clin Pathol, 46, 37–40. Erkinjuntti, T. (1997). Vascular dementia: challenge of clinical diagnosis. Int Psychogeriatr (Suppl), 1, 51–8. (2002). Subcortical vascular dementia. Cerebrovasc Dis, 13 Suppl 2, 58–60. Erkinjuntti, T., Haltia, M., Palo, J., Sulkava, R. & Paetau, A. (1988). Accuracy of the clinical diagnosis of vascular dementia: a prospective clinical and post mortem neuropathological study. J Neurol Neurosurg Psychiatry, 51, 1037–44. Erkinjuntti, T., Ostbye, T., Steenhuis, R. & Hachinski, V. (1997). The effect of different diagnostic criteria on the prevalence of dementia. N Engl J Med, 337, 1667–74. Erkinjuntti, T., Inzitari, D., Pantoni, L. et al. (2000). Limitations of clinical criteria for the diagnosis of vascular dementia in clinical trials. Is a focus on subcortical vascular dementia a solution? Ann N Y Acad Sci, 903, 262–72. Esiri, M. M. (2000). Which vascular lesions are important in vascular dementia? Ann NY Acad Sci, 903, 239–43. Esiri, M. & Wilcock, G. K. (1986). Cerebral amyloid angiopathy in dementia and old age. J Neurol Neurosurg Psychiatry, 49, 1221–6. Esiri, M. M., Wilcock, G. K. & Morris, J. H. (1997). Neuropathological assessment of the lesions of significance in vascular dementia. Neurol Neurosurg Psychiatry, 63, 749–53. Esiri, M. M., Nagy, Z., Smith, M. Z. et al. (1999). Cerebrovascular disease and threshold for dementia in the early stages of Alzheimer’s disease. Lancet, 354, 919–20. Esler, W. P., Kimberly, W. T., Ostaszewski, B. L. et al. (2002). Activitydependent isolation of the presenilin- -secretase complex reveals nicastrin and a substrate. Proc Natl Acad Sci, 99, 2720–5. Fein, G., Di Sclafani, V., Tanabe, J. et al. (2000). Hippocampal and cortical atrophy predict dementia in subcortical ischaemic vascular disease. Neurology, 55, 1626–35. Feldman, E. (1994). Intracerebral hemorrhage. In Clinical Atlas of Cerebrovascular Disorders. ed. M. Fisher, pp. 11–17, Chicago, IL: Wolfe.

Vascular dementias

Ferro, J. M. (1998). Vasculitis of the central nervous system. J. Neurol, 245, 766–76. Fieschi, C., Rasura, M., Anzini, A. & Beccia, M. (1998). Central nervous system vasculitis. J Neurol Sci, 153, 159–71. Fisher, C. M. (1969). The arterial lesions underlying lacunes. Acta Neuropath Berl, 12, 1–15. (1982). Lacunar strokes and infarcts: a review. Neurology, 32, 871–6. (1989). Binswanger’s encephalopathy: a review. J Neurol, 236, 65–79. Francis, C. & Piette, J. C. (2000). The mystery of Sneddon syndrome: relationship with antiphospholipid syndrome and systemic lupus erythematosis. J Autoimmun, 15, 139–43. Frank, A., Diez-Tejedor, E., Bullido, M. J., Valdivieso, F. & Baneiro, P. (2002). APOE genotype in cerebrovascular disease and vascular dementia. J Neurol Sci, 203–204, 173–6. Fratiglioni, L., De Ronchi, D. & Aguero-Torres, H. (1999). Worldwide prevalence and incidence of dementia. Drugs Aging, 15, 365–75. Fratiglioni, L., Launer, L. J., Andersen, K. et al. (2000). Incidence of dementia and major subtypes in Europe: A collaborative study of population based cohorts. Neurology, 54 (Suppl. 5), S10–S15. Geschwind, D. H., FitzPatrick, M., Mischel, P. S. & Cummings, J. L. (1995). Sneddon’s syndrome is a thrombotic vasculopathy: neuropathologic and neuroradiologic evidence. Neurology, 45, 557–60. Gil-Nagel, A., Wilcox, K. J., Stewart, J. M. et al. (1995). Familial cerebral cavernous angioma: clinical analysis of a family and phenotypic classification. Epilepsy Res, 21, 27–36. Gold, G., Giannakopoulos, P. M., Montes-Paixao Junior, C. et al. (1997). Sensitivity and specificity of newly proposed clinical criteria for possible vascular dementia. Neurology, 49, 690–4. Gold, G., Bouras, C., Canuto, A. et al. (2002). Clinicopathological validation study of four sets of clinical criteria for vascular dementia. Am J Psychiatry, 59, 82–7. Goto, M., Kimura, T., Hagio, S. et al. (1995). Neuropathological analysis of dementia in a Japanese leprosarium. Dementia, 6, 157–61. Graff-Redford, N. R., Damasio, H., Yamada, T., Eslinger, P. J. & Damasio, A. R. (1985). Nonhaemorrhagic thalamic infarction. Clinical, neuropsychological and electrophysiological findings in four anatomical groups defined by computerized tomography. Brain, 108, 485–516. Gutiérrez-Molina, M., Caminero-Rodr´ıquez, A. & Mart´ınez Garc´ıa, C. et al. (1994). Small arterial granular degeneration in familial Binswanger’s disease. Acta Neuropathol (Berl), 87, 98–105. Hachinski, V. C. (1987). Leukoaraiosis. Arch Neurol, 44, 21–3. (1991). Binswanger’s disease: neither Binswanger’s nor a disease. J Neurol Sci, 103, 1. Hachinski, V. C., Lassen, N. A. & Marshall, J. (1974). Multi-infarct dementia: a cause of mental deterioration in the elderly. Lancet, 2, 207–10. Hademenos, G. J., Alberts, M. J., Awad, I. et al. (2001). Advances in the genetics of cerebrovascular disease and stroke. Neurology, 56, 997–1008.

Hadjivassiliou, M., Tooth, C. L., Romanowski, C. A. et al. (2001). Aneurysmal SAH: cognitive outcome and structural damage after clipping or coiling. Neurology, 56, 1672–7. Hanley, J. G., Walsh, N. M. G. & Sangalang, V. (1992). Brain pathology in systemic lupus erythematosis. J Rhematol, 19, 732–41. Hanson, V. G., Horowitz, M., Rosenbluth, D., Speira, H. & Puszkin, S. (1992). Systemic Lupus Erythematosus patients central nervous system involvement show antibodies to a 50 kD neuronal membrane protein. J Exp Med, 176, 565–73. Harrison, P. E., Jr. (1976). Granulomatous angiitis of the central nervous system. Case report and review. J Neurol Sci, 29, 335–41. Hashiguchi, S., Mine, H., Ide, M. & Kawachi, Y. (2000). Watershed infarction associated with dementia and cerebral atrophy. Psychiatry Clin Neurosci, 54, 163–8. Hébert, R., Lindsay, J., Verreault, R., Rockwood, K., Hill, G. & Dubois, M.-F. (2000). Vascular dementia: incidence and risk factors in the Canadian Study of Health and Aging. Stroke, 31, 1487. Heiss, W. D. (1992). Experimental evidence of ischemic thresholds and functional recovery. Stroke, 23, 1668–72. Henon, H., Pasquier, F., Durieu, I. et al. (1997). Preexisting dementia in stroke patients. Baseline frequency, associated factors, and outcome. Stroke, 28, 2429–36. Henon, H., Durieu, I., Guerouaou, D. et al. (2001). Poststroke dementia: incidence and relationship to prestroke cognitive decline. Neurology, 57, 1216–22. Holmes, C., Cairns, N., Lantos, P. & Mann, D. A. (1999). Validity of current clinical criteria for Alzheimer’s disease, vascular dementia and dementia with Lewy bodies. Br J Psychiatry, 174, 45–50. Howard, G. F., Ho, S. U., Kim, K. S. & Wallach, J. (1984). Bilateral corotid artery occlusion resulting from giant cell arteritis. Ann Neurol, 15, 204–7. Hulette, C. D., Nochlin, D., McKeel, D. et al. (1997). Clinical– neuropathologic findings in multi-infarct dementia: a report of six autopsied cases. Neurology, 48, 668–72. Hurst, R. W., Bagley, L. J., Galetta, S. et al. (1998). Dementia resulting dural arterio-venous fistulas: the pathologic findings of venous hypertensive encephalopathy. Am J Neuroradiol, 19, 1267–73. Ikeda, M., Hokoishi, K., Maki, N. et al. (2001). Increased prevalence of vascular dementia in Japan: a community based epidemiological study. Neurology, 57, 839–44. Inafuku, T., Watanabe, M., Takagi, M. et al. (1998). Giant cell arteritis with bilateral obstruction of the internal carotid artery – report of an autopsy case. Rinsho Shinkeigaku, 38, 323–8. Ince, P. G., McArthur, F. K., Bjertness, F., Torvik, A., Candy, J. M. & Edwardson, J. A. (1995). Neuropathological diagnoses in elderly patients in Oslo: Alzheimer’s disease, Lewy body disease, vascular lesions. Dementia, 6, 162–8. Inglesias, S., Chapon, F. & Baron, J. C. (2000). Familal occipital calcifications, haemorrhagic strokes, leukoencephalopathy, dementia and external carotid dysplasia. Neurology, 55, 1661–7. Inzitari, D., Erkinjuntti, T., Wallin, A., Del Ser, T., Romanelli, M. & Pantoni, L. (2000). Subcortical vascular dementia as a specific target for clinical trials. Ann N Y Acad Sci, 903, 510–21.

323

324


Ironside, J. W., Martin, J. F., Richmond, J. & Timperley, W. (1984). Lymphomatoid granulomatosis with cerebral involvement. Neuropathol Appl Biol, 10, 397–406. Ishikawa, E., Yanaka, K., Sugimoto, K., Ayuzawa, S. & Nose, T. (2002). Reversible dementia in patients with chronic subdural hematomas. J Neurosurg, 96, 680–3. Jafar, J. J., Menoni, R., Feinberg, H., LeBreton, G. & Crowell, R. M. (1989). Selective platelet deposition during focal ischaemia in cats. Stroke, 20, 664–7. Jellinger, K. (1977). Giant cell granulomatous angiitis of the central nervous system. J Neurol, 215, 175–90. Jellinger, K. (2001). Neuropathologic substrates of Vascular Dementia. J Neuropath Exp Neurol, 60, 658–9 (letter). Jellinger, K. A. (2002). The pathology of ischemic–vascular dementia: an update. J Neurol Sci, 203–204, 153–7. Jellinger, K., Danielczyk, W., Fischer, P. & Gabriel, P. (1990). Clinicopathological analysis of dementia disorders of the elderly. J Neurol Sci, 95, 239–58. Jennekens, F. G. & Kater, L. (2002). The central nervous system in systemic lupus erythematosus. Part 2. Pathogenetic mechanisms of clinical syndromes: a literature investigation. Rheumatology (Oxford), 41, 619–30. Joachim, C. L., Morris, J. H. & Selkoe, D. J. (1988). Clinically diagnosed Alzheimer’s disease: Autopsy results in 150 cases. Ann Neurol, 24, 50–6. Johnson, R. T. & Richardson, E. P. (1968). The neurological manifestations of Systemic Lupus Erythematosus. A clinical pathological study of 24 cases and a review of the literature. Medicine (Baltimore), 47, 337–69. Jorm, A. F. (1991). Cross national comparison of the occurrence of Alzheimer and vascular dementias. Eur Arch Psychiatry Clin Neurosci, 240, 218–22. Jorm, A. F. & Jolley, D. (1998). The incidence of dementia: a meta-analysis. Neurology, 51, 728–33. Joutel, A., Corpechot, C., Ducros, A. et al. (1996). Notch3 mutations in CADASIL, a hereditary late-onset condition causing stroke and dementia. Nature, 383, 707–10. Joutel, A., Vahedi, K., Corpechot, C. et al. (1997). Strong clustering and stereotyped nature of Notch3 mutations in CADASIL patients. Lancet, 350, 1511–15. Joutel, A., Andreux, F., Gaulis, S. et al. (2000). The ectodomain of the Notch3 receptor accumulates within the cerebrovasculature of CADASIL patients. J Clin Invest, 105, 597–605. Joutel, A., Favrole, P., Labauge, P. et al. (2001). Skin biopsy immunostaining with a Notch3 monoclonal antibody for CADASIL diagnosis. Lancet, 358, 2049–51. Kageyama, Y., Kodama, Y., Yamamoto, S. et al. (2000). A case of multiple intracranial cavernous angiomas presented with dementia and parkinsonism – clinical and MRI study for 10 years. Rinsho Shinkeigaku, 40, 1105–9. Kalaria, R. N. (1996). Cerebral vessels in ageing and Alzheimer’s disease. Pharm Therap, 72, 193–214. (2001). Advances in molecular genetics and pathology of cerebrovascular disorders. Trends Neurosci, 24, 392– 400.

Kalaria, R. N., Cookson, N. J., Mizuno, T. et al. (2000). Microvascular pathology in familial multiinfarct dementia or CADASIL: implications for cerebral blood flow. Ischemic Blood Flow in the Brain. Tokyo: Springer, pp. 156–166. Kalimo, H., Kaste, M. & Haltia, M. (2002a). Vascular diseases. In Greenfield’s Neuropathology eds Graham DI, Lantos PL, 7th edn, pp. 281–355. London, UK: Arnold. Kalimo, H., Ruchoux, M. M., Viitanen, M. & Kalaria, R. N. (2002b). CADASIL: a common form of hereditary arteriopathy causing brain infarcts and dementia. Brain Pathol, 12, 371–84. Kalimo, H., Viitanen, M., Carlström, C., Nordberg, A. & Bronge, L. (2002c). CADASIL in monozygotic twins. Neuropathol Appl Neurobiol, 28, 151. Kariya, S., Kawahara, M. & Suzumura, A. (2000). A case of multiple cavernous angioma with dementia. Rinsho Shinkeigaku, 40, 1003–7. Katzenstein, A. -L. A., Carrington, C. B. & Liebow, A. A. (1979). Lymphomatoid granulomatosis: a clinicopathologic study of 152 cases. Cancer, 43, 360–73. Kawamura, J., Meyer, J. S., Ichijo, M., Kobari, M., Terayama, Y. & Weathers, S. (1993). Correlations of leuko-araiosis with cerebral atrophy and perfusion in elderly normal subjects and demented patients. J Neurol Neurosurg Psychiatry, 56, 182–7. Kawano, H., Ueda, K. & Fujishima, M. (1990). Prevalence of dementia in a Japanese community (Hisayama): morphological reappraisal of the type of dementia. Jpn J Med, 29, 261–5. Kerr, R. S. C., Hughes, J. T., Blamires, T. & Teddy, P. J. (1987). Lymphomatoid granulomatosis apparently confined to one temporal lobe. J Neurosurg, 67, 612–15. Kimberly, W. T., Zheng, J. B., Guenette, S. Y. & Selkoe, D. J. (2001). The intracellular domain of the beta-amyloid precursor protein is stabilized by Fe65 and translocates to the nucleus in a notch-like manner. J Biol Chem, 276, 40288–92. Knopman, D. S., DeKosky, S. T., Cummings, J. L. et al. (2001). Practice parameter: diagnosis of dementia (an evidence-based review). Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology, 56, 1143–53. Korczyn, A. D. (2002). The complex nosological concept of vascular dementia. J Neurol Sci, 203–204, 3–6. Kraepelin, E. (1910). Das senile und präsenile Irresein. In: Psychi¨ atrie: Ein Lehrbuch fur pp. 533–632. ¨ Studierende und Artzte, Leipzig: Johann Ambrosius Barth. Kramer, J. H., Reed, B. R., Mungas, D., Weiner, M. W. & Chui, H. C. (2002). Executive dysfunction in subcortical ischaemic vascular disease. J Neurol Neurosurg Psychiatry, 72, 217–20. Kreiter, K. T., Copeland, D., Bernardini, G. L. et al. (2002). Predictors of cognitive dysfunction after subarachnoid hemorrhage. Stroke, 33, 200–8. Krendal, D. A., Ditter, S. M., Frankel, M. R. & Ross, W. K. (1990). Biopsy proven cerebral vasculitis associated with cocaine abuse. Neurology, 40, 1092–4. Lammie, G. A. (2002). Hypertensive small vessel disease and stroke. Brain Pathol, 12, 358–70. Larner, A. J., Kidd, D., Elkington, P. et al. (1999). Spatz-Lindenberg disease: a rare cause of vascular dementia. Stroke, 30, 687–9.

Vascular dementias

Law, I., Iida, H., Holm, S. et al. (2000). Quantitation of regional cerebral blood flow corrected for partial volume effect using O-15 water and PET: II. Normal values and gray matter blood flow response to visual activation. J Cereb Blood Flow Metab, 20, 1252–63. Lee, D. Y., Lee, J. H., Ju, Y.-S. et al. (2002). The prevalence of dementia in older people in an urban population of Korea. The Seoul study. J Am Geriatr Soc, 50, 1233–9. Lesnik Oberstein, S. A., van den Boom, R., van Buchem, M. A. et al. (2001). Cerebral microbleeds in CADASIL. Neurology, 57, 1066–70. Levine, S. R., Brust, J. C. M., Futrell, N. et al. (1990). Cerebrovascular complications of the use of the “crack” form of alkoidal cocaine. N Engl J Med, 323, 699–704. LeWitt, P. A., Forno, L. S. & Brant-Zawadski, M. (1983). Neoplastic angioendotheliomatosis: a case with spontaneous regression and radiographic appearance of cerebral arteritis. Neurology, 33, 39–44. Lie, J. T. (1988). Thromboangiitis revisited. Pathol Annu, 23 (II), 257. Liebow, A. A., Carrington, C. R. B. & Friedman, P. J. (1972). Lymphomatoid granulomatosis. Hum Pathol, 3, 457–558. Lim, L., Ron, M. A., Omerod, I. E. et al. (1988). Psychiatric and neurological mainifestations in systemic lupus erythematosis. Q J Med, 66, 27–38. Lin, J. X., Tomimoto, H., Akiguchi, I. et al. (2000). Vascular cell components of the medullary arteries in Binswanger’s disease brains: a morphometric and immunoelectron microscopic study. Stroke, 31, 1838–42. ¨ Lindenberg, R. & Spatz, H. (1939). Uber die thromboendarteriitis obliterans der Hirngefäe (Cerebrale form der v. Winiwarter-Buergerschen Krankheit). Virchows Arch (A), 305, 531–57. Liu, C. K., Miller, B. L., Cummings, J. L. et al. (1992). A quantitative MRI study of vascular dementia. Neurology, 42, 138–43. Lobo, A., Launer, A. J., Fratiglioni, L. et al. (2000). Prevalence of dementia and major subtypes in Europe: A collaborative study of population based cohorts. Neurology, 54 (Suppl. 5), S4–9. Loeb, C. (1990). Vascular dementia. Dementia, 1, 175–84. (2000). Binswanger’s disease is not a single entity. Neurol Sci, 21, 343–8. Longstreth, W. T., Jr, Manolio, T. A., Arnold, A. et al (1996). Clinical correlates of white matter findings on cranial magnetic resonance imaging of 3301 elderly people. The Cardiovascular Health Study. Stroke, 27, 1274–82. Majamaa, K., Finnilä, S., Turkka, J. & Hassinen, I. E. (1998). Mitochondrial DNA haplogroup U as a risk factor for occipital stroke in migraine. Lancet, 352, 455–6. Marie, P. (1901). Des foyers lacunaires de désintégration de différents autres e´ tats cavitaires du cerveau. Rev Méd, 21, 281–98. Markesbery, W. R. (1995). Vascular dementia. Neuropathology of Dementing Disorders, ed. W. R. Markesbery, pp. 293–311. London: Arnold. Mast, H., Tatemichi, T. K. & Mohr, J. P. (1995). Chronic brain ischemia: the contributions of Otto Binswanger and Alois

Alzheimer to the mechanisms of vascular dementia. J Neurol Sci, 132, 4–10. Mayer, S. A., Kreiter, K. T., Copeland, D. et al. (2002). Global and domain-specific cognitive impairment and outcome after subarachnoid hemorrhage. Neurology, 59, 1750–8. ¨ Mellies, J. K., Bäumer, T., Muller, J. A. et al. (1998). SPECT study of a German CADASIL family. A phenotype with migraine and progressive dementia only. Neurology, 50, 1715–21. Mendez, M. F., Stanley, T. M., Medel, N. M. et al. (1997). The vascular dementia of Fabry’s disease. Dement Geriatr Cogn Disord, 8, 252–7. Meyer, J. S., Rauch, G. M., Rauch, R. A., Haque, A. & Crawford, K. (2000). Cardiovascular and other risk factors for Alzheimer’s Disease and Vascular Dementia. Ann N Y Acad Sci, 903, 411–23. Millikan, C. & Futrell, N. (1990). The fallacy of the lacune hypothesis. Stroke, 21, 1251–57. Mitchell, I., Hughes, R. A. C., Maisey, M. & Cameron, S. (1994). Cerebral Lupus. Lancet, 343, 579–82. Mitsias, P. & Levine, S. R. (1994). Large cerebral vessel occlusive disease in systemic lupus erythematosus. Neurology, 44, 385–93. Mölsä, P. K., Paljärvi, L., Rinne, U. K. & Säkö, E. (1985). Validity of clinical diagnosis of Dementia: a prospective clinicopathological study. J Neurol Neurosurg Psychiatry, 48, 1085–90. Moody, D. M., Bell, M. A. & Challa, V. R. (1990). Features of the cerebral vascular pattern that predict vulnerability to perfusion or oxygenation deficiency: an anatomic study. Am J Neuroradiol, 11, 431–9. Moore, P. M. (1989). Diagnosis and management of isolated angiitis of the central nervous system. Neurology, 39, 167–73. Moore, P. M. & Cupps, T. R. (1983). Neurological complications of vasculitis. Ann Neurol, 14, 155–67. Moroney, J. T., Bagiella, E., Desmond, D. W. et al. (1997). Metaanalysis of the Hachinski Ischemic Score in pathologically verified dementias. Neurology, 49, 1096–105. Moroney, J. T., Tang, M. X., Berglund, L. et al. (1999). Low density lipoprotein cholesterol and the risk of dementia with stroke. J Am Med Assoc, 282, 254–60. Morrow, P. L. & McQuillen, J. B. (1993). Cerebral vasculitis associated with cocaine abuse. J Forens Sci, 38, 732–8. Munch, G., Schinzel, R., Loshe, C., et al. (1998). Alzheimer’s disease – synergistic effects of glucose deficit, oxidative stress and advanced glycation endproducts. J Neural Transm, 105, 439–61. Mungas, D., Reed, B. R., Ellis, W. G. & Jagust, W. J. (2001). The effects of age on rate of progression of Alzheimer disease and dementia with associated cerebrovascular disease. Arch Neurol, 58, 1243–7. Nishino, H., Rubino, F. A., DeRemee, R. A., Swanson, J. W. & Parisi, J. E. (1993). Neurological involvement in Wegener’s granulomatosis: an analysis of 234 consecutive patients at the Mayo Clinic. Ann Neurol, 33, 4–9. Nolan, K. A., Lino, M. M., Seligmann, A. W. & Blass, J. P. (1998). Absence of vascular dementia in an autopsy series from a dementia clinic. J Am Geriatr Soc, 46, 597–604. Okeda, R. (1973). Morphometrische Vergleichsuntersuchungen an Hirnarterien bei Binswangerscher Encephalopathie und

325

326


Hochdruckencephalopathie. Acta Neuropathol (Berl.), 26, 23–43. Ophoff, R. A., DeYoung, J., Service, S. K. et al. (2001). Hereditary vascular retinopathy, cerebroretinal vasculopathy, and hereditary endotheliopathy with retinopathy, nephropathy, and stroke map to a single locus on chromosome 3p21. 1-p21. 3. Am J Hum Genet, 69, 447–53. O’Sullivan, M., Lythgoe, D. J., Pereira, A. C. et al. (2002). Patterns of cerebral blood flow reduction in patients with ischemic leukoaraiosis. Neurology, 59, 321–6. Pantoni, L. & Garcia, J. H. (1995). The significance of Cerebral White Matter Abnormalities 100 years after Binswanger’s Report. Stroke, 26, 1293–301. (1997). Pathogenesis of leukoaraiosis. Stroke, 28, 652–9. Pantoni, L., Garcia, J. H. & Brown, G. G. (1996). Vascular pathology in three cases of progressive cognitive deterioration. J Neurol Sci, 135, 131–9. Pantoni, L., Leys, D., Fazekas, F. et al. (1999). Role of white matter lesions in cognitive impairment of vascular origin. Alzheimer Dis Assoc Disord, 13, Suppl 3, S49–54. Pantoni, L., Basile, A. M., Romanelli, M. et al. (2001). Abulia and cognitive impairment in two patients with capsular genu infarct. Acta Neurol Scand, 104, 185–90. Petito, C. K., Gottleib, G. J., Dougherty, J. H. & Petito, F. A. (1978). Neoplastic angioendotheliosis: ultrastructural study and review of the literature. Ann Neurol, 3, 393–9. Pohjasvaara, T., Mäntylä, R., Ylikoski, R. et al. (2000). Comparison of different clinical criteria (DSM-III, ADDTC ICD-10, NINDS-AIREN, DSM-IV) for the diagnosis of vascular dementia. National Institute of Neurological Disorders and Stroke-Association Internationale pour la Recherche et L’Enseignment en Neurosciences. Stroke, 31, 2952–7. Poirier, J. & Derouesne, C. (1984). Cerebral lacunae. A proposed new classification. Clin Neuropathol, 3, 266. (1998). Distinguishing lacunar infarcts from dilatations of the perivascular space. J Neurol, 245, 813–14. Posner, H. B., Tang, M.-X., Luchsinger, J., Lantigua, R., Stern, Y. & Mayeux, R. (2002). The relationship of hypertension in the elderly in AD, vascular dementia, and congnitive function. Neurology, 58, 1175–81. Reinglass, J. L., Muller, J., Wissman, S. & Wellman, H. (1977). Central nervous system angioendotheliosis: a treatable multiple infract dementia. Stroke, 8, 218–21. Rockwood, K., Howard, K., MacKnight, C. & Darvesh, S. (1999). Spectrum of disease in vascular cognitive impairment. Neuroepidemiology, 18, 248–54. Rollnik, J. D., Brandis, A., Dehghani, K. et al. (2001). Die primäre Angiitis des ZNS (PACNS). Nervenarzt 72, 798–801. Roman, G. C. (1999). A historical review of the concept of vascular dementia: lessons from the past for the future. Alzheimer Dis Assoc Disord, 13, (Suppl 3), S4–S8. (2001). Diagnosis of vascular dementia and Alzheimer’s disease. Int J Clin Pract (Suppl.) 120, 9–13. (2002). Vascular dementia revisited: diagnosis, pathogenesis, treatment, and prevention. Med Clin North Am, 86, 477–99.

Roman, G. C., Tatemichi, T. K. Erkinjuntti, T. et al. (1993). Vascular dementia: diagnostic criteria for research studies. Report of the NINDS-AIREN International workshop. Neurology, 43, 250–60. Romanul, F. C. A. (1970). Examination of the brain and spinal cord. In: Neuropathology. Methods and Diagnosis, ed. C. G. Tedeschi, 1st edn. pp. 131–214. Boston, MA: Little, Brown & Co. Rosén, W. G., Terry, R. D. & Fuld, P. A. et al. (1980). Pathologic verification of ischemic score in differentiation of dementias. Ann Neurol, 7, 486–8. Rowley, H. A., Lowenstein, D. H., Rowbotham, M. C. & Simon, R. P. (1989). Thalamomesencephalic strokes after cocaine abuse. Neurology, 39, 428–30. Ruchoux, M. M. & Maurage, C. A. (1997). CADASIL: cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy. J Neuropathol Exp Neurol, 56, 947–64. Ruchoux, M. M., Chabriat, H., Bousser, M. G., Baudrimont, M. & Tournier-Lasserve, E. (1994). Presence of ultrastructural arterial lesions in muscle and skin vessels of patients with CADASIL. Stroke, 25, 2291–2. Sacco, R. L. (1994). Classification of stroke. In Clinical Atlas of Cerebrovascular Disorders, ed. M. Fisher, pp. 2.2–2.25, London: Wolfe. Scheltens, P. & Hijdra, A. H. (1998). Diagnostic criteria for vascular dementia. Haemostasis, 28, 151–7. Schmidt, B. J., Meagher-Villemure, K. & Del Carpio, J. (1984). Lymphomatoid granulomatosis with isolated involvement of the brain. Ann Neurol, 15, 478–81. Schröder, J. M., Sellhaus, B. & Jorg, J. (1995). Identification of the characteristic vascular changes in a sural biopsy of a case with cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL). Acta Neuropathol (Berl), 89, 116–21. Scolding, N. J. & Joseph, F. G. (2002). The neuropathology and pathogenesis of systemic lupus erythematosis. Neuropathol Appl Neurobiol, 28, 173–89. Seno, H., Ishino, H., Inagaki, T., Iijima, M., Kaku, K. & Inata, T. (1999). A neuropathological study of dementia in nursing homes over a 17 year period, in Shimane Prefecture, Japan. J Gerontol A Biol Sci Med Sci, 54, M312–14. Sheibani, K., Battifora, H., Winberg, C. D. et al. (1986). Further evidence that ‘malignant angioendotheliomatosis’ is an angiotropic large cell lymphoma. N Engl J Med, 314, 943–8. Shen, Y. & Yu, X. (2001). Epidemiology of vascular dementia in China. In: Cerebrovascular Disease and Dementia, ed. E. Chiu, L. Gustafson, D. Ames, M. F., Folstein, pp. 47–54. UK: Martin Dunitz Ltd. Simon, R. H., Abeles, M., Farber, N. J., Grunnet, M. & Brennan, T. (1981). Lymphomatoid granulomatosis with multiple intracranial lesions. J Neurosurg, 55, 293–8. Skoog, I. (1998). Status of risk factors for vascular dementia. Neuroepidemiology, 17, 2–9. Sonninen, V. & Savontaus, M. L. (1987). Hereditary multi-infract dementia. Eur Neurol, 27, 209–15.

Vascular dementias

Sourander, P. & W˚alinder, J. (1977). Hereditary multi-infarct dementia. Acta Neuropath (Berl.), 39, 247–54. Spinner, N. B. (2000). CADASIL: notch signaling defect or protein accumulation problem? J Clin Invest, 105, 561–2. Steinberg, D. (2002). Atherogenesisin perspective: hypercholesterolemia and inflammation as partners in crime. Nat Med, 8, 1211–17. Suh, G. H. & Shah, A. (2001). A review of the epidemiological transition in dementia – cross national comparisons of the indices related to Alzheimer’s disease and vascular dementia. Acta Psychiatr Scand, 104, 4–11. Taillia, H., Chabriat, H., Kuetz, A. et al. (1998). Cognitive alterations in non-demented CADASIL patients. Cerebrovasc Dis, 8, 97–101. Tanahashi, C., Nakayama, A., Yoshida, M., Ito, M., Mori, N. & Hashizume, Y. (2000). MELAS with the mitochondrial DNA 3243 point mutation: a neuropathological study. Acta Neuropathol (Berl), 99, 31–8. Tanoi, Y., Okeda, R. & Budka, H. (2000). Binswanger’s encephalopathy: serial sections and morphometry of the cerebral arteries. Acta Neuropathol, 100, 347–55. Tatemichi, T. K. (1990). How acute brain failure becomes chronic: A view of the mechanisms of dementia related to stroke. Neurology, 40, 1652–9. Tatemichi, T. K., Desmond, D. W., Mayeux, R. et al. (1992). Dementia after stroke: baseline frequency, risks, and clinical features in a hospitalised cohort. Neurology, 42, 1185–93. Tatemichi, T. K., Desmond, D. W., Stern, Y. et al. (1994). Cognitive impairment after stroke: frequency, patterns, and relationship to functional abilities. J Neurol Neurosurg Psychiatry, 57, 202–7. Tatemichi, T. K., Desmond, D. W. & Prohovnik, I. (1995). Strategic infarcts in vascular dementia. A clinical and brain imaging experience. Arzneimittelforschung, 45, 371–85. Tomlinson, B. E., Blessed, G. & Roth, M. (1970). Observations on the brains of demented old people. J Neurol Sci, 11, 205–42. Tournier-Lasserve E, Joutel A, Melki, J. et al. (1993). Cerebral Autosomal Dominant Arteriopathy with Subcortical Infarcts and Leukoencephalopathy maps to chromosome 19q12. Nature Genet, 3, 256–9. Travkov, L., Rigaud, A. S., Baudic, S., Smagghe, A., Boller, F. & Forette, F. (2002). Apolipoprotein Eε4 allele frequency in demented and cognitively impaired patients with and without cerebrovascular disease. J. Neurol. Sci, 203–204, 177–81. Treves, T. A., Gadoth, N., Blumen, S. & Korczya, A. D. (1995). Intravascular malignant lymphomatosis: a cause of subacute dementia. Dementia, 6, 286–93. Tuominen, S., Juvonen, V., Amberla, K. et al. (2001). Phenotype of a homozygous CADASIL patient in comparison to 9 agematched heterozygous patients with the same R133C Notch3 mutation. Stroke, 32, 1767–74. Ueda, K., Kawano, H., Hasuo, Y. & Fujishima, M. (1992). Prevalence and etiology of dementia in a Japanese community. Stroke, 23, 798–803.

Ulrich, J., Probst, A. & Wuest, M. (1986). The brain diseases causing senile dementia. J Neurol, 233, 118–22. van Bogaert, L. (1955). Encephalopathie sous-corticale progressive (Binswanger) a evolution rapide chez deuz soeurs. Med Hellen, 24, 961–72. Vinters, H. V., Ellis, W. G., Zarow, C. et al. (2000). Neuropathologic substrates of Ischaemic Vascular Dementia. J Neuropath Exp Neurol, 59, 931–45. von Strauss, E., Viitanen, M., De Ronchi, D. et al. (1999). Aging and the occurrence of dementia: findings from a population based cohort with a large sample of nonagenarians. Arch Neurol, 56, 587–92. Wade, J. P., Mirsen, T. R., Hachinski, V. C., Fisman, M., Lau, C. & Merskey, H. (1987). The clinical diagnosis of Alzheimer’s disease. Arch Neurol, 44, 24–9. Wang, W., Wu, S., Cheng, X. et al. (2000). Prevalence of Alzheimer’s disease and other dementing disorders in an urban community of Beijing, China. Neuroepidemiology, 19, 194–200. Wang, W., Prince, C., Mou, Y. & Pollman, M. J. (2002). Notch3 signaling in vascular smooth muscle cells induces c-FLIP expression via ERK/MAPK activation: Resistance to FasLinduced apoptosis. J Biol Chem, 277, 21723–9. World Health Organization (1992). ICD10 Classification of Mental and Behavioural Disorders: Clinical Descriptions and Diagnostic Guidelines. Geneva: WHO 50–51. Worster-Draught, C., Hill, T. R. & McMenemey W. H. (1933). Familial presenile dementia with spastic paraparalysis. J Neurol Psychopathol, 14, 27–34. Wright, R. A. & Kokmen, E. (1999). Gradually progressive dementia without discrete cerebrovascular events in a patient with Sneddon’s syndrome. Mayo Clin Proc, 74, 57–61. Yanagawa, S., Ito, N., Arima, K. & Ikeda, S. (2002). Cerebral autosomal recessive arteriopathy with subcortical infarcts and leukoencephalopathy. Neurology, 58, 817–20. Zhan, S. S., Beyreuther, K. & Schmitt, H. P. (1993). Vascular dementia in Spatz–Lindenberg’s disease (SLD): cortical synaptophysin immunoreactivity as compared with dementia of Alzheimer type and non-demented controls. Acta Neuropathol (Berl), 86, 259–64. Zhang, W. W. & Olsson, Y. (1997). The angiography of subcortical arteriosclerotic encephalopathy (Binswanger’s disease): immunohistochemical studies using markers for components of extracellular matrix, smooth muscle actin and endothelial cells. Acta Neuropathol (Berl), 93, 219–24. Zhang, W. W., Badonic, T., Hoog, A. et al. (1994). Structural and vasoactive factors influencing intracerebral arterioles in case of vascular dementia and other cerebrovascular disease: a review. Immunohistochemical studies in expression of collagens, basal lamina components and endothelin-1. Dementia, 5, 153–62. Zuber, M., Blustajn, J., Arquizan, C., Trystram, D., Mas, J. L. & Meder, J. F. (1999). Angiitis of the central nervous system. J Neuroradiol, 26, 101–17.

327

328


Proforma Cerebrovascular disease : 0 = No 1. Vascular morbid anatomy

1 = Yes

Atherosclerosis – % occlusion none | =0

20% | =1

50% =3

| 80% | =5

100% =7 Left | Right

Location:

ACA MCA PCA Carotid Vertebral Carotid bifurcation Basilar

27 28 29 30 31 32

Other surface vascular pathology If Yes, Describe 2.

23

Focal lesions a. Infarcts Infarct present

23

Number: + > 10 mm diameter

24 Left | Right

Arterial territory:

ACA MCA PCA Vertebro-basilar Watershed Other

27 28 29 30 31 32

% Location:

b.

Frontal Parietal Temporal Hippocampus Occipital Thalamus Other

27 28 29 30 31 32 32

Lacunes: Lacune(s) (< 10 mm diameter) present

22

Location: Dp.Gr|WhMtr|Bstem|Othr|Mult =1 Number:

=2

Left | Right Age % Age

=3

=4

1–4 | 5–9 | 10 > =1

=3

33

=5 =5

34

Vascular dementias

c.

Haemorrhages Haemorrhage present

24

Number:

Single|Multiple =1

35

=3

< 5 mm |6–10 mm| > 10 mm

Size of largest:

=1

=3

Loc. of largest: CTx |WhMtr|Dp.Gr|Bstem| Cbm =1 3.

=2

=3

=4

36

=5 37 =5

Diffuse small vessel disease a. Grey matter Mild = 1, Moderate = 3, Severe = 5 27 28 29

b.

Arteriolar sclerosis Virchow-Robin space expansion Perivascular gliosis and attentuation White matter

27 28 29

f.

Arteriolar sclerosis Virchow–Robin space expansion Diffuse myelin loss and WM gliosis Other micro vascular disease

BG

Thal

Pons

Front

Par

Occ

47

If yes, described: Neuropathologic diagnosis: 1. Character Infarcts only Multiple lacunes ‘Diffuse small vessel disease’ Binswanger’s disease Haemorrhage only Other If other please specify Aetiology Hypertension Atherosclerosis Embolism Ischaemic (other) If other, please specify

27 28 29 30 31 32

2.

27 28 29 30

329

14 Familial and sporadic cerebral amyloid angiopathies associated with dementia and the BRI dementias Gordon T. Plant,1 Jorge Ghiso,2 Janice L. Holton,3 Blas Frangione2 and Tamas Revesz3 1 2 3

Department of Neurology, National Hospital for Neurology and Neurosurgery, Queen Square, London WC1N 3BG, UK Departments of Pathology and Psychiatry, New York University School of Medicine, New York, New York, USA Queen Square Brain Bank, Department of Molecular Pathogenesis and Division of Neuropathology, Institute of Neurology, London WC1N 3BG, UK

Introduction In this chapter we review a number of conditions which have in common cerebral amyloid angiopathy (CAA) (Table 14.1). However, and particularly in the context of dementia, it is now possible to make a clear distinction in this group of diseases. The distinction relates to the pathological processes which primarily lead to dementia. Both sporadic and familial CAA from any cause can lead to dementia as a consequence of the angiopathy itself. The discussion regarding the causes of dementia in such instances may be closely related to the discussion in arteriosclerotic vascular dementia (Chapter 13). CAA occurring in Alzheimer’s disease (AD) may contribute to the dementia by similar mechanisms. The clinico-pathological consequences of the CAA may be in some cases insignificant; in others there may be diffuse white matter rarefaction very similar to hypertensive Binswanger’s encephalopathy (Chapter 13); or there may be recurrent haemorrhage. The haemorrhages themselves are usually large and lobar – with characteristic distribution – but occasionally smaller, deep haemorrhages of the type associated with hypertensive lacunar disease may occur. Finally areas of infarction may be seen but this rarely dominates the clinical picture. Lobar cerebral haemorrhage seems to occur particularly in those cases in which fibrinoid necrosis and microaneurysm formation occur – these are the hallmarks of severe amyloid angiopathy (Vonsattel et al., 1991). Ischaemic white matter changes (leukoencephalopathy or leukoariosis) can occur even when the vessels in the affected white matter do not show evidence of angiopathy, the ischaemia presumably being secondary to involvement of the penetrating cortical

vessels (Dubas et al., 1985; Vonsattel et al., 1991). Yoshimura et al. (1992) have carried out an autopsy study of a group of 20 patients with cerebral haemorrhage due to CAA (almost all sporadic) 75% of whom were known to have become demented. Of the demented patients some had changes of AD while others showed leukoencephalopathy. However, in this group of disorders dementia can also result from a primary neurodegeneration. Since the first edition of this book was published there have been considerable advances in the understanding of two conditions then referred to as ‘Familial CAA with non-neuritic plaque formation (British type)’ and ‘Familial CAA with deafness and ocular haemorrhage (Danish type)’. These two conditions are now known to result from mutations of the BRI2 gene (see below) and may be grouped together as the ‘BRI dementias’. The severity of the CAA in these cases (more widespread than in any other disorder, familial or sporadic) had led to their being classified with other examples of familial CAA. However, in both of these conditions the major cause of dementia is neurofibrillary degeneration and neuronal loss in limbic structures and as such they may arguably bear a closer relationship to AD than, for example, the Dutch type of hereditary cerebral haemorrhage with amyloidosis (HCHWA-D) despite the latter being a disease characterized, in common with AD, by the deposition of A. Hence in the BRI dementias there is striking evidence of pathological changes resulting from the CAA itself but this is no more than a contributing factor to the overall dementing process. Amyloid itself is the result of protein aggregation in a particular structure (the -pleated sheet) to form highly insoluble fibrils. A large number of peptides each with quite different amino acid sequences have been found to make up


330

Table 14.1. Sporadic and hereditary cerebral amyloid angiopathies (CAAs) Sporadic CAAs

Principal clinical features

Gene Chromosome Precursor molecule Precursor function Amyloid protein

Familial CAAs

SCAA

AD

HCHWA-D

FAD

HCHWA-I

FAP/MVA

FAF

PrP-CAA

FBD

Cerebral haemorrhage Dementia due to coexistent AD

Dementia Cerebral haemorrhage rare

Cerebral haemorrhage Dementia may occur

Dementia Cerebral haemorrhage rare Spasticity and ataxia rare

Cerebral haemorrhage

Cerebral features rare in FAP Strokes, spasticity in MVA

Cerebral features rare

Dementia, ataxia No strokes

Progressive dementia, spasticity and ataxia Strokes rare

AβPP 21 A precursor protein (APP)

AβPP 21 A precursor protein (APP)

AβPP, PS1,PS2 21, 14, 1 A precursor protein (APP)

CYST C 20 Cystatin C (Cyst C)

TTR 18 Transthyretin (TTR)

GEL 9 Gelsolin (Gel)

PRNP 20 Prion protein (PrP)

Unknown

Unknown

A

A

Protease inhibitor ACys

Transport protein ATTR

Actin-binding Unknown protein AGel APrP

Unknown A

FDD

Progressive dementia spasticity and ataxia Cataract, deafness, ocular haemorrhage Strokes rare BRI2 BRI2 13 13 ABri precursor ADan precursor protein protein (ABriPP) (ADanPP) Unknown Unknown ABri

ADan

Note: SCAA = sporadic cerebral amyloid angiopathy; AD = Alzheimer’s disease; HCHWA-D = hereditary cerebral hemorrhage with amyloidosis – Dutch type; FAD = familial Alzheimer’s disease; HCHWA-I = hereditary cerebral hemorrhage with amyloidosis – Icelandic type; FAP/MVA = familial amyloid polyneuropathy/meningo-vascular amyloidoses; FAF = familial amyloidosis Finnish type; PrP-CAA = prion disease with cerebral amyloid angiopathy; FBD = familial British dementia; FDD = familial Danish dementia; APP = amyloid-β precursor protein gene; PS1 = presenilin-1 gene; PS2 = presenilin-2 gene; CYST C = cystatin C gene; TTR = transthyretin gene; GEL = gelsolin gene; PRNP = prion protein gene; BRI2 = BRI2 gene; A = amyloid- protein; ACys = amyloid-cystatin C; ATTR = amyloid-transthyretin; AGel = amyloid–gelsolin; APrP = amyloid–prion protein; ABri = amyloid-Bri; ADan = amyloid-Dan.

332

G. T. Plant, J. Ghiso, J. L. Holton, B. Frangione and T. Revesz

identical amyloid fibrils in a variety of disorders including those associated with dementia and CAA. In some instances a wild-type protein may be potentially amyloidogenic, but amyloid deposition only occurs if there is over-production or deficient metabolism. In other cases a single amino acid substitution may render fragments of a protein amyloidogenic or amyloid may be produced from a mutated elongated precursor protein. An important question in all of these conditions is the extent to which the amyloid deposition is itself responsible for the clinical features or whether it is epiphenomenal. In Gerstmann–Sträussler–Scheinker disease (GSS) for example, the disease may occur with or without prominent parenchymal amyloid deposition and CAA is rare. In cases with marked CAA there is a clear direct pathological consequence of the amyloid deposition – namely vascular damage – but in AD and in the BRI dementias the role of amyloidogenesis itself in the neurodegenerative aspects of the pathology is controversial: there may be an independent neurotoxic effect of the amyloidogenic proteins (which are also found in blood, CSF and tissue in non-amyloid or pre-amyloid forms) or there may be a parallel mechanism which leads to neurodegeneration. Such considerations are clearly of major importance in planning treatment strategies. Since the first edition of this book a larger number of hereditary disorders have been shown to be associated with CAA but the conditions in which it is especially significant remain AD, the Icelandic and Dutch varieties of hereditary CAA with cerebral haemorrhage and the two conditions now referred to as the BRI dementias.

Historical review Scholz (1938) described in the brains of elderly patients the deposition of amyloid within the media of cortical and meningeal blood vessels. He also coined the term ‘drusige Entartung’ to describe the appearance of the amyloid in the parenchyma, which appeared to have spread through the walls of the capillaries. (This is often erroneously cited ¨ as ‘drusige Entartung’, which would be translated as ‘glandular degeneration’ while the term derives from ‘Druse’, a geode, the same word is used to describe hyaline deposits at the optic nerve head and in the retina). Figure 14.10 illustrates the aptness of this description. Scholz’ interest was vascular disease but Divry (1927), working on the causes of dementia, had already reported the occurrence of amyloid in the plaques found in AD and senile dementia. This led him to the discovery of amyloid angiopathy in the same material (Divry, 1941). Other reports of CAA in cases of dementia followed and were considered to be either atypical

(Van Bogaert et al., 1940; Luërs, 1947; Corsellis & Brierley, 1954) or typical examples of AD (Benedek & McGovern, 1949). The original reports of the Familial British Dementia (Worster-Drought et al., 1940, 1944) are the earliest English language descriptions of CAA of any kind but the authors did not recognize it as such and described the vascular changes as ‘hyaline degeneration’. It was the discovery by Corsellis and Brierley (1954) of familial cases of dementia with CAA, and the similarity of the pathological changes in the vessels to the earlier descriptions in the French and German literature (Scholz, 1938; Divry, 1941) that led to a review of the pathological material from Worster-Drought’s cases and the appropriate stains were then carried out (see McMenemey, 1970). Interest in CAA as being of clinical significance increased in the 1970s when it was appreciated that it may be responsible for cerebral haemorrhage in patients without hypertension. One of the earliest reports was a familial example (Gudmundsson et al., 1972) but it became clear that sporadic cases were much more common and the debate was opened particularly following the influential review by Vinters (1987). It was recognized that CAA may account for 5–10% of primary cerebral haemorrhage and a much higher proportion of those which are lobar in elderly patients without hypertension. The significance of CAA in AD, either with respect to the pathogenesis or the clinical features is an area of current widespread research activity (see Chapter 9). One thing is certain, it provided an important technical breakthrough because it is easier to extract amyloid protein from vessels than from parenchymal plaques and it was in material from amyloid angiopathy that A was first identified in AD by Glenner and Wong (1984). The first amyloid protein to be identified in CAA had been ACys in HCHWA-I, reported in 1983 and, as will be seen in what follows, this approach has remained a powerful tool.

Morphological aspects of CAA The most common pattern of involvement in CAA is an acellular thickening of leptomeningeal arterioles and of small and medium sized cortical arterioles. However, in severe examples capillaries and veins can be affected and the changes can be extremely widespread. Positive staining of amyloid-laden blood vessels using Congo red and polarized light or Thioflavin S or T and fluorescence are specific as both methods depend upon the presence of beta-pleated sheet-rich structures. Vonsattel et al. (1991) have suggested a grading system for the severity of amyloid angiopathy. In ‘mild’ involvement amyloid is present in the media of

Cerebral amyloid angiopathies

Fig. 14.1. A sporadic case of cerebral amyloid angiopathy. (a) A cortical arteriole with thickened wall and fine spicules penetrating into the surrounding neuropil. Amyloid-laden blood vessels are stained with Congo red (b) with typical apple green birefringence in polarized light (c). (d–e) Progressive degenerative changes in amyloid-laden blood vessels are highlighted by corresponding loss of smooth muscle cells (a: haematoxylin and eosin, × 120; (b) and (c): Congo red, × 120; (d ), (e), ( f ) smooth muscle actin immunohistochemistry, × 120).

morphologically normal vessels; in ‘moderate’ CAA the smooth muscle cells are lost; and in ‘severe’ disease the vascular architecture is severely disrupted with ‘doublebarrelling’ (Fig. 14.9) and microaneurysm formation. It is in severe CAA that fibrinoid necrosis and evidence of perivascular leakage of blood is found. Ultrastructural studies have revealed filaments 10 nm in diameter arranged in a disorderly manner. The earliest deposits are seen in the outer aspect of the basement membrane around smooth muscle cells in smaller vessels and at the media–adventitia border in arteries. As the deposition progresses smooth muscle cells are lost and eventually the bood vessel wall may be entirely replaced by amyloid fibrils which may radiate into the surrounding parenchyma (Fig. 14.1).

cortical small arteries and arterioles) with a predilection for the occipital lobes and in both cases the amyloid protein is A and principally A40 with lower amounts of A42 (Gravina et al., 1995). The most important clinical consequence of sporadic CAA is cerebral haemorrhage and this applies also to AD where evidence of CAA related haemorrhage is seen in over 5% (Ellis et al., 1996). The ApoE ε4 allele is a risk factor for both sporadic CAA as well as AD-related CAA (for review see McCarron & Nicoll, 2000), while inheritance of the ApoE ε2 allele is associated with haemorrhage in both SCAA and AD (Greenberg, 1995). CAA-associated cerebral vasculitis has been reported (Ginsberg et al., 1988) as has Binswanger-type pathology. The evidence is against a correlation between the severity of the amyloid angiopathy and the degree of dementia (Greenberg et al., 2000) and as far as dementia is concerned the commonest cause of dementia in sporadic CAA is co-existent AD (see Chapter 9).

Sporadic forms of CAA Vinters and Gilbert (1983) found CAA in 36% of individuals over 60 years of age and 46% of those over 70. This is now referred to as Sporadic CAA (SCAA) but it is also associated with sporadic AD. CAA is present in over 80% of cases of AD and there is a correlation between the incidence of haemorrhagic and ischaemic lesions and the severity of the angiopathy (Ellis et al., 1996). Sporadic CAA and the CAA seen in AD are very similar in distribution (leptomeningeal and

Hereditary cerebral haemorrhage with amyloidosis Icelandic type (HCHWA-I) Clinical features ` Arnason in his doctoral dissertation (1935) reported a familial occurrence of cerebral haemorrhage in Iceland, and the pathological finding of CAA was demonstrated by

333

334


Gudmundsson et al. (1972) who described a pedigree of 116 individuals in which 18 had had cerebral haemorrhage. Seven pedigrees including information on progenitors born up to 200 years ago have more recently been described (Jensson et al., 1987) from small rural communities. All the families may have originated from an area surrounding a large fjord in the West of Iceland but cases are now found throughout Iceland. Dominant transmission with full penetrance has been observed although some carriers may have survived into old age without symptoms. The onset of the disease is relatively early – ranging from 20 to 40 years of age and first strokes proven to be due to CAA are rare in these pedigrees after the age of 50. Death at the first stroke is less common than in the Dutch families described below. Over half may make a good recovery after the first insult. In most there is a protracted course of multiple strokes and several years may elapse without incident. Progressive cognitive decline has sometimes been observed: such patients are usually bed-ridden or in a wheel chair ` at the time of the fatal stroke. Arnason (1935) had noted 3 of 10 patients in whom the progression of their illness was slow and dementia was more prominent than paralysis in those cases. Blöndal et al. (1989) reported on the pathological changes in 19 cases with dementia out of a total of 52 cases studied pathologically. In those 19 cases the duration of the dementia ranged from one to 17 years before death with the mean age of onset being 27.3 years. There was often a history of headache prior to the first vascular accident and the average number of strokes during the course of the disease was 3.2. Twenty five per cent of the patients developed epilepsy.

Neuropathology Neuropathological findings were first described by Gudmundsson et al. (1972) and later by Blöndal et al. (1989). Very extensive amyloid deposition in small arteries as well as arterioles in grey and white matter of the cerebrum, basal ganglia, brainstem and cerebellum underlies the propensity to multiple severe cerebral haemorrhage. The vascular changes are most pronounced in the leptomeningeal and cortical arterioles, where a double lumen, segmental fibrinoid necrosis and aneurysmal dilatation of arteries are common. Perivascular amyloid deposition is seen in the cortex, in the subpial layer and in the basal ganglia but neither argyrophilic plaque nor neurofibrillary tangle formation has been found in the brain. Multiple cortical areas of infarction and haemorrhage of different ages are apparent. The authors are not aware of any published reports describing any white matter pathology, or indicating whether or not there is spinal cord involvement.

Fig. 14.2. Hereditary cerebral haemorrhage with amyloidosis Icelandic type. Leptomeningeal and cerebral cortical blood vessels showing severe cerebral amyloid angiopathy due to deposition of mutated cystatin C (Cystatin C immunohistochemistry, × 70).

Amyloid chemistry and genetics The protease inhibitor cystatin-C (ACys) has been biochemically and immunohistochemically identified in the amyloid lesions (Cohen et al., 1983; see also Fig. 14.2). The deposited ACys amyloid contains a single amino acid substitution, glutamine for leucine, at position 68 (Ghiso et al., 1986) due to a point mutation in the CYSTATIN C gene (Levy et al., 1989), which is located on chromosome 20 (Abrahamson et al., 1989). This mutation has been found in a number of separate pedigrees (Abrahamsson et al., 1992). Cystatin C levels in the cerebrospinal fluid are low (Grubb et al., 1984; Shimode et al., 1991). Using immunohistochemical techniques the presence of the cystatin C has been demonstrated in skin (Blöndal et al., 1989) and more recently in other peripheral tissues such as lymph glands, which has been shown to be at least in part in the form of amyloid (Benedikz et al., 1990). It has therefore been suggested that ‘hereditary ACys amyloidosis’ may be a more accurate term but the use of HCHWA-I is likely to continue because the clinical features, the major pathological feature and the relationship to the Dutch cases are emphasized.


There have been some reports suggesting cystatin C reactivity in sporadic CAA coexisting with A protein reactivity associated with cerebral haemorrhage, (Fujihara et al., 1989; Maruyama et al., 1990) or leukoaraiosis (Vinters et al., 1990). Van Duinen et al. (1987) and Coria et al. (1987) did not find evidence of this in either aged or Dutch cases of CAA and Yamada et al. (1989) failed to find any evidence of cystatin C deposition in aged Japanese cases of sporadic CAA. More recently Levy et al. (2001) have shown evidence of colocalization of cystatin C with A in AD (both vascular and parenchymal) and the pathological importance of this finding is reinforced by the observation that a polymorphism in the cystatin C gene may be associated with a greater susceptibility to AD (Crawford et al., 2000). Therefore cystatin C may be involved in amyloid angiopathy and even parenchymal amyloid deposition in AD.

Hereditary cerebral haemorrhage with amyloidosis – Dutch type (HCHWA-D) Clinical features It has been recognized for some time that there is a familial occurrence of cerebral haemorrhage in the Netherlands and pedigrees from the villages of Katwijk & Scheveningen have been described (Luyendijk & Schoen, 1964; Luyendijk et al., 1986). Approximately 500 individuals in Holland are at risk of developing the disease. The presentation is usually between the ages of 45 and 60 years as an acute cerebral haemorrhage often with spread to the subarachnoid space giving rise to typical features: headache, vomiting and focal neurological deficit. Around two-thirds will die as a result of this first episode or soon after, but in others there may be a more prolonged and protracted course in which several minor strokes occur due to cerebral haemorrhage and such patients may indeed develop dementia. Pathologically it is known that cerebral infarction occurs as well as haemorrhage but it is recurrent haemorrhage that dominates the clinical picture in those patients who survive the first episode. Other associated features may be preexisting attacks of headache without neurological deficit, of uncertain origin, and some patients develop epilepsy late in the course of the disease. Other neurological features seem simply to reflect the residual damage from the acute episodes, there is no suggestion of a progressive spasticity or ataxia as is seen in the British family described below. There is also a suggestion that there may be a gradual decline in cognitive functions in some cases and dementia may very occasionally be the presenting feature of the disease (Wattendorf et al., 1982; case 3 of Haan et al., 1989).

The existence and characteristics of dementia in this disease is of considerable interest because of the pathological similarities with and differences from AD, (reviewed below) but it is clearly difficult to decide whether or not dementia has occurred as a result of any process other than the recurrent strokes. Haan et al. (1990b) examined 16 patients between 6 months and 12 years after their first cerebral haemorrhage and found cognitive impairment in 75% of the cases. The presence of dementia showed a correlation with the number of focal lesions on CT scan but not with white matter hypodensity. In two cases serial psychological testing revealed progressive deterioration without evidence of further strokes or the acquisition of new lesions on CT scanning. It was not possible to decide on the basis of neuropsychological features whether a dementia of a type occurring in AD was present in addition to a ‘multi-infarct’ dementia but the major contribution to the dementia was from the vascular lesions.

Neuro-imaging A CT study of 24 patients (Haan et al., 1990a) revealed 99 focal lesions in 59 scans of which 13% were ischaemic and 87% haemorrhagic. Cerebellar and brainstem lesions were not seen. A comparison with lobar haemorrhage in other conditions led to a conclusion that the majority of these lesions were likely to be primary cerebral haemorrhage rather than haemorrhagic transformation of ischaemic strokes, but it was noted that the shape of the haemorrhages was more often irregular rather than the smooth, rounded shape of cerebral haemorrhage due to other causes, and a progression of symptoms after the acute onset was more likely to occur. The haemorrhages in the Dutch families are often multiple. A comparison was also made with cases of sporadic CAA presenting with cerebral haemorrhage. In HCHWA-D the haemorrhages tend to spare the frontal lobes whereas this was the most frequent location of the haematomas in sporadic CAA and this correlates with the differences in the distribution of the amyloid angiopathy in the two conditions. Periventricular white matter lesions were seen in both demented and non-demented patients in the CT study and any contribution to the dementia was uncertain. However, MR imaging in a smaller group of patients showed periventricular white matter changes in all cases (Haan et al., 1990c). The MR studies also demonstrated the different ages of the multiple haemorrhages, most patients having a number of lesions showing different signal characteristics reflecting the age of the lesions (Fig. 14.3). Bornebroek et al. (1996) have reviewed more recent data concerning the clinical and radiological aspects of HCHWA-D and point out

335

336


Fig. 14.3. The upper two panels show magnetic resonance (MR) images showing a left temporo-parietal haemorrhage in a case of hereditary cerebral haemorrhage (Dutch type). There is high signal intensity both in the T1 - and T2 -weighted images (left and right panels, respectively). The haemorrhage had occurred two weeks prior to the scan. The frequency of haemorrhages in this condition is equally distributed over both hemispheres sparing basal ganglia and cerebellum. The lower three panels are MR images of another case of the disease showing multiple haemorrhages of different ages. These show as low intensity regions in the right parietal and left frontal lobes on the T1 -weighted image (left panel) and as high intensity on the T2 -weighted image (centre). Another old haemorrhage is seen high in the right parietal lobe in the T2 -weighted image on the right. This is surrounded by a rim of low signal, indicating the presence of haemosiderin. In addition to the evidence of previous haemorrhage, there is white matter high intensity in the central panel indicating leukoaraiosis. (From Haan et al., 1990c).

that dementia can preceed the first stroke and that White Matter Hyperintensities (WMHs) are seen on MR before a clinical event such as a stroke has occurred.

Neuropathology Luyendijk and Bots (1980) first demonstrated that the disease was due to extensive amyloid deposition in small leptomeningeal and cortical (both cerebral and cerebellar) arterioles. A number of conventional neuropathological studies have been reported (Wattendorf et al., 1982; Luyendijk et al., 1986, 1988; Van Duinen et al., 1987). Brains were of normal or slightly reduced weight and tended to show thickened leptomeninges externally. Cortical atrophy was not marked. Fixed brain slices showed recent large haemorrhages and remnants of older ones, mainly in the subcor-

tical white matter. In cortex and subcortical white matter recent and old haemorrhagic infarcts were found. These lesions were predominantly located in the temporal, parietal and occipital regions. Extracranial arteries and arteries at the base of the brain showed no marked abnormality. Maat-Schieman et al. (1996) have recently reviewed the histopathological aspects of HCHWA-D. Histologically, in addition to the presence of multiple vascular lesions there are small to moderate numbers of cortical A peptide deposits resembling the diffuse plaques seen in AD. These deposits were missed in the earliest studies because only more sensitive immunohistochemical staining procedures demonstrated them. Dense plaque cores and neurofibrillary tangles are not seen. Although dystrophic neurites are not prominent some ubiquitin positive structures suggestive of dystrophic neurites are occasionally


present in the diffuse amyloid deposits (Tagliavini et al., 1993). Cortical and hippocampal neuronal populations outside the infarcts appear well preserved. Cerebral and cerebellar white matter show in varying degrees oedema and demyelination with sparing of the ‘U’ fibres.

Amyloid chemistry and genetics Van Duinen et al. (1987) examined six brains immunohistochemically using an antibody raised against a region of the A peptide of AD and Down’s syndrome. Both the amyloid angiopathy and the plaque-like lesions showed positive staining (see also Fig. 14.4). The amyloid protein was also isolated from leptomeningeal vessels and amino acid sequencing has shown it to be A (Van Duinen et al., 1987). Coria et al. (1987) showed that the positive immunostaining spread out of the blood vessels into the surrounding parenchyma and found the same immunoreactivity in sporadic CAA and CAA due to aging and coined the term ‘the -amyloid diseases’ (‘BAD’). A deposits in blood vessel walls were largely confined to leptomeningeal small arteries and cortical arterioles. Among the cortical arterioles affected were the deep penetrating vessels that supply the white matter. The narrowed lumens of these vessels could plausibly account for the incomplete, diffuse infarction in the white matter. Vascular amyloid deposits, but not the diffuse parenchymal plaques, reacted not only for A but also for its precursor protein (Tagliavini et al., 1990). This suggests that the vascular compartment may be the main source of A that accumulates in the brain in this disease. Parenchymal deposits are in the main a ‘halo’ around the vessel, a feature less constantly seen in AD. A point mutation has been identified in the gene encoding the A precursor protein (AβPP gene) at codon 693, causing a single amino acid substitution of glutamine instead of glutamic acid at position 22 in the A peptide (Levy et al., 1990). The vascular amyloid deposits in HCHWA-D are composed of both the wild type A peptide and the A-Q22 variant (Prelli et al., 1990) which has been shown both to form amyloid fibrils at an accelerated rate (Wisniewski et al., 1991) and to be toxic to cerebral smooth muscle (Davis et al., 1999) and endothelial cells (Miravalle et al., 2000). Maat-Schieman et al. (2000) and Yamagucchi et al. (2000) have recently studied A deposition in the frontal cortex of 24 cases of HCHWA-D. A42 immunostaining was positive in all parenchymal deposits – classified as clouds, fine diffuse and dense diffuse deposits and as coarse and homogeneous plaques. A40 reactivity was observed principally in homogeneous plaques and, in a subset of these, degenerating neurites without tau deposition or amyloid cores were seen. Amyloid fibrils (identified by electron

Fig. 14.4. Hereditary cerebral haemorrhage with amyloidosis Dutch type. Deposition of A in blood vessels and diffuse plaques. (A immunohistochemistry, × 40).

microscopy) were not present in fine diffuse plaques. Small bundles of fibrils were identified in dense diffuse and homogeneous plaques and amyloid masses in coarse plaques. ‘Clouds’ were seen only in young patients, coarse plaques only in the most aged and the ratio of fine to dense diffuse plaques decreased with age. As plaque density did not increase with age the authors suggested that this represents a model of plaque evolution. A42 deposition on the cell surface plasma membrane appears to be an initial event in diffuse plaques developing into amorphous/fibrillary amyloid between cell processes. Vascular amyloid was composed of some pure A42 deposits – assumed to be recent – but mostly stained for both species.

Other examples of familial A CAA The mutation of the AβPP gene underlying HCHWA-D discussed above was the first mutation described in that gene. It is within the region encoding for the A peptide. Susbequent studies on other A related familial disorders have shown that the clinical phenotype of those mutations found within the coding region of A includes strokes while those outside this region are associated with

337

338


a more typical AD-type clinical picture (Ghiso & Frangione 2001; Revesz et al., 2002). Thus other mutations which affect codons 692–694 of the AβPP gene have been described which are associated with extensive CAA. The following are recently described mutations in which the clinical phenotypes are characterized by dementia, cerebral haemorrhage and leukoenceophalopathy in differing proportions all showing prominent CAA: the A692G Flemish mutation (Hendriks et al., 1992); the E693K (Italian) mutation (Tagliavini et al., 1999); the E693G (Arctic) mutation (Nilsberth et al., 2001); and the D694N (Iowa) mutation (Grabowski et al., 2001). Severe CAA can also occur in pedigrees of AD due to mutations in the presenilin-1 (PS1) and in the presenilin-2 (PS-2) genes. It has been suggested that if mutations occur after codon 200 of the PS1 gene CAA is prominent in the phenotype (Mann et al., 2001). In cases with 9 and I83/M84 mutations of the PS1 gene the phenotype is a variant AD with spastic paraparesis and cotton wool plaques together with extensive CAA (Crook et al., 1998; Houlden et al., 2000). There are in the older literature examples of familial disorders in which CAA is found, some others may also be examples of familial AD and were often described as such ¨ in the original reports (Van Bogaert et al., 1940; Luers, 1947, cases 1 and 2 (siblings); Corsellis & Brierley, 1954, Case 1; Gerhard et al., 1972; Aikawa et al., 1985). In addition to dementia these cases had spasticity and sometimes ataxia with death occurring in the range 40–60 years. Indeed a recent review of the pathology in Family CO originally described by Van Bogaert et al. (Van Bogaert et al., 1940) showed that the characteristic changes included cotton wool plaques suggesting that the disease in this family represents variant AD (Houlden et al., 2000).

Familial British dementia (FBD; familial BRI dementia-British type) Clinical features This disorder was first described by Worster-Drought and colleagues as discussed above. In subsequent publications the disorder was referred to as an example of atypical AD (e.g. Aikawa et al., 1985); as an example of GSS (Masters et al., 1981; Baraitsar, 1990); as an atypical form of GSS (Courten-Myers & Mandybur, 1987; Baraitsar, 1990) and as a form of primary CAA classified with the Icelandic and Dutch types of HCHWA, (Vinters, 1987); and with the hereditary spastic parapareses (Baraitsar, 1990). In the years following the original Worster-Drought cases a number of similar families were described. Griffiths et al.

(1982) described very similar pathological changes in two siblings from Oxford, England and following a review of the literature suggested that they were dealing with a disorder that was ‘clearly the same’ as that which had affected the brother and two sisters described by Worster-Drought. Their conclusion was affirmed by Plant et al. (1990), who demonstrated that both sets of siblings were descended from a common ancestor, a British woman who died in 1883. The link between the two families was established by consulting a family history taken by a House Physician in 1924 when one of the affected family members was admitted to the National Hospital for Nervous Diseases in Queen Square under the care of Dr MacDonald Critchley. In 1982 an autopsied case of ‘familial cerebellar ataxia with amyloid angiopathy’ was published from the National Hospital, although in life the case had been under the care of St Bartholomew’s Hospital (Love & Duchen, 1982). We have long suspected that this patient suffered from the same disease – despite there being no mention of dementia in the case report – and said as much in the first edition of this book. The descendents of the Love and Duchen case have provided a family tree which we were able to trace back to a couple born around 1780. One of that individual’s 11 children is a common ancestor to the Love and Duchen pedigree and another is a common ancestor to the WorsterDrought pedigree. We have also established from the family that dementia was a feature of that case’s illness, although not recorded in the Love and Duchen paper (Mead et al., 2000). A pedigree of 372 individuals has so far been published (Mead et al., 2000) but at the time of writing many more descendents have been identified. A second pedigree, in which just two siblings are affected, has not so far been linked to the main Worster-Drought family (Mead et al., 2000). We have information on over 50 individuals at risk of developing the disease. Descendants are living in Australia, New Zealand, South Africa, Canada and the United States as well as in the United Kingdom. The descendants of two children of the common ancestor (III 5 and 6 in Fig. 1 of Plant et al., 1990), who both died of the disease in the 1930s (and had between them 15 children) have not been traced. It is highly likely that other affected and at risk individuals exist among their descendants. The disorder is so distinctive that we believe these cases will come to light before long. Clinical details of 6 living affected patients and 35 historical cases have been analysed. The median age of onset of symptoms is 48 years (range 40–60) and the median age of death 56 years (range 48–70). Mode of inheritance is autosomal dominant and penetrance complete by the age of 60. Psychometric assessment early in the course of the disease has shown marked impairment of memory progressing ultimately to a global dementia. Many of the cases developed


Fig. 14.5. Moderately T2 -weighted image of a case of familial British dementia. Extensive leukoaraiosis is shown as periventricular high intensity. White matter changes are more extensive in this disorder than in HCHWA-I and HCHWA-D and evidence of cerebral haemorrhage is rare. Familial Danish dementia has very similar MR scan appearances.

personality change as an early manifestation, becoming irritable or in some cases depressed. The spastic paralysis is far more profound than is seen in GSS or in AD. Pseudobulbar palsy and dysarthria are universal and all patients have progressed to a chronic vegetative state; mute, unresponsive, quadraplegic and incontinent. The question of the relative contributions of the vascular pathology and the plaque formation to the clinical course is uncertain but only 26% of the cases have sustained stroke-like episodes and there has been only one example of a catastrophic cerebral haemorrhage (Case 2B of the second pedigree, Mead et al., 2000). One case collapsed in the street and was unconscious for 24 hours, she never left hospital and the story was a slowly progressive one from then on but the initial event was certainly compatible with an acute infarct or cerebral haemorrhage and there was some evidence for minor haemorrhage in the autopsy studies. As well as nystagmus some cases developed other brainstem symptoms such as diplopia. One had a craniotomy because she was thought to have a posterior fossa tumour. Mead et al. (2000) carried out a neuropsychological assessment in 12 at-risk individuals who were not symptomatic. Two subjects showed evidence of marked impairment of recall and recognition memory with no evidence of a decline in general intelligence, naming problems, frontal lobe dysfunction nor perception. The most consistent impairment in presymptomatic cases was in delayed-recall memory tasks.

Neuro-imaging Figure 14.5 shows axial images of a magnetic resonance scan of an affected individual at the age of 60, three years following the onset of symptoms. The major features are the moderate ventricular dilatation and relatively preserved cortex with extensive periventricular white matter changes compatible with leukoariosis. We had been aware that symptomatic cases at presentation had dramatic changes on neuro-imaging of this type and suspected that MRI changes would be an early feature of the disease. This was confirmed in the study of presymptomatic, at risk individuals carried out by Mead et al. (2000). White matter hyperintensities (WMHs) were found distributed throughout the cerebral white matter in such cases (Fig. 14.6). Evidence of lacunar infarction was found in some cases but no evidence of haemorrhage. The corpus callosum was affected by punched out lesions on T1 -imaging and atrophy. The white matter changes become confluent in more severely affected cases.

Amyloid chemistry and genetics The greatest advance since the first edition of this book has been the identification of the amyloid protein and underlying genetic defect in this disorder. When suitable pathological material from the British family became available in the mid 1980s immunohistochemical studies were carried

339

340


Fig. 14.6. Early changes are shown on presymptomatic cases of familial British dementia. The top panel shows T2 -weighted axial images, there are patchy ischaemic changes in white matter. Later these changes become confluent (see Fig. 14.5). Corpus callosum lesions are seen early (bottom).


Fig. 14.7. The amyloidogenic peptides ABri and ADan are here compared. The two different mutations each extend the wild type BRI protein (top) by 11 amino acids to form the two precursor proteins ABriPP 277 and ADanPP 277. Why the break occurs between amino acids 243 and 244 is not known but the result in each case is a peptide of 34 amino acids. (Fig. 7 in Vidal et al., PNAS 97:4920–4925, 2000, reprinted with permission. Copyright (2000) National Academy of Sciences, USA).

out but failed to identify the nature of the amyloid deposits using antibodies specific for a panel of amyloidogenic molecules (Ghiso et al., 1995). The amino acid sequence of the amyloid subunit was eventually obtained by combining partial sequence data retrieved from internal peptides generated via trypsin digestion and homology searches in EST data banks. The amyloid subunit so identified consists of 34 amino acids with no sequence identity to any known amyloid molecule. This peptide has been designated ABri to identify it by its association with the British family from which the isolation was originally carried out. The EST data bank indicated the presence of a larger precursor protein and this has been cloned and

sequenced (Vidal et al., 1999). The gene BRI located on the long arm of chromosone 13 encodes a 266-amino acid protein. The gene encoding the precursor protein was then sequenced in a number of members of the British family and it was shown that the disease results from a single nucleotide substitution in the stop codon at codon 267 which becomes AGA instead of TGA. The open reading frame is increased to 277 amino acids instead of 266 and the ABri amyloid peptide is formed from the last 34 amino acids of the mutated precursor protein ABriPP277 (Fig. 14.7). Following this discovery it was decided to rename the disease familial British dementia to emphasize the fact

341

342


recognized by Worster-Drought and colleagues in the original description that the disease is a primary dementia as well as showing severe CAA. We wished to distinguish the disorder from the other familial and sporadic conditions in which CAA occurs but in which dementia occurs largely or entirely as a result of recurrent haemorrhage or ischaemia. However, it was clear that, if other disorders resulting from different mutations of the BRI gene were described, it would not be appropriate to refer to them all as familial British dementias and it would be preferable to clasify any such disorders as BRI dementias. Three isoforms of the BRI gene are now described and it is the BRI2 gene on chromosome 13 that is mutated in FBD (Vidal et al., 2001).

Neuropathology Detailed neuropathology has been reported in a number of cases of familial British dementia (Worster-Drought et al., 1940, 1944; McMenemey 1952, 1970; Griffiths et al., 1982; Love & Duchen, 1982; Plant et al., 1990; Revesz et al., 1999; Mead et al., 2000; Holton et al., 2001). Although the pathological changes were originally thought to be confined to the central nervous system (Plant et al., 1990), deposition of the amyloid protein ABri in systemic organs has been added to the list of morphological abnormalities that occur in FBD (Ghiso et al., 2001). The brains are of normal or slightly reduced overall weight with leptomeningeal thickening and mild to moderate diffuse atrophy of both the cerebral and cerebellar hemispheres. The cerebral cortex is relatively well preserved in slices of the fixed brain but the white matter is reduced in bulk, diffusely discoloured and contains small cystic infarcts (Fig. 14.8). The brainstem and spinal cord, when examined, appear normal to the naked eye. Histologically FBD is remarkably similar to AD as the pathological changes include severe, widespread CAA, parenchymal amyloid plaques and neurofibrillary degeneration (Fig. 14.11(a)–(e)). Ischaemic white matter damage is also described in the majority of the FBD cases examined pathologically. CAA affects blood vessels in the leptomeninges, both grey and white matter throughout the central nervous system with only a few anatomical areas remaining unaffected (Holton et al., 2001). The majority of the blood vessels affected by CAA are smaller than 300 m. Degenerative changes of the blood vessels associated with CAA, including double barrelling or complete obstruction, are seen (Fig. 14.9). Occasional small perivascular haemorrhages or slight inflammation may be seen around the affected vessels and spicules of amyloid radiating from capillaries into the surrounding neuropil (drusige

Fig. 14.8. British-type CAA: macroscopic view of brain section to show leukoencephalopathy and small white matter infarcts.

Entartung) are common (Fig. 14.10). The blood vessels of the retina also show severe CAA. All vessels affected by amyloid deposition stain positively with an antibody recognizing ABri, which also reveals that ABri frequently deposits in perivascular plaques around blood vessels with CAA (Fig. 14.11(c)–(d )). Such perivascular plaques occur in many areas of the CNS and are particularly prominent in the cerebellar cortex, which shows evidence of severe degeneration together with evidence of previous ischaemia (Fig. 14.11(b)). Argyrophilic amyloid plaques, strongly labelled with an anti-ABri antibody, are most commonly found in limbic areas (Fig. 14.11(a)), while silver and Congo rednegative, ABri-positive ‘diffuse deposits’ occur in several regions including the entorhinal cortex and fusiform gyrus where they are the main parenchymal lesion type (Holton et al., 2001). The amyloid plaques are either large, about 150 m across, with or without a central congophilic core or are relatively small, with appearances like the cores of the large plaques without a rim (Figs. 14.12 and 14.13). There is severe neurofibrillary degeneration in FBD and the tau pathology shows a close topographic association with both fibrillar and non-fibrillar ABri deposition. Originally,


Fig. 14.9. CAA in FBD: amyloid angiopathy to show double lumen formation. PAS × 200.

the amyloid plaques were described as ‘non-neuritic’, but both distended dystrophic neurites occurring around small and large plaques and finer neuritic profiles permeating large amyloid plaques could be demonstrated by recent studies using more sensitive tau immunohistochemistry (Revesz et al., 1999; Holton et al., 2001) (Fig. 14.11(E )). Abnormal neurites are also found around some of the blood vessels with CAA. In FBD neurofibrillary tangles are ultrastructurally composed of paired helical filaments and immunoblotting of insoluble tau extracted from one affected brain has shown an electrophoretic migration pattern identical to abnormal tau in AD (Revesz et al., 1999; Holton et al., 2001). The severe white matter changes are similar to those seen in Binswanger’s encephalopathy and are considered to be secondary to the amyloid deposition in arterioles penetrating through the cortex to supply the white matter.

Fig. 14.10. CAA in FBD: Drusige entartung. Electron micrograph (× 8400) of an arteriole showing a thickened wall composed of amyloid fibrils extending from the capaillary wall into the parenchyma. The inset shows a cortical capillary with amyloid (PAS) showing again the extension of the amyloid into the parenchyma.

Familial Danish dementia (FDD; heredopathia ophthalmo-oto-encephalica; familial BRI dementia-Danish type) Clinical features Strömgren (1981) described 9 cases in 3 generations of a dominantly inherited disorder originating in the region of ˚Arrhus in Jutland, Denmark. The clinical picture is unique, there is early (around age 20) cataract formation followed by deafness (around 30) and a progressive ataxia and dementia beginning in the fifth and sixth decades. Progression to death occurs less than 10 years following the onset of neurological symptoms. Stroke-like episodes are rare and once the neurological disease becomes established the clinical manifestations are very similar to the British family described above. Some patients have recurrent intra-ocular

343

344


Fig. 14.11. Familial British dementia. Numerous hippocampal (a) and cerebellar (b) ABri deposits are characteristic. (c) An affected leptomeningeal vessel showing ABri deposition with enhanced peripheral staining. (d ) A small cerebral parenchymal vessel with staining of the vessel wall and also surrounding parenchyma. (e) Tau immunohistochemistry showing a neurofibrillary tangle (double arrow), neuropil threads (arrow) and ABri plaque-associated abnormal neurites (arrow head). (a)–(d) ABri immunohistochemistry, (a) × 6, (b) × 30, (c), (d ) and (e): × 200). (Reproduced with permission from the American Journal of Pathology, Holton et al. 158:515–516, 2001).

haemorrhage, before the neurological disease becomes apparent.

Neuro-imaging Neuro-imaging studies in advanced cases show extensive white matter ischaemic change similar to the British family (Fig. 14.14).

Amyloid chemistry and genetics Appreciating as we did the similarities between this condition and the Worster-Drought pedigree no time was lost in sequencing the BRI gene in this family. FDD is associated with a 10-nt duplication (TTTAATTTGT) occurring between codons 265 and 266 of the BRI2 gene (Vidal et al., 2000; Fig. 14.7). This also results in an out of frame stop codon and an extended precursor protein

(ADanPP277). The constituent amyloid protein, designated ADan, matched exactly the last 34 amino acids of this mutated precursor. Thus it can be seen, illustrated in Fig. 14.7, that the two mutations which give rise to ABri and ADan both result in precursor molecules (ABriPP277 and ADanPP277 respectively) which are 277 instead of 266 amino acids in length and both give rise to peptides which are formed by cleavage between codons 243 and 244, are 34 amino acids in length, have an identical N-terminal amino acid sequence (22 amino acids) but completely different C-terminal sequences (12 amino acids). We are now left with the challenge of explaining both the similarities and differences between these two familial conditions, which we are referring to as the BRI dementias, based upon the biological properties of the two 34 amino acid peptides ABri and ADan. Should more BRI gene mutations be identified it would be preferable to refer to familial BRI dementia – British type, familial BRI


Fig. 14.12. CAA in FBD: large non-neuritic amyloid plaque, cerebellum. There is a central core PAS × 300.

dementia – Danish type, etc. Of particular importance in terms of studies which may increase our understanding of mechanisms of neurodegeneration such as animal models is the fact that neither of these two peptides have any sequence in common with any known naturally occurring peptide. This unique feature of this pair of neurodegenerative disorders may also have significant implications in the development of treatment strategies, such as those involving vaccination.

Neuropathology The first published report (Strömgren, 1981) noted accumulation of lipid in the vessel walls and parenchyma but, repeating the earlier Worster-Drought story, amyloid stains were not carried out. However, recent studies have established that there is extensive CAA. Detailed neuropathological data are available in only a small number of FDD cases (Holton et al., 2002). In these cases the brain weight is slightly reduced and the ventricles are enlarged. The cortical ribbon may show thinning and the subcortical white

Fig. 14.13. CAA in FBD: small non-neuritic amyloid plaques, hippocampus. There is no central core. PAS × 400.

matter, hippocampus, brainstem and cerebellum are reduced in bulk. The major histological features of FDD are similar to those seen in FBD and these include widespread CAA, labelled with antibodies to ADan, and neurofibrillary degeneration (Fig. 14.15(a)–(e)). There are, however, also interesting differences in the neuropathological abnormalities between these two closely related diseases, which include that, in contrast to FBD, in which the extracellular ABri deposits form amyloid plaques, in FDD the numerous hippocampal parenchymal ADan deposits are silver and Congo red-negative suggesting that ADan is primarily in pre-amyloid (non-fibrillar) rather than amyloid (fibrillar) conformation in these lesions. In FDD, like in FBD, the neurofibrillary pathology, including neurofibrillary tangles, neuropil threads and abnormal neurites, is severe in the limbic structures, and is also present in neocortical areas. Abnormal neurites mainly cluster around CAA, and are absent around non-fibrillar ‘diffuse’ ADan parenchymal deposits (Fig. 14.15(d ), (e)). Immunohistochemical, electron microscopic and biochemical analysis of neurofibrillary tangles confirms that the cytoskeletal pathology in FDD is comparable to that seen in FBD

345

346


deposition, but less prominent glial response in areas where ABri and ADan deposit as preamyloid). As has been described in AD the vascular and parenchymal amyloid lesions in both FBD and FDD contain amyloid-associated proteins and also components of the classical and alternative complement pathways including C1q, C3d, C4d, Bb and C5b-9 suggesting in situ complement activation (Rostagno et al., 2002; Revesz et al., 2002). The retinal involvement in FDD presumably accounts for the intraocular haemorrhage in some cases and this feature, together with the deafness and cataract clearly distinguish the disease from all of the families described above. The eye is involved in a number of hereditary amyloidoses – corneal lattice dystrophy occurs in the Finnish type of familial amyloid polyneuropathy first described by Merejota (1969) in which the amyloid protein is gelsolin. Vitreous opacities occur in various TTR amyloidoses and ocular microangiopathy has been described in TTR amyloidosis (Ando et al., 1992). Since the first edition of this volume both TTR and gelsolin families have been described with prominent CAA, now to be discussed.

CAA and transthyretin

Fig. 14.14. T2 -weighted MR scan from an advanced case of familial Danish dementia showing extensive white matter ischaemic changes.

and AD. The retinal changes with marked ADan amyloid angiopathy and parenchymal damage, are more severe in FDD than in FBD (Holton et al., 2002). Immunoelectron microscopic examination of the vascular and parenchymal lesions confirms that ABri and ADan deposit in both fibrillar (amyloid) and non-fibrillar (preamyloid) configurations. In all the cases of FDD which have so far been examined neuropathologically, there is also deposition of variable amounts of A peptide in blood vessels and brain parenchyma, which can occur either in isolation or in combination with ADan deposition (Fig. 14.15(i), ( j)). The A parenchymal deposits reflect the deposition of ADan in that they are most frequent in the limbic structures but they are also found in the neocortex where they exceed ADan deposits. The mechanism and significance of such a codeposition of ADan and A remains to be elucidated. In both FBD and FDD there is a striking microglial and astrocytic response to vascular and parenchymal amyloid

A number of systemic late onset autosomal dominant disorders have been described due to mutations of the transthyretin (TTR) gene located on chromosome 18 (Benson, 1996). In one example of a familial oculoleptomeningeal amyloidosis, due to the deposition of a variant TTR, the involvement of meningeal vessels has led to cerebral haemorrhage and myelopathy but not specifically to dementia (Uitti et al., 1988). A Japanese family (the ‘K’ family) has been reported with TTR related type 1 familial amyloid polyneuropathy (FAP) in which there was, in addition to a sensorimotor and autonomic peripheral neuropathy, cerebellar ataxia and pyramidal tract signs without dementia or stroke (Ikeda et al., 1989). The leptomeningeal and pia-arachnoid vessels were principally involved and the brain parenchyma spared. Most vessels were observed to be free of amyloid shortly after entering the cerebral cortex (Ushiyama et al., 1991). Characteristically the amyloid is deposited in the pia-arachnoid membranes and choroid plexus. CAA of this type may in fact be common in type 1 FAP even without evidence of CNS involvement (Ushiyama et al., 1991) and the variant TTR has been identified in the amyloid fibrils in the ‘K’ family (Kametani et al., 1992). More recently two pedigrees have been described in which dementia is a prominent feature of the phenotype. In the Hungarian (D18G; Garzuly et al., 1996; Vidal et al., 1996) and Ohio (V30G; Petersen et al., 1997) families there


Fig. 14.15. Familial Danish dementia. (a)–(c) Hippocampal blood vessels showing cerebral amyloid angiopathy. (d) and (e) Numerous neurofibrillary tangles (double arrow), neuropil threads (arrow) and abnormal neurites (arrow head) clustering around amyloid laden blood vessels (asterix) are present in the hippocampal formation. ( f ) Structures containing ADan outline the hippocampus, which are either plaque-like (arrow) or ill-defined and confluent. (g) Cerebellar leptomeningeal and parenchymal blood vessels containing (h) ADan amyloid (arrowhead: subpial deposition, arrow: vascular deposition, double arrow: perivascular deposition of ADan). Deposition of A peptide in temporal neocortex (i) and hippocampus ( j ). The whole circumference of some of the blood vessels is stained (arrow), while only an outer rim is stained in some others (double arrow). (a) and (g) PAS, × 30, (b) and (c) Congo red, × 120, (d ) Bielschowsky’s silver, × 60, (e) AT8 immunohistochemistry, × 60, ( f ) and (h) ADan immunohistochemistry, ( f ) × 5, H: × 30, (i) and ( j ) A immunohistochemistry, (i) × 15, ( j ) × 60). (Reproduced with permission from the Journal of Neuropathology and Experimental Neurology, Holton et al. 61:254–267, 2002).

347

348


(CJD), kuru, variant CJD and the Gerstmann–Sträussler– Scheinker syndrome (GSS) – CAA is distinctly uncommon. CAA was noted in one case in six members of the ‘W’ family (Adam et al., 1982) and care must be taken in attributing CAA in GSS and Creutzfeldt–Jakob disease to PrP deposition without immunohistochemical confirmation as A CAA may coexist with sporadic and familial spongiform encephalopathy (Roberts et al., 1988; Tateishi et al., 1992; Watanabe & Duchen, 1993). There is now definitive documentation of deposition of PrP within blood vessels causing CAA in one family with GSS, characterized by a T to G mutation at codon 145 of the PRNP gene resulting in a premature stop codon and a shorter PrP product (Ghetti et al., 1996).

Fig. 14.16. Familial meningocerebrovascular amyloidosis, Hungarian type. Amyloid deposition in spinal leptomeninges and both leptomeningeal and parenchymal blood vessels (haematoxylin and eosin, × 70). Inset showing amyloid deposits staining positive for transthyretin (× 70). (Courtesy of Ferenc Garzuly and Thomas Wisniewski.)

is marked CAA involving the brain parenchyma (Fig. 14.16). Indeed in the Hungarian family the peripheral nerves, peripheral organs and the eye are spared.

CAA and gelsolin Gelsolin is an actin binding protein and the gene is found on chromosome 9. Gelsolin amyloid fibrils are composed of gelsolin fragments spanning positions 173–243 or 173–225 with amino acid substitutions which are highly amyloidogenic (Ghiso et al., 1990; Levy et al., 1990b; Maury & Baumann, 1990, de la Chapelle et al., 1992). These disorders are known collectively as Familial amyloidosis – Finnish type or FAF). It is now clear that in the gelsolin amyloidoses the deposition of gelsolin amyloid occurs in basement membranes and amyloid angiopathy can be seen in a large number of organs including the brain. Kiuru et al. (1999) have studied Finnish cases with the G654A mutation and found extensive CAA in the meninges, brain and spinal cord. To our knowledge, dementia has not been a feature in any FAF families.

CAA and prion protein (PRP) The prion diseases are considered elsewhere in this volume (Chapter 17). Whilst many of the human prion diseases are associated with dementia – Creutzfeldt–Jakob disease

Differential diagnosis and neuropathological recommendations These conditions should be considered particularly in cases in which there is a positive family history of dementia and in which dementia is accompanied by ataxia and spasticity (Masters & Beyreuther, 2001). The differential diagnosis includes other familial causes of dementia, the most common of which is AD (Chapter 9), but frontotemporal dementias including Pick’s disease (Chapter 11), a wide range of metabolic diseases (Chapter 23) and GSS (Chapter 17) should also be considered. Only the latter condition and variant AD with spastic paraparesis (Crook et al., 1998; Houlden et al., 2000) are likely to closely mimic clinically FBD and FDD. Cases with stroke-like episodes need to be distinguished from other causes of ischaemic dementia and occasionally normal pressure hydrocephalus or paraneoplastic syndromes may overlap clinically. When the naked eye appearances of the brain in a case of familial CAA are considered, the most likely differential diagnosis includes other forms of vascular dementia. If possible, the spinal cord as well as the brain should be examined, since this is unlikely to be abnormal (with the exception of Wallerian degeneration) in other forms of vascular dementia. Histological sampling needs to be widespread, including multiple areas of cerebral cortex, hippocampus, cerebral white matter, basal ganglia, thalamus and multiple levels of the brain stem, cerebellum and spinal cord. It is essential to carry out stains for amyloid if the true nature of these conditions is not to be overlooked and it is, in fact, recommended that an amyloid stain be carried out on all cases of vascular dementia (Chapter 13). Once the amyloid has been detected, although the morphology of any parenchymal amyloid deposits considerably assists in distinguishing one form of familial CAA from another, its biochemical nature should be investigated with immunostains for amyloid


proteins including A, ACys, ATTR, AGel, PrP, ABri, ADan and λ and κ light chains. In conclusion, these rare familial conditions cause dementia partly on the basis of neurodegeneration (for example, AD, FBD and FDD) and partly as a result of recurrent focal ischaemia, recurrent haemorrhage or leucoaraiosis or a combination of these changes. Haemorrhage seems to predominate in those families in which the CAA is less extensive, affecting mainly the leptomeningeal and cortical vessels (HCHWA-I and -D). When large and small vessels throughout the brain are involved, the vascular consequence is leukoaraiosis rather than haemorrhage (FBD and FDD). This distinction cannot be attributed to the sparing of larger vessels in the British and Danish families as vessels of all sizes are extensively involved. Although rare in themselves, the identification and study of familial CAA is proving of great importance in increasing our understanding of common neurodegenerative conditions including AD.

Acknowledgements The work described in this chapter which has been carried out by the authors was supported by NIH grants AG05891 and AG08721, by grants from the Alzheimer Association, the Brain Research Trust and the CRDC of RF and UCMS/UCLH. We are also grateful to Hans Braendgaard and Marie Bojsen-Møller for invaluable assistance.

REFERENCES Abrahamson, M., Islam, M. Q., Szpirer, J. et al. (1989). The human Cystatin-C gene (CST3), mutated in hereditary Cystatin-C amyloid angiopathy, is located on chromosome 20. Hum Genet, 82, 223–6. Abrahamsson, M., Jonsdottir, S., Olafsson, I. et al. (1992). Hereditary Cystatin-C amyloid angiopathy: identification of the disease-causing mutation and specific diagnosis by polymerase chain reaction based analysis. Hum Genet, 89, 377–80. Adam, J., Crow, T. J., Duchen, L. W. et al. (1982). Familial cerebral amyloidosis and spongiform encephalopathy. Journal of Neurology Neurosurgery and Psychiatry, 45, 37–45. Aikawa, H., Suzuki, K., Iwasaki, Y. & Iizuka, R. (1985). Atypical Alzheimer’s disease with spastic paresis and ataxia. Annals of Neurology, 17, 297–300. Ando, E., Ando, Y., Maruokas, S. et al. (1992). Ocular microangiopathy in familial amyloidotic polyneuropathy, type 1. Graefes Archives of Clinical Experimental Ophthalmology, 230, 1–5. ` Arnason, A. (1935). Apoplexie und ihre Vererbung. Acta Psychiatr Neurol Suppl VII.

Baraitsar, M. (1990). The Genetics of Neurological Disease, pp. 93, 129, and 284, Oxford: Oxford Medical Publications. Benedek, S. & McGovern, V. J. (1949). A case of Alzheimer’s disease with amyloidosis of the cerebral cortex. Medical Journal of Australia, 2, 429–430. Benedikz, E., Blondal, H. & Gudmundsson, G. (1990). Skin deposits in hereditary cystatin C amyloidosis. Virchow’s Arch Ann Pathol Anat Histopathol, 417, 325–31. Benson, M. D. (1996). Leptomeningeal amyloid and variant transthyretins. Am Pathol, 148, 351–4. Blöndal, H., Gudmundsson, G., Benedikz, E. & Johannesson, G. (1989). Dementia in hereditary Cystatin-C amyloidosis. Progr Clin Biol Res, 317, 157–64. Bornebroek, M., Haan, J., Maat-Schieman, M. L. et al. (1996). Hereditary cerebral hemorrhage with amyloidosis-Dutch type (HCHWA-D): I–A review of clinical, radiologic and genetic aspects. Brain Pathol, 6, 111–14. de la Chapelle, A., Kere, J., Sack, G. H. Jr., Tolvanen, R. & Maury, C. P. (1992). Familial amyloidosis, Finnish type: G654 – a mutation of the gelsolin gene in Finnish families and an unrelated American family. Genomics, 13, 898–901. Cohen, D. H., Feiner, H., Jensson, O. & Frangione, B. (1983). Amyloid fibril in hereditary cerebral haemorrhage with amyloidosis (HCHWA) is related to the gastroentero-pancreatic neuroendocrine protein, -trace. J Exp Med, 158, 623–8. Coria, F., Castano, E. M. & Frangione, B. (1987). Brain amyloid in normal aging and cerebral amyloid angiopathy is antigenically related to Alzheimer’s disease beta-protein. Am J Pathol, 129, 422–8. Corsellis, J. A. N. & Brierley, J. B. (1954). An unusual type of pre-senile dementia: (atypical Alzheimer’s disease with amyloid vascular change). Brain, 77, 571–87. Courten-Myers, G. & Mandybur, T. I. (1987). Atypical Gerstmann– Sträussler syndrome or familial spino cerebellar ataxia and Alzheimer’s disease? Neurologyl, 37, 269–75. Crawford, F. C., Freeman, M. J., Schinka, J. A. et al. (2000). A polymorphism in the cystatin C gene is a novel risk factor for late-onset Alzheimer’s Disease. Neurology, 55, 763–8. Crook, R., Verkkoniemi, A., Perez-Tur, J. et al. (1998). A variant of Alzheimer’s disease with spastic paraparesis and unusual plaques due to deletion of exon 9 of presenilin. 1. Nat Med, 4, 452–5. Davis, J., Cribbs D. H., Cotman C. W. & Van Nostrand W. E. (1999). Pathogenic amyloid beta-protein induces apoptosis in cultured human cerebrovascular smooth muscle cells. Amyloid, 6, 157–64. ´ Divry, P. (1927). Etude histo-chimique des plaques sénile. J Belge Neurol Psychiatr, 27, 643–57. (1941). De l’amylöidose vasculaire cérébral et méningée (méningiopathie amylöide) dans la démence sénile. J Belge Neurol Psychiat, 42, 141–58. Dubas, F., Gray, F., Roullet, E. & Escourolle, R. (1985). Leucoencephalopathy in diffuse haemorrhagic cerebral amyloid angiopathy. Ann Neurol, 18, 54–9. Duchen, L. W. (1992). Current status review: cerebral amyloid. Int J Exp Pathol, 73, 535–50.

349

350


Ellis, R. J., Olichney, J. M., Thal, L. J. et al. (1996). Cerebral amyloid angiopathy in the brains of patients with Alzheimer’s Disease: the CERAD experience, Part XV. Neurology, 46, 1592–6. Fujihara, S., Shimode, K., Nakamura, M. et al. (1989). Cerebral amyloid angiopathy with the deposition of Cystatin-C ( trace) and protein. In Alzheimer’s Disease and Related Disorders: Progress in Clinical and Biological Research, ed. K. Iqbal, H. M. Wisniewski & B. Winblad. Vol 317, pp. 939–44. New York: Alan R Liss. Garzuly, F., Vidal, R., Wisniewski, T. et al. (1996). Familial meningocerebrovascular amyloidosis, Hungarian type, with mutant transthyretin (TTR Asp18Gly) Neurology, 47, 1562–7. Gerhard, L., Bergener, M. & Homayun, S. (1972). Angiopathie bei Alzheimerscher Krankheit. Z Neurol, 201, 43–61. Ghetti, B., Piccardo, P., Spillantini, M. G. et al. (1996). Vascular variant of prion protein cerebral amyloidosis with tau-positive neurofibrillary tangles: the phenotype of the stop codon 145 mutation in PRNP. Pro Nat Acad Sci USA, 93, 744–8. Ghiso, J. & Frangione, B. (2001). Cerebral amyloidosis, amyloid angiopathy and their relationship to stroke and dementia. J Alz Dis, 3, 65–73. Ghiso, J., Jensson, O. & Frangione, B. (1986). Amyloid fibrils in hereditary cerebral haemorrhage with amyloidosis of Icelandic type is a variant of grace basic protein (cystatin-C). Pro Natl Acad Sci, 83, 2974–8. Ghiso, J., Haltia, M., Prelli, F., Novello, J. & Frangione, B. (1990). Gelsolin variant (Asn-187) in familial amyloidosis Finnish type. Biochem J, 272, 827–30. Ghiso, J., Plant, G. T., Révész, T. et al. (1995). Familial cerebral amyloid angiopathy (British type) with nonneuritic amyloid plaque formation may be due to a novel protein. J Neurol Sci, 129, 74–5. Ghiso, J., Holton, J., Miravalle, L. et al. (2001). Systemic amyloid deposits in familial British dementia. J Biol Chem, 276, 43909– 14. Ginsberg, L., Geddes, J. & Valentine, A. (1988). Amyloid angiopathy and granulamatous angiitis of the central nervous system: a case responding to corticosteroid treatment. J Neurol, 235, 438–40. Glenner, G. G. & Wong, C. W. (1984). Alzheimer’s disease: initial report of the purification and characterisation of a novel cerebrovascular amyloid protein. Biochem Biophys Res Communi, 120, 885–90. Grabowski, T. J., Cho, H. S., Vonsattel, J. P. et al. (2001). Novel amyloid precursor protein mutation in an Iowa family with dementia and severe cerebral amyloid angiopathy. Ann Neurol, 49, 697–705. Gravina, S. A., Ho, L., Eckman, C. B. et al. (1995). Amyloid beta protein (A) in Alzheimer’s Disease brain. Biochemical and immunocytochemical analysis with antibodies specific to forms ending at A40 or A42. J Biol Chem, 270, 7013–16. Greenberg, S. M. (2000). Clinical aspects and diagnostic criteria of sporadic CAA-related haemorrhage. In Cerebral Amyloid Angiopathy in Alzheimer’s Disease and Related Disorders, ed. M. M. Verbeek, R. M. W. de Waal, H. V. Vinters, pp. 3–19. Dordrecht: Kluwer Academic Publishers.

Greenberg, S. M., Rebeck, G. W., Vonsattel, J. P. et al. (1995). Apolipoprotein E epsilon 4 and cerebral haemorrhage associated with amyloid angiopathy. Ann Neurol, 38, 254–9. Griffiths, R. A., Mortimer, T. F., Oppenheimer, D. R. & Spalding, J. M. K. (1982). Congophilic angiopathy of the brain: a clinical and pathological report on two siblings. J Neurol Neurosurg Psychiatry, 45, 396–408. Grubb, A., Jensson, O., Gudmundson, G. et al. (1984). Evidence that abnormal metabolism of g-trace is the basic defect in hereditary cerebral haemorrhage with amyloidosis. N Engl J Med, 311, 1547–9. Gudmundsson, G., Hallgrimsson, J., Jonasson, T. A. & Bjarnason, O. (1972). Hereditary cerebral haemorrhage with amyloidosis. Brain, 95, 387–404. Haan, J., Roos, R. A. C., Briët, P. E. et al. (1989). Hereditary cerebral haemorrhage with amyloidosis – Dutch type: clinical characteristics. Clin Neurol Neurosurg, 91, 285–90. Haan, J., Algra, P. R. & Roos, R. A. C. (1990a). Hereditary cerebral haemorrhage with amyloidosis – Dutch type: clinical and CT analysis of 24 cases. Arch Neurol, 47, 649–53. Haan, J., Lanser, J. B. K., Zijdervald, I. et al. (1990b). Dementia in hereditary cerebral haemorrhage with amyloidosis – Dutch type. Arch Neurol, 47, 965–7. Haan, J., Roos, R. A. C., Algra, P. R. et al. (1990c). Hereditary cerebral haemorrhage with amyloidosis – Dutch type: Magnetic resonance imaging findings of seven patients. Brain, 113, 1251–67. Hendriks, L., van Duijn, C. M., Cras, P. et al. (1992). Presenile dementia and cerebral haemorrhage linked to a mutation at codon 692 of the beta-amyloid precursor protein gene. Nat Genet, 1, 218–21. Holton, J. L., Ghiso, J., Lashley, T. et al. (2001). Regional Distribution of amyloid-Bri deposition and its association with neurofibrillary degeneration in familial British dementia. Am J Pathol, 158, 515–26. Holton, J. L., Lashley, T., Ghiso, J. et al. (2002). Familial Danish dementia: a novel form of cerebral amyloidosis associated with deposition of both amyloid-Dan and amyloid-beta. J Neuropathol Exp Neurol, 61, 254–67. Houlden, H., Baker, M., McGowan, E. et al. (2000). Variant Alzheimer’s disease with spastic paraparesis and cotton wool plaques is caused by PS-1 mutations that lead to exceptionally high amyloid-beta concentrations. Ann Neurol, 48, 806–8. Ikeda, S-I., Allsop, D. & Glenner, G. G. (1989). The morphology and distribution of plaque and related deposites in the brain of Alzheimer’s Disease and controls: an immunohistochemical study using amyloid -protein antibody. Laborat Invest, 60, 113–22. ` Jensson, O., Gudmundsson, G., Arnason, A. et al. (1987). Hereditary Cystatin-C (g-trace) amyloid angiopathy of the CNS causing cerebral haemorrhage. Acta Neurol. Scand, 76, 102–14. Kametani, F., Ikeda, S., Yanagisawa, N. et al. (1992). Characterisation of a transthyretin-related amyloid fibril protein from cerebral amyloid angiopathy in type 1 familial amyloid polyneuropathy. J Neurol Sci, 108, 178–83. Kiuru, S., Salonen, O. & Haltia, M. (1999). Gelsolin-related spinal and cerebral amyloid angiopathy. Ann Neurol, 45, 305–11.


Levy, E., Lopez-Otin, C., Ghiso, J. et al. (1989). Stroke in Icelandic patients with hereditary amyloid angiopathy is related to a mutation in the Cystatin-C gene, an inhibitor of cysteine proteases. J Exp Med, 169, 1771–8. Levy, E., Carmen, M. D., Fernandez-Madrid, I. J. et al. (1990a). Mutation of the Alzheimer’s disease amyloid gene in hereditary cerebral haemorrhage, Dutch type. Science, 248, 1124–26. Levy, E., Haltia, M., Fernandez-Madrid, I., Kouvinen, O., Ghiso, J., Prelli, F. & Frangione, B. (1990b). Mutation in gelsolin gene in Finnish hereditary amyloidosis. J Exp Med, 172, 1865–7. Levy, E., Sastre, M. & Kumar, A. (2001). Codeposition of cystatin C with amyloid-beta protein in the brain of Alzheimer disease patients. J Neuropathol Exp Neurol, 60, 94–104. Love, S. & Duchen, L. W. (1982). Familial cerebellar ataxia with cerebrovascular amyloid. J Neurol Neurosurg Psychiat, 45, 271–3. ¨ Luërs, T. (1947). Uber die familiäre juvenile Form der Alzheimerischen Krankheit mit neurologischen Herderscheinungen. Arch fur ¨ Psychiat Nervenkrankheiten, 179, 132–45. Luyendijk, W. & Bots, G. T. A. M. (1980). Familial incidence of ICH. In: Pia, H. W., Langmaid C., Zierski, J. (eds). Spontaneous Intracerebral Haematomas. Advances in Diagnosis and Therapy, pp. 50–5. Berlin/Heidelberg/New York: Springer-Verlag. Luyendijk, W. & Schoen, J. H. R. (1964). Intracerebral haematomas. A clinical study of 40 surgical cases. Psychiat Neurol Neurochir, 67, 445–68. Luyendijk, W., Vegter-an der Vlis, M. & Went, L. N. (1986). Epidemiological and clinical aspects of hereditary cerebral haemorrhage in Dutch families. Acta Neurol Scand, 73, 311–12. Luyendijk, W., Bots, G. T. A. M., Vegter-an der Vlis, M., Went, L. N. & Frangione, B. (1988). Hereditary cerebral haemorrhage caused by cortical amyloid angiopathy. J Neurol Sci, 85, 267–80. Maat-Schieman, M. L., van Duinen, S. G., Bornebroek, M., Haan, J. & Roos, R. A. (1996). Hereditary cerebral hemorrhage with amyloidosis-Dutch type (HCHWA-D): II – A review of histopathological aspects. Brain Pathol, 6, 115–20. Maat-Schieman, M. L., Yamaguchi, H., van Duinen, S. G., Natte, R. & Roos, R. A. (2000). Age-related plaque morphology and Cterminal heterogeneity of amyloid beta in Dutch-type hereditary cerebral hemorrhage with amyloidosis. Acta Neuropathol (Berl), 99, 409–19. McCarron, M. O. & Nicoll, J. A. R. (2000). ApoE genotype in relation to sporadic and Alzheimer-related CAA. In: Verbeek M. M., de Waal R. M. W., Vinters H. V. (eds). Cerebral Amyloid Angiopathy in Alzheimer’s Disease and Related Disorders, pp. 81–102. Dordrecht: Kluwer Academic Publishers. McMenemey, W. H. (1952). Discussion. In: Proceedings of the First International Congress of Neuropathology. Rome, Volume 2. Rosenberg and Sellier, Turin, pp. 432–6. McMenemey, W. H. (1970). Discussion. In: Wolstenholme, G. E. W., O’Connor, M. (eds). Alzheimer’s Disease and Related Conditions: A Ciba Foundation Symposium, pp. 132–3. London: J. and A. Churchill. Mann, D. M., Pickering-Brown, S. M., Takeuchi, A. & Iwatsubo, T. (2001). Amyloid angiopathy and variability in amyloid beta deposition is determined by mutation position in presenilin-1linked Alzheimer’s disease. Am J Pathol, 158, 2165–75.

Maruyama, K., Ikeda, S., Ishihara, T., Allsop, D. & Yanagisawa, N. (1990). Immunohistochemical characterisation of cerebrovascular amyloid in 46 autopsied cases using antibodies to beta protein and Cystatin-C. Stroke, 21, 397–403. Masters, C. L. & Beyreuther, K. (2001). The Worster-Drought syndrome and other syndromes of dementia with spastic paraparesis: the paradox of molecular pathology. J Neuropathol Exp Neurol, 60, 317–19. Masters, C. L., Gajdusek, D. C. & Gibbs, C. J. (1981). The familial occurrence of Creutzfeldt-Jacob disease and Alzheimer’s disease. Brain, 104, 535–58. Maury, C. & Baumann, M. (1990). Isolation and characterization of cardiac amyloid in familial amyloid polyneuropathy type IV (Finnish): relation of the amyloid protein to variant gelsolin. Biochem Biophys Acta, 1096, 84–6. Mead, S., James-Galton, M., Revesz, T. et al. (2000). Familial British dementia with amyloid angiopathy: early clinical, neuropsychological and imaging findings. Brain, 123, 975–91. Merejota, J. (1969). Familial systemic paramyloidosis with lattice dystrophy of the cornea, progressive cranial neuropathy, skin changes and various internal symptoms. Ann Clin Res, 1, 314–24. Miravalle, L., Tokuda, T., Chiarle, R. et al. (2000). Substitutions at codon 22 of Alzheimer’s Abeta peptide induce diverse conformational changes and apoptotic effects in human cerebral endothelial cells. J Biol Chem, 275, 27110–16. Nilsberth, C., Westlind-Danielsson, A., Eckman, C. B. et al. (2001). The ‘Arctic’ APP mutation (E693G) causes Alzheimer’s disease by enhanced Abeta protofibril formation. Nat Neurosci, 4, 887–93. Petersen, R. B., Goren, H., Cohen, M. et al. (1997). Transthyretin amyloidosis: a new mutation associated with dementia. Ann Neurol, 41, 307–13. Plant, G. T., Révész, T., Barnard, R. O., Harding, A. E. & GautierSmith, P. C. (1990). Familial amyloid angiopathy with nonneuritic amyloid plaque formation. Brain, 113, 721–47. Prelli, F., Levy, E., van Duinen, S. G., Bots, G. T., Luyendijk, W. & Frangione, B. (1990). Expression of a normal and variant Alzheimer’s beta-protein gene in amyloid of hereditary cerebral haemorrhage, Dutch type: DNA and protein diagnostic assays. Biochem Biophys Res Commun, 170, 301–7. Revesz, T., Holton, J. L., Doshi, B., Anderton, B. H., Scaravilli, F. & Plant, G. T. (1999). Cytoskeletal pathology in familial cerebral amyloid angiopathy (British type) with non-neuritic amyloid plaque formation. Acta Neuropathol (Berl), 97, 170–6. Revesz, T., Holton, J. L., Lashley, T. et al. (2002). Sporadic and familial cerebral amyloid angiopathies. Brain Pathol, 12, 343–57. Roberts, G. W., Lofthouse, R., Allsop, D., Landon, M., Kidd, M. & Prusiner, S. B. (1988). CNS amyloid proteins in neurodegenerative diseases. Neurology, 38, 1534–40. Rostagno, A., Revesz, T., Lashley, T. et al. (2002). Complement activation in chromosome 13 dementias: Similarities with Alzheimer’s disease. J Biol Chem 277: 49782–90.

351

352


Scholtz, W. (1938). Studien zur Pathologie der Hirngefässe; die drusige Entartung der Hirnarterien und Capillaren. (eine Form seniler Gefässerkrankung). Zeitschrift fur ¨ die Gesamte Neurologie und Psychiatrie, 162, 694–715. Shimode, K., Fujihara, S., Nakamura, M. & Kobayashis Tsunematsu, T. (1991). Diagnosis of cerebral amyloid angiopathy by enzyme linked immunoabsorbant assay of cystatin-C in cerebrospinal fluid. Stroke, 22, 860–6. Strömgren, E. (1981). Heredopathia opthalmo-oto-encephalica. In: Vinken P. J. & Bruyn, G. W. (eds). Handbook of Neurology Vol 42. North Holland Publishing Co., Amsterdam, pp. 150–2. Tagliavini, F., Ghiso, J., Timmers, W. F., Giaccone, G., Bugiani, O. & Frangione, B. (1990). Coexistence of Alzheimer’s amyloid precursor protein and amyloid protein in cerebral vessel walls. Lab Invest, 62, 761–7. Tagliavini, F., Giaccone, G., Orso Bugiani & Frangione, B. (1993). Ubiquinated neurites are associated with preamyloid and cerebral amyloid- deposits in patients with HCHWA-D. Acta Neuropathologica, 85, 267–71. Tagliavini, F., Rossi, G., Padovani, A. et al. (1999). A new PP mutation related to hereditary cerebral haemorrhage. Alz Reports, 2, S28. Tateishi, J., Kitamoto, T., Doh-ura, K., Boellard, J. W., & Pfeiffer, J. (1992). Creutzfeld-Jacob disease with amyloid angiopathy: diagnosis by immunological analysis and transmission experiments. Acta Neuropathol (Berl), 83, 559–63. Uitti, R. J., Donat, J. R., Rozdilsky, B., Schneider, R. J. & Koeppen, A. H. (1988). Familial oculoleptomeningeal amyloidosis: report of a new family with unusual features. Arch Neurol, 45, 1118–22. Ushiyama, M., Ikeda, S. & Yanagisawa, N. (1991). Transthyretintype cerebral amyloid angiopathy in type 1 familial amyloid polyneuropathy. Acta Neuropath, 81, 524–8. Van Bogaert, L., Maere, M. & Smedt, E. de. (1940). Sur les formes familiales précoces de la maladie d’Alzheimer. Monatsschr Psychiat Neurol, 102, 249–301. Van Broeckhoven, C., Haan, J., Bakker, E. et al. (1990). Amyloid protein precursor gene and hereditary cerebral haemorrhage with amyloidosis (Dutch) Science, 248, 1120–2. ˜ E. M., Prelli, F., Bots, G. T. A. B., Van Duinen, S. G., Castano, Luyendijk, W. & Frangione, B. (1987). Hereditary cerebral haemorrhage with amyloidosis in patients of Dutch origin is related to Alzheimer disease. Proc Natl Acad Sci USA, 84, 5991–4. Vidal, R., Garzuly, F., Budka, H. et al. (1996). Meningocascular amyloidosis associated with a novel transthyretin missense mutation at codon 18 (TTRD18G). Am J Pathol, 148, 361–6. Vidal, R., Frangione, B., Rostagno, A. et al. (1999). A stop-codon mutation in the BRI gene associated with familial British dementia. Nature, 399, 776–81.

Vidal, R., Revesz, T., Rostagno, A. et al. (2000). A decamer duplication in the 3’ region of the BRI gene originates an amyloid peptide that is associated with dementia in a Danish kindred. Proc Natl Acad Sci USA, 97, 4920–5. Vidal, R., Calero, M., Revesz, T., Plant, G., Ghiso, J. & Frangione, B. (2001). Sequence, genomic structure and tissue expression of human BRI3, a member of the BRI gene family. Gene, 266, 95–102. Vinters, H. V. (1987). Cerebral amyloid angiopathy: a critical review. Stroke, 18, 311–24. Vinters, H. V. & Gilbert, J. J. (1983). Cerebral amyloid angiopathy: incidence and complications in the aging brain. II. The distribution of amyliod vascular changes. Stroke, 14, 924–8. Vinters, H. V., Secor, D. L., Partridge, W. M. & Gray, F. (1990). Immunohistochemical study of cerebral amyloid angiopathy. III. Widespread Alzheimer A4 peptide in cerebral microvessel walls colocalises with -trace in patients with leucoencephalopathy. Ann Neurol, 28, 34–42. Vonsattel, J. P. G., Myers, R. H., Hedley-White, E. T., Ropper, A. H., Bird, E. D. & Richardson, E. P. (1991). Cerebral amyloid angiopathy without and with cerebral haemorrhages: a comparative histological study. Ann Neurol, 30, 637–49. Watanabe, R. & Duchen, L. W. (1993). Cerebral amyloid in human prion diseases. Neuropath Appl Neurobiol, 19, 253–60. Wattendorf, A. R., Bots, G. T. A. M., Went, L. N. & Endtz, L. J. (1982). Familial cerebral amyloid angiopathy presenting as recurrent cerebral haemorrhage. J Neurol Sci, 55, 121–35. Wisniewski, T., Ghiso, J. & Frangione, B. (1991). Peptides homologous to the amyloid protein of Alzheimer’s disease containing a glutamine for glutamic acid substitution have accelerated amyloid fibril formation. Biochem Biophys Res Commun, 179, 1247–54. Worster-Drought, C., Greenfield, J. G. & McMenemey, W. H. (1940). A form of familial presenile dementia with spastic paralysis: (including the pathological examination of a case) Brain, 63, 237–54. Worster-Drought, C., Greenfield, J. G. & McMenemey, W. H. (1944). A form of familial presenile dementia with spastic paralysis Brain, 67, 38–43. Yamada, M., Tsukagoshi, H. & Wada, Y. et al. (1989). Absence of the Cystatin-C amyloid in the cerebral amyloid angiopathy, senile plaque and extra-CNS amyloid deposits of aged Japanese. Acta Neurol Scand, 79, 504–9. Yamaguchi, H., Maat-Schieman, M. L., van Duinen, S. G. et al. (2000). Amyloid beta protein (Abeta) starts to deposit as plasma membrane-bound form in diffuse plaques of brains from hereditary cerebral hemorrhage with amyloidosis-Dutch type, Alzheimer disease and nondemented aged subjects. J Neuropathol Exp Neurol, 59, 723–32. Yoshimura, M., Yamanouchi, H., Kuzuhara, S. et al. (1992). Dementia in central amyloid angiopathy: a clinico-pathological study. Lab Invest, 60, 113–22.

15 Parkinson’s disease, dementia with Lewy bodies, multiple system atrophy and the spectrum of diseases with -synuclein inclusions Benoit I. Giasson, Virginia M.-Y. Lee and John Q. Trojanowski Center for Neurodegenerative Disease Research, University of Pennsylvania School of Medicine, Philadelphia, PA, USA

Introduction Synucleinopathies are a group of neurodegenerative disorders sharing in common the presence of intracellular fibrillar inclusions composed of polymerized -synuclein (-syn). Although the biological function of this presynaptic protein remains largely unknown, overwhelming evidence indicates that its self-aggregation in neurons and/or oligodendrocytes is associated with the impairment of nervous system function and cellular demise leading to disease. Pathological inclusions composed of -syn have been the focus of enumerable clinical and pathological studies with the earliest dating almost to the beginning of the twentieth century. However, the finding that these inclusions are comprised of -syn only became apparent in the past 5 years following the identification of a mutation in the -syn gene in families with Parkinson’s disease. -Syn inclusions are the defining characteristics of several disorders including Parkinson’s disease (PD), dementia with Lewy bodies (DLB), and multiple system atrophy (MSA). However, they are also found in a significant percentage of other neurodegenerative disorders including neurodegeneration with brain iron accumulation type-1 (NBIA-1), Down’s syndrome, and familial and sporadic Alzheimer’s disease (AD) (Table 15.1). The understanding of the role of -syn in neurodegenerative disease requires the determination of the normal function and properties of this protein, the clinicopathological assessment of the relationship between -syn inclusions and disease, and the elucidation of the mechanism of pathogenesis. In this chapter, we will focus on the most recent developments in these aspects of synucleinopathies, but the description of these diseases also requires an overview of the most significant

features that have been documented for over almost 100 years.

The synuclein protein family Four syn proteins have been identified in humans. They are all relatively small proteins (123–143 amino acids), but their most striking features are the presence of 5–6 imperfect repeats (KTKEGV) distributed throughout most of the amino-terminal half of the protein and an acidic carboxylterminal region (Fig. 15.1). -Syn was first cloned from the electric ray Torpedo california by screening an expression library with an antiserum raised against cholinergic vesicle (Maroteaux et al., 1988). This protein was named syn, because of its initial localization within neuronal nuclei and presynaptic terminals. However, localization of -syn to the nuclei was not consistently confirmed in subsequent studies in mammalian systems, suggesting that this initial observation may be due to cross-reactivity or that it is a specific property of -syn in the electric ray. -Syn is also referred to as the non-amyloid component of senile plaques precursor protein (NACP) (Ueda et al., 1993), SYN1 (Maroteaux & Scheller, 1991), or synelfin (George et al., 1995). It is a heat-stable protein that is ‘natively unfolded’, such that it has no apparent defined structure (Davidson et al., 1998; Weinreb et al., 1996), and it is predominantly expressed in the central nervous system (CNS) neurons, where it is localized at presynaptic terminals (George et al., 1995; Iwai et al., 1995a; Jakes et al., 1994; Withers et al., 1997). However, in some mammalian species, -syn can also be normally found in the perikarya of dorsal root ganglion neurons (Giasson et al., 2001b). The function


353

354

B. I. Giasson, V. M.-Y. Lee and J. Q. Trojanowski

Table 15.1. Summary of -synucleinpathies Diseases defined by -syn inclusions

Diseases associated with -syn inclusions

Parkinson’s disease

Neurodegeneration with brain iron accumulation type-1 Down’s syndrome Parkinsonism-dementia complex of Guam Familial Alzheimer’s disease Sporadic Alzheimer’s disease Pure autonomic failure

Dementia with Lewy bodies Lewy body variant of Alzheimer’s disease Multiple system atrophy

of -syn still remains unknown, but several clues suggest that it is involved in modulating synaptic transmission and neuronal plasticity. Electron microscopy studies have localized -syn in close proximity to synaptic vesicles at axonal termini (George et al., 1995; Iwai et al., 1995a), and biochemical analysis has revealed that a small fraction may be associated with vesicular membranes, but the majority is likely cytoplasmic (George et al., 1995; Irizarry et al., 1996). It is predicted that -syn can form amphipathic helices that can associate with vesicular membranes (George et al., 1995), and indeed an increase in -helical structure is observed upon binding to small synthetic unilamellar vesicles in vitro (Davidson et al., 1998). The observation that the expression pattern of -syn in neuronal populations involved in male zebra finch song learning during the critical development period when singing is acquired led to the suggestion that it may be involved in neuronal plasticity (George et al., 1995). However, it does not appear to play a role in initial synaptic formation since it localizes

to synapses after they are formed (Murphy et al., 2000; Withers et al., 1997). Ablation of -syn by engineering a null-mutation in mice is not associated with any overt phenotype, but results in a subtle alteration in the time period of recovery following induced repetitive synaptic vesicular release (Abeliovich et al., 2000). The reduction of -syn levels in cultured hippocampal neurons results in a significant reduction of the distal pool of synaptic vesicle (Murphy et al., 2000). Hence, these results demonstrate that -syn can modulate vesicular synaptic function, but the precise mechanism is unclear. The other three syn proteins have been studied less extensively, perhaps because they have not been implicated directly in human disease. -syn, initially named phosphoprotein-14, is also a heat stable protein predominantly expressed in neuronal axonal terminals (Giasson et al., 2001b; Jakes et al., 1994; Nakajo et al., 1990, 1994; Shibayama-Imazu et al., 1993), and it has a high degree of primary sequence homology with -syn (Fig. 15.1), suggesting that it may have a similar function. Both proteins have been shown to have chaperone activity in vitro (Souza et al., 2000b), and it is possible that they may be involved in assisting changes in protein folding that occur during vesicular release, re-uptake or cycling. -syn (also termed persyn) was cloned as a gene product expressed in breast tumors leading to another alternative name, breast cancer-specific gene 1 (BCSG1) (Ji et al., 1997). Further investigation demonstrated that -, - and -syn were expressed in a significant percentage of breast and ovarian tumours, and -syn is frequently found (70%) in high-grade breast tumors (Bruening et al., 2000). Overexpression of -syn in breast cancer cells augments cell

Fig. 15.1. Amino sequence alignment of human syn proteins. The imperfect repeats of the type KTKEGV are identified above the amino acid sequences. The red font highlights amino acid residues that are conserved between all four proteins. The arrows delineate the hydrophobic amino acid stretch in -syn, but missing in -syn, that is necessary for fibrillogenesis. Asterisks mark the position of amino acid residues that are mutated in kindreds with PD. The sequences for - and -syns were obtained from Jakes et al., 1994, -syn and synoretin were obtained from Ji et al., 1997 and Surguchov et al., 1999, respectively.

Parkinson’s disease, dementia with Lewy bodies

motility, invasiveness, and metastasis (Jia et al., 1999). It is also expressed in several neuronal populations, and although it is also a cytoplasmic protein, its distribution is not localized to axonal terminals (Buchman et al., 1998; Giasson et al., 2001b). Synoretin has close homology to -syn (Fig. 15.1), but it is predominantly expressed in retina (Surguchov et al., 1999).

Synucleinopathies Synucleinopathies: a group of neurodegenerative disorders with -synuclein pathological inclusions The term synucleinopathies was recently coined following the stunning realization that pathological inclusions comprised of -syn were a unifying feature of a spectrum of neurodegenerative diseases. The first report, which implicated -syn in a neurodegenerative disorder, described the isolation of a fragment of -syn from partially purified sodium dodecyl sulphate (SDS)-insoluble fractions from AD brain (Ueda et al., 1993). Antibodies raised against this isolated fragment of -syn were reported to recognize the extracellular amyloid component of senile plaques in AD brain (Ueda et al., 1993), and this fragment corresponding to amino acid residues 61–95 in -syn was termed the non-amyloid component of senile plaques (NAC). Hence, -syn is often referred to as the NAC precursor protein or NACP. The ability to label senile plaques with antibodies to NAC was substantiated in several studies that further suggested that a significant percentage of both diffuse and mature plaques contained NAC (Iwai et al., 1995b; Masliah et al., 1996; Takeda et al., 1998a). However, it is now widely believed that NAC is not a constitutive component of senile plaques, since more recent studies with better characterized antibodies demonstrated that NAC is not detected within senile plaques (Bayer et al., 1999; Culvenor et al., 1999). It is likely that the initial biochemical isolation of insoluble -syn was due to the presence of -syn inclusions in some AD brains. These inclusions could have been in the form of cortical (Lewy bodies) (LBs) (see below for more details) or dystrophic -syn neurites that are intertwined, but spatially distinct, from the extracellular amyloid component of senile plaques. Furthermore, Culvenor and colleagues noted that the labelling of senile plaques with the first generation of antibodies to NAC could be explained by cross-reactivity with the amyloid- peptide, which has amino acid sequence homology with the peptide used to raised these antibodies (Culvenor et al., 2000). The breakthrough that directly implicated -syn in neurodegenerative diseases came from genetics. Polymeropoulos and colleagues reported an autosomal

dominant mutation (A53T) in -syn resulting from a G to A transition at position 209 of the -syn gene (Polymeropoulos et al., 1997). The A53T mutation was initially identified in a large Italian family (the Contursi kindred) and three small Greek families (Polymeropoulos et al., 1997), but it was thereafter identified in an additional 8 families (Athanassiadou et al., 1999; Markopoulou et al., 1999; Papadimitriou et al., 1999; Spira et al., 2001). Another autosomal dominant mutation (A30P) in -syn has also been identified in a German kindred (Kruger et al., 1998). Shortly following the report of the A53T mutation in syn, a series of studies demonstrated and confirmed that -syn was the major component of several types of pathological inclusions characteristic of many disorders including PD, DLB, and MSA (Irizarry et al., 1998; Spillantini et al., 1997, 1998a, b; Takeda et al., 1998b; Tu et al., 1998). Furthermore, -syn pathology is also present in a subset of patients afflicted by several other neurodegenerative disorders, although in these cases the distribution can be restricted to certain neuroanatomical regions (see below: ‘Other synucleinpathies’).

Parkinson’s disease PD is the most common movement disorder with a prevalence of 0.6 % at 65 years of age, but the risk of developing PD increases with age such that its prevalence can reach 4–5% by the age of 85 (de Rijk et al., 1997). PD is a progressive disorder characterized by bradykinesia, resting tremor, and cogwheel rigidity. The presentation of at least 2 of these features in addition to responsiveness to levodopa are considered to be essential requirements for the clinical diagnosis of this disorder (Simuni & Hurtig, 2000) to minimize the likelihood of being confounded with other disorders, such as MSA, progressive supranuclear palsy, or other diseases that also exhibit a parkinsonism syndrome (Hughes et al., 2001). The neuronal circuitry involved in coordinated movement is complex (Obeso et al., 2000), but the disabling symptoms of PD are predominantly due to a profound reduction in striatal dopamine content caused by the demise of dopaminergic neurons in the substantia nigra pars compacta (SNpc) and its projections to the striatum (Damier et al., 1999; Pakkenberg et al., 1991). In addition to dopaminergic neurons in the SNpc, there can be a reduction, albeit to a lesser extent, in other populations of neurons in the brainstem and basal forebrain also leading to neurotransmitter deficits. For example, norepinephrine, serotonin and acetylcholine are variably decreased in PD due to the loss of neurons in the locus ceruleus, raphe nuclei and the nucleus basalis of Meynert, respectively (Mayeau et al., 1984; Whitehouse et al., 1983). Indeed, in

355

356


Fig. 15.2. (a) Micrographs of a coronal section of the midbrain at the level of the SN in a patient with PD demonstrating the severe loss of pigmented cells (courtesy of Dr M. Forman). (b), (c) Appearance of nigral LBs (arrows) stained with hematoxylin and eosin in the SNpc of a patient with PD. Note that some LBs can have a more pronounced corona (b), while the corona can also be narrow (c). (d) Immunoelectron microscopy of a brain stem LB immunolabeled with an antibody to -syn and visualized by an immuno-peroxidase method (From Baba et al., 1998, with permission). (e) Two LBs in a pigmented neurons in the SNpc of PD patient immunolabelled with anti--syn Mab Syn 202. (f ) Immunogold-labelled -syn filaments assembled in vitro from human recombinant protein and incubated with an antibody (Syn 303) specific for -syn. The dark spheres are 5 nm colloidal gold particles. (g), (h) Immuno-labelling of LNs and neuroaxonal spheroids, respectively, in the SNpc of a patient with PD with anti -syn antibody Syn 505. Tissue section in E, G, and H were counterstained with hematoxylin. Scale bar = 40 m in (b), (c); 4 m in (d ); 28 m in (e); 170 nm in (f ); 60 m in (g), (h).

the decades following the initial descriptions of the classic clinical and pathological features of PD, a large body of literature has developed from research designed to establish the neuropathological correlates of the different clinical manifestations of PD (for reviews see Dunnett & Björkland, 1999; Lang & Lozano, 1998). Patients with PD can exhibit many other clinical manifestations in addition to extrapyramidal signs. For example, autonomic dysfunction, depression, seborrhoea, sleep disturbance, sensory symptoms and dementia are frequently observed (Hughes et al., 1982, 1993; Simuni & Hurtig, 2000; Snider et al., 1976). Dementia in PD has been the focus of many studies and the reported co-prevalence of dementia in PD can vary widely between 10 and 80%, although a more consistent estimate is around 25–30% (Aarland et al., 1996; Brown & Marsden, 1984; Hietanen & Teravainen, 1988; Marttila & Rinne, 1976; Mayeau et al., 1988, 1990; Molsa et al., 1984; Rajput & Rozdilsky, 1975). Since the most common cause of dementia in the elderly is AD, and the most important risk factor for developing PD and AD is age (de Rijk et al., 1997; Zabar & Kawas, 2000),

it is not that surprising that a significant percentage of patients will be affected by both disorders, especially when the prevalence of AD can reach > 30% in individuals by the age of 80 (Zabar & Kawas, 2000). However, patients clinically diagnosed with PD are still 4–6 times more likely to become demented than control subjects (Mayeau et al., 1990; Rajput et al., 1987a; Riggs, 1993), indicating that there could be a common mechanism linking AD and PD or that PD patients are more vulnerable to develop a second neurodegenerative disorder. Furthermore, another confounding factor in trying to interpret these clinical data is the finding that many of these patients do not have ‘pure’ AD and PD, but rather have DLB (see DLB section below). The definite diagnosis of PD necessitates the clinical presentation of parkinsonism as well as characteristic neuropathological features. PD brains display extensive loss of dopaminergic neurons in the SNpc (Fig. 15.2(a), the presence of intracytoplasmic inclusions known as LBs in some of the remaining dopaminergic neurons, variable extracellular melanin released from degenerating neurons and gliosis (Cronford et al., 1995; Fearnley & Lees, 1991; Forno,


1996; McGeer et al., 1989). Friederich Lewy first described LBs in the cholinergic neurons of the substantia innominata in 1912 (see Gibb & Poewe, 1986), but their presence in dopaminergic neurons of the SNpc was first noted by Trétiakoff in 1919. These pathological lesions now known as ‘classical’ LBs are eosinophilic and usually have a distinctive laminated spheroid appearance (Figs. 15.2(b), (c)). Ultrastructurally, they are composed of a halo of radiating fibrils (7–25 nm in diameter) often referred to as the ‘corona’ surrounding a matted meshwork of filaments intertwined with amorphous material at the ‘cores’ (Fig. 15.2(d )) (Duffy & Tennyson, 1965; Forno, 1996; Hill et al., 1991; Schmidt et al., 1991). LBs can range from 15–30 M in diameter, and single or multiple LBs can be observed in neurons (Fig. 15.2(e)). LBs often displace or deform organelles such as mitochondria, endoplasmic reticulum and Golgi apparatus as well as nucleus. LBs are usually located within the perikarya, but they can occasionally be observed in the proximal processes. LBs are not restricted to the substantia nigra in PD as they can also present in many other brain stem nuclei and diencephalic regions including the locus ceruleus, the raphe nucleus, the basal nucleus of Meynert, the dorsal nucleus of vagus, the ventral tegmental nucleus, the pedunculo-pontine nucleus, the thalamus and the hypothalamus (Forno, 1996). Initial attempts to identify the primary components of LBs relied predominantly on immunocytochemistry, where it was demonstrated that many proteins are present in at least a subset of LBs (see Gomez-Tortosa et al., 1998, and references therein). More elaborate studies attempted to purify LBs (Galvin et al., 1997; Iwatsubo et al., 1996). Some studies suggested that neurofilaments, which appeared to be aberrantly processed, might be the elusive building blocks of the filamentous component of LBs (Galvin et al., 1997; Hill et al., 1991; Schmidt et al., 1991). However, soon after the A53T mutation in the -syn gene was discovered (Polymeropoulos et al., 1997) and attention was drawn to this protein, it became apparent that -syn is the major fibrillary component of LBs. The evidence supporting this notion includes biochemical and in situ analysis of human specimens, as well as in vitro characterization of recombinant -syn expressed from bacteria and transgenic animal studies: (i) Antibodies to -syn detect LBs more intensely and more consistently than any other antibodies (Fig. 15.2(e)), and antibodies to -syn and -syn do not detect LBs (Baba et al., 1998; Irizarry et al., 1998; Spillantini et al., 1997, 1998b; Takeda et al., 1998b); (ii) Immuno-electron microscopic studies have demonstrated that LB fibrils are labelled intensely with -syn antibodies in situ (Fig. 15.2(d )) (Arima et al., 1998a, b; Baba et al., 1998; Wakabayashi et al., 1998a); (iii) Pale bodies, which

are putative precursors of LBs, are labelled intensely with antibodies to -syn (Irizarry et al., 1998; Takeda et al., 1998b; Wakabayashi et al., 1998a); (iv) Sarcosyl-insoluble filaments decorated with antibodies to -syn have been observed in extracts of PD and DLB brains (Crowther et al., 2000; Spillantini et al., 1998b); (v) -syn accumulates as detergent-insoluble and cross-linked species in brains containing Lewy pathology, but not in control brains (Fig. 15.3) (Baba et al., 1998; Campbell et al., 2000; Giasson et al., 2000a, b; Gwinn-Hardy et al., 2000); (vi) Recombinant human -syn can readily assemble in vitro into elongated homopolymers with similar widths as sarcosyl-insoluble

Fig. 15.3. Biochemical fractionation and analysis of -syn in diseased brains. (a) Cingulate gyrus from control (C) and DLB (D) brains or (b) frontal cortex from control (c) and NBIA-1 (N) brains were fractionated biochemical as previously described (Giasson et al., 2000). Samples of high salt-soluble (HS-sol) and Triton X-100-insoluble (TX-ins) fractions were loaded on 12% SDS-polyacryamide and analysed by Western blotting using antibodies specific to -syn. -syn appears as a ∼ 17 kDa band (depicted by arrows) in the HS fractions from control and diseased brains. In DLB brain, -syn accumulates in the Triton X-100-insoluble fraction as SDS-soluble monomers and SDS-resistent aggregates (indicated by asterisks). In NBIA-1 brain, -syn accumulated as SDS-resistant aggregates in both the HS-soluble and Triton-X-100 insoluble fractions. The electrophoretic mobilities of molecular mass markes are indicated on the left.

357

358


fibrils or filaments visualized in situ in pathological inclusions (Fig. 15.2(f ) (Conway et al., 1998; El Agnaf et al., 1998; Giasson et al., 1999; Hashimoto et al., 1998; Narhi et al., 1999); (vii) Transgenic mice expressing wild-type and mutant -syn develop LB-like -syn inclusions associated with impairment of neuronal function (see section on ‘Transgenic animal model’ below). Although LBs are often emphasized in the description of PD, formation of -syn inclusions in neuronal processes is usually even more abundant. The presence of aberrant inclusions in neuronal processes in PD brain was first described as dystrophic ubiquitin-positive neurites termed Lewy neurites (LNs) (Dickson et al., 1991; Braak et al., 1994). LNs are also composed of filamentous -syn (Fig. 15.2(g) (Irizarry et al., 1998; Spillantini et al., 1997, 1998b; Takeda et al., 1998a), and immuno-labelling with antibodies to -syn demonstrated that -syn can also accumulate within even larger, aberrantly distended, neuronal processes known as spheroids (Fig. 15.2(h)).

Dementia with Lewy Bodies Cortical LBs, first described by Okazaki in 1961 (Okazaki et al., 1961), are also comprised of -syn fibrils (Baba et al., 1998; Spillantini et al., 1997, 1998b), but they have a less clearly defined morphology that may account for their inconspicuous appearance by hematoxylin and eosin staining compared to nigral LBs (Forno, 1996). Cortical LBs are smaller than the classical LBs found in the brainstem, they are typically more irregular in shape and they are not laminated (i.e. they do not have a corona and core)(Fig. 15.4(a)). In affected brains, cortical LBs typically have an uneven distribution with the highest density in the cingulate gyrus, entorhinal cortex, amygdaloid nucleus, temporal cortex and insular cortex (Kosaka, 1990; Perry et al., 1990; Rezaie et al., 1996). They are less abundant in the hippocampus proper and frontal and parietal cortical lobes. The occipital lobe usually displays the lowest density and the cerebellum is typically spared. The terminology that has been used in the literature to describe patients with cortical LBs is complex; depending on variable emphasis on the clinical versus pathological findings as well as on differences in diagnostic criteria. Attempts to classify patients with cortical LBs into disease entities have given rise to a list of appellations, which, for example, included diffuse or cortical LB dementia (Gibb et al., 1987; Lennox et al., 1989), senile dementia of the LB type (Perry et al., 1990), the LB variant of AD (Hansen et al., 1990) and LB disease (Kosaka et al., 1984). In order to unify all these studies and for the purpose of this review, the term dementia with LBs (DLB) most commonly refers to patients

with dementia associated with widespread brain stem, diencephalic and cortical LBs. Most of the earliest reports of DLB were from Japanese researchers (Kosaka, 1990, 1993), where the patients described presented with early to mid-adult or late adult onset, but most cases reported in America and Europe were thought to have a late onset. The initial clinical features of DLB are usually dementia with subsequent parkinsonism, although a significant percentage (∼25–40%) of cases initially display parkinsonism followed by dementia (Kosaka, 1990; Lennox et al., 1989). Prior to the late 1980s there were only a few reports from American and European groups describing cortical LBs (Burkhardt et al., 1988). The interest in cortical LBs increased greatly as the use of ubiquitin immunocytochemistry significantly improved their detection (Lennox et al., 1989). This led to a series of studies demonstrating the high frequency of widespread LBs (20– 30%) in the brains of patients clinically followed with presumed AD (Bergeron & Pollanen, 1989; Galasko et al., 1994; Hansen et al., 1990; Perry et al., 1990). In fact, DLB is the second most common neurodegenerative dementing illness of the elderly after AD (McKeith et al., 1996). DLB can present in a ‘pure’ form where LBs are the major pathological lesions, however, these cases are typically less common since the majority of DLB brains typically have sufficient AD pathology (i.e. neurofibrillary tangles and senile -amyloid plaques) for a concurrent diagnosis of AD and DLB (Kosaka, 1990). The term LB variant of AD (LBVAD) is typically used to describe these patients (Hansen et al., 1990; Katzman et al., 1995; Samuel et al., 1996), although this terminology can be confusing since it implies that this disorder is an altered form of AD. Moreover, patients with LBVAD usually have mainly AD pathology in the form of senile plaques and less abundant neurofibrillary tangles (NFTs) (Dickson et al., 1989; Lennox et al., 1989; Perry et al., 1990). DLB may account for 75% of the so-called ‘plaque-only’ or ‘plaquepredominant’ form of AD in which NFTs are either sparse or absent (Hansen et al., 1993). It is unknown why these patients develop concurrent AD pathology and LBs, and it is still not clear if they represent a different disease entity. However, it is possible that these patients may have a defect in modulating and/or eliminating aberrant proteinaceous aggregates resulting in the accumulation of various forms of inclusions. Alternatively, one type of inclusion may affect cellular function leading to the formation of inclusions comprised of a different protein. The involvement of cortical LBs in the pathogenesis of cognitive decline is suggested by the correlations between severity of dementia and LB density in patients with DLB (Haroutunian et al., 2000; Hurtig et al., 2000; Lennox et al., 1989; Mattila et al., 1998; Samuel et al., 1996).


Fig. 15.4. (a) Immunostaining of LB (arrow) and LNs (arrowheads) in the cingulate and (b) LNs (arrowheads) in the CA2/3 hippocampal region of a DLB patient. (c) Immunostaining of LNs and dystrophic neurites containing aggregated -syn within a senile plaque in the cingulate of a patient with DLB. (d ) Appearance of classical LBs (arrows) stained with hematoxylin and eosin in the locus ceruleus of an individual from the Contursi kindred (courtesy of Dr J. Duda). (e) Immuno-detection of abundant LNs and neuroaxonal spheroids in the nucleus basalis of Meynert in a member of the Contursi kindred. ( f ) Immuno-labeling of a classical LB (arrow) and perikaryal accumulation of -syn (arrowhead) in the SN of a patient with NBIA-1. (g) Immuno-detection of -syn inclusions in the cell bodies and neuronal processes in the globus pallidus of a patient with NBIA-1. (h) Intense accumulation of -syn in neuroaxonal spheroids in the SNpc of a patient with NBIA-1. (i) LBs in the amygdala of an AD patient with a mutation (N141I) in presenilin-2. Tissue sections were immunostained with anti--syn antibody Syn 505 (a), (c), ( f ), (g), (h), (i), Syn 202 (b) or Syn 303 (e). Tissue sections were counterstained with hematoxylin. Scale bar = 40 m in (a), (b), (c), (d ), (i); 80 m in (e); 55 m in ( f ), (g), (h).

The awareness that a high percentage of demented patients are affected by DLB and that the clinical assessment of these patients can be confounded with AD spearheaded attempts to identify diagnostic clinical differences between these disorders. A number of potential neuropsychiatric features of DLB have been suggested in several case series, including visual hallucinations, delusions, parkinsonian

symptoms, neuroleptic sensitivity, marked fluctuation of cognitive abilities, syncope and falls (Crystal et al., 1990; Galasko et al., 1994; Gibb et al., 1987; Hansen et al., 1990; Louis et al., 1997; Mega et al., 1996; Perry et al., 1990). These findings and the need for a universal set of diagnostic criteria led to an international meeting giving rise to DLB consensus criteria (McKeith et al., 1996). By definition, the

359

360


progressive deterioration of cognition and memory leading to impaired functional ability is a mandatory feature of DLB. Fluctuation in cognitive function, visual hallucinations and parkinsonism were proposed as three major clinical identifiers of DLB. Although these consensus criteria are helpful in providing guidelines for classifying patients, their accuracy is limited by the inconsistency in clinical presentation. For example, some patients with DLB do not progress to develop parkinsonism symptoms (Kosaka, 1990). DLB can account for about one-quarter of PD patients that are also demented; however, the pathological changes that explain the dementia in many PD patients remain enigmatic (Hughes et al., 1993). In an extensive evaluation of 31 PD patients that became demented, 26% had abundant cortical LBs and 29% had the pathological changes associated with AD, but about half of the patients did not have an identifiable cause for dementia (Hughes et al., 1993). It is possible that certain brain nuclei that typically are affected by neuronal loss in PD, for example the nucleus basalis of Meynert, may be involved in the loss of cognitive function, but this notion remains speculative. Although LBs are the diagnostic marker of LB diseases, LNs that always co-exist with LBs may have a more profound deleterious impact, since neuritic -syn pathology rather than LBs may be the most abundant form of -syn pathology in PD and other LB diseases (see Fig. 15.4(a)) (Braak et al., 1999; Spillantini et al., 1998b). The characterization of novel anti--syn antibodies that are selective for pathological -syn has further revealed an unanticipated high abundance of LNs (Duda et al., 2002a). The presence of LNs is usually under-appreciated with conventional anti-syn antibodies, because detection is obscured by the abundance of -syn in the neuropil and the smaller size of LNs. These novel antibodies specific for pathological -syn have also revealed abundant LN pathology in the striatum of the majority of DLB patients and a subset of PD patients (Duda et al., 2002a). This accumulation of aggregated -syn in the striatum may contribute to dysfunction of the nigrostriatal pathways associated with parkinsonism. LNs, initially detected with ubiquitin immunocytochemistry (Dickson et al., 1991), are also abundant in the CA2/CA3 region of the hippocampus in DLB brains (Fig. 15.4(b)). LNs and swollen -syn dystrophic neurites are also frequently observed around senile plaques (Fig. 15.4(c)). In DLB and PD brains, the accumulation of both - and syn was observed in the mossy fibre terminals that synapse on hilar neurons (Galvin et al., 1999). However, this alteration in the properties of syn proteins is different from -syn pathological lesions since it is not defined by the present of fibrillar -syn. The accumulation of - and -syn

in synaptic terminals is likely due to the degeneration of presynaptic terminals independent of -syn fibrillogenesis, a notion that is consistent with the accumulation of other synaptic proteins such as syaptophysin, synapsin and synaptobrevin in the hilum (Galvin et al., 1999). The recent pathological re-examination of a member the Contursi kindred with the A53T -syn mutation and the report of an Australian kindred also harbouring this mutation suggested that these patients do not have typical PD pathology since they have a far more widespread -syn pathology than previously described (Duda et al., 2002b; Spira et al., 2001). Clinically, these patients initially presented with parkinsonism, but usually developed additional clinical features including progressive decline in cognitive function. Occasional classical LBs were observed in the SNpc (Fig. 15.4(d )), but the striking pathological change was the extensive -syn neuritic pathology throughout the brainstem and subcortical nuclei, the limbic system, and the cerebral cortex. Some areas, such as the dorsal vagal nucleus and the nucleus basalis of Meynert, contained abundant spheroids comprised of aggregated -syn (Fig. 15.4(e)). LB disorders likely constitute a spectrum of diseases. At one end of the extreme, LBs are restricted to brainstem nuclei typical of PD and at the other end of the spectrum LBs have a widespread distribution as in DLB. This notion has been supported by the analysis of 100 patients with PD that revealed the presence of at least some cortical LBs in all PD brains. Moreover, the spectrum of LB diseases can extend to rare cases with abundant LBs in the cerebral cortex and amygdala, but a paucity of PD pathology in the brainstem (Hansen et al., 1990; Kosaka et al., 1996).

Multiple system atrophy MSA is an adult onset neurodegenerative disease characterized by varying degrees of parkinsonism features, cerebellar ataxia, and autonomic dysfunction (Wenning et al., 1994, 1997). MSA was a termed used by Graham and Oppenheimer to group three syndromes: olivopontocerebellar atrophy (OPCA), striatonigral degeneration and Shy–Drager syndrome (Graham & Oppenheimer, 1969). The notion that MSA is a unique clinicopathological entity was consolidated by the description of glial cytoplasmic inclusions (GCIs) in oligodendrocytes of patients clinically diagnosed with this disorder (Kato et al., 1991; Papp et al., 1989). It is now widely accepted that MSA is a single disease entity defined pathologically by the presence of GCIs. The clinical assessment of patients with MSA has been the focus of a consensus criteria meeting, and it was proposed that the classification of patients with MSA should be


Fig. 15.5. Micrographs depicting coronal sections of the cerebellum from an MSA (a) and a control (b) patient. Note the severe atrophy and discolouration of the white matter. (c) Detection of GCIs in the cerebellar white matter of a patient with MSA using Gallyus silver staining. (d ) Immuno-labelling of GCIs with anti--syn antibody Syn 202 in the cerebellar white matter of a patient with MSA.The tissue section in (d ) was counterstained with haematoxylin. Scale bar = 40 m in (c) and (d ).

simplified to MSA-C (cerebellar ataxia) or MSA-P (parkinsonism) depending on the relative predominance of clinical and pathological abnormalities (Gilman et al., 1999). Under this new classification OPCA is referred as MSA-C, and striatonigral degeneration as MSA-P. The use of the term Shy–Drager, which was originally characterized by prominent autonomic dysfunction, was discouraged since this feature is common to all MSA patients.

MSA brains show varying degrees of atrophy of the cerebellum (Fig. 15.5(a)), pons, and medulla, as well as the loss of pigmented cells in the SNpc (Wenning et al., 1994, 1997). Demyelination can often be observed grossly by the ‘pallor’ or ‘loss of whiteness’ of the white matter resulting in a yellowish tint (Fig. 15.5(a)). In fact, the use of antibodies to a ‘buried’ epitope of basic myelin protein demonstrated that myelin degeneration in MSA is more widespread than

361

362


visible with conventional histological stains (Matsuo et al., 1998). GCIs usually appear as flame- or sickle-inclusions in oligodendrocytes that can be readily detected by Gallyas silver staining (Fig. 15.5(c)) (Papp et al., 1989). GCIs can be found throughout the white matter, but the greatest abundance of these inclusions occurs in the basal ganglia, the substantia nigra, the pontine nucleus, medulla and cerebellum (Arima et al., 1992; Duda et al., 2000b; Lantos, 1997). GCIs are composed ultrastructurally of a meshwork of randomly arranged, loosely packed filaments with cross-sectional diameters of 15 to 30 nm (Abe et al., 1992; Kato et al., 1991; Kato & Nakamura, 1990; Nakazato et al., 1990; Papp et al., 1989). The filaments are coated with electron dense granules or ‘fuzzy’ material, which increases the effective cross-sectional diameter to approximately 40–50 nm. Using immunological, ultrastructural and biochemical approaches, it has been demonstrated that these fibrils are composed of polymerized -syn. By immunocytochemistry, antibodies to -syn label all GCIs (Fig. 15.5(d )), but -syn and -syn are not present in these inclusions (Arima et al., 1998a; Duda et al., 2000b; Tu et al., 1998; Wakabayashi et al., 1998a, b). At the ultrastructural level, antibodies to -syn intensely and directly decorate the fibrillar component of GCIs. Moreover, the abundance of GCIs coincides with the accumulation of insoluble and SDS-resistant -syn aggregates (Dickson et al., 1999; Duda et al., 2000b; Tu et al., 1998). Whilst most pathological inclusions in MSA are in oligodendrocytes, protein aggregates can also be observed in the form of neuronal cytoplasmic inclusions (NCIs), most of which are indistinguishable from LBs, and in neuritic processes, especially in pontine nuclei and striatum (Arima et al., 1992; Kato et al., 1991). Infrequent neuronal nuclear inclusions (NNIs) and glial nuclear inclusions (GNIs) have also been reported (Papp & Lantos, 1992). NCIs, NNIs and neuritic aggregates also appear to be comprised of fibrillar -syn, but the composition of GCIs is still unresolved (Arima et al., 1998a; Spillantini et al., 1998a; Tu et al., 1998; Wakabayashi et al., 1998a).

Other synucleinopathies Neurodegeneration with brain iron accumulation type 1 (NBIA1) (previously known as Hallervorden–Spatz disease (Shevell, 1996)) is a rare progressive neurodegenerative disorder that occurs in both familial and sporadic forms (Dooling et al., 1974). Clinically, this disease is characterized by parkinsonian features, pyramidal signs, seizures and mental deterioration that can culminate in a dementia (Dooling et al., 1974). The histopathological findings defining NBIA-1 are neuroaxonal spheroids (i.e. 20–100 m wide

axonal swelling), and intense iron deposits in the globus pallidus and SNpc. The familial form of the disease is a recessive trait linked to chromosome 20p12.3–p13 (Taylor et al., 1996), and the gene defect involved was identified recently as pantothenate kinase 2 (PANK2) (Zhou et al., 2001). Missense mutations, nonsense mutations, insertion and deletions resulting in frameshift mutations and nucleotide substitutions resulting in aberrant splicing have been described in the PANK2 gene in patients with NBIA-1. PANKs are essential enzymes for the biosynthesis of coenzyme A, which is the major acyl carrier and plays a central role in fatty acid metabolism. The pathogenesis associated with PANK2 defects is still unclear, especially since at least three other PANKs are expressed in humans. However, PANK2 appears to be the predominant PANK expressed in certain brain regions (Zhou et al., 2001), suggesting that reduction in PANK activity in these areas may lead eventually to the accumulation of toxic product and/or the depletion of necessary biomolecules. LBs have been reported in the majority of the brains of NBIA1 patients, but some brains do not appear to display these lesions (see Saito et al., 2000 and references therein). Following the finding that -syn is a major component of pathological inclusions, several groups have reexamined NBIA-1 brains with anti--syn antibodies. It was demonstrated that LBs (Fig. 15.4(f )), LNs (Fig. 15.4(g)), and rare GCIs composed of -syn, can be widely distributed throughout the cortex, subcortical areas and brainstem of NBIA patients (Arawaka et al., 1998; Saito et al., 2000; Tu et al., 1998; Wakabayashi et al., 1999). However, the accumulation of -syn in axonal swellings has been more controversial. Some studies reported abundant and intense -syn immunoreactivity in neuroaxonal spheroids (Fig. 15.4(h)) (Arawaka et al., 1998; Galvin et al., 2000; Tu et al., 1998), while others noted only faint immunoreactivity within rare spheroids (Wakabayashi et al., 1999). This controversy was addressed further by Newell and colleagues in a study examining several disorders characterized by axonal spheroids (i.e. NBIA1, infantile neuroaxonal dystrophy, diffuse axonal injury, methotrexate-induced leukocephalopathy, Niemann–Pick disease-type C, and dementia pugilistica) (Newell et al., 1999). In this latter study, it was demonstrated that -syn is a common component of neuroaxonal spheroids regardless of the etiology. A subset of neuroaxonal spheroids, but not LBs or GCIs, also accumulates - and -syn (Galvin et al., 2000). The presence of LBs and LNs, albeit with a more restricted distribution that usually includes the amygdala, has been reported in a high percentage (40–60%) of other neurodegenerative diseases, including familial AD (due to mutations in the amyloid- precursor protein or presenilin1 and -2) (Fig. 15.4(i)) (Lippa et al., 1998), Down’s syndrome


(Lippa et al., 1999), and parkinsonism–dementia complex of Guam (Yamazaki et al., 2000; Forman et al., 2002). LBs and LNs were also observed specifically in the sympathetic and parasympathetic nervous system of patients with a rare disorder termed pure autonomic failure (Arai et al., 2000; Kaufmann et al., 2001).

Etiology of Parkinson’s disease Environmental factors in Parkinson’s disease Environmental factors have long been implicated in the etiology of PD. The pandemic of von Economo’s encephalitis and its sequela, postencephalitic parkinsonism (PEP), during the first half of the twentieth century suggested that an infectious agent could be involved in PD. The specific infectious agents causative of PEP have not been identified, and although an estimated 60% of those infected with von Economo encephalitis between 1916 to the 1930s developed PEP within 15 years of the viral illness (Duvoisin & Yahr, 1965), virtually no new cases have been reported since the mid-1950s (Rajput et al., 1984). More recent interest in environmental causes of PD was predominantly driven by the discovery that 1,2,3,6tetrahydropyridine (MPTP), a substance inadvertently produced during the attempt to synthesize a meperidine analogue, induced parkinsonism with concomitant selective damage to the nigrostriatal dopaminergic system in addicts who consumed what they thought was a ‘designer drug’ (Langston et al., 1983). The active form of MPTP, 1methyl-4-phenylpyridinium ion (MPP+), is selectively imported into dopamineregic neurons by the plasma membrane dopamine transporter, and MPP+ accumulates in mitochondria where it inhibits complex I (Przedborski & Jackson-Lewis, 1998). Although MPTP toxicity in monkeys recapitulates many of the features of PD, LBs are not observed in affected humans or in animal models of MPTPinduced parkinsonism (Forno et al., 1996; Langston et al., 1999). Thus, it is likely that the mechanisms of MPTP toxicity, although resulting in a selective demise of the nigrostriatal pathway, are different from those in synucleinopathies, and may not reflect the etiology of idiopathic PD. Further evidence for the possible involvement of environmental factors in PD comes from epidemiological studies suggesting that there is a greater prevalence of parkinsonism in rural areas (Hubble et al., 1993; Koller et al., 1990; Rajput et al., 1986, 1987b; Tanner et al., 1987). These findings are still controversial, since they were not supported by other studies (Seidler et al., 1996; Semchuk et al., 1991; Tanner et al., 1989), perhaps due to geographical or lifestyle differences in the population studied. However, it

is still proposed that chemicals in drinking water or exposure to pesticides or herbicides can increase the risk of PD. Indeed, some studies have indicated that the exposure to fertilizers, pesticides, and wood preservatives may pose an environmental risk for PD, but no specific agents have been identified consistently as causative in these substances (Barbeau et al., 1987; Koller et al., 1990; Seidler et al., 1996; Semchuk et al., 1992, 1993). Perhaps the best evidence for the possible role of environmental factors in causing PD is a recently described model where rats were exposed to rotenone by intravenous injection (Betarbet et al., 2000). Rotenone is a pesticide that recently has been used mostly to control fish populations, and similar to MPP+, it is a mitochondrial complex I inhibitor. The administration of rotenone to rats resulted in many features which closely parallel those seen in PD patients, including such cardinal phenotypic features as bradykinesia, posture instability and an unsteady gait. Moreover, the nigrostriatal dopaminergic system selectively degenerated, and surviving substantia nigra neurons developed intracytoplasmic -syn-rich inclusions that resembled LBs by light and electron microscopy. Furthermore, some inclusions had a distinct core surrounded by a fibrillar halo similar to authentic human LBs, and other inclusions resembled ‘pale bodies’, presumptive LB precursor lesions. As noted by Bertarbet et al. (2000), the levels of rotenone in the brain of treated rats appear insufficient for the inhibition of the electron transfer chain to be the sole mechanism of LB formation and cell death in this model. Alternatively, increased generation of free radicals resulting from mitochondrial inhibition has been suggested as a possible mechanism of cellular damage. The notion that oxidative injury plays an important role in PD as well as other neurodegenerative diseases has been supported by numerous reports (for reviews see Beal, 1996; Giasson et al., 2002b). In particular, relevant to PD are the permanent heightened levels of free radicals in dopaminergic neurons due to dopamine metabolism and autoxidation (Graham, 1978; Graham et al., 1978; Hastings et al., 1996; Smythies & Galzigna, 1998). This property is thought to render dopaminergic neurons of the SNpc more vulnerable to oxidative insults, which may be associated with the exposure to environmental toxins.

Genetics of Parkinson’s disease Although the vast majority of PD cases appear to be sporadic, several gene loci have been linked to kindreds with PD or a parkinsonism syndrome (Table 15.2). Furthermore, mutations in the tau gene also are pathogenic for parkinsonism associated with frontotemporal dementia (see Chapter 12). In addition, some controversial findings

363

364


Table 15.2. Summary of genetic loci linked to PD

Locus/gene

Inheritance

Onset

Pathology

Chromosomal position

Gene

PARK 1a PARK 2a PARK 3A PARK 4B UCH-L1a PARK 6C PARK 7D PARK 8E

dominant recessive dominant dominant dominant recessive recessive dominant

40s 20–40s 60s 30s ∼ 50 ∼ 40 30–40s 50s

Nigral degeneration; Lewy pathology Nigral degeneration; no Lewy pathologya Nigral degeneration; Lewy pathology Nigral degeneration; Lewy pathology Unknown Unknown Unknown Nigral degenration; no Lewy pathology

4q21 6q25.2–q27 2p13 4p16 4p14 1p35–37 1p36 12p11.2–q13.1

−syn Parkin ? ? UCH-L1 ? DJ-1 ?

a

See text; A–Gasser et al., (1998); B–Farrer et al., (1999); Gwinn-Hardy et al., (2000); C–Valente et al., (2001); Bentivoglio et al., (2001); D–van Duijn et al., (2001); E–Funayama et al., (2002).

suggest that polymorphisms in the genes for ApoE and the detoxifying enzyme cytochrome p450 2D6 (CYP2D6) may increase or decrease the risk for sporadic PD, or may influence the age of onset (Atkinson et al., 1999; Bon et al., 1999; Bordet et al., 1994; Christensen et al., 1998; Kosel et al., 1996; Kruger et al., 1999; Sabbagh et al., 1999). Other polymorphisms, for example, a substitution in mitochondrial tRNA (A4336G) has been cited as a potential risk factor for both AD and PD (Egensperger et al., 1997). Recently, an amino acid substitution (G336S) was identified in the midsize neurofilament subunit in a proband with PD, but this substitution, if causal of disease, would have a low penetrance (25%) (Lavedan et al., 2002). The identification of precise genetic mutations causal of movement disorders has provided new insights into the pathogenic mechanisms of PD. Two mutations (A53T and A30P) have been identified in the -syn (SNCA) gene. The clinical and pathological features associated with the A53T mutation were recently re-examined (see section on DLB), and although these patients have the characteristics of PD, they also accumulate widespread -syn inclusions, which is more similar to DLB pathology. Although dementia is not a prominent feature, perhaps because they were not the focus of intense study in this kindred, these patients are affected by progressive cognitive decline. However, the pathology in patients harbouring the A53T mutation is distinctive for typical DLB patients as they have a much greater abundance of -syn neuroaxonal spheroids. In vitro studies of fibrillogenesis and transgenic mouse models support the idea that the pathogenicity of the A53T mutation is due to increased propensity to form fibrils (see sections on ‘Fibrillogenesis’ and ‘Animal models’ below). Clinically patients with the A30P SNCA mutation develop parkinsonism (Kruger et al., 1998), but due to incomplete

post-mortem pathological assessment, it is still unknown if they have definite PD with -syn pathology. The possible mechanisms of disease pathogenesis associated with this mutation are also unclear (see section of ‘Fibrillogenesis’ below). Genetic alterations resulting in the disruption of the parkin gene were initially identified in patients with a disease entity termed autosomal recessive juvenile parkinsonism (AR-JP) (Kitada et al., 1998). These patients display the typical clinical features of parkinsonism, but they also often present with foot dystonia, show symptomatic benefit from sleep, and display a marked response to levodopa therapy. Onset can be as early as the first decade of life, but for most cases it is later than what could be considered ‘juvenile’. In fact, onset is best described as before the age of 40, although some rare cases can have onset as late as the sixth decade of life (Klein et al., 2000). Neuropathological examination demonstrates that neuronal loss and gliosis are restricted to the SN and locus ceruleus (Hayashi et al., 2000a, b; Mori et al., 1998; Takahashi et al., 1994). In several cases LB pathology was not observed (Hayashi et al., 2000a, b), leading to the speculation that the etiology of AR-JP differs significantly from PD. However, the recent autopsy of a patient with a 40 base pair deletion in exon 3 in one allele of parkin and a R275W mutation in the other allele revealed the presence -syn positive LB pathology in the typical regions affected in PD (Farrer et al., 2001). Mutations in the parkin gene may account for the majority of patients with early onset parkinsonism. Almost all types of genetic alterations in the parkin gene have been associated with disease. Many different nonsense mutations, missense mutations and frameshift mutations resulting from the insertion of one, two or five base pairs have been reported (Giasson & Lee, 2001; Lucking et al., 2000). However, deletions of one or several exons and duplications and


triplications are also common (Kitada et al., 1998; Lucking et al., 2000). Parkin functions as an E3 ligase (Shimura et al., 2000; Zhang et al., 2000), a component of a protein complex involved in the ubiquitination of proteins resulting in their targeting for degradation by the proteasome (Hershko & Ciechanover, 1998). Several parkin substrates has been identified (for review see Giasson & Lee, 2001); however, the possibility that -syn may be a substrate for this ligase suggested that parkin may be involved in idiopathic PD. Shimura et al. suggested that parkin may recognize and promote the ubiquitination of a newly identified Oglycosylated isoform of -syn (Sp22) that contains complex monosaccharide chains (Shimura et al., 2001), which is highly unusual for a cytoplasmic protein. Although the levels of Sp22 are higher in AR-JP brains compared to control brains, where it is very scarce, the importance of Sp22 is still unclear since it constitutes only a very minor fraction of total -syn (Shimura et al., 2001). Furthermore, the specificity of parkin for the glycosylated isoform of -syn is very stringent, as non-gycosylated -syn is completely inert to parkin (Chung et al., 2001; Shimura et al., 2001). Shimura et al. proposed that ubiquitination of Sp22 may be a prerequisite to the formation of LBs, since it was believed that AR-JP patients do not have these inclusions. The recent report of an AR-JP patient with -syn positive LBs and LNs (Farrer et al., 2001) demonstrates that some of these patients can develop these inclusions, but it does not completely negate the hypothesis put forward by Shimura et al. The patient described by Farrer et al. harbours one null allele and a R275W mutation in the other allele, which reduces the catalytic activity of parkin, but it still has substantial activity (Chung et al., 2001). Previous autopsies of AR-JP patients demonstrating no -syn positive pathology were performed on individuals with homologous null mutations. Nevertheless, further investigation will be necessary to resolve whether parkin is required for the formation of Lewy pathology and whether parkin is involved in idiopathic PD. Genetic analysis has also linked another protein involved in ubiquitin metabolism with PD. A nucleotide substitution resulting in the amino acid change I93M in ubiquitin carboxy-terminal hydrolase (UCH)-L1 was found in two siblings with parkinsonism (Leroy et al., 1998). Since this amino acid change or other substitution in UCHL-1 has not been identified in other PD patients, it is still not entirely clear if the I93M substitution is a dominant mutation causal of disease or an extremely rare polymorphism that co-segregates with disease. These uncertainties notwithstanding, the I93M mutation in UCHL-1 results in a partial loss of catalytic activity (Leroy et al., 1998). However, UCHs

constitute a large family of de-ubiquitinating enzymes that can cleave polymeric ubiquitin into monomers, and it is puzzling that the partial reduction in activity of a single allele could be pathogenic. Nevertheless, these enzymes are thought to be important to reduce the ubiquitin-tagged degradation of specific proteins and in the regeneration of free and re-usable ubiquitin following protein degradation (Hershko and Ciechanover, 1998).

Mechanisms of -syn polymerization and inclusion formation In vitro modelling of -syn fibrillogenesis -Syn can readily assemble in vitro into elongated homopolymers with similar widths as sarcosyl-insoluble fibrils isolated from human diseased brains or filaments visualized directly in LBs and GCIs (see Fig. 15.2(d )) (Conway et al., 1998; El Agnaf et al., 1998; Giasson et al., 1999; Narhi et al., 1999). Polymerization is associated with a concomitant change in secondary structure from random coil to anti-parallel -sheet structure (Conway et al., 2000a; Narhi et al., 1999; Serpell et al., 2000) consistent with the Thioflavine-S reactivity of these filaments (Conway et al., 2000a; Narhi et al., 1999). One of the rate limiting steps in the polymerization of -syn appears to be the formation of ‘seeds’ or protofibrils (see Fig. 15.6; Wood et al., 1999; Conway et al., 2000b), and although controversial, it has also been suggested that some types of protofibrils may have toxic properties (Volles et al., 2001). The A53T SNCA mutation may accelerate fibrillogenesis, as recombinant A53T -syn has a greater propensity to polymerize into fibrils than wild-type -syn (Conway et al., 1998; El Agnaf et al., 1998; Giasson et al., 1999; Narhi et al., 1999). This mutation also affects the ultrastructure of the polymers in that the filaments are slightly wider and more twisted in appearance, as if assembled from two protofilaments (Conway et al., 1998; El Agnaf et al., 1998; Giasson et al., 1999). The A30P mutation may also modestly increase the propensity of -syn to polymerize (El Agnaf et al., 1998; Narhi et al., 1999), but this finding has not been reported consistently (Giasson et al., 1999; Serpell et al., 2000). However, these discrepancies may be due to technical differences in methods for assessing -syn polymerization, since the A30P mutation may increase the propensity to form spherical protofibrils rather than filaments (Conway et al., 2000b). The pathological effects of this mutation also may be related to its reduced binding to vesicles (Jensen et al., 1998), but this notion has been challenged (Perrin et al., 2000).

365

366


Fig. 15.6. Model of the formation of filamentous -syn inclusions. (a) Under normal physiological conditions, -syn exists predominantly as a random coil monomer (open circle). Small amounts of -syn exist in a -pleated sheet confirmation (open rhombus). The latter monomers may polymerize into short oligomers and protofibrils. Further polymerization results in elongated filaments. Under normal conditions, kinetics favour the presence of random coiled monomers. Chaperone and proteolytic activities may act to prevent the formation of elongated filaments. (b) Under pathological conditions the equilibrium may be shifted so that the formation of inclusions can occur. Increased expression of -syn may result in an increase in monomers (1). Reduced proteolytic (2) or chaperone (3) activities may result in diminished clearance of -syn protein with -pleated sheet conformation. Other alteration in the cytoplasmic milieu and modification of -syn may increase the propensity of -syn to convert into a -pleated sheet confirmation (4). The accumulation of protofibrils will drive the formation of elongated filaments, but some types of protofibrils also may be toxic. Elongated filaments will intertwine and organize to form pathological lesions. Oxidative/nitrative-induced cross-links (V) will irreversibly stabilize polymers, fibrils and filament bundles promoting the formation of pathological inclusions.

The ability to study -syn fibrillogenesis in vitro has allowed investigators to identify conditions that could induce or inhibit this process. For example, di- and trivalent metals (Uversky et al., 2001a), large polymers (i.e. polyethylene glycol) (Shtilerman et al., 2002), glycosaminoglycans (Cohlberg et al., 2002) and some pesticides (Uversky et al., 2001b) can induce fibrillogenesis. Phosphorylation of Ser-129 in -syn, which is a substrate for casein kinase II, also can promote the assembly of -syn into fibrils in vitro. (Fujiwara et al., 2002). Further supporting the notion that this post-translational modification can promote inclusion formation in vivo are studies showing that detergent-insoluble -syn in diseased brain appears to be extensively phosphorylated (∼90%), while

only a small fraction of soluble -syn in normal brain is modified. Oxidation-induced alterations can also influence -syn polymerization, but the outcomes are more complex. Oxidative processes can promote the formation of aggregates (Hashimoto et al., 1999a, b), but these events may not culminate in the formation of filamentous inclusions, since the oxidation of specific amino acid residues or the generation of oxidatively-induced adducts can inhibit the assembly of mature filaments (Conway et al., 2001; Uversky et al., 2002b). On the other hand, if oxidative/ nitrative alterations occur after -syn polymerization, covalent cross-linking can stabilize intermediate-sized polymers and filaments driving the process towards elongated filaments (Souza et al., 2000a). Furthermore, cross-linking


also can promote the bundling of fibrils promoting the formation of inclusions. The widespread nitrative damage evidenced by most -syn pathological lesions in all synucleinopathies analyzed (Duda et al., 2000a; Giasson et al., 2000a) suggests that oxidative/nitrative alterations are involved in promoting inclusion formation. Tissue culture experimentations have demonstrated that oxidative or nitrative injury can induce the formation of intracellular -syn aggregates (Ostrerova-Golts et al., 2000; Paxinou et al., 2001). Ablation of the carboxyl-terminal region of -syn increases the propensity of the protein to form fibrils (Crowther et al., 1998; Serpell et al., 2000). Although the pathological implications of the latter finding are still unclear, it is possible that the aberrant proteolysis of -syn may promote the formation of ‘seeds’ that could initiate -syn filament assembly. Further structure–function analysis of -syn demonstrated that the middle hydrophobic region of the protein (see Fig. 15.1) is required for filament formation. -syn lacks the ability to form fibrils under the same conditions that result in -syn fibrillization, because it is deficient in an 11 amino acid stretch within this region (Biere et al., 2000; Giasson et al., 2001b; Serpell et al., 2000). In contrast, -syn, which has a similar hydrophobic region to -syn (Fig. 15.1), can polymerize at a much slower rate (Biere et al., 2000). Furthermore, -syn cannot co-assemble with -syn, but a four-fold molar excess of -syn appears to reduce the rate of -syn polymerization (Uversky et al., 2002a). A model of hypothetical steps leading to -syn fibrillization and inclusion formation is shown in Fig. 15.6.

Transgenic animal models of synucleinopathies Expression of human wild-type or mutant -syn in Drosophila results in the formation of filamentous -syn inclusions concomitant with the demise of dorsomedial dopaminergic neurons and impairment of locomotor function (Feany & Bender, 2000). The over-expression of the chaperone heat shock protein 70 (HSP70) can suppress dopaminergic neuronal loss without a notable reduction in intracellular inclusions (Auluck et al., 2002). Perhaps, the level of -syn expressed in flies is too high for the expression of HSP70 to prevent the formation of inclusions or HSP70 chaperone activity may not be involved in -syn folding. -Syn pathological inclusions in flies and in humans can sequester HSP70, suggesting that the depletion of functional HSP70 from normal cellular activity by these inclusions may contribute to neuronal demise. The first described transgenic animal model of synucleinopathies was mice overexpressing wild-type hu-

man -syn driven by the PDGF- promoter. The expression of -syn in these mice resulted in the formation of amorphous, non-filamentous -syn aggregates associated with a mild impairment of motor function and reduction in striatal dopaminergic axonal terminals (Masliah et al., 2000). The PDGF--syn mice were bred with another transgenic line expressing human amyloid precursor protein, and which develop extracellular -amyloid plaques, to generate double transgenic lines. These double transgenic mice demonstrated earlier motor impairment and developed deficits in learning and memory associated with the accumulation of filamentous -syn inclusions (Masliah et al., 2001). Moreover, the cross-breeding of the -syn transgenic lines with transgenic lines over-expressing human -syn ameliorates the motor impairments concomitant with a reduction in -syn inclusions (Hashimoto et al., 2001). Expression of wild-type and A53T or A30P mutant -syn in transgenic mice driven by the Thy-1 promoter resulted in the appearance of perikaryal and neuritic accumulations, but without notable differences between wild-type and mutant proteins (Kahle et al., 2000; van der Putten et al., 2000). In one study, mice expressing wild-type or A53T mutant -syn developed an early onset motor impairment (> 3 weeks of age) as measured by rotating rod performance, and this phenotype was associated with axonal degeneration in the spinal roots and muscle denervation (van der Putten et al., 2000). A subset of -syn inclusions in these mice was argyrophilic and immunoreactive for ubiquitin, although they lacked the filamentous structures that are characteristic of authentic human -syn inclusions. These mice also accumulated detergent-insoluble -syn similar to human brains with pathological inclusions (Kahle et al., 2001). A more recent transgenic mouse study using the PrP promoter to express human wild-type and A53T -syn demonstrated the dramatic differences in the pathogenicity of these proteins (Giasson et al., 2002a). Mice expressing wild-type -syn remained healthy without any notable pathological changes. However, mice expressing A53T syn developed a progressive early-adult onset complex motor dysfunction leading to paralysis and death. This phenotype coincided with the appearance of widely distributed neuronal pathological inclusions comprised of -syn in most regions of the neuroaxis, although the SNpc and hippocampus were spared. The morphological, biochemical, histological (e.g. amyloidogenic), and ultrastructural (i.e. they are composed of 10–16 nm -syn fibrils) properties of these inclusions closely parallelled those of human pathological inclusions. -Syn inclusions in transgenic mice might lead to the congestion of, or impaired, intracellular trafficking in cell bodies and axons. One of

367

368


the pathological consequences of these inclusions is the obstruction of axonal transport resulting in Wallerian-type degeneration. Collectively, it can be surmised from these animal models that -syn inclusions can lead to neuron dysfunction and death. Mice with filamentous inclusions develop the worst phenotype leading to premature early-adult onset death suggesting that these inclusions, and not protofibrils, are pathological. However, both modes of toxicity are not mutually exclusive, and the notion that some protofibrils may be harmful cannot be overlooked.

Future direction and therapeutic intervention It is still unclear what initiates the formation of -syn pathological inclusions in the majority of affected patients, although genetic components, especially mutations in -syn, can enhance this process in some individuals. In MSA, it is tempting to speculate that an aberrant increase in expression and/or a reduction in the degradation of -syn, which is normally only expressed at low levels in oligodendrocytes (Richter-Landsberg et al., 2000), may eventually lead to the accumulation of -syn promoting the formation of inclusions. Although the same mechanisms could be involved in neuronal inclusions, other factors such as changes in cellular enzymatic activities or post-translational modification of -syn may be involved (Fig. 15.6). Transgenic -syn models suggest that the formation of pathological inclusions leads to the demise of neurons, likely by acting as molecular obstacles preventing normal cytoplasmic transport and morphology. -Syn inclusions also can behave as ‘sieves’, trapping other macromolecules and perturbing cellular homeostasis, axonal transport and synaptic transmission. The disruption of axonal transport can result in the dying back of axons and eventually neuronal death. Many important aspects of -syn biology and biophysical properties of polymerization remain to be elucidated in order to assess the best course of action for novel therapeutics. For example, the putative toxicity of protofibrils will have to be further investigated. More sensitive and selective diagnostic methods are required for effective intervention aimed at preventing the formation of syn inclusions. Since early detection preceding the formation of inclusions may be difficult, it will be important to determine if the formation of -syn inclusions is reversible as has been demonstrated in transgenic models of other neurodegenerative diseases associated with proteinaceous inclusions (Schenk et al., 1999; Yamamoto et al., 2000). The expeditious progress in the fundamental understanding of synucleinopathies within just the last 5 years suggests that

novel therapeutic approaches may be developed in the near future.

Acknowledgements This work was supported by the National Institutes of Health and by a Pioneer Award from the Alzheimer’s Association. V. M.-Y. L. is the John H. Ware III Professor in Alzheimer’s disease research. B. I. G. is the recipient of a fellowship from the Canadian Institutes of Health Research. We are grateful to Drs. Douglas C. Miller and Lawrence I. Golbe for providing tissue from a patient of the Contursi kindred.

REFERENCES Aarland, D., Tangberg, E., Larsen, J. P. & Cummings, J. L. (1996). Frequency of dementia in Parkinson’s disease. Arch. Neurol., 53, 538–42. Abe, H., Yagishita, S., Amano, N., Iwabuchi, K., Hasegawa, K. & Kowa, K. (1992). Argyrophilic glial intracytoplasmic inclusions in multiple system atrophy: immunocytochemical and ultrastructural study. Acta Neuropathol. (Berl.), 84, 273–7. Abeliovich, A., Schmitz,Y., Farinas, I. et al. (2000). Mice lacking alpha-synuclein display functional deficits in the nigrostriatal dopamine system. Neuron, 25, 239–52. Arai, K., Kato, N., Kashiwado, K. & Hattori,T. (2000). Pure autonomic failure in association with human alpha-synucleinopathy. Neurosci. Lett., 296, 171–3. Arawaka, S., Saito, Y., Murayama, S. & Mori, H. (1998). Lewy body in neurodegeneration with brain iron accumulation type 1 is immunoreactive for alpha-synuclein. Neurology, 51, 887–9. Arima, K., Murayama, S., Mukoyama, M. & Inose, T. (1992). Immunocytochemical and ultrastructural studies of neuronal and oligodendroglial cytoplasmic inclusions in multiple system atrophy. Acta Neuropathol. (Berl.), 83, 453–60. Arima, K., Ueda, K., Sunohara, N. et al. (1998a). NACP/alphasynuclein immunoreactivity in fibrillary components of neuronal and oligodendroglial cytoplasmic inclusions in the pontine nuclei in multiple system atrophy. Acta Neuropathol. (Berl.), 96, 439–44. Arima, K., Ueda, K. et al. (1998b). Immunoelectron-microscopic demonstration of NACP/alpha-synuclein epitopes on the filamentous component of Lewy bodies in Parkinson’s disease and in dementia with Lewy bodies. Brain Res., 808, 93–100. Athanassiadou, A., Voutsinas, G., Psiouri, L. et al. (1999). Genetic analysis of families with Parkinson disease that carry the Ala53Thr mutation in the gene encoding alpha-synuclein. Am. J. Hum. Genet., 65, 555–8. Atkinson, A., Singleton, A. B., Steward, A. et al. (1999). CYP2D6 is associated with Parkinson’s disease but not with dementia with Lewy Bodies or Alzheimer’s disease. Pharmacogenetics, 9, 31–5.


Auluck, P. K., Chan, E., Trojanoswki, J. Q., Lee, V. M.-Y., & Bonini, N. M. (2002). Chaperon suppression of -synuclein toxicity in a Drosophila model of Parkinson’s disease. Science, 295, 865–8. Baba, M., Nakajo, S., Tu, P. H. et al. (1998). Aggregation of alphasynuclein in Lewy bodies of sporadic Parkinson’s disease and dementia with Lewy bodies. Am. J. Pathol., 152, 879–84. Barbeau, A., Roy, M., Bernier, G., Campanella, G. & Paris, S. (1987). Ecogenetics of Parkinson’s disease: prevalence and environmental aspects in rural areas. Can. J. Neurol. Sci., 14, 36–41. Bayer, T. A., Jakala, P., Hartmann, T. et al. (1999). Alpha-synuclein accumulates in Lewy bodies in Parkinson’s disease and dementia with Lewy bodies but not in Alzheimer’s disease beta-amyloid plaque cores. Neurosci. Lett., 266, 213–16. Beal, M. F. (1996). Mitochondria, free radicals, and neurodegeneration. Curr. Opin. Neurobiol., 6, 661–6. Bentivoglio, A. R., Cortelli, P., Valente, E.M. et al. (2001). Phenotypic characterisation of autosomal recessive PARK6-linked parkinsonism in three unrelated Italian families. Mov Disord., 16, 999– 1006. Bergeron, C. & Pollanen, M. (1989). Lewy bodies in Alzheimer disease – one or two diseases? Alzheimer Dis. Assoc. Disord., 3, 197–204. Betarbet, R., Sherer, T. B., MacKenzie, G., Garcia-Osuna, M., Panov, A. V. & Greenamyre, J. T. (2000). Chronic systemic pesticide exposure reproduces features of Parkinson’s disease. Nat. Neurosci., 3, 1301–6. Biere, A. L., Wood, S. J., Wypych, J. et al. (2000). Parkinson’s diseaseassociated alpha-synuclein is more fibrillogenic than - and synuclein and cannot cross-seed its homologs. J. Biol. Chem., 275, 34574–9. Bon, M. A., Jansen Steur, E. N., de Vos, R. A. & Vermes, I. (1999). Neurogenetic correlates of Parkinson’s disease: apolipoprotein-E and cytochrome P450 2D6 genetic polymorphism. Neurosci. Lett., 266, 149–51. Bordet, R., Broly, F., Destee, A. & Libersa, C. (1994). Genetic polymorphism of cytochrome P450 2D6 in idiopathic Parkinson disease and diffuse Lewy body disease. Clin. Neuropharmacol., 17, 484–8. Braak, H., Braak, E., Yilmazer, D. e al. (1994). Amygdala pathology in Parkinson’s disease. Acta Neuropathol. (Berl.), 88, 493–500. Braak, H., Sandmann-Keil, D., Gai, W. & Braak, E. (1999). Extensive axonal Lewy neurites in Parkinson’s disease: a novel pathological feature revealed by alpha-synuclein immunocytochemistry. Neurosci. Lett., 265, 67–9. Brown, R. G. & Marsden, C. D. (1984). How common is dementia in Parkinson’s disease? Lancet, 2, 1265. Bruening, W., Giasson, B. I., Klein-Szanto, A. J., Lee, V. M.-Y., Trojanowski, J. Q. & Godwin, A. K. (2000). Synucleins are expressed in the majority of breast and ovarian carcinomas and in preneoplastic lesions of the ovary. Cancer, 88, 2154–63. Buchman, V. L., Hunter, H. J., Pinon, L. G. et al. (1998). Persyn, a member of the synuclein family, has a distinct pattern of expression in the developing nervous system. J. Neurosci., 18, 9335–41.

Burkhardt, C. R., Filley, C. M., Kleinschmidt-DeMasters, B. K., de la Monte, S., Norenberg, M. D. & Schneck, S. A. (1988). Diffuse Lewy body disease and progressive dementia. Neurology, 38, 1520–8. Campbell, B. C., Li, Q. X., Culvenor, J. G. et al, (2000). Accumulation of insoluble alpha-synuclein in dementia with Lewy bodies. Neurobiol. Dis., 7, 192–200. Christensen, P. M., Gotzsche, P. C. & Brosen, K. (1998). The sparteine/debrisoquine (CYP2D6) oxidation polymorphism and the risk of Parkinson’s disease: a meta-analysis. Pharmacogenetics, 8, 473–9. Chung, K. K., Zhang, Y., Lim, K. L. et al. (2001). Parkin ubiquitinates the alpha-synuclein-interacting protein, synphilin- 1: implications for Lewy-body formation in Parkinson disease. Nat. Med., 7, 1144–50. Cohlberg, J. A., Li, J., Uversky, V. N. & Fink, A. L. (2002). Heparin and other glycosaminoglycans stimulate the formation of amyloid fibrils from alpha-synuclein in vitro. Biochemistry, 41, 1502–11. Conway, K. A., Harper, J. D. & Lansbury, P. T. (1998). Accelerated in vitro fibril formation by a mutant alpha-synuclein linked to early-onset Parkinson disease. Nat. Med., 4, 1318–20. (2000a). Fibrils formed in vitro from alpha-synuclein and two mutant forms linked to Parkinson’s disease are typical amyloid. Biochemistry, 39, 2552–63. Conway, K. A., Lee, S. J., Rochet, J. C., Ding, T. T., Williamson, R. E. & Lansbury, P. T. (2000b). Acceleration of oligomerization, not fibrillization, is a shared property of both alpha-synuclein mutations linked to early-onset Parkinson’s disease: implications for pathogenesis and therapy. Proc. Natl Acad. Sci. USA, 97, 571–6. Conway, K. A., Rochet, J. C., Bieganski, R. M. & Lansbury, P. T., Jr. (2001). Kinetic stabilization of the alpha-synuclein protofibril by a dopamine–alpha-synuclein adduct. Science, 294, 1346–9. Cronford, M. E., Chang, L. & Miller, B. L. (1995). The neuropathology of Parkinsonism: an overview. Brain Cogn., 28, 321–41. Crowther, R. A., Jakes, R., Spillantini, M. G. & Goedert, M. (1998). Synthetic filaments assembled from C-terminally truncated alpha- synuclein. FEBS Lett., 436, 309–12. Crowther, R. A., Daniel, S.E. & Goedert, M. (2000). Characterisation of isolated alpha-synuclein filaments from substantia nigra of Parkinson’s disease brain. Neurosci. Lett., 292, 128–30. Crystal, H. A., Dickson, D. W., Lizardi, J. E., Davies, P. & Wolfson, L. I. (1990). Antemortem diagnosis of diffuse Lewy body disease. Neurology, 40, 1523–8. Culvenor, J. G., McLean, C. A., Cutt, S. et al. (1999). Non-Abeta component of Alzheimer’s disease amyloid (NAC) revisited. NAC and alpha-synuclein are not associated with Abeta amyloid. Am. J. Pathol., 155, 1173–81. Culvenor, J. G., McLean, C. A., Campbell, B. C. V., Masters, C. L. & Li, Q.-X. (2000). The role of NAC in amyloidogenesis in Alzheimer’s disease. Am. J. Pathol., 156, 735. Damier, P., Hirsch, E. C., Agid, Y. & Graybiel, A. M. (1999). The substantia nigra of the human brain. II. Patterns of loss of

369

370


dopamine-containing neurons in Parkinson’s disease. Brain 122 (8), 1437–48. Davidson, W. S., Jonas, A., Clayton, D. F. & George, J. M. (1998). Stabilization of alpha-synuclein secondary structure upon binding to synthetic membranes. J. Biol. Chem., 273, 9443–9. de Rijk, M. C., Tzourio, C., Breteler, M. M. et al. (1997). Prevalence of parkinsonism and Parkinson’s disease in Europe: the EUROPARKINSON Collaborative Study. J. Neurol. Neurosurg. Psychiatry, 62, 10–15. Dickson, D. W., Crystal, H., Mattiace, L. A. et al. (1989). Diffuse Lewy body disease: light and electron microscopic immunocytochemistry of senile plaques. Acta Neuropathol. (Berl.), 78, 572–84. Dickson, D. W., Ruan, D., Crystal, H. et al. (1991). Hippocampal degeneration differtiate diffuse Lewy body disease (DLBD) from Alzheimer’s disease. Neurology, 41, 1402–9. Dickson, D. W., Liu, W., Hardy, J. et al. (1999). Widespread alterations of alpha-synuclein in multiple system atrophy. Am. J. Pathol., 155, 1241–51. Dooling, E. C., Schoene, W. C. & Richardson, E. P., Jr. (1974). Hallervorden-Spatz syndrome. Arch. Neurol., 30, 70–83. Duda, J. E., Giasson, B. I., Chen, Q. et al. (2000a). Widespread nitration of pathological inclusions in neurodegenerative synucleinopathies. Am. J. Pathol., 157, 1439–45. Duda, J. E., Giasson, B. I., Gur, T. L. et al. (2000b). Immunohistochemical and biochemical studies demonstrate a distinct profile of alpha-synuclein permutations in multiple system atrophy. J. Neuropathol. Exp. Neurol., 59, 830–41. Duda, J. E., Giasson, B. I., Mabon, M. E., Lee, V. M.-Y. & Trojanoswki, J. Q. (2002a). Novel antibodies to oxidized -synuclein reveal abundant neuritic pathology in Lewy body disease. Ann. Neurol., 52, 205–10. Duda, J. E., Giasson, B. I., Mabon, M. E. et al. (2002b). Concurrence of -synuclein and tau brain pathology in the Contursi kindred. Acta Neuropathol. (Berl.), 104, 7–11. Duffy, P. E. & Tennyson, V. M. (1965). Phase and electron microscopy observations of Lewy bodies and melamin granules in the substantia nigra and locus ceruleus in Parkinson’s disease. J. Neuropathol. Exp. Neurol., 24, 398–413. Dunnett, S. B. & Björkland, A. (1999). Prospects for new restorative and neuroprotective treatments in Parkinson’s disease. Nature, 399, A32–9. Duvoisin, R. C. & Yahr, M. D. (1965). Encephalitis and parkisonism. Arch. Neurol., 12, 227–239. Egensperger, R., Kosel, S., Schnopp, N. M., Mehraein, P. & Graeber, M. B. (1997). Association of the mitochondrial tRNA(A4336G) mutation with Alzheimer’s and Parkinson’s diseases. Neuropathol. Appl. Neurobiol., 23, 315–21. El Agnaf, O. M., Jakes, R., Curran, M. D. & Wallace, A. (1998). Effects of the mutations Ala30 to Pro and Ala53 to Thr on the physical and morphological properties of alpha-synuclein protein implicated in Parkinson’s disease. FEBS Lett., 440, 67–70. Farrer, M., Gwinn-Hardy, K., Muenter, M. et al. (1999). A chromosome 4p haplotype segregating with Parkinson’s disease and postural tremor. Hum. Mol. Genet., 8, 81–5.

Farrer, M., Chan, P., Chen, R. et al. (2001). Lewy bodies and parkinsonism in families with parkin mutations. Ann. Neurol., 50, 293–300. Feany, M. B. & Bender, W. W. (2000). A Drosophila model of Parkinson’s disease. Nature, 404, 394–8. Fearnley, J. & Lees, A. J. (1991). Parkinson’s disease: neuropathology. Brain, 114, 2283–301. Forman, M. S., Schmidt, M. L., Kasturi, S., Perl., D. P., Lee, V. M.-Y. & Trojanowski, J. Q. (2002). Tau and -synuclein pathology in amygdala of parkinsonism-dementia complex patients of Guam. Am. J. Pathol., 160, 1725–31. Forno, L. S. (1996). Neuropathology of Parkinson’s disease. J. Neuropathol. Exp. Neurol., 55, 259–72. Forno, L. S., DeLanney, L. E., Irwin, I. & Langston, J. W. (1996). Electron microscopy of Lewy bodies in the amygdalaparahippocampal region. Comparison with inclusion bodies in the MPTP-treated squirrel monkey. Adv. Neurol., 69, 217–28. Fujiwara, H., Hasegawa, M., Dohmae, N. et al. (2002). -Synuclein is phosphorylated in synucleinopathy lesions. Nat.Cell Biol., 4, 160–4. Funayama, M., Hasegawa, K., Kowa, H., Saito, M., Tsuji, S. & Obata, F. (2002). A new locus for Parkinson’s disease (PARK8) maps to chromosome 12p11.2–q13.1. Ann. Neurol., 51, 296–301. Galasko, D., Hansen, L. A., Katzman, R. et al. (1994). Clinical– neuropathological correlations in Alzheimer’s disease and related dementias. Arch. Neurol., 51, 888–95. Galvin, J. E., Lee, V. M. -Y., Baba, M. et al. (1997). Monoclonal antibodies to purified cortical Lewy bodies recognize the mid-size neurofilament subunit. Ann. Neurol., 42, 595–603. Galvin, J. E., Uryu, K., Lee, V. M. -Y. & Trojanowski, J. Q. (1999). Axon pathology in Parkinson’s disease and Lewy body dementia hippocampus contains alpha-, beta-, and gamma-synuclein. Proc. Natl. Acad. Sci. USA, 96, 13450–5. Galvin, J. E., Giasson, B., Hurtig, H. I., Lee, V. M.-Y. & Trojanowski, J. Q. (2000). Neurodegeneration with brain iron accumulation, type 1 is characterized by alpha-, beta-, and gamma-synuclein neuropathology. Am. J. Pathol., 157, 361–8. Gasser, T., Muller-Myhsok, B., Wszolek, Z. K. et al. (1998). A susceptibility locus for Parkinson’s disease maps to chromosome 2p13. Nat. Genet., 18, 262–5. George, J. M., Jin, H., Woods, W. S. & Clayton, D. F. (1995). Characterization of a novel protein regulated during the critical period for song learning in the zebra finch. Neuron, 15, 361–72. Giasson, B. I. & Lee, V. M.-Y. (2001). Parkin and the molecular pathways of Parkinson’s disease. Neuron, 31, 885–8. Giasson, B. I., Uryu, K., Trojanowski, J. Q. & Lee, V. M.-Y. (1999). Mutant and wild type human alpha-synucleins assemble into elongated filaments with distinct morphologies in vitro. J. Biol. Chem., 274, 7619–22. Giasson, B. I., Duda, J. E., Murray, I. V. et al. (2000a). Oxidative damage linked to neurodegeneration by selective alpha-synuclein nitration in synucleinopathy lesions. Science, 290, 985–9. Giasson, B. I., Duda, J. E., Forman, M. S., Lee, V. M.-Y. & Trojanoswki, J. Q. (2001b). Prominent perikaryal expression of - and -synuclein in neurons of dorsal root ganglion and in medullary neurons. Exp. Neurol., 172, 354–62.


Giasson, B. I., Murray, I. V., Trojanowski, J. Q. & Lee, V. M.-Y. (2001c). A hydrophobic stretch of 12 amino acid residues in the middle of alpha- synuclein is essential for filament assembly. J. Biol. Chem., 276, 2380–6. Giasson, B. I., Duda, J. E., Quinn, S. M., Zhang, B., Trojanoswki, J. Q. & Lee, V. M.-Y. (2002a). Neuronal -synucleinopathy with severe movement disorder in mice expressing A53T human -synuclein. Neuron, 34, 521–33. Giasson, B. I., Ischiropoulos, H., Lee, V. M.-Y. & Trojanoswki, J. Q. (2002b). The relationship between oxidative/nitrative stress and pathological inclusions in Alzheimer’s and Parkinson’s diseases. Free Radic. Biol. Med., 32, 1264–75. Gibb, W. R. G. & Poewe, V. V. H. (1986). The centenary of Friederich H. Lewy. Neuropathol. Appl. Neurobiol., 12, 217–21. Gibb, W. R. G., Esiri, M. M. & Lees, A. J. (1987). Clinical and pathological features of diffuse Lewy body disease (Lewy body dementia). Brain, 110, 1131–53. Gilman, S., Low, P. A., Quinn, N. et al. (1999). Consensus statement on the diagnosis of multiple system atrophy. J. Neurol. Sci., 163, 94–8. Gomez-Tortosa, E., Ingraham, A. O., Irizarry, M. C. & Hyman, B. T. (1998). Dementia with Lewy bodies. J. Am. Geriatr. Soc., 46, 1449–58. Graham, D. G. (1978). Oxidative pathways for catecholamines in the genesis of neuromelanin and cytotoxic quinones. Mol. Pharmacol., 14, 633–43. Graham, D. G., Tiffany, S. M., Bell, W. R., Jr. & Gutknecht, W. F. (1978). Autoxidation versus covalent binding of quinones as the mechanism of toxicity of dopamine, 6-hydroxydopamine, and related compounds toward C1300 neuroblastoma cells in vitro. Mol. Pharmacol., 14, 644–53. Graham, J. G. & Oppenheimer, D. R. (1969). Orthostatic hypertension and nicotine insensitivity in a case of mutliple system atrophy. J. Neurol. Neurosurg. Psychiatry, 32, 28–34. Gwinn-Hardy, K., Mehta, N. D., Farrer, M. et al. (2000). Distinctive neuropathology revealed by alpha-synuclein antibodies in hereditary parkinsonism and dementia linked to chromosome 4p. Acta Neuropathol. (Berl.), 99, 663–72. Hansen, L. A., Salmon, D., Galasko, D. et al. (1990). The Lewy body variant of Alzheimer’s disease: a clinical and pathologic entity. Neurology, 40, 1–8. Hansen, L. A., Masliah, E., Galasko, D. & Terry, R. D. (1993). Plaqueonly Alzheimer disease is usually the Lewy body variant, and vice versa. J. Neuropathol. Exp. Neurol., 52, 648–54. Haroutunian, V., Serby, M., Purohit, D. P. et al. (2000). Contribution of Lewy body inclusions to dementia in patients with and without Alzheimer disease neuropathological conditions. Arch. Neurol., 57, 1145–50. Hashimoto, M., Hsu, L. J., Sisk, A. et al. (1998). Human recombinant NACP/alpha-synuclein is aggregated and fibrillated in vitro: relevance for Lewy body disease. Brain Res., 799, 301–6. Hashimoto, M., Hsu, L. J., Xia, Y. et al. (1999a). Oxidative stress induces amyloid-like aggregate formation of NACP/alphasynuclein in vitro. Neuroreport, 10, 717–21. Hashimoto, M., Takeda, A., Hsu, L. J., Takenouchi, T. & Masliah, E. (1999b). Role of cytochrome c as a stimulator of

alpha-synuclein aggregation in Lewy body disease. J. Biol. Chem., 274, 28849–52. Hashimoto, M., Rockenstein, E., Mante, M., Mallory, M. & Masliah, E. (2001). -Synuclein inhibits -synuclein aggregation: a possible role as an anti-parkinsonian factor. Neuron, 32, 213–23. Hastings, T. G., Lewis, D. A. & Zigmond, M. J. (1996). Role of oxidation in the neurotoxic effects of intrastriatal dopamine injections. Proc. Natl Acad. Sci. USA, 93, 1956–61. Hayashi, S., Wakabayashi, K., Ishikawa, A. et al. (2000a). An autopsy case of autosomal-recessive juvenile parkinsonism with a homozygous exon 4 deletion in the parkin gene. Mov Disord., 15, 884–8. Hayashi, S., Wakabayashi, K., Ishikawa, A. et al. (2000b). An autopsy case of autosomal-recessive juvenile parkinsonism with a homozygous exon 4 deletion in the parkin gene. Mov Disord., 15, 884–8. Hershko, A. & Ciechanover, A. (1998). The ubiquitin system. Annu. Rev. Biochem., 67, 425–79. Hietanen, M. & Teravainen, H. (1988). The effect of age of disease onset on neuropsychological performance in Parkinson’s disease. J. Neurol. Neurosurg. Psychiatry, 51, 244–9. Hill, W. D., Lee, V. M.-Y., Hurtig, H. I., Murray, J. M. & Trojanowski, J. Q. (1991). Epitopes located in spatially separate domains of each neurofilament subunit are present in Parkinson’s disease Lewy bodies. J. Comp. Neurol., 309, 150–60. Hubble, J. P., Cao, T., Hassanein, R. E., Neuberger, J. S. & Koller, W. C. (1993). Risk factors for Parkinson’s disease. Neurology, 43, 1693–7. Hughes, A. J., Daniel, S. E., Kilford, L. & Lees, A. J. (1982). Accuracy of clinical diagnosis of idiopathic Parkinson’s disease-a cliocopathological study of 100 cases. J. Neurol. Neurosurg. Psychiatry, 55, 181–4. Hughes, A. J., Daniel, S. E., Blankson, S. & Lee, A. J. (1993). A clinicopathological study of 100 cases of Parkinson’s disease. Arch. Neurol., 50, 140–8. Hughes, A. J., Daniel, S. E. & Lees, A. J. (2001). Improved accuracy of clinical diagnosis of Lewy body Parkinson’s disease. Neurology, 57, 1497–9. Hurtig, H. I., Trojanowski, J. Q., Galvin, J. et al. (2000). Alphasynuclein cortical Lewy bodies correlate with dementia in Parkinson’s disease. Neurology, 54, 1916–21. Irizarry, M. C., Kim, T. W., McNamara, M. et al. (1996). Characterization of the precursor protein of the non-A beta component of senile plaques (NACP) in the human central nervous system. J. Neuropathol. Exp. Neurol., 55, 889–95. Irizarry, M. C., Growdon, W., Gomez-Isla, T. et al. (1998). Nigral and cortical Lewy bodies and dystrophic nigral neurites in Parkinson’s disease and cortical Lewy body disease contain alpha-synuclein immunoreactivity. J. Neuropathol. Exp. Neurol., 57, 334–7. Iwai, A., Masliah, E., Yoshimoto, M. et al. (1995a). The precursor protein of non-A beta component of Alzheimer’s disease amyloid is a presynaptic protein of the central nervous system. Neuron, 14, 467–75.

371

372


Iwai, A., Yoshimoto, M., Masliah, E. & Saitoh, T. (1995b). NonA component of Alzheimer’s disease amyloid (NAC) is amyloidogenic. Biochemistry, 34, 10139–45. Iwatsubo, T., Yamaguchi, H., Fujimuro, M. et al. (1996). Purification and characterization of Lewy bodies from the brains of patients with diffuse Lewy body disease. Am. J. Pathol., 148, 1517–29. Jakes, R., Spillantini, M. G. & Goedert, M. (1994). Identification of two distinct synucleins from human brain. FEBS Lett., 345, 27–32. Jensen, P. H., Nielsen, M. S., Jakes, R., Dotti, C. G. & Goedert, M. (1998). Binding of alpha-synuclein to brain vesicles is abolished by familial Parkinson’s disease mutation. J. Biol. Chem., 273, 26292–4. Ji, H., Liu, Y. E., Jia, T. et al. (1997). Identification of a breast cancer-specific gene, BCSG1, by direct differential cDNA sequencing. Cancer Res., 57, 759–64. Jia, T., Liu, Y. E., Liu, J. & Shi, Y. E. (1999). Stimulation of breast cancer invasion and metastasis by synuclein gamma. Cancer Res., 59, 742–7. Kahle, P. J., Neumann, M., Ozmen, L. et al. (2000). Subcellular localization of wild-type and Parkinson’s disease-associated mutant alpha-synuclein in human and transgenic mouse brain. J. Neurosci., 20, 6365–73. Kahle, P. J., Neumann, M., Ozmen, L. et al. (2001). Selective insolubility of -synuclein in human Lewy body diseases is recapitalated in a transgenic mouse model. Am. J. Pathol., 159, 2215–25. Kato, S. & Nakamura, H. (1990). Cytoplasmic argyrophilic inclusions in neurons of pontine nuclei in patients with olivopontocerebellar atrophy: immunohistochemical and ultrastructural studies. Acta Neuropathol. (Berl.), 79, 584–94. Kato, S., Nakamura, H., Hirano, A., Ito, H., Llena, J. F. & Yen, S. H. (1991). Argyrophilic ubiquitinated cytoplasmic inclusions of Leu-7-positive glial cells in olivopontocerebellar atrophy (multiple system atrophy). Acta Neuropathol. (Berl.), 82, 488–93. Katzman, R., Galasko, D., Saitoh, T., Thal, L. J. & Hansen, L. (1995). Genetic evidence that the Lewy body variant is indeed a phenotypic variant of Alzheimer’s disease. Brain Cogn., 28, 259–65. Kaufmann, H., Hague, K. & Perl, D. (2001). Accumulation of alpha-synuclein in autonomic nerves in pure autonomic failure. Neurology, 56, 980–1. Kitada, T., Asakawa, S., Hattori, N. et al. (1998). Mutations in the parkin gene cause autosomal recessive juvenile parkinsonism. Nature, 392, 605–8. Klein, C., Pramstaller, P. P., Kis, B. et al. (2000). Parkin deletions in a family with adult-onset, tremor-dominant parkinsonism: expanding the phenotype. Ann. Neurol., 48, 65–71. Koller, W., Vetere-Overfield, B., Gray, C. et al. (1990). Environmental risk factors in Parkinson’s disease. Neurology, 40, 1218–21. Kosaka, K. (1990). Diffuse Lewy body disease in Japan. J. Neurol., 237, 197–204. (1993). Dementia and neuropathology in Lewy body disease. Adv. Neurol., 60, 456–63. Kosaka, K., Yoshimura, M., Ikeda, K. & Budka, H. (1984). Diffuse type of Lewy body disease: progressive dementia with abundant cortical Lewy bodies and senile changes of varying degree-A new disease? Clin. Neuropathol., 3, 185–92.

Kosaka, K., Iseki, E., Odawara, T. & Yamamoto, T. (1996). Cerebral type of Lewy body disease. Neuropathology, 16, 32–5. Kosel, S., Lucking, C. B., Egensperger, R., Mehraein, P. & Graeber, M. B. (1996). Mitochondrial NADH dehydrogenase and CYP2D6 genotypes in Lewy-body parkinsonism. J. Neurosci. Res., 44, 174–83. Kruger, R., Kuhn, W., Muller, T. et al. (1998). Ala30Pro mutation in the gene encoding alpha-synuclein in Parkinson’s disease. Nat. Genet., 18, 106–8. Kruger, R., Vieira-Saecker, A. M., Kuhn, W. et al. (1999). Increased susceptibility to sporadic Parkinson’s disease by a certain combined alpha-synuclein/apolipoprotein E genotype. Ann. Neurol., 45, 611–17. Lang, A. E. & Lozano, A. M. (1998). Parkinson’s disease. N. Engl. J. Med. 339, 1130–43. Langston, J. W., Ballard, P., Tetrud, J. W. & Irwin, I. (1983). Chronic Parkinsonism in humans due to a product of meperidineanalog synthesis. Science, 219, 979–980. Langston, J. W., Forno, L. S., Tetrud, J., Reeves, A. G., Kaplan, J. A. & Karluk, D. (1999). Evidence of active nerve cell degeneration in the substantia nigra of humans years after 1-methyl-4phenyl-1,2,3,6-tetrahydropyridine exposure. Ann. Neurol., 46, 598–605. Lantos, P. L. (1997). Multiple system atrophy. Brain Pathol., 7, 1293–7. Lavedan, C., Buchholtz, S., Nussbaum, R. L., Albin, R. L. & Polymeropoulos, M. H. (2002). A mutation in the human neurofilament M gene in Parkinson’s disease that suggests a role for the cytoskeleton in neuronal degeneration. Neurosci. Lett., 322, 57–61. Lennox, G., Lowe, J., Landon, M., Byrne, E. J., Mayer, R. J. & GodwinAusten, R. B. (1989). Diffuse Lewy body disease: correlative neuropathology using anti-ubiquitin immunocytochemistry. J. Neurol. Neurosurg. Psychiatry, 52, 1236–47. Leroy, E., Boyer, R., Auburger, G. et al. (1998). The ubiquitin pathway in Parkinson’s disease. Nature, 395, 451–2. Lippa, C. F., Fujiwara, H., Mann, D. M. et al. (1998). Lewy bodies contain altered alpha-synuclein in brains of many familial Alzheimer’s disease patients with mutations in presenilin and amyloid precursor protein genes. Am. J. Pathol., 153, 1365–70. Lippa, C. F., Schmidt, M. L., Lee, V. M.-Y. & Trojanowski, J. Q. (1999). Antibodies to alpha-synuclein detect Lewy bodies in many Down’s syndrome brains with Alzheimer’s disease. Ann. Neurol., 45, 353–7. Louis, E. D., Klatka, L. A., Liu, Y. & Fahn, S. (1997). Comparison of extrapyramidal features in 31 pathologically confirmed cases of diffuse Lewy body disease and 34 pathologically confirmed cases of Parkinson’s disease. Neurology, 48, 376–80. Lucking, C. B., Durr, A., Bonifati, V. et al. (2000). Association between early-onset Parkinson’s disease and mutations in the parkin gene. French Parkinson’s Disease Genetics Study Group. N. Engl. J. Med., 342, 1560–7. McGeer, P. L., Itagaki, S., Akiyama, H. & McGeer, E. G. (1989). Rate of cell death in parkinsonsim indicates active neuropatholoigcal process. Ann. Neurol., 24, 574–6.


McKeith, I. G., Galasko, D., Kosaka, K. et al. (1996). Consensus guidelines for the clinical and pathologic diagnosis of dementia with Lewy bodies (DLB): report of the consortium on DLB international workshop. Neurology, 47, 1113–24. Markopoulou, K., Wszolek, Z. K., Pfeiffer, R. F. & Chase, B. A. (1999). Reduced expression of the G209A alpha-synuclein allele in familial Parkinsonism. Ann. Neurol., 46, 374–81. Maroteaux, L. & Scheller, R. H. (1991). The rat brain synucleins; family of proteins transiently associated with neuronal membrane. Brain Res. Mol. Brain Res., 11, 335–43. Maroteaux, L., Campanelli, J. T. & Scheller, R. H. (1988). Synuclein: a neuron-specific protein localized to the nucleus and presynaptic nerve terminal. J. Neurosci., 8, 2804–15. Marttila, R. J. & Rinne, U. K. (1976). Dementia in Parkinson disease. Acta Neurol. Scand., 54, 431–41. Masliah, E., Iwai, A., Mallory, M., Ueda, K. & Saitoh, T. (1996). Altered presynaptic protein NACP is associated with plaque formation and neurodegeneration in Alzheimer’s disease. Am. J. Pathol., 148, 201–10. Masliah, E., Rockenstein, E., Veinbergs, I. et al. (2000). Dopaminergic loss and inclusion body formation in alpha-synuclein mice: implications for neurodegenerative disorders. Science, 287, 1265–9. Masliah, E., Rockenstein, E., Veinbergs, I. et al. (2001). -Amyloid peptides enhance alpha synuclein accumulation and neuronal deficits in a transgenic mouse model linking Alzheimer’s disease and Parkinson’s disease. Proc. Natl. Acad. Sci. USA, 98, 12245–50. Matsuo, A., Akiguchi, I., Lee, G. C., McGeer, E. G., McGeer, P. L. & Kimura, J. (1998). Myelin degeneration in multiple system atrophy detected by unique antibodies. Am. J. Pathol., 153, 735–44. Mattila, P. M., Röyttä, M., Torikka, H., Dickson, D. W. & Rinne, J. O. (1998). Cortical Lewy bodies and Alzheimer-type changes in patients with Parkinson’s disease. Acta Neuropathol. (Berl.), 95, 576–82. Mayeau, R., Stern, Y., Cote, L. & Williams, J. B. (1984). Altered serotonin metabolism in depressed patients with Parkinson’s disease. Neurology, 34, 642–6. Mayeau, R., Stern, Y., Rosenstein, R. et al. (1988). An estimate of the prevalence of dementia of idiopathic Parkison’s disease. Arch. Neurol., 45, 260–2. Mayeau, R., Chen, J., Mirabello, E. et al. (1990). An estimate of the incidence of dementia in idiopathic Parkinson’s disease. Neurology, 40, 1513–17. Mega, M. S., Masterman, D. L., Benson, D. F. et al. (1996). Dementia with Lewy bodies: reliability and validity of clinical and pathologic criteria. Neurology, 47, 1403–9. Molsa, P. K., Marttila, R. J. & Rinne, U. K. (1984). Extrapyramidal signs in Alzheimer’s disease. Neurology, 34, 1114–6. Mori, H., Kondo, T., Yokochi, M. et al. (1998). Pathologic and biochemical studies of juvenile parkinsonism linked to chromosome 6q. Neurology, 51, 890–2. Murphy, D. D., Rueter, S. M., Trojanowski, J. Q. & Lee, V. M.-Y. (2000). Synucleins are developmentally expressed, and alphasynuclein regulates the size of the presynaptic vesicular pool in primary hippocampal neurons. J. Neurosci., 20, 3214–20.

Nakajo, S., Omata, K., Aiuchi, T. et al. (1990). Purification and characteization of a novel brain-specific 14 kDa protein. J. Neurochem., 55, 2031–8. Nakajo, S., Shioda, S., Nakai, Y. & Nakaya, K. (1994). Localization of phosphoneuroprotein 14 (PNP 14) and its mRNA expression in rat brain determined by immunocytochemistry and in situ hybridization. Brain Res. Mol. Brain Res., 27, 81–6. Nakazato, Y., Yamazaki, H., Hirato, J., Ishida, Y. & Yamaguchi, H. (1990). Oligodendroglial microtubular tangles in olivopontocerebellar atrophy. J. Neuropathol. Exp. Neurol. 49, 521– 30. Narhi, L., Wood, S. J., Steavenson, S. et al. (1999). Both familial Parkinson’s disease mutations accelerate alpha-synuclein aggregation. J. Biol. Chem., 274, 9843–6. Newell, K. L., Boyer, P., Gomez-Tortosa, E. et al. (1999). Alphasynuclein immunoreactivity is present in axonal swellings in neuroaxonal dystrophy and acute traumatic brain injury. J. Neuropathol. Exp. Neurol., 58, 1263–8. Obeso, J. A., Rodriguez-Oroz, M. C., Rodriguez, M. et al. (2000). Pathophysiology of the basal ganglia in Parkinson’s disease. Trends Neurosci., 23, S8–19. Okazaki, H., Lipkin, L. E. & Aronson, S. M. (1961). Diffuse intracytoplasmic inclusions (Lewy type) assocaited with prgressive dementia and quadraparesis. J. Neuropathol. Exp. Neurol., 20, 237–44. Ostrerova-Golts, N., Petrucelli, L., Hardy, J., Lee, J. M., Farrer, M. & Wolozin, B. (2000). The A53T alpha-synuclein mutation increases iron-dependent aggregation and toxicity. J. Neurosci., 20, 6048–54. Pakkenberg, B., Moller, A., Gundersen, H. J., Mouritzen, D. A. & Pakkenberg, H. (1991). The absolute number of nerve cells in substantia nigra in normal subjects and in patients with Parkinson’s disease estimated with an unbiased stereological method. J. Neurol. Neurosurg. Psychiatry, 54, 30–3. Papadimitriou, A., Veletza, V., Hadjigeorgiou, G. M., Patrikiou, A., Hirano, M. & Anastasopoulos, I. (1999). Mutated alphasynuclein gene in two Greek kindreds with familial PD: incomplete penetrance? Neurology, 52, 651–4. Papp, M. I. & Lantos, P. L. (1992). Accumulation of tubular structures in oligodendroglial and neuronal cells as the basic alteration in multiple system atrophy. J. Neurol. Sci., 107, 172–82. Papp, M. I., Kahn, J. E. & Lantos, P. L. (1989). Glial cytoplasmic inclusions in the CNS of patients with multiple system atrophy (striatonigral degeneration, olivopontocerebellar atrophy and Shy–Drager syndrome). J. Neurol. Sci., 94, 79–100. Paxinou, E., Chen, Q., Weisse, M. et al. (2001). Induction of alpha-synuclein aggregation by intracellular nitrative insult. J. Neurosci., 21, 8053–61. Perrin, R. J., Woods, W. S., Clayton, D. F. & George, J. M. (2000). Interaction of human alpha-synuclein and Parkinson’s disease variants with phospholipids J. Biol. Chem., 275, 34393–8. Perry, R. H., Irving, D., Blessed, G., Fairbairn, A. & Perry, E. K. (1990). Senile dementia of Lewy body type. A clinically and neuropathologically distinct form of Lewy body dementia in the elderly. J., Neurol., Sci., 95, 119–39.

373

374


Polymeropoulos, M. H., Lavedan, C., Leroy, E. et al. (1997). Mutation in the alpha-synuclein gene identified in families with Parkinson’s disease. Science, 276, 2045–7. Przedborski, S. & Jackson-Lewis, V. (1998). Mechanisms of MPTP toxicity. Mov. Disord., 13 Suppl 1, 35–8. Rajput, A. H. & Rozdilsky, B. (1975). Parkinsonism and dementia: effects of L-dopa. Lancet, 1, 1084. Rajput, A. H., Offord, K. P., Beard, C. M. & Kurland, L. T. (1984). Epidemiology of parkinsonism: incidence, classification, and mortality. Ann. Neurol., 16, 278–82. Rajput, A. H., Uitti, R. J., Stern, W. & Laverty, W. (1986). Early onset Parkinson’s disease in Saskatchewan – environmental considerations for etiology. Can. J. Neurol. Sci., 13, 312–16. Rajput, A. H., Offord, K. P., Beard, C. M. & Kurland, L. T. (1987a). A case control study of smoking habits, dementia, and other illness in idiopathic Parkinson’s disease. Neurology, 37, 226–32. Rajput, A. H., Uitti, R. J., Stern, W. et al. (1987b). Geography, drinking water chemistry, pesticides and herbicides and the etiology of Parkinson’s disease. Can. J. Neurol. Sci., 14, 414–18. Rezaie, P., Cairns, N. J., Chadwick, A. & Lantos, P. L. (1996). Lewy bodies are located preferentially in limbic areas in diffuse Lewy body disease. Neurosci. Lett., 212, 111–14. Richter-Landsberg, C., Gorath, M., Trojanowski, J. Q. & Lee, V. M.-Y. (2000). Alpha-synuclein is developmentally expressed in cultured rat brain oligodendrocytes. J. Neurosci. Res., 62, 9–14. Riggs, J. E. (1993). A controlled study of dementia in Parkinson’s disease. J. Neurol. Neurosurg. Psychiatry, 45, 969–74. Sabbagh, N., Brice, A., Marez, D. et al. (1999). CYP2D6 polymorphism and Parkinson’s disease susceptibility. Mov. Disord., 14, 230–236. Saito, Y., Kawai, M., Inoue, K. et al. (2000). Widespread expression of alpha-synuclein and tau immunoreactivity in Hallervorden– Spatz syndrome with protracted clinical course. J. Neurol. Sci., 177, 48–59. Samuel, W., Galasko, D., Masliah, E. & Hansen, L. A. (1996). Neocortical lewy body counts correlate with dementia in the Lewy body variant of Alzheimer’s disease. J. Neuropathol. Exp. Neurol., 55, 44–52. Schenk, D., Barbour, R., Dunn, W. et al. (1999). Immunization with amyloid-beta attenuates Alzheimer-disease-like pathology in the PDAPP mouse. Nature, 400, 173–7. Schmidt, M. L., Murray, J., Lee, V. M.-Y., Hill, W. D., Wertkin, A. & Trojanowski, J. Q. (1991). Epitope map of neurofilament protein domains in cortical and peripheral nervous system Lewy bodies. Am. J. Pathol., 139, 53–65. Seidler, A., Hellenbrand, W., Robra, B. P. et al. (1996). Possible environmental, occupational, and other etiologic factors for Parkinson’s disease: a case-control study in Germany. Neurology, 46, 1275–84. Semchuk, K. M., Love, E. J. & Lee, R. G. (1991). Parkinson’s disease and exposure to rural environmental factors: a population based case-control study. Can. J. Neurol. Sci., 18, 279–86. (1992). Parkinson’s disease and exposure to agricultural work and pesticide chemicals. Neurology, 42, 1328–35.

(1993). Parkinson’s disease: a test of the multifactorial etiologic hypothesis. Neurology, 43, 1173–80. Serpell, L. C., Berriman, J., Jakes, R., Goedert, M. & Crowther, R. A. (2000). Fiber diffraction of synthetic alpha-synuclein filaments shows amyloid- like cross-beta conformation. Proc. Natl. Acad. Sci. USA, 97, 4897–902. Shevell, M. (1996). Naming of syndromes. Lancet, 348, 1662. Shibayama-Imazu, T., Okahashi, I., Omata, K. et al. (1993). Cell and tissue distribution and developmental change of neuron specific 14 kDa protein (phosphoneuroprotein 14). Brain Res., 622, 17–25. Shimura, H., Hattori, N., Kubo, S. et al. (2000). Familial Parkinson disease gene product, parkin, is a ubiquitin-protein ligase. Nat. Genet., 25, 302–305. Shimura, H., Schlossmacher, M. G., Hattori, N. et al. (2001). Ubiquitination of a new form of alpha-synuclein by parkin from human brain: implications for Parkinson’s disease. Science, 293, 263–9. Shtilerman, M. D., Ding, T. T. & Lansbury, P. T., Jr. (2002). Molecular crowding accelerates fibrillization of alpha-synuclein: could an increase in the cytoplasmic protein concentration induce Parkinson’s disease? Biochemistry, 41, 3855–60. Simuni, T. & Hurtig, H. I. (2000). Parkinson’s disease: the clinical picture. In Neurodegenerative Dementias, pp. 193–203. ed. C. M. Clark and J. Q.Trojanoswki. New York, NY: McGraw-Hill. Smythies, J. & Galzigna, L. (1998). The oxidative metabolism of catecholamines in the brain: a review. Biochim. Biophys. Acta, 1380, 159–62. Snider, S. R., Fahn, S., Isgreen, W. P. & Cote, L. J. (1976). Primary sensory symptoms in parkinsonism. Neurology, 26, 423–9. Souza, J. M., Giasson, B. I., Chen, Q., Lee, V. M.-Y. & Ischiropoulos, H. (2000a). Dityrosine cross-linking promotes formation of stable alpha-synuclein polymers. Implication of nitrative and oxidative stress in the pathogenesis of neurodegenerative synucleinopathies. J. Biol. Chem., 275, 18344–49. Souza, J. M., Giasson, B. I., Lee, V. M.-Y. & Ischiropoulos, H. (2000b). Chaperone-like activity of synucleins. FEBS Lett., 474, 116–19. Spillantini, M. G., Schmidt, M. L., Lee, V. M.-Y., Trojanowski, J. Q., Jakes, R. & Goedert, M. (1997). Alpha-synuclein in Lewy bodies. Nature, 388, 839–40. Spillantini, M. G., Crowther, R. A., Jakes, R., Cairns, N. J., Lantos, P.L. & Goedert, M. (1998a). Filamentous alpha-synuclein inclusions link multiple system atrophy with Parkinson’s disease and dementia with Lewy bodies. Neurosci. Lett., 251, 205–8. Spillantini, M. G., Crowther, R. A., Jakes, R., Hasegawa, M. & Goedert, M. (1998b). Alpha-synuclein in filamentous inclusions of Lewy bodies from Parkinson’s disease and dementia with lewy bodies. Proc. Natl. Acad. Sci. USA, 95, 6469–73. Spira, P. J., Sharpe, D. M., Halliday, G., Cavanagh, J. & Nicholson, G. A. (2001). Clinical and pathological features of a Parkinsonian syndrome in a family with an Ala53Thr alpha-synuclein mutation. Ann. Neurol., 49, 313–19. Surguchov, A., Surgucheva, I., Solessio, E. & Baehr, W. (1999). Synoretin – a new protein belonging to the synuclein family. Mol. Cell Neurosci., 13, 95–103.


Takahashi, H., Ohama, E., Suzuki, S. et al. (1994). Familial juvenile parkinsonism: clinical and pathologic study in a family. Neurology, 44, 437–41. Takeda, A., Hashimoto, M., Mallory, M. et al. (1998a). Abnormal distribution of the non-Abeta component of Alzheimer’s disease amyloid precursor/alpha-synuclein in Lewy body disease as revealed by proteinase K and formic acid pretreatment. Lab. Invest., 78, 1169–77. Takeda, A., Mallory, M., Sundsmo, M., Honer, W., Hansen, L. & Masliah, E. (1998b). Abnormal accumulation of NACP/alphasynuclein in neurodegenerative disorders. Am. J. Pathol., 152, 367–72. Tanner, C. M., Chen, B., Wang, W. Z. et al. (1987). Environmental factors in the etiology of Parkinson’s disease. Can. J. Neurol. Sci., 14, 419–23. Tanner, C. M., Chen, B., Wang, W. et al. (1989). Environmental factors and Parkinson’s disease: a case-control study in China. Neurology, 39, 660–4. Taylor, T. D., Litt, M., Kramer, P. et al. (1996). Homozygosity mapping of Hallervorden–Spatz syndrome to chromosome 20p12.3–p13. Nat. Genet., 14, 479–81. Tu, P. H., Galvin, J. E., Baba, M. et al. (1998). Glial cytoplasmic inclusions in white matter oligodendrocytes of multiple system atrophy brains contain insoluble alpha-synuclein. Ann. Neurol., 44, 415–22. Ueda, K., Fukushima, H., Masliah, E. et al. (1993). Molecular cloning of cDNA encoding an unrecognized component of amyloid in Alzheimer disease. Proc. Natl Acad. Sci. USA, 90, 11282–6. Uversky, V. N., Li, J. & Fink, A. L. (2001a). Metal-triggered structural transformations, aggregation, and fibrillation of human alpha-synuclein. J. Biol. Chem., 276, 44284–96. Uversky, V. N., Li, J. & Fink, A. L. (2001b). Pesticides directly accelerate the rate of alpha-synuclein fibril formation: a possible factor in Parkinson’s disease. FEBS Lett., 500, 105–8. Uversky, V. N., Li, J., Souillac, P. et al. (2002a). Biophysical properties of the synucleins and their propensites to fibrillate. J. Biol. Chem., 277, 11970–8. Uversky, V. N., Yamin, G., Souillac, P. O., Goers, J., Glaser, C. B. & Fink, A. L. (2002b). Methionine oxidation inhibits fibrillation of human -synuclein in vitro. FEBS Lett., 517, 239–44. Valente, E. M., Bentivoglio, A. R., Dixon, P. H. et al. (2001). Localization of a novel locus for autosomal recessive early-onset parkinsonism, PARK6, on human chromosome 1p35–p36. Am. J. Hum. Genet., 68, 895–900. van der Putten, H., Wiederhold, K. H., Probst, A. et al. (2000). Neuropathology in mice expressing human alpha-synuclein. J. Neurosci., 20, 6021–9. van Duijn, C. M., Dekker, M. C., Bonifati, V. et al. (2001). Park7, a novel locus for autosomal recessive early-onset parkinsonism, on chromosome 1p36. Am. J. Hum. Genet., 69, 629–34. Volles, M. J., Lee, S.-J., Rochet, J.-C. et al. (2001). Vesicle permeabilization by protofibrillar -synuclein: implications for the pathogenesis and treatment of Parkinson’s disease. Biochemistry, 40, 7812–19.

Wakabayashi, K., Hayashi, S., Kakita, A. et al. (1998a). Accumulation of alpha-synuclein/NACP is a cytopathological feature common to Lewy body disease and multiple system atrophy. Acta Neuropathol. (Berl.), 96, 445–52. Wakabayashi, K., Yoshimoto, M., Tsuji, S. & Takahashi, H. (1998b). Alpha-synuclein immunoreactivity in glial cytoplasmic inclusions in multiple system atrophy. Neurosci. Lett., 249, 180–2. Wakabayashi, K., Yoshimoto, M., Fukushima, T. et al. (1999). Widespread occurrence of alpha-synuclein/NACP-immunoreactive neuronal inclusions in juvenile and adult-onset Hallervorden– Spatz disease with Lewy bodies. Neuropathol. Appl. Neurobiol., 25, 363–8. Weinreb, P. H., Zhen, W., Poon, A. W., Conway, K. A. & Lansbury, P. T. (1996). NACP, a protein implicated in Alzheimer’s disease and learning, is natively unfolded. Biochemistry, 35, 13709–15. Wenning, G. K., Ben Shlomo, Y., Magalhaes, M., Daniel, S. E. & Quinn, N. P. (1994). Clinical features and natural history of multiple system atrophy. An analysis of 100 cases. Brain, 117 (4), 835–45. Wenning, G. K., Tison, F., Ben Shlomo, Y., Daniel, S. E. & Quinn, N. P. (1997). Multiple system atrophy: a review of 203 pathologically proven cases. Mov. Disord., 12, 133–47. Whitehouse, P. J., Hedreen, J. C., White, C. L. & Price, D. L. (1983). Basal forebrain neurons in the dementia of Parkinson’s disease. Ann. Neurol., 13, 243–8. Withers, G. S., George, J. M., Banker, G. A. & Clayton, D. F. (1997). Delayed localization of synelfin (synuclein, NACP) to presynaptic terminals in cultured rat hippocampal neurons. Brain Res. Dev. Brain Res., 99, 87–94. Wood, S. J., Wypych, J., Steavenson, S., Louis, J.-C., Citron, M. & Biere, A. L. (1999). -Synuclein fibrillogenesis is nucleationdependent. J. Biol. Chem. 274, 19509–12. Yamamoto, A., Lucas, J. J. & Hen, R. (2000). Reversal of neuropathology and motor dysfunction in a conditional model of Huntington’s disease. Cell, 101, 57–66. Yamazaki, M., Arai, Y., Baba, M. et al. (2000). Alpha-synuclein inclusions in amygdala in the brains of patients with the parkinsonism–dementia complex of Guam. J. Neuropathol. Exp. Neurol., 59, 585–91. Zabar, Y. & Kawas, C. H. (2000). Epidemiology and clinical genetics of Alzheimer’s disease. In Neurodegenerative Dementias, pp. 79–94. ed. C. M.Clark and J. Q.Trojanoswki. New York, NY: McGraw-Hill. Zhang, Y., Gao, J., Chung, K. K., Huang, H., Dawson, V. L. & Dawson, T. M. (2000). Parkin functions as an E2-dependent ubiquitin protein ligase and promotes the degradation of the synaptic vesicle-associated protein, CDCrel-1. Proc. Natl Acad. Sci. USA, 97, 13354–9. Zhou, B., Westaway, S. K., Levinson, B., Johnson, M. A., Gitschier, J. & Hayflick, S. J. (2001). A novel pantotheante kinase gene (PANK2) is defective in Hallervorden-Spatz syndrome. Nat. Genet., 28, 345–9.

375

In memory of Dr Edward Peirson Richardson, Jr

16 Huntington’s disease Jean Paul G. Vonsattel and Maxim Lianski New York Brain Bank, The Taub Institute, Columbia University, New York, USA

Huntington’s disease (HD) is an autosomal dominant illness, usually with mid-life onset, of psychiatric, cognitive and motor symptoms. Death occurs 12–15 years from the time of symptomatic onset (Folstein, 1989; Harper et al., 1991). The HD mutation consists of an unstable expansion of CAG (trinucleotide) repeats within the coding region of the gene ‘IT15’ (for ‘Interesting transcript’ referred to as HD-IT15 CAG repeats). This gene, on chromosome 4 (4p63), encodes the 350 kDa protein huntingtin of unknown function (Group et al., 1993). An expanded polyglutamine residue (polyQ) distinguishes the mutated huntingtin (with about 37 to 250 polyQ) from the wild type (with 8 to about 34–36 polyQ) (Nance et al., 1999; Rubinsztein et al., 1996). The disease occurs when the critical threshold of about 37 polyQ is exceeded. This phenomenon is observed in a group of inherited, neurodegenerative diseases caused by polyQ extension, which is increasingly referred to as polyglutaminopathies (Paulson, 1999). The pathogenicity of the mutant huntingtin is unknown. The mutant huntingtin retains some functions of the wild type huntingtin since individuals homozygous for the HD gene produce only mutant huntingtin and are clinically indistinguishable from heterozygous patients (Wexler et al., 1987; Myers et al., 1989). The mutant huntingtin is present in all organs, yet the brunt of the changes of HD identified so far occurs in the brain (Sathasivam et al., 1999). The degeneration involves initially the striatum (neuronal loss, gliosis) and cortex, and eventually may appear throughout the brain as a constellation of the toxic effect of the mutation and the ensuing secondary changes. Features that HD shares with other polyglutaminopathies are sparse, ubiquitinated, neuronal, nuclear inclusions, and dystrophic neurites; and neuronal loss in regions more or less distinctive for each disease of

this group (DiFiglia et al., 1997). Among the theories for the selective cellular damage in HD, the most compelling involve impaired energy metabolism and excitotoxicity, and relative, selective endotoxicity perhaps resulting from misfolding of mutant huntingtin. The availability of a mouse model of HD allows defining the pathological thresholds of polyQ expansion and the susceptibility of subclasses of neurons to degenerate.(Hemachandra et al., 1998; Mangiarini et al., 1996). The transgenic mice, the evaluations of the aggregation and interaction of mutant huntingtin with known and novel proteins provide opportunities for identifying factors responsible of the neuronal loss, which is crucial for the development of therapy.

Clinical and genetic features of Huntington’s disease The clinical, diagnostic criteria include: (i) a family history of chorea, (ii) gradual motor disability with chorea or rigidity of no other cause, and (iii) psychiatric disturbance with slowly developing dementia of no other cause. The disease is universal. The estimated occurrence of HD patients in North America and Europe is 5–10/100 000 (Conneally, 1984). It is highest in populations of western European origin, and lowest in African or Asian populations. New mutations rarely occur (Myers et al., 1993). Most persons with the HD mutation develop and function normally into early adulthood. Involuntary movements may begin any time after infancy. Chorea is common in adults while rigidity prevails in patients with juvenileonset. Deficits in attention and memory are often present


376


Table 16.1. CAG repeats of IT15 gene on chromosome 4 (4p63) with or without Huntington’s disease Individuals

(CAG)n

General population (both alleles) Critical threshold Possible HD carriers; some HD patients

6–34

The majority of HD patients HD patients with juvenile onset (6% of all HD patients)

40–55 (*)

35–37 35– 39 (*)

70 or more (*)

Comment

Diagnostically uncertain range. Occurrence in apparently normal elderly individuals(8) Fig. 16.1 Expansion exceeding 100 is rare. Usually paternal transmission.

HD: Huntington disease. (*) Usually heterozygotes, one allele affected.

at the time of onset of motor dysfunction, and gradually worsen. Hayden (1981), Folstein (1990) and Harper (1996) provide comprehensive reviews of dementia in HD. The mean age of patients at onset of movement disorders is 40 years. In 9% of the patients symptoms are present before the age of 20. Twenty-five percent of subjects remain asymptomatic until age 50 or later (Myers et al., 1985). Young et al. (1986) reported that 94% of the patients were adult-onset, and 6% were juvenile-onset in the Venezuelan cohort. In 90% of the patients with age of onset under 10 years, the gene is paternally transmitted (Harper, 1996). The HD-IT15 CAG segment of the huntingtin gene is inherited in a mendelian fashion, is polymorphic, unsta-

ble, and undergoes changes during meiosis, including increases or decreases of 1–5 CAG repeat units (Duyao et al., 1993; Hersch et al., 1994; Telenius et al., 1995). Larger increases may occur, particularly in paternal transmissions. Table 16.1 summarizes the HD-IT15 CAG length repeats or polyQ observed in the general and in the HD populations. Individuals with 34 or fewer polyQ on the longest allele will not develop HD, those with 35 to 39 may or may not develop HD, and those with 40 or more will develop the disease. Thus, the critical length is 35–37 polyQ (pathological threshold) beyond which the disease is likely to occur if the carrier lives long enough. Indeed, the repeats in HD nonagenarians are usually in the lower range (38–40), thus carriers with repeats in this range may develop HD only beyond a certain age (Karluk et al., 1999). Most patients with adult onset of symptoms have expansions ranging from 40– 55 polyQ (Fig. 16.1) (Albin & Tagle, 1995; Duyao et al., 1993; Gómez-Tortosa et al., 1998; Kremer et al., 1994; MacMillan et al., 1993). Expansions of 70 or more repeats usually occur in patients with juvenile onset of symptoms. Rarely, patients have expansions exceeding 100 repeats. The number of repeats correlates inversely with age of onset or with age at death. More than 99% of patients with the clinical and pathological hallmarks of HD have an expanded huntingtin allele (Hersch et al., 1994; Kremer et al., 1994; Persichetti et al., 1994). Rare individuals with the phenotype of HD have been reported to have HD-IT15 CAG repeat lengths in the normal range (Group et al., 1993; Margolis et al., 2001; Persichetti et al., 1994; Rosenblatt et al., 1998). Furthermore, apparently normal elderly individuals were found to have 36–39 repeats (Rubinsztein et al., 1996). Nevertheless, evaluation of the HD-IT15 CAG repeats is a powerful diagnostic test available to clinicians (Kremer et al., 1994). Such a test, however, should be offered only if a set

Fig. 16.1. Frequency in percent of HD-IT15 CAG repeats units (polyQ) in heterozygous patients. (a), normal allele; (b), abnormal allele.

377

378

J. P. G. Vonsattel and M. Lianski

of recommended provisions be followed (IHA and WFN, 1994).

Organization of the basal ganglia system Nomenclature The basal ganglia consist of the corpus striatum and the amygdaloid nucleus (Carpenter & Sutin, 1983). Because of their connections, the subthalamic nucleus and substantia nigra are often included among the basal ganglia. The corpus striatum consists of the neostriatum (caudate nucleus and putamen) and paleostriatum (globus pallidus). The globus pallidus (GP) or paleostriatum is divided into external (GPe) and internal (GPi) segments. The neostriatum is commonly referred to as the striatum. The substantia nigra (SN) has two main zones: the pars reticulata (SNr), and the pars compacta (SNc) the gradual pigmentation of which becomes visible on gross examination at puberty.

Pathways The striatum collects inputs from the entire neocor¨ tex (Kunzle, 1975, 1977; Yeterian & Van Hoesen, 1978). It processes the signals and then sends them through other parts of the basal ganglia to areas of frontal cortex that have been implicated in motor planning and execution (Graybiel et al., 1994). Albin et al. (1989) proposed a model of the functional anatomy of disorders of the basal ganglia. The model is a practical conception of basal ganglia pathophysiology with emphasis on chorea, parkinsonism, and hemiballism. According to this model, the basal ganglia concerned with motor functions have two compartments, one for input, and one for output. The input compartment consists of the caudate nucleus (CN) and putamen, which receive inputs from the cerebral cortex, intralaminar thalamic nuclei (centromedian– parafascicular nuclear complex), and the SNc. The output compartment includes the subthalamic nucleus, SNr and GPi. The target nuclei of the output compartment are in the thalamus, which has an excitatory action upon the cortex. Two major pathways (a direct and an indirect) integrate the input compartment with the output compartment. The direct (monosynaptic) striatal pathway projects to the GPi. The indirect pathway passes first to the GPe, subthalamic nucleus and SNr, and then to the GPi, which sends projections to the thalamus. These two efferent systems of the

striatum have apparently opposing effects upon the output nuclei and thalamic target nuclei (Alexander & Crutcher, 1990). The disruption of these striatal efferent pathways leads to the development of motor dysfunction. A selective loss of striatal neurons that give rise to the indirect pathway reduces the inhibitory action of the GPe upon the subthalamic nucleus. The subthalamic nucleus then becomes hypofunctional and causes reduction of the inhibitory action of the GPi upon the thalamus. This subsequent disinhibition of the thalamus leads to chorea. Albin et al. hypothesized that chorea might result from preferential loss of striatal neurons projecting to the GPe, and that rigidity – akinesia might be due to the additional loss of striatal neurons projecting to the GPi (Albin et al., 1990). However, data suggest that dyskinesia does not result only from an imbalance of activity between the two pallidal segments (GPe – hyperactivity; GPi – hypoactivity) but also from imbalance within each pallidal segment (Chesselet & Delfs, 1996; Matsumura et al., 1995). Interestingly, Rosas et al. (2001), using high-resolution T1 -weighted scans, found that the left striatum is larger than the right in neurologically normal adults in contrast to that of individuals with HD.

The striosome – matrix compartments The primate neostriatum is heterogeneously organized based on levels of acetylcholinesterase (AChE) activity. The intensity of histochemical staining for AChE is weak in the 300– to 600 m-wide striosomes and strong in the matrix (Goldman-Rakic, 1982; Holt et al., 1997). Other molecular markers, including huntingtin, exhibit an uneven distribution corresponding to the striosome – matrix compartments (Ferrante et al., 1997). Afferent and efferent connections of the striatum contribute to the striosome – matrix configuration. Afferents to the striosomes originate in the SNc, prefrontal cortex, and limbic system. Efferents from the striosomes terminate in the SNc. Afferents to the matrix originate in the motor and somatosensory cortices, and in the parietal, occipital and frontal cortices. Efferents from the matrix terminate in the GPe, SNr and GPi (Gerfen et al., 1987; Graybiel, 1990). The striosome – matrix organization detected by AChE activity is relatively preserved in the HD neostriatum (Ferrante et al., 1989). Neuronal loss and gliosis occur in both compartments and appear first in the striosomes indicating that the neurons in striosomes may be more vulnerable at an early stage of HD than those in the matrix (Hedreen & Folstein, 1995). Matrix neurons projecting to the GPe appear to degenerate before matrix neurons projecting to the GPi.


account for more than 90% of neostriatal neurons and all contain gamma-aminobutyric acid (GABA). They are the main input and output neurons of the striatum. Subsets of spiny neurons that project to the GPe express enkephalin and neurotensin. Subsets of spiny neurons that project to the GPi or SNr express substance P (SP) and dynorphin. Thus, enkephalin is a reliable marker for the indirect pathway while SP is a reliable marker for the direct pathway. Aspiny neurons are interneurons with local connections. Medium aspiny neurons colocalize nicotinamide adenine dinucleotide phosphate diaphorase (NADPH-d), somatostatin (SS), neuropeptide Y (NPY), and nitric oxide synthase (NOS). Other medium aspiny neurons contain cholecystokinin (CCK), or the calcium-binding protein parvalbumin. The large aspiny neurons utilize ACh (Graybiel & Ragsdale, 1983).

Glutamate and dopamine (DA) neurotransmission in the striatum

Fig. 16.2. Microphotograph including a field of the head of the caudate nucleus of a neurologically normal, 73-year-old individual. Note the presence of one large neuron. The neuropil is relatively homogeneous. Bar = 15m.

Classification of neostriatal neurons Two groups of neostriatal neurons can be distinguished with cresyl violet. One group consists of small- or mediumsized neurons, and a second consists of large neurons (40 m in diameter or larger) (Fig. 16.2). The ratio of smallor medium to large neurons averages 175:1 (range 128– 258:1). Golgi and ultrastructural studies identify at least six categories of neurons (Braak & Braak, 1982; DiFiglia et al., 1976). The two main categories consist of neurons with spiny dendrites (spiny neurons – the most vulnerable in HD), and neurons with smooth dendrites (aspiny neurons – relatively preserved in HD) (DiFiglia et al., 1976). Both the spiny and aspiny neurons are represented by smallor medium-sized neurons, and large neurons (Graveland et al., 1985b). Spiny neurons are projection neurons that

Cortico-striatal projections use glutamate as a neurotransmitter, which is the major excitatory neurotransmitter in the brain. Glutamate activates ionotropic glutamate receptors (iGluR), which control ion channels, and metabotropic glutamate receptors (mGluR), which control the activity of membrane enzymes via G-proteins (Graybiel & Ragsdale, 1983; Nicoletti et al., 1996). Glutamate activates the iGluR that are N-methyl-d-aspartate (NMDA), alpha-amino-3hydroxy-5-methyl-4-isoxazole-propionic acid (AMPA), and kainate receptors. So far, five subtypes of NMDA receptors (NMDAR) can be distinguished. The neostriatal spiny neurons contain predominantly NMDAR-1 and NMDAR-2B. Aspiny neurons contain mainly NMDAR-2D. Among the three known subtypes of AMPA receptors, GluR1 predominates in the striosomes and in aspiny interneurons. The mGluR family includes three groups of G-protein coupled receptors, which modulate excitatory synaptic transmission. Individual mGluR subtypes mediate distinct, facilitatory (group I subtype) or inhibitory (group II and group III subtypes) actions on neuronal degenerative processes. Their activation might lead either to neurotoxicity or neuroprotection. Neostriatal spiny neurons express mainly mGlu5 (group I) and mGlu3 (group II) receptors. Excitotoxic mechanisms are thought to play a major role in the pathophysiology of HD. Overstimulation of iGluR (notably NMDAR) increases the neuronal, cytoplasmic concentration of Ca2+ causing cell death. Overstimulation of group I mGluR opens voltage-operated Ca2+ channels and facilitates glutamate release resulting in neurotoxicity.

379

380


Dopaminergic innervation of neostriatal neurons originates in the SNc. The five subtypes of DA receptors (D1 to D5) identified in the neostriatum can be subdivided into D1-like (D1 and D5), and D2-like (D2, D3, and D4) receptors as per pharmacological criteria. D1 receptors are localized in striatal spiny neurons that give rise to the direct pathway. D2 receptors are present in spiny neurons of the indirect pathway. There is a marked loss of D1 and D2 receptors in asymptomatic mutation carriers of HD (Weeks et al., 1996). This suggests that altered dopamine receptors contribute to the pathophysiology of HD.

tional methods; morphometric approaches, histochemical, immunohistochemical, and in situ hybridization techniques. These techniques revealed among others the presence of ubiquitinated, nuclear inclusions and cytoplasmic aggregates containing mutant huntingtin (DiFiglia et al., 1997; Gutekunst et al., 1999; Maat-Schieman et al., 1999). Distinct topographic and cellular alterations, notably in the striatum and cerebral cortex, are characteristic of HD. A grading system that stages the extent of striatal degeneration has been developed and widely used as a research tool.

General features

Neuropathology Historical view Anton initially observed an association between bilateral atrophy of the putamen and choreic movements in the presence of an apparently normal cerebral cortex and spinal cord (Anton, 1896). Jelgersma correlated the atrophy of the caudate nucleus with chorea in HD (Jelgersma, 1908). Alzheimer attributed the chorea of HD mainly to atrophy of the striatum (Alzheimer, 1911). There was disagreement in early reports about the extent of involvement of the claustrum (Bruyn, 1968; Forno & Jose, 1973; Lewy, 1923; McCaughey, 1961); hypothalamus (Bruyn, 1973); hypothalamic lateral tuberal nucleus (Kremer et al., 1991); amygdala (Bruyn et al., 1979; Davison et al., 1932); hippocampal formation (Braak & Braak, 1992; Forno & Jose, 1973); thalamus (Hallervorden, 1957; Lewy, 1923; McCaughey, 1961); subthalamic nucleus (Lewy, 1923; Spielmeyer, 1926); red nucleus (Lange, 1981); substantia nigra especially pars reticulata (Campbell et al., 1961; Hallervorden, 1957; Lewy, 1923; Richardson, 1990; Schroeder, 1931; Spielmeyer, 1926); nucleus ceruleus (Zweig et al., 1992); superior olivary nucleus (Forno & Jose, 1973; Spielmeyer, 1926; Weisschedel, 1938, 1939); pons and medulla oblongata (McCaughey, 1961; Zweig et al., 1989); cerebellum (Lewy, 1923; Terplan, 1924; Tokay, 1930); and spinal cord (Forno & Jose, 1973; Hallervorden, 1957; Jéquier, 1947; Spielmeyer, 1926; Terplan, 1924). The discrepancies between early reports on the changes of HD are due, in part, to the wide spectrum of pathological features that can exist across HD brains and to the lack of a large series of HD brains available for study by the same group of investigator(s). In our HD research centre, the opportunity to systematically evaluate more than 1200 HD brains during the past 20 years has enhanced our knowledge of the changes that occur in HD. Current understanding of HD changes is based on information gathered from the use of standardized, conven-

On external examination, 80% of HD brains show atrophy of the frontal lobes, and 20% are apparently normal. The mean brain weight at postmortem examination, after fixation (n = 163) was 1067 g (normal about 1350 g) with a sample modal weight of 1140 g (Vonsattel et al., 1985). Halliday et al. measured the total brain volumes of seven HD brains (one grade 4, five grade 3, and one grade 2) and recorded a 19 percent loss compared to controls (n = 12). (Halliday et al., 1998). Examination of coronal sections reveals bilateral atrophy of the striatum in 95% of the HD brains (Figs. 16.3, 16.4 and 16.5). The striatal atrophy is prominent in 80%, mild in 15%, and subtle, if at all in 5% of the brains (Vonsattel & DiFiglia, 1998). The striatal atrophy is apparently symmetric, although recent, morphometric data obtained in vivo revealed that the atrophy is more prominent on the left than on the right side (Rosas et al., 2001). Non-striatal regions show atrophy of variable severity or have normal appearance. As a rule, the brain is diffusely smaller than normal in the late stage of disease. Increased atrophy may occur in non-striatal regions of HD brain with superimposed morbidity. For example, enhanced atrophy of the limbic system (cingulate gyrus, amygdala, hippocampal formation) with severe widening of the temporal horn of the lateral ventricle may occur when HD coexists with Alzheimer disease. Morphometric analyses using five standardized coronal sections each from 30 graded brains revealed a 21–29% cross-sectional area loss in the cerebral cortex, a 28% loss in the thalamus, a 57% loss in the CN, and a 64% loss in the putamen (de la Monte et al., 1988). The white matter showed a 29–34% loss in area supporting observations made as early as 1914 or 1927 (Kiesselbach, 1914; Dunlap, 1927) (Fig. 16.4(b)). With volumetric analyses from six HD brains, Lange found a 20% loss of the ‘hemisphere and cortex’, 58% loss of the ‘striatum’, 57% loss of the GPe, 50% loss of the GPi, and a 24% loss of the subthalamic nucleus


(Lange et al., 1976). Hence Lange’s suggestion that HD is a polytopic disorder with primary and secondary involvements of the brain. Interestingly, in vivo measurements of the striatum of 24, genetically tested, HD individuals showed 41% reduction of the caudate nucleus, and 49% reduction of the putamen compared to controls (Rosas et al., 2001). In HD, the striatum is probably the only site where neuronal loss is associated with ‘active’ reactive, fibrillary astrocytosis (Fig. 16.6). An increased density of oligodendrocytes, up to twice that of controls, is observed, notably in the anterior neostriatum (Forno & Jose, 1973; Hallervorden, 1957; McCaughey, 1961; Myers et al., 1991). Data gathered in the tail of the caudate nucleus of presymptomatic, gene carrier individuals suggest that an increased oligodendrocytic density may precede the onset of symptoms for many years (Gómez-Tortosa et al., 2001). Usually, by conventional methods of evaluation, there is no visible reactive gliosis in the non-striatal parts of the HD brain even when there is atrophy. Scattered, reactive microgliocytes are present, and can be detected with appropriate antibodies within the striatum, neocortex and white matter (Sapp et al., 1999a). There is no lymphoplasmacytic infiltration. The occurrence of prominent, nuclear inclusions in neurons, and scattered glial cells in HD transgenic mice contributed to the identification of nuclear inclusions in scarce neurons in human HD brains (Figs. 16.7, 16.8). (Davies et al., 1997; DiFiglia et al., 1997). These inclusions are not visible in tissue sections stained with hematoxylin and eosin (HE) or with Luxol-fast-blue counterstained with HE (LHE), but are labelled with antibodies directed against ubiquitin or against mutant huntingtin (Fig. 16.8). Interestingly, these inclusions can be detected long before the onset of symptoms in otherwise apparently normal brains of presymptomatic gene carriers (Gómez-Tortosa et al., 2001). Ubiquitinated, neuronal, nuclear inclusions are a common feature of polyQ diseases or polyglutaminopathies.

Fig. 16.3. Coronal sections passing through the left (a), control), and right (b), HD patient) nucleus accumbens. (a). Shows no abnormality and is from a 34-year-old man (suicide, brain weight 1680 g). (b). Shows diffuse atrophy, the brunt of which involves the neostriatum of a 48-year-old man with HD (brain weight 1100 g, grade 3).

Grading of striatal neuropathology The hallmark of HD is the gradual atrophy of the neostriatum (Fig. 16.9). Neostriatal degeneration has an ordered and topographical distribution (Birnbaum, 1941; Dunlap, 1927; Forno & Jose, 1973; Hallervorden, 1957; Kiesselbach, 1914; Lewy, 1923; McCaughey, 1961; Neustaedter, 1933; Roos et al., 1985; Schroeder, 1931; Terplan, 1924; Vonsattel et al., 1985). The tail of the caudate nucleus (TCN) shows more degeneration than the body (BCN), which in turn is more involved than the head (HCN). Similarly the caudal portion of the putamen is more degenerated than the

Fig. 16.4. Coronal sections passing through the left (a), and right (b) anterior commissure of same control (a), and same patient (b) as Fig. 16.3. (a). Shows no abnormality. (b). Shows severe atrophy of the centrum semiovale and striatum, and to a lesser extent of cortex compared to the control (a).

381

382


Fig. 16.5. Coronal sections passing through the left (a), and right (b) lateral geniculate body of the same control (a), and same patient (b) as Fig. 16.3. A. Shows no abnormality. (b) shows diffuse atrophy of centrum semiovale, and corpus callosum. Both the tail and body of the caudate nucleus are almost indistinguishable.

rostral portion (Fig. 16.10). Along the coronal (or dorsoventral) axis of the neostriatum, the dorsal neostriatal regions (Fig. 16.11) are more involved than the ventral ones (Fig. 16.12). Along the medio-lateral axis, the paraventricular half of the CN is more involved than the paracapsular half. With the progression of the disease, neostriatal degeneration appears to simultaneously move in a caudo-rostral direction, in a dorso-ventral direction, and in a mediolateral direction (Fig. 16.10). Fibrillary astrogliosis parallels the loss of neurons along the caudo-rostral and dorsoventral gradients of decreasing severity. Most remaining neostriatal neurons in the postmortem brains have normal morphology but contain more lipofuscin and may be smaller than normally expected. In addition, scattered atrophic neurons with a tendency to form small groups stain darker with LHE or HE or Cresyl violet than the apparently healthy neurons. Thus, they are referred to as neostriatal dark neurons (NDN). These neurons have a scalloped cellular membrane, a granular dark cytoplasm, and a nucleus with condensed chromatin (Fig. 16.13). They are scarce in both the atrophic and in the relatively preserved zones. However, their density increases in the intermediary zone, which lies between the two other zones. About 20% of NDN are labelled with TdT-mediated dUTP-biotin nick end labelling (TUNEL) methods suggesting that they may be undergoing apoptosis. Observations made using a cel-

Fig. 16.6. Dorsal third of the head of the caudate nucleus showing reactive astrocytes labelled with glial fibrillary acidic protein (GFAP). The neostriatum, and to a lesser extent the external segment of globus pallidus are the sites where reactive astrocytosis occurs in HD. Bar 40 = m.

lular model of HD support this hypothesis (Saudou et al., 1998). Less than 5% of the HD brains show unusual microscopical changes especially in the anterior neostriatum. They consist of one to five (rarely more) discrete, round islets of relatively intact parenchyma (Fig. 16.14). The crosssections of the islets measure 0.5–1.0 mm, and thus are larger than striosomes. The density of neurons in islets is the same as or slightly lower than that of the normal neostriatum, but the density of astrocytes is increased. (Vonsattel et al., 1992). Islets are found more frequently in patients with juvenile than adult onset of clinical symptoms. The reason for this is unclear. Vonsattel et al. (1985) developed a grading system the framework of which is the distinctive, temporospatial


Fig. 16.7. Microphotograph of a mouse transgenic for exon 1 of the human HD gene (line R6/2, polyQ 145). Prominent, ubiquitin-labelled nuclear inclusions involving the neurons and scant, glial cells. Cx, cortex (frontal, parasagittal); WM, white matter; CA1, field CA1 or Sommer sector of hippocampus. Bar = 15 m.

pattern of degeneration in the HD striatum (Fig. 16.9, Fig. 16.10). The assignment of a grade of neuropathological severity is based on gross and microscopic findings using conventional methods of examination obtained from three standardized, coronal sections that include the striatum (1. at the level of the nucleus accumbens (Fig. 16.3), 2. just caudal edge of the anterior commissure (Fig. 16.4), and 3. at the level of the lateral geniculate body (Fig. 16.5).) This system has five grades (0–4) of severity of striatal involvement. Grade 0 comprises less than 1% of all HD brains (n = 1200). Gross examination shows features indistinguishable from normal brains. However, further evaluations includ-

ing cell counts indicate a 30–40% loss of neurons in the HCN, and no visible reactive astrocytosis. As alluded to before, a study using sections including the tail of the caudate nucleus of three, presymptomatic, gene carriers revealed nuclear inclusions in all three brains including one individual, with 37 polyQ, who died three decades before the expected age for onset of symptoms. In addition, cell counts of the TCN revealed an increased density of oligodendrocytes among the presymptomatic HD gene carriers (Gómez-Tortosa et al., 2001). Grade 1 comprises 4% of all HD brains. The TCN is much smaller than normal and atrophy of the BCN may also be present. Neuronal loss and astrogliosis are evident in the TCN, and less so in the BCN, dorsal portion of both the head and nearby dorsal putamen. Cell counts show 50% or greater loss of neurons in the HCN. A careful examination of the entire length of the TCN is necessary for assignment of grade 1 since the body and head of the caudate nucleus, and putamen may appear normal on gross examination (Fig. 16.16(a)). The TCN of neurologically normal subjects may show variations including periodic constrictions or segmentations (Fig. 16.15). In contrast to HD, the normal variations are focal, and therefore, likely to be apparent in only one or two coronal sections. Brains assigned grade 2 comprise 16%, those assigned grade 3 comprise 52%, and those assigned grade 4 comprise 28% of all HD brains. Gross striatal atrophy is mild to moderate in grade 2 (the medial outline of the HCN is only slightly convex but still bulges into the lateral ventricle), and severe in grade 3 (the medial outline of the HCN forms a straight line or is slightly concave medially (Fig. 16.3(a)). Thus, the microscopical changes in grade 2, and grade 3 are more severe than in grade 1, and less than in grade 4 brains. In grade 4, the striatum is severely atrophic (the medial contour of the HCN is concave, as is the anterior limb of internal capsule (Fig. 16.16(b)). The neostriatum has lost 95% or more neurons. In at least 50% of grade 4 brains the underlying nucleus accumbens remains relatively preserved.

Relationship of neostriatal degeneration to changes in other brain regions In general, a good correlation exists between the grade of striatal disease and the atrophy of other brain regions than the striatum. In grades 1 and 2, non-striatal structures of the brain are unremarkable or show only mild atrophy unless there is age related volumetric loss or a superimposed disease (Fig. 16.16(a)). However, in grades 3 and 4, non-neostriatal structures including globus pallidus or paleostriatum, neocortex, thalamus, subthalamic nucleus,

383

384


Fig. 16.8. Ubiquitin immunoreactive nuclear inclusions of neurons from frontal cortex (a) and head of the caudate nucleus (b) of a 55-year-old woman with minimal chorea, and with 37 HD-IT15 CAG repeats. These inclusions are not visible with HE- or LHE-stained sections. The inclusions may be detectable with appropriate antibodies long before symptom onset in otherwise unremarkable brains from presymptomatic individuals. Bar = 12 m.

Fig. 16.9. Phenotypic spectrum including 11 HD neostriata (from grade 1 to grade 4) each at the level of the nucleus accumbens compared to a control neostriatum (circled, and centrally located). Note the gradual flattening of the medial border of the head of the caudate nucleus, which forms a straight or a slightly convex line in grade 3, and which is concave in grade 4.


SN, white matter, and cerebellum are slightly to markedly smaller than expected normally (Fig. 16.4(b), 16.5(b)). As further detailed below, these structures may show minimal or marked neuronal loss usually without reactive astrocytosis. Similarly, the white matter atrophy may be severe, yet without any microscopical abnormality recognized by conventional methods.

Globus pallidus The globus pallidus shows atrophy in grades 3 and 4 with the external segment much more involved than the internal segment (Fig. 16.4(b)). In grade 4, there is a 50% volume reduction of the GP. Microscopically, the GP is less abnormal than would be expected from the degree of macroscopic atrophy. The neurons are more densely packed than normal in grade 3, and even more so in grade 4, suggesting that, although tissue bulk decreases, neurons are relatively preserved. According to Lange et al. (1976) the absolute number of pallidal neurons decreases up to 40%, but the neuronal density is up to 42% higher than normal in the Gpe, and 27% higher in the GPi. Thus, the pallidal atrophy is apparently chiefly due to loss of neuropil, and hence of striatal fibre connections and fibre passage, and to a lesser extent to loss of neurons (Schroeder, 1931; Neustaedter, 1933;. McCaughey, 1961; Campbell et al., 1961; Vonsattel et al., 1985). Reactive gliosis is usually confined to the external segment, and is visible in grade 4, and to a lesser extent in grade 3. The ansa lenticularis is thinner than normal in grades 3, and 4.

Cerebral cortex Atrophy of the cerebral cortex may or may not be pronounced in grades 3, and 4 (Figs. 16.3(b), 16.4(b), 16.5(b), and 16.16(a) and (b)). Lange found that the volume loss of the cerebral cortex in HD was more severe in the occipital lobe than in the other lobes (Lange, 1981). In support to this observation is the recent in vivo study on eleven HD patients using high-resolution MR-images, which revealed specific, heterogeneous thinning of the cortical ribbon, which probably proceeds in a caudo-rostral direction (Rosas et al., 2002). Halliday et al. found a correlation between the extent of the striatal atrophy and the relatively uniform cortical atrophy in seven brains including one grade 4, five grade 3, and one grade 2 (Halliday et al., 1998). Forno and Jose (1973) stated that changes in the cerebral cortex were often subtle and difficult to evaluate on histological examination, so that to the naked eye the thinning of the cortical ribbon was the more reliable finding. Even when atrophy is marked, neuronal loss in the HD cerebral

Fig. 16.10. Diagram summarizing the grades, and the topographical variation of striatal neuronal loss, and gliosis characteristic of HD. There are decreasing gradients of neuronal loss and gliosis along the caudo-rostral, medio-lateral, and dorso-ventral axes of the striatum. These gradients are often blurred in grade 4 because of the extent of the changes. Neocortical, and neostriatal, neuronal, nuclear inclusions can be detected with antibodies directed against expanded polyQ or against ubiquitin. These inclusions may be visualized years before symptom onset in gene carriers.

cortex is hard to appreciate on general survey of histological sections resulting in contradictory statements in the literature. On the one hand, Dunlap found the HD cortex (n = 29) to be ‘slightly thinner than in the controls (n = 30) but the difference was very little’, and he found no cell loss (Dunlap, 1927). On the other hand, Terplan claimed that neocortical, neuronal loss is severe in HD. Terplan compared the normal cortex from a 20-year-old executed convict with that of the severely atrophic cortex from a 38-year-old HD woman who died after a 10-year history of chorea, and who was found to have lung gangrene at autopsy (Terplan, 1924). Zalneraitis et al. (1981) observed little or no neuronal loss, normal astrocytes, and a relatively normal content of glial fibrillary acidic protein (GFAP) in the cortex of 14 HD brains. Sotrel et al. (1991) performed morphometric studies of 81 prefrontal cortices, and showed a loss of large, pyramidal neurons in layers III, V, and VI in grades 2–4, with the greatest loss in grade 4. There was no astrogliosis; however, cortical oligodendrocytic density was increased (Sotrel et al., 1991). Cudkowicz and Kowall (1990) found a loss of cortical, long projecting neurons but a normal number of local circuit neurons in graded brains. Contrary to Sotrel et al., they did not find a correlation between the grades and the severity of cortical, neuronal loss. Selemon et al. (1995) measured

385

386


Fig. 16.11. Microphotograph of a field at midpoint between the ependymal and the medial border of internal capsule of the dorsal third of the head of the caudate nucleus (HD, grade 3). The neuropil is loose-textured, and both neuronal loss and reactive gliosis are severe in the dorsal neostriatum in contrast to the relatively preserved nucleus accumbens (Figure 16.12). LHE. Bar = 15 m.

the cortical thickness of Brodmann area (BA) 9 from nine grade 3 or grade 4 brains. They noticed an overall 28% reduction of the cortical thickness with a range from normal thickness to severe thinning (up to 46%), and they observed that cortical, neuronal density was increased in three of them. In a later study, using eight of the nine HD brains, Selemon et al. recorded a 35% increase in neuronal density, a 61% increased glial density, and 30% cortical thinning in BA46 (Selemon et al., 1998). Furthermore, Rajkowska et al. (1998) evaluated BA9, and BA17 of seven HD brains of the same series as that used by Selemon and reported a 50 to 80% reduction of density of ‘extra large neurons’ in layers I, III, V, and VI. In addition, Rajkowska et al. observed, in layer VI, a 23% decrease in the density of large neurons associ-

Fig. 16.12. Microphotograph of a field of the nucleus accumbens at midpoint along the medio-lateral axis of the same section, thus same patient as Fig. 16.11. Neuronal loss and gliosis are mild in contrast to the severely involved dorsal part of the head of the caudate nucleus (Fig. 16.11). LHE. Bar = 15m.

ated with a 150% increase in the density of small neurons. Hedreen et al. (1991) found a 57% neuronal loss in layer VI, and a 71% loss in layer V in BA10 of five grade 4 brains. Macdonald et al. (1997) recorded up to 55% loss of pyramidal neurons in the angular gyrus without detecting any correlation with the grade of striatal atrophy in six brains. Braak and Braak (1992) detected ‘a layer-specific’ loss of neurons without increase of astrocytes in the entorhinal cortex, and subiculum of seven patients with grades 3 or 4 striatal atrophy. The extent of the cerebral, cortical degeneration in HD depends in part on the stage of the disease at the time of the evaluation. It apparently varies from area to area and it is difficult to assess on general survey. Probably a reliable assessment of the cerebral cortical degeneration in HD requires morphometric evaluations.


Fig. 16.13. Neostriatal dark neurons (NDN) in the middle dorso-ventral third of the head of the caudate nucleus, grade 3. With TUNEL methods many but not all NDN are labelled. NDN are probably neurons that were undergoing degeneration at the time of the patient’s demise. The neuron in the upper, right corner is apparently normal in contrast to the three NDN that are aligned along the diagonal passing from the centre to the lower, left corner. LHE. Bar = 12 m.

Fig. 16.14. Huntington’s disease, grade 4 (juvenile). Discrete, relatively preserved, round islet of rostral neostriatum, which is surrounded by loose-textured, gliotic parenchyma. LHE. Bar = 20 mm.

umetric loss) and the scarcity of reactive astrocytes (Lange et al., 1976).

Cerebellum Thalamus, substantia nigra and subthalamic nucleus In the thalamus, neuronal loss with or without astrocytosis involving the centrum medianum is regularly observed in grade 4, and to a lesser extent in grade 3 brains; otherwise the thalamus is apparently normal in lower grades. There is a loss of neurons in the SNr (Lewy, 1923; Spielmeyer, 1926; Campbell et al., 1961; Hallervorden, 1957; Richardson, 1990; Schroeder, 1931). The SNc is thinner than normal, yet its number of neurons is apparently normal in all grades giving the impression of an increased density of pigmented neurons. (Campbell et al., 1961; Richardson, 1990). In the subthalamic nucleus, there is a discrepancy between the marked atrophy present in grades 3, and 4 (up to 25% vol-

There is no consensus on cerebellar findings in HD. Dunlap’s comprehensive report on 29 patients with chronic chorea (17 with proven family history) found only one patient with HD who had cerebellar atrophy. He identified the fraction of the weight of cerebrum/cerebellum to be 5.8/1 in HD compared to 7.2/1 in controls (Dunlap, 1927) (originally quoted as 1/5.8 in HD compared to 1/7.2 in controls). McCaughey found ‘possible patchy loss of Purkinje’s cells’ in six brains and loss of neurons in the dentate nucleus in nine of his series of 21 HD brains. (McCaughey, 1961). In a series of ‘about 300’ HD brains Rodda found only three with ‘severe atrophy of the cerebellum’ at postmortem examination (Rodda, 1981). Two of those three patients had

387

388


Fig. 16.15. Schematic, sagittal representation of the caudate nucleus. Variations of the tail of the caudate nucleus are occasionally seen in subjects without neurological disorder. The variations include segmental constriction, or segmental absence of the tail of the caudate nucleus, which mimics HD findings in grade 1 or less likely in grade 2. A careful examination of the entire length of the tail is necessary for assignment of grade 1 because the anterior striatum can appear normal on gross examination while the tail may be atrophic.

Fig. 16.16. Coronal sections passing through the nucleus accumbens of a nonagenarian woman with 37 HD-IT15 CAG repeats, grade 1 (a), and of an 18-year-old with 82 HD-IT15 CAG repeats, grade 4 (b). On gross examination the anterior neostriatum of the nonagenarian is apparently normal (a); however, the tail, and to a lesser extent the body of the caudate nucleus of the nonagenarian were markedly atrophic (not shown). In contrast, the neostriatum of the 18-year-old patient (b) is severely atrophic, its medial border is concave as is the anterior limb of internal capsule. The corpus callosum of the nonagenarian is more atrophic than that of the young patient suggesting age-related changes.

adult onset symptoms, and no definite family history of HD. The third patient had epilepsy and a family history of HD, and died at the age of 6 years. Jeste et al. (1984) conducted a quantitative study of the cerebellar cortex of 17 HD patients, two of whom had epilepsy. There was no cerebellar atrophy noticed on gross examination. They found a decrease (up to 50%) of the density of Purkinje cells but normal thickness of granular and molecular layers. The Purkinje cell loss was variable in its extent in different patients. Cerebellar atrophy is often reported in patients with juvenile onset. The four patients with juvenile onset, and severe cerebellar atrophy reported by Jervis all had epilepsy (Jervis, 1963). Markham and Knox (1965) reported a 9-yearold patient with epilepsy, severe cerebellar atrophy but ‘no focal atrophy in Sommer’s sector’. Byers et al. (1973) reported four juvenile patients all with severe cerebellar atrophy. The hippocampal formation was available in three of the four patients. Of these three hippocampi, two showed neuronal loss and reactive gliosis suggesting that to some extent the cerebellar atrophy may have been secondary to remote hypoxic–ischaemic events. Juvenile HD patients are prone to seizures, which may account for some cerebellar or hippocampal neuronal loss, two sites notably vulnerable to hypoxic–ischaemic events. In our collection of about 1200 HD brains, we found that the cerebellum is smaller than normally expected in grades, 3 and 4 but relatively less atrophic than the cerebral cortex, and globus pallidus. Despite the clear presence of atrophy, neuronal density in the cerebellar cortex frequently appears within normal limits. Segmental loss of Purkinje cells with or without Bergmann gliosis may occur; however, these changes are inconsistent and seem not to be specific for HD. Extensive loss of neurons in the cerebellar cortex was rarely encountered in our series, then concomitant neuronal changes as those caused by ischaemia were found in the Sommer sector or in neocortical watershed territories. The cerebellum of all of our 11 patients with juvenile onset was smaller than normal, and the Purkinje cell density was either apparently normal or slightly decreased except in one case where it was marked. The advent of antiepileptic drugs prevents hypoxic episodes and may explain why cerebellar changes in juvenile patients are less severe now than before despite the possible toxicity of phenytoin on Purkinje cells. Quantitative studies are needed to determine whether the cerebellum is a site of primary degeneration in HD.

Relative vulnerability of neostriatal neurons As mentioned before, medium, spiny, projection neurons are especially vulnerable in HD. Evidences of relative, selective degeneration of the spiny neurons include curved


dendrites, and decreased or increased density of their spines, the shape of which is altered (Graveland et al., 1985a). Variable rates of degeneration among different types of spiny neurons occur. Enkephalin-containing spiny neurons projecting to the GPe are more affected than substance P-containing neurons projecting to the Gpi (Reiner et al., 1988; Richfield et al., 1995; Sapp et al., 1995). The NADPH-d aspiny interneurons are relatively resistant to degeneration in HD perhaps because of the types of glutamate receptors they express (Ferrante et al., 1987). Interestingly, infusion of the endogenous NMDAR agonist quinolinic acid in rat striatum causes loss of spiny neurons with sparing of NADPH-d aspiny neurons. This experimental, selective, neuronal vulnerability resembles the one that occurs in HD. Furthermore, the intrastriatal infusion of the non-NMDA receptor agonist, kainate or quisqualate, causes loss of both spiny, and NADPH-d aspiny neurons, yet, local injection of s-4-carboxy-3-hydroxyphenylglycine, which is both an agonist of group II, and an antagonist of group I mGluR, protects striatal neurons against quinolinic acid (Beal et al., 1986; Koh et al., 1986; Nicoletti et al., 1996). Thus, NADPH-d-enriched neurons may lack NMDAR and instead have a preponderance of kainate receptors. Likewise, subtypes of mGluR may have protective effects. These observations indicate that the selective, neuronal vulnerability, and variable rates of neuronal degeneration may be a function of specific, glutamate receptors either iGluR or mGlurR or both. In short, striatal projection neurons degenerate in a specific temporal sequence: Striato-SNc followed by striato-GPe, and striato-SNr, then followed by striato-GPi (Albin, 1995). The extreme striatal atrophy and the loss of neurons in grade 4 indicates that both spiny and aspiny neurons are vulnerable at the end stage of the disease. That both medium-sized and large neostriatal neurons eventually degenerate in HD is well documented (Alzheimer, 1911; Hallervorden, 1957; Lewy, 1923; Schroeder, 1931; Terplan, 1924; Lange, 1981; Oyanagi & Ikuta, 1987). Impaired energy metabolism may also play a role in the selective vulnerability of medium spiny neurons. Indeed, regional and selective, neostriatal, neuronal loss (projection neurons > interneurons) can be induced by subacute, systemic injection of 3-nitropropionic acid, which causes irreversible inhibition of succinate dehydrogenase – complex II of the mitochondrial respiratory chain (Beal et al., 1993a). NMDA antagonists can block the toxic effect of 3-nitropropionic acid. Gu et al. reported increased cerebral lactic acid levels in patients, and decreased succinate dehydrogenase activity in postmortem striata, which further supports the hypothesis of impairment of energy metabolism in HD (Gu et al., 1996). Impaired energy metabolism results in decreases in high-energy phos-

phate stores and a deteriorating membrane potential. Under these conditions, the voltage-sensitive Mg2+ block of NMDA receptors is relieved, allowing glutamate to activate the receptors, which might result in a slow ongoing excitotoxicity with eventual neuronal death (Beal et al., 1993b).

The grading system in Huntington disease – research Since 1985 most investigations using HD postmortem tissue include correlation with the grade of neuropathological severity. The use of the grading system has helped to identify the earliest histopathological, and biochemical changes in HD (Augood et al., 1996; Beal et al., 1988; Gourfinkel-An et al., 1998; Hedreen & Folstein, 1995; Landwehrmeyer et al., 1995; Reiner et al., 1988). The analysis of low-grade striatum showed that immunoreactive enkephalin-containing neurons projecting to the GPe were more affected than the SP containing neurons projecting to the Gpi. Histochemical studies of NADPH-d labelled aspiny neurons showed that these cells were relatively spared in low and high grades of striata (Ferrante et al., 1987). Some alterations in the HD cortex are not synchronized with those involving the gradient of striatal degeneration (Mazurek et al., 1997). For example, there was no relationship found between elevated peptide levels in the cortex and the grade. The HD-polyQ lengths correlate with the grades of neuropathological severity (Furtado et al., 1996; Halliday et al., 1998; Penney et al., 1997). Correlation of polyQ lengths and grades using 310 brains from clinically diagnosed HD patients showed that repeat stretches of 37–40 units occurred in all grades but that the largest alleles (> 50 units) occurred only in grades 3 or 4 (Persichetti et al., 1994). Furtado et al. (1996) found that greater polyQ were associated with higher grades or greater neuronal loss in both the caudate nucleus and putamen. Gutekunst et al. observed that aggregates of mutant huntingtin increase in size in higher grades (Gutekunst et al., 1999). Sapp et al. (1999a) reported a grade-dependent increase in reactive microglia in both neocortex and striatum. Interestingly Rosas et al. (2001) provided data obtained in vivo that polyQ lengths govern in part the rate of striatal atrophy in individuals with HD and that the volume loss of the striatum is more prominent on the left than on the right side.

Characteristics of wild type and mutant huntingtin Huntingtin is an ubiquitous, 350 kD protein of unknown function, which is expressed in neurons throughout the

389

390


brain regardless of their relative vulnerability (Gutekunst et al., 1995; Landwehrmeyer et al., 1995). In the normal striatum, huntingtin immunoreactivity is weak in the striosomes and strong in the matrix with variable intensity among medium sized neurons (Ferrante et al., 1997). Huntingtin is found in the cytoplasm of neurons, dendrites, and axon terminals (DiFiglia et al., 1997;. Gutekunst et al., 1999). Huntingtin associates with vesicle membranes and microtubules and, therefore, may function in vesicle transport (DiFiglia et al., 1995). Huntingtin is probably essential during embryogenesis since deletion of both huntingtin alleles in mice is lethal in utero (Duyao et al., 1995; White et al., 1997). However, embryonic stem cells may differentiate to functional neurons in the absence of huntingtin (Metzler et al., 1999). Mutant huntingtin with its elongated NH2-terminal polyglutamine stretch can be distinguished from the wild type by its slower migration in SDS-page, and detection on Western blot with antihuntingtin antisera (DiFiglia et al., 1997; Gutekunst et al., 1999). Mutant huntingtin codistributes with wild type in all regions, and in both grey and white matter. Mutant huntingtin is also detected with the wild type in cortical synaptosomes of HD heterozygotes’ brain tissue indicating that the abnormal protein is expressed and transported with the normal one to nerve endings (Aronin et al., 1995). In the HD brain, huntingtin immunoreactivity is increased in neurons compared to control brains. Labelling appears diffusely elevated in the nucleus and cytoplasm, and in multivesicular organelles of affected neurons (Sapp et al., 1997). The N-terminal region of mutant huntingtin accumulates in the nucleus (Fig. 16.8) and cytoplasm of the neocortex, entorhinal cortex, subiculum, pyramidal neurons of the hippocampus, neostriatum, but apparently not in the globus pallidus, cerebellum, and pars compacta of the substantia nigra (DiFiglia et al., 1997; Gutekunst et al., 1999; Maat-Schieman et al., 1999). Cortical neurons with nuclear or cytoplasmic inclusions or both are more prevalent in the brains of patients with juvenile than with adult onset of symptoms (DiFiglia et al., 1997). Indeed, brains of patients with adult onset of symptoms and either grade 1 or grade 2 show only scant nuclear inclusions and cytoplasmic aggregates of mutant huntingtin in degenerating corticostriatal axons (Sapp et al., 1999b). These findings suggest that cortical projections to the striatum may be among the earliest affected by mutant huntingtin. As mentioned, nuclear inclusions may be detectable long before onset of symptoms as observed in one individual, with 37 repeats, who died three decades before the expected age for onset of symptoms (Gómez-Tortosa et al., 2001). The labelling of neurons with antibodies directed against mutant

huntingtin apparently does not correlate with their relative vulnerability (Kuemmerle et al., 1999). Mice transgenic for exon 1 of the HD gene encoding a highly expanded polyQ tract develop nuclear inclusions before the onset of motor deficits (Fig. 16.7) (Bates et al., 1998; Davies et al., 1997; Mangiarini et al., 1996). Studies in cultured cells show that the presence of mutant protein in the nucleus is required to cause cell death, but not the formation of nuclear inclusions (Kim et al., 1999; Saudou et al., 1998). Other studies in cultured cells and in transgenic mice suggest that the cytoplasmic accumulation of mutant huntingtin may be sufficient to cause neurotoxicity (Hodgson et al., 1999; Kim et al., 1999; Reddy et al., 1998). Extensive ongoing research is exploring the roles of expanded polyQ not only in HD but also in the other, known, seven polyglutaminopathies.

Putative pathogenic mechanisms involving mutant huntingtin Interestingly, heterozygous ‘knock out’ mice with only one wild type huntingtin allele develop a normal phenotype (Duyao et al., 1995). As mentioned before, the mechanism whereby mutant huntingtin is variably harmful to many classes of neurons is mysterious. Several hypotheses of the pathogenesis of mutant huntingtin have been or are being tested. This brief overview addresses a few of them. The mutation through the ensuing abnormal protein confers a toxic gain of function. Accordingly, the pathologic allele gains a deleterious effect, possibly unrelated to the function of the normal allele or normal protein. That homozygous patients have the same phenotypes as heterozygous patients is evidence against a loss of function being causative of HD. Data gathered so far suggest that cleavage products including the expanded polyQ of mutant huntingtin are toxic to susceptible neurons or functionally disruptive with gradually cumulative damages. Indeed, delay of onset of symptoms implies either a progressivity of protein aggregates or injury to susceptible cells (Paulson, 1999; Wanker, 2000). Many theories have been proposed to explain the formation of insoluble aggregates (Persichetti et al., 1999; Perutz et al., 1994). The co-distribution of ubiquitin with mutant huntingtin in the aggregates suggests that ubiquitin-dependent proteolysis of mutant huntingtin is incomplete. Yet, whether the presence of the aggregates contributes to the altered cell function is not known (Kuemmerle et al., 1999). Putatively, the aggregates might alter the regulation of gene transcription, protein interactions, and protein transport within the nucleus and


cytoplasm, and vice versa (Huang et al., 1998; Persichetti et al., 1999; Wheeler et al., 2000). Mutant huntingtin with its expanded polyQ shares similarities with known transcription factors. The expanded polyQ might induce a gain of function resulting in binding of the mutant huntingtin to DNA with possible disruption of transcription (Cha, 2000). Thus, the current growing interest to determine whether transcriptional dysregulation might drive the pathogenesis of HD. A gain of function might involve a conformational change of mutant huntingtin with polymerization into fibrillar structure (Scherzinger et al., 1997; Huang et al., 1998). Maybe the configuration resulting from the expansion of polyQ modifies the solubility of huntingtin or its interactions with other proteins. The repertoire of either wild-type huntingtin or mutant huntingtin interacting proteins keeps expanding. Burke et al. (1996) found that huntingtin interacts through its polyglutamines with the enzyme glyceraldehyde-3phosphate dehydrogenase (GAPDH), and they postulate that inhibition of GAPDH results in a loss of energy metabolism. This enzymatic inhibition supports the hypothesis of impairment of energy metabolism as a possible mechanism of the selective neurodegeneration. Li et al. (1995) identified a protein (huntingtin-associated protein or HAP-1) of unknown function, which binds to huntingtin. The expanded polyQ enhanced this binding, which may induce a change of function. Mutant huntingtin binds tightly to huntingtin interactors containing WW domains (protein interaction motifs named for their critically spaced tryptophan (W) residues) and prevents interactions with wild-type huntingtin, which may contribute critically to the pathogenesis of HD. (Passani et al., 2000). Possible activation of the apoptotic pathway was recorded in 13 graded HD brains. (Portera-Cailliau et al., 1995). Saudou et al. (1998), using an in vitro model of HD that exhibits features of neurodegeneration of HD in vivo, reported cytopathological features of apoptosis in degenerating striatal neurons transfected by mutant huntingtin. Furthermore, they demonstrated that striatal but not hippocampal neurons were susceptible to the effects of mutant huntingtin. Wheeler et al. (2000) hypothesized that the effect of the glutamine tract of mutant huntingtin may act by altering interaction with a critical cellular constituent or by depleting a form of huntingtin essential to medium spiny neurons. Finally, vesicle transport may be affected since mutant and wild-type huntingtins associate with vesicle membranes and co-distribute in axon endings (DiFiglia et al., 1995). Recent attempts using animal models or cell cultures to reduce the pathogenicity of mutant huntingtin provide

encouragement. In a transgenic mouse model of HD, expression of a dominant-negative caspase-1 mutant was found to extend survival and delay both symptomatic onset and appearance of neuronal inclusions. Furthermore, in this model mice, receiving intra-ventricular caspase inhibitors delayed onset of symptoms, reduced symptoms, and increased life expectancy compared to those without administration of caspase inhibitors (Ona et al., 1999). Histone deacetylase inhibitors were found to reduce lethality in two Drosophila models of polyQ disease (Steffan et al., 2001). Cell-specific factors may be crucial in determining the selective neuronal and regional toxicity of mutant huntingtin in the striatum and cortex. Given the clinical and pathological spectra of HD, however, additional genetic factors are likely to influence disease phenotype.

Concomitant neuropathological findings Huntington’s disease does not preclude simultaneous occurrence of cerebral abnormalities other than those characteristic for HD. The reactive, gemistocytic gliosis is apparently indistinguishable from that expected in non-HD brains when, in addition to the HD changes, another lesion such as infarct or metastasis simultaneously involves the neostriatum. With increasing age, the human brain undergoes changes that are commonly ascribed to usual ageing if the individual was without neurological symptoms. The changes are qualitatively similar in Alzheimer’s disease (AD) except that they are more prominent, widespread, and symptomatic. These changes include cerebral atrophy, enlargement of the ventricular system, neuronal loss, amyloid deposition, cerebral amyloid angiopathy, formation of neuritic, immature and diffuse plaques; formation of neurofibrillary tangles of Alzheimer, granulovacuolar degeneration, and accumulation of Marinesco, or Hirano bodies (Chapter 7). These features span a continuum between physiological ageing and neurodegeneration and require a certain threshold of severity or frequency for individuals to become symptomatic, provided they live long enough. The evaluation of changes ascribed to usual aging or to AD are challenging in the HD population for the following reasons. First, HD brains from older patients are rare since the life expectancy of the HD population is shorter than the non HD population, suggesting that AD would be unusual in the HD population (Forno & Jose, 1973; McIntosh et al., 1978; Reyes & Gibbons, 1985; Sax & Vonsattel, 1992). Second, the symptoms due to the HD mutation are likely to predominate and mask those of AD, thus the

391

392


diagnosis of coexistent AD depends on neuropathological evaluation. We identified Alzheimer changes insufficient to meet the neuropathological criteria of AD in 13% of HD brains from patients of all ages. The morphologic and topographic features of Alzheimer changes in the HD brains were identical to those observed in non-HD individuals. Those HD patients with Alzheimer changes were older (mean 72 years) than those without (mean 57 years).

Coexistence of Huntington’s disease and Alzheimer’s disease Among 4050 brains, we found 18 with co-existing HD and AD. We assigned the additional diagnosis of Alzheimer’s disease only to those HD brains with Alzheimer-related cortical lesions corresponding to stages V or VI, or isocortical stages according to Braak et al. (1993). The group with coexisting HD and AD consisted of 13 women and five men (mean age 75.7, ± SD 8.8 years). There was only one who carried the two diagnoses intra-vitam. Indeed, the coexistence of HD with AD is almost never identified intra vitam (McIntosh et al., 1978; Moss et al., 1988; Reyes & Gibbons, 1985; Vonsattel et al., 1985). However, dementia was documented in the rest of the patients and was thought to have been secondary to HD. Details on dementia were lacking in the clinical notes to enable distinction between cortical vs. subcortical dementia. The mean expanded HD-IT15 CAG allele of this group was 41.1 ± 1.89 repeats (shortest 38, longest 44 repeats). In our series, the changes of HD coexists with that of AD in 2% of the HD patients irrespective of their age at death, or in 5% of those who were 65 years or older at death. Therefore, the frequency of AD in the HD population resembles that of the general population since an estimated 3–5% of people older than 65 years are thought to have AD (CoreyBloom et al., 1995; Van Broeckhoven, 1995).

of 35–39 (Table 16.1). Thus, the evaluation of the neostriatal sites involved early in the course of the disease may be crucial to determine whether the mutation caused the changes. For example, in grade 1, the tail of the caudate nucleus is more atrophic than both the body and the head (Fig. 16.10). We have identified six HD brains from nonagenarians. Their HD-IT15 CAG long allele ranged between 38 and 40. Three of them had minimal or no cognitive impairment (Fig. 16.16 (A)). Preservation of cognition even 10 to 30 years following the onset of chorea may occur in late onset HD, which exemplifies the variation occurring in this disease (Britton et al., 1995). One of the three demented HD nonagenarians had coexistent AD, one had coexistent Pick disease (with Pick bodies and Pick cells); and one had hippocampal sclerosis. This series of HD nonagenarians conveys three messages. First, there is a risk of attributing a case of HD to a new mutation in an individual whose carrier parent had mild symptoms with pathologic changes ascribed to ageing or who died before symptom onset. Second, individuals with an HD-IT15 CAG expansion up to 40 or even 41 repeats may have far longer life expectancies than patients with higher repeat sequences. The variability of symptoms and striatal atrophy seen in this group of patients support the hypothesis that extra genes influence the pathophysiology of HD as suggested as early as 1935 (Patzig, 1935). Third, the severity of striatal atrophy in HD nonagenarians may range from mild to severe. When subtle, the HD changes are visible only in sites involved early in the course of the disease and may be obscured by concomitant changes such as neuritic plaques, neurofibrillary tangles of Alzheimer or gliosis (Karluk et al., 1999). The situation whereby old subjects with long-standing, slowly worsening, hereditary chorea, and mild, striatal atrophy is at times referred to as status subchoreaticus or subchorea (Lange et al., 1976; Patzig, 1935). Striatal lesions due to vasculopathies are observed in older individuals and may be superimposed upon the HD changes, thus masking it. The presence of ubiquitin labelled nuclear inclusions in scattered cortical and neostriatal neurons support the diagnosis of HD.

Nonagenarians with Huntington’s disease The symptoms of HD in older individuals (old–old (75 to 85 year), and oldest-old (85-year-old or older)) may be enigmatic even to clinicians familiar with the disease. Coexisting cerebral changes are frequent in nonagenarians, which may mask the clinical and neuropathological features of HD. Furthermore, the cerebral HD changes may be subtle since patients of this group may have HD-IT15 CAG repeat expansion that are in the diagnostically uncertain range

Clinicopathological discrepancies and differential diagnoses The accuracy of clinical diagnosis of HD is high. The morbid course of the disease and its symptoms are quite characteristic. The family history is a contributory diagnostic factor. At the time of death of the patients, clinical evidence of HD is usually identified. The family history might be


misleading, however. We evaluated five brains from patients with the clinical diagnosis of HD, while the neuropathological findings were characteristic of those of AD in four of them, and of Pick disease in the other. Caution must be exercised when determining the presence of the gradients of increasing severity of neuronal loss and gliosis along the ventro-dorsal, latero-medial, and antero-posterior axes of the striatum in the early stages of the disease to mistakenly avoid ruling out HD (Fig. 16.10). These gradients of increasing severity of pathological changes are probably pathognomonic for HD and should be evaluated in all brains from patients with movement disorders.

Progressive supranuclear palsy On gross examination, changes of the lenticular nucleus in progressive supranuclear palsy (PSP) (Chapter 11) can mimic those of HD. The discolouration of the GP, particularly of the GPi, is more pronounced in PSP than in HD. Blurring of the pallidal lateral and medial medullary laminae occurs rather in PSP than in HD, although these laminae are less distinct in grades 3 or 4 than normally expected. In PSP, the atrophy of the GP is more pronounced than that involving the putamen and CN, whereas in HD the brunt of the atrophy involves the CN and putamen. The brain of one patient with the clinical diagnosis of HD in our series showed atrophy of the striatum with the gradients of neuronal loss and gliosis characteristic for HD, and severe atrophy with gliosis and neuronal loss involving the GPe and GPi, subthalamic nucleus, red nucleus, SNc and SNr, dentate nucleus and inferior olivary nucleus. In addition, there were scattered argyrophilic neurons in the atrophic areas, and scant glial cytoplasmic inclusions. Only large neostriatal neurons were argyrophilic. Thus, this brain simultaneously displayed the changes of HD and those of PSP.

Pick disease In a series of 32 neuropathologically evaluated brains with circumscribed atrophy, 69% exhibited atrophy of the CN, 56% displayed pallidal atrophy, and 59% showed nigral lesions (Tissot et al., 1975). The atrophy of the Pick striatum topographically matches that of the cortex; therefore, in Pick disease (Chapter 11), the anterior portions of the striatum are more involved than the posterior, which is in contrast to HD (Spatz, 1952). Furthermore, in Pick disease, the HCN is often more involved than the putamen, and the nucleus accumbens is severely atrophic, in contrast to HD ¨ (Binetti et al., 1998; Luers & Spatz, 1957). Usually, both the

SNc and SNr are involved in Pick disease, whereas there is ¨ relative preservation of the SNc in HD (Luers & Spatz, 1957; Richardson, 1990).

Multiple system atrophy and corticobasal degeneration or corticodentatonigral degeneration with neuronal achromasia Evaluating the topography of the striatal atrophy helps distinguishing between multiple system atrophy or corticobasal degeneration, and HD. In multiple system atrophy (Chapter 15) the putamen is usually more involved than the CN, neuronal loss and gliosis are more conspicuous in the lateral portion of the putamen than medio-dorsally. The paracapsular portions of the HCN and BCN are often involved while the TCN is preserved. In corticobasal degeneration (Chapter 11) the middle third of the striatum along the antero-posterior axis of the cerebral hemisphere bears the brunt of the changes. Usually the atrophy is more pronounced medially than laterally. In addition, the SNc shows neuronal loss and gliosis (Gibb et al., 1989; Rebeiz et al., 1968; Riley et al., 1990). According to our experience, these changes involving the SNc are rare in pure HD. In contrast to multiple system atrophy and corticobasal degeneration glial cytoplasmic inclusions do not occur in HD (Chin & Goldman, 1996; Feany & Dickson, 1995). While the changes are fairly symmetric in HD, they tend to be asymmetric in corticobasal degeneration (Boeve et al., 1999).

Neuroacanthocytosis (chorea-acanthocytosis, McLeod syndrome) Patients with neuroacanthocytosis (also referred to as familial amyotrophic chorea with acanthocytosis, familial neuroacanthocytosis or Levine–Critchley syndrome, chorea-acanthocytosis) have acanthocytes (erythrocytes with spiky, peripheral projections [acantha = thorn]) with or without lipoprotein, and movement disorders such as dystonia, ataxia, parkinsonism, chorea, orolingual tic-like movements, seizures, dementia, and muscular atrophy (Hardie et al., 1991; Peppard et al., 1990; Rinne et al., 1994a,b; Sakai et al., 1985; Yamamoto et al., 1982). Neuroacanthocytosis is either sporadic or familial, and although frequent, chorea may be lacking (Folstein, 1989). Choreaacanthosis was linked to chromosome 9q21 (Rubio et al., 1997). The gene encodes chorein perhaps a structural protein. A 260-bp deletion was found in the coding region of the cDNA sequence in both alleles of patients resulting in a truncated form of chorein (Ueno et al., 2001). McLeod syndrome is an X-linked form of neuroacanthocytosis with movement disorders of insidious onset, in the

393

394


fourth decade, including facial tics, limb chorea, and dystonia of trunk and neck (Danek et al., 2001). The patients have weak expression of the antigens for the Kell blood group system (Ho et al., 1994; Witt et al., 1992). Often the patients have epilepsy, splenomegaly, and muscular changes. The muscle biopsy usually shows alterations that are mixed, neuropathic (fibre type grouping), and myopathic. Myopathic changes are subtle and include central nucleation, fibre size variation; and rare basophilic, or necrotic fibres. The most consistent abnormalities described in neuroacanthocytosis are gross atrophy of the striatum and widening of the frontal horns of the lateral ventricles. The published pictures of coronal sections show severe atrophy of the nucleus accumbens. Neuronal loss and gliosis are severe in the CN and slightly less so in the putamen; neuronal loss is also described in the GP, probably in both external and internal segments (Bird et al., 1978; Hardie et al., 1991; Iwata et al., 1984; Rinne et al., 1994b). Details of the neostriatal distribution of neuronal loss and gliosis are not available in the literature, with the exception of the report by Iwata et al. (1984). These authors found the putamen to be involved ventrally while the dorsal portion was barely affected. Neuronal loss and gliosis were occasionally found in the medial thalamus (Rinne et al., 1994b). The substantia nigra was reported as being normal or as being remarkable for neuronal loss. In some cases the SNc was more involved than the SNr or vice versa (Bird et al., 1978; Hardie et al., 1991; Rinne et al., 1994a). According to Rinne et al. (1994a) neuronal loss does occur in the substantia nigra, especially in patients who had parkinsonism; loss of pigmented neurons or of tyrosine hydroxylase immunoreactive neurons was documented throughout the substantia nigra with the ventrolateral portion most severely involved. The cerebral and cerebellar cortices, subthalamic nucleus, brainstem and medulla oblongata are all apparently within normal limits. Our experience with neuroacanthocytosis is limited to the review of slides of two cases including one reported by Bird et al. (1978), and has left us with the impression that the neostriatal gradients of neuronal loss and gliosis so characteristically present in HD are lacking in neuroacanthocytosis. In addition, it seems that the nucleus accumbens is severely atrophic in neuroacanthocytosis while it is relatively spared in HD as compared to the dorsal half of the HCN or dorsal putamen. Another difference is that gliosis in the GPi is frequently described in neuroacanthocytosis but occurs rarely in HD. Finally, the SNc shows neuronal loss in neuroacanthocytosis with parkinsonism in contrast to HD. The pathological spectrum of the neostriatum in neuroacanthocytosis is incompletely described; systematic evalu-

ation providing topographic data (rostral vs. caudal, dorsal versus ventral, medial versus lateral, GPe vs. GPi) will be helpful in making the diagnosis of neuroacanthocytosis or of HD.

Dentato-rubro-pallido-luysian atrophy Dentato-rubro-pallido-luysian atrophy (DRPLA) including Haw River syndrome is a rare autosomal dominant illness, which like HD belongs to the group of the polyglutaminopathies (Paulson, 1999; Robitaille et al., 1997; Singer, 1992). Thus, DRPLA and HD share some clinical and pathological features. The defect in DRPLA is located in the coding sequence of a gene on chromosome 12p that produces a 190 kDa protein referred to as atrophin the function of which is unknown (Nagafuchi et al., 1994; Schulz & Dichgans, 1999). The same defect causes the Haw River syndrome, a DRPLA-like disorder involving an African American kindred in North Carolina (Burke et al., 1994; Farmer et al., 1989). Like HD, DRPLA displays normal alleles that are polymorphic, and abnormal, expanded alleles that are unstable (Koide et al., 1994). The normal alleles of DRPLA gene have 8 to 35 CAG repeats while the abnormal alleles have 49 to 70 CAG repeats (Becher et al., 1997; Potter et al., 1995). Like in HD, the longer the polyQ expansion, the earlier and the more severe are the symptoms (Ikeuchi et al., 1995). Paternal transmission is often associated with the most severe form of the disease. Movement disorders predominate in DRPLA. Based on their nature three clinical types of DRPLA are distinguished (Iizuka & Hirayama, 1986). Ataxia prevails in the ataxochoreoathetoid type. Chorea prevails in the pseudoHuntington type, which is often misdiagnosed as HD. Myoclonus, epilepsy, and gradual dementia dominate in the myoclonus epilepsy type, which usually affects younger patients than the other types. The brunt of the neuronal loss and gliosis involves the dentatofugal and pallidofugal systems including the external segment of globus pallidus (Becher et al., 1997; Tsuchiya et al., 1998). This pattern of degeneration occurs in each of the three clinical subtypes. Thus, the clinical heterogeneity contrasts with the relative consistency of the sites of predilection for degeneration. The salient neuropathological differences between DRPLA and HD are as follows: In DRPLA neuronal loss and gliosis in the neostriatum are usually much less conspicuous than in the external segment of globus pallidus, which is in contrast to HD. The involvement of the subthalamic nucleus and red nucleus, although variable, is prominent in DRPLA and much less in HD. Myelin loss involving the hilus of the dentate nucleus is seen in DRPLA in contrast to


HD. Neuronal loss involving dentate nucleus may or may not be prominent in DRPLA while neuronal density is often apparently normal or only slightly decreased in HD. The inferior olivary nucleus may show changes reminiscent of pseudohypertrophy in DRPLA and is usually unremarkable in HD. In addition to the changes usually occurring in DRPLA, the following features are observed in brains from individuals with Haw River syndrome: myelin loss of the subcortical white matter, calcification of the basal ganglia and the presence of neuroaxonal dystrophy (Farmer et al., 1989; Robitaille et al., 1997). Neuronal, nuclear inclusions in genetically proven DRPLA were found throughout the brain including spinal cord, and dorsal root ganglia, and in visceral organs (Becher et al., 1998; Yamada et al., 2001). Of interest are the changes involving the brain of a patient with spontaneous oral-facial dyskinesia (Altrocchi & Forno, 1983). Atrophy of the CN was slight on gross examination. However, severe, uneven neuronal loss and fibrillary astrocytosis was limited to the dorsal half of the CN and putamen, giving a mosaic appearance. The lateral half of the tail of the caudate nucleus was similarly involved. Mild gliosis was also present in the dorsal portion of the GPe. The other deep grey nuclei, cortex, white matter, brainstem, and cerebellum were normal. We encountered instances in which, on gross examination of the coronal sections, the atrophy of the striatum mimicked that usually occurring in HD, e.g. Creutzfeldt– Jakob disease (Growdon & Vonsattel, 1993), subacute sclerosing panencephalitis, and neuronal lipofuscinosis. As mentioned before, occasionally HD-like changes can occur in the absence of expanded CAG repeats (Margolis et al., 2001; Persichetti et al., 1994). The identification of such brains following careful evaluations including saving fresh frozen samples for nucleic acid studies, and genotyping family members could unveil an as yet unknown subcategory of striatal degeneration. Clarification of this putative subcategory of brains would be invaluable in the process of understanding the mechanism of striatal degeneration.

Acknowledgements We thank E. D. Bird, MD for his immense and continuous support. We thank the pathologists who referred case material to the New York Brain Bank, Columbia University, New York (NY); and Harvard Brain Tissue Resource Center, McLean Hospital, Belmont, MA. We are grateful to the families of the patients for providing brain tissue for research. This work was supported in part by NIH grants NS 16367 (Huntington’s Disease Center Without Walls).

REFERENCES Albin, R. L. (1995). Selective neurodegeneration in Huntington’s disease. Annals of Neurology, 38, 835–6. Albin, R. L. & Tagle, D. A. (1995). Genetics and molecular biology of Huntington’s disease. Trends in Neurosciences, 18, 11–14. Albin, R. L., Young, A. B. & Penney, J. B. (1989). The functional anatomy of basal ganglia disorders. Trends in Neurosciences, 12, 366–75. Albin, R. L., Reiner, A., Anderson, K. D., Penney, J. B. & Young, A. B. (1990). Striatal and nigral neuron subpopulations in rigid Huntington’s disease: implications for the functional anatomy of chorea and rigidity-akinesia. Annals of Neurology, 27, 357– 65. Alexander, G. E. & Crutcher, M. D. (1990). Functional architecture of basal ganglia circuits: neural substrates of parallel processing. Trends in Neurosciences, 13, 266–71. Altrocchi, P. H. & Forno, L. S. (1983). Spontaneous oral-facial dyskinesia: neuropathology of a case. Neurology, 33, 802–5. ¨ Alzheimer, A. (1911). Uber die anatomische Grundlage der Huntingtonschen Chorea und der choreatischen Bewegungen ¨ uberhaupt. Neurologisches Centralblatt, 30, 891–2. ¨ Anton, G. (1896). Uber die Beteiligung der grossen basalen Gehirnganglien bei Bewegungstörungen und insbesondere bei Chorea. Jahrbucher fur ¨ ¨ Psychiatrie und Neurologie (Lpz), 14, 141–81. Aronin, N., Chase, K., Young, C. et al. (1995). CAG expansion affects the expression of mutant huntingtin in the Huntington’s disease brain. Neuron, 15, 1193–201. Augood, S. J., Faull, R. L. M., Love, D. R. & Emson, P. C. (1996). Reduction in enkephalin and substance P messenger RNA in the striatum of early grade Huntington’s disease: A detailed cellular in situ hybridization study. Neuroscience, 72, 1023–36. Bates, G. P., Mangiarini, L. & Davies, S. W. (1998). Transgenic mice in the study of polyglutamine repeat expansion diseases. Brain Pathology, 8, 699–714. Beal, M. F., Kowall, N. W., Ellison, D. W., Mazurek, M. F., Swartz, K. J. & Martin, J. B. (1986). Replication of the neurochemical characteristics of Huntington’s disease by quinolinic acid. Nature, 321, 168–71. Beal, M. F., Ellison, D. W., Mazurek, M. F. et al. (1988). A detailed examination of substance P in pathologically graded cases of Huntington’s disease. Journal of the Neurological Sciences, 84, 51–61. Beal, M. F., Brouillet, E., Jenkins, B. G. et al. (1993a). Neurochemical and histologic characterization of striatal excitotoxic lesions produced by the mitochondrial toxin 3-nitropropionic acid. The Journal of Neuroscience, 13, 4181–92. Beal, M. F., Hyman, B. T. & Koroshetz, W. (1993b). Do defects in mitochondrial energy metabolism underlie the pathology of neurodegenerative diseases? Trends in Neurosciences, 16, 125– 31. Becher, M. W., Rubinsztein, D. C., Leggo, J. et al. (1997). Dentatorubral and pallidolyusian atrophy (DRPLA). Clinical and neuropathological findings in genetically confirmed North

395

396


American and European Pedigrees. Movement Disorders, 12, 519–30. Becher, M. W., Kotzuk, J. A., Sharp, A. H. et al. (1998). Intranuclear neuronal inclusions in Huntington’s disease and dentatorubropallidoluysian atrophy: correlation between the density of inclusions and IT 15 CAG triplet repeat length. Neurobiology of Disease, 4, 387–97. Binetti, G., Growdon, J. H. & Vonsattel, J.-P. G. (1998). Pick’s disease. In Blue Books of Practical Neurology. The dementias (19) (Boston, Butterworth-Heinemann), pp. 7–44. Bird, T. D., Cederbaum, S., Valpey, R. W. & Stahl, W. L. (1978). Familial degeneration of the basal ganglia with acanthocytosis: a clinical, neuropathological, and neurochemical study. Annals of Neurology, 3, 253–8. Birnbaum, G. (1941). Chronisch-progressive Chorea mit Kleinhirnatrophie. Archiv fur ¨ Psychiatrie und Nervenkrankheiten, 114, 160–82. Boeve, B. F., Maraganore, D. M., Parisi, J. E. et al. (1999). Pathologic heterogeneity in clinically diagnosed corticobasal degeneration. Neurology, 53, 795–800. Braak, H. & Braak, E. (1982). Neuronal types in the striatum of man. Cell and Tissue Research, 227, 319–42. (1992). Allocortical involvement in Huntington’s disease. Neuropathology and Applied Neurobiology, 18, 539–47. Braak, H., Braak, E. & Bohl, J. (1993). Staging of Alzheimer related cortical destruction. European Neurology, 33, 403–8. Britton, J. W., Uitti, R. J., Ahlskog, J. E., Robinson, R. G., Kremer, B. & Hayden, M. R. (1995). Hereditary late-onset chorea without significant dementia: genetic evidence for substantial phenotypic variation in Huntington’s disease. Neurology, 45, 443–7. Bruyn, G. W. (1968). Huntington’s chorea; historical, clinical and laboratory synopsis. In Handbook of Clinical Neurology. Vol. 6, pp. 298–378. Amsterdam, Elsevier Science Publishers V. B. Bruyn, G. W. (1973). Neuropathological changes in Huntington’s chorea. In Huntington’s chorea 1872–1972. Advances in Neurology. Volume 1, pp. 399–403. New York: Raven. Bruyn, G. W., Bots, G. T. A. M. & Dom, R. (1979). Huntington’s chorea: Current neuropathological status. In Advances in Neurology, vol. 23, Huntington’s disease (New York, Raven), pp. 83–93. Burke, J. R., Wingfield, M. S., Lewis, K. E. et al. (1994). The Haw River Syndrome: dentatorubropallidoluysian atrophy (DRPLA) in an African-American family. Nature Genetics, 7, 521–4. Burke, J. R., Enghild, J. J., Martin, M. E. et al. (1996). Huntingtin and DRPLA proteins selectively interact with the enzyme GAPDH. Nature Medicine, 2, 347–50. Byers, R. K., Gilles, F. H. & Fung, C. (1973). Huntington’s disease in children. Neurology, 23, 561–9. Campbell, A. M. G., Corner, B., Norman, R. M. & Urich, H. (1961). The rigid form of Huntington’s disease. Journal of Neurology, Neurosurgery, and Psychiatry, 24, 71–7. Carpenter, M. B. & Sutin, J. (1983). Human Neuroanatomy, 8th edn. Baltimore/London: Williams & Wilkins. Cha, J.-H. (2000). Transcriptional dysregulation in Huntington’s disease. Trends in Neurosciences, 23, 387–92.

Chesselet, M.-F. & Delfs, J. M. (1996). Basal ganglia and movement disorders: an update. Trends in Neurosciences, 19, 417–22. Chin, S., S.-M. & Goldman, J. E. (1996). Glial inclusions in CNS degenerative diseases. Journal of Neuropathology and Experimental Neurology, 55, 499–508. Conneally, P. M. (1984). Huntington disease: genetics and epidemiology. American Journal of Human Genetics, 36, 506–26. Corey-Bloom, J., Thal, L. J., Galasko, D. et al. (1995). Diagnosis and evaluation of dementia. Neurology, 45, 211–18. Cudkowicz, M. & Kowall, N. W. (1990). Degeneration of pyramidal projection neurons in Huntington’s disease cortex. Annals of Neurology, 27, 200–4. Danek, A., Rubio, J. P., Rampoldi, L. et al. (2001). McLeod neuroacanthocytosis: genotype and phenotype. Annals of Neurology, 50, 755–64. Davies, S. W., Turmaine, M., Cozens, B. A. et al. (1997). Formation of neuronal intranuclear inclusions underlies the neurological dysfunction in mice transgenic for the HD mutation. Cell, 90, 537–48. Davison, C., Goodhart, S. P. & Shlionsky, H. (1932). Chronic progressive chorea. The pathogenesis and mechanism; a histopathologic study. Archives of Neurology and Psychiatry (Chicago), 27, 906–28. de la Monte, S. M., Vonsattel, J. P. & Richardson, E. P., Jr. (1988). Morphometric demonstration of atrophic changes in the cerebral cortex, white matter, and neostriatum in Huntington’s disease. Journal of Neuropathology and Experimental Neurology, 47, 516–25. DiFiglia, M., Pasik, P. & Pasik, T. (1976). A Golgi study of neuronal types in the neostriatum of monkeys. Brain Research, 114, 245– 56. DiFiglia, M., Sapp, E., Chase, K. et al. (1995). Huntingtin is a cytoplasmic protein associated with vesicles in human and rat brain neurons. Neuron, 14, 1075–81. DiFiglia, M. et al. (1997). Aggregation of huntingtin in neuronal intranuclear inclusions and dystrophic neurites in brain. Science, 277, 1990–3. Dunlap, C. B. (1927). Pathologic changes in Huntington’s chorea. Archives of Neurology and Psychiatry, (Chicago), 18, 867–943. Duyao, M., Ambrose, C., Myers, R. et al. (1993). Trinucleotide repeat length instability and age of onset in Huntington’s disease. Nature Genetics, 4, 387–92. Duyao, M. P., Auerbach, A. B., Ryan, A. et al. (1995). Inactivation of the mouse Huntington’s disease gene homolog (Hdh ). Science, 269, 407–10. Farmer, T. W., Wingfield, M. S., Lynch, et al. (1989). Ataxia, chorea, seizures, and dementia. Pathologic features of a newly defined familial disorder. Archives of Neurology, 46, 774–9. Feany, M. B. & Dickson, D. W. (1995). Widespread cytoskeletal pathology characterizes corticobasal degeneration. American Journal of Pathology, 146, 1388–96. Ferrante, R. J., Kowall, N. W., Beal, M. F. et al. (1987). Morphologic and histochemical characteristics of a spared subset of striatal neurons in Huntington’s disease. Journal of Neuropathology and Experimental Neurology, 46, 12–27.


Ferrante, R. J., Kowall, N. W. & Richardson, E. P., Jr. (1989). Cellular composition of striatal patch and matrix compartments in Huntington’s disease. Society of Neuroscience, 15, 935. Ferrante, R. J., Gutekunst, C.-A., Persichetti, F. et al. (1997). Heterogeneous topographic and cellular distribution of huntingtin expression in the normal human neostriatum. The Journal of Neuroscience, 17, 3052–63. Folstein, S. E. (1989). The diagnosis of Huntington’s disease. In Huntington’s Disease. A Disorder of Families. pp. 125–48. Baltimore: The Johns Hopkins University Press. (1990). Huntington’s Disease. A Disorder of Families. London, The Johns Hopkins University Press Ltd. Forno, L. S. & Jose, C. (1973). Huntington’s chorea: a pathological study. In Advances in Neurology, vol.1, Huntington’s chorea pp. 453–70. New York: Raven. Furtado, S., Suchowersky, O., Rewcastle, B., Graham, L., Klimek, M. L. & Garber, A. (1996). Relationship between trinucleotide repeats and neuropathological changes in Huntington’s disease. Annals of Neurology, 39, 132–6. Gerfen, C. R., Herkenham, M. & Thibault, J. (1987). The neostriatal mosaic: II. Patch- and matrix-directed mesostriatal dopaminergic and non-dopaminergic systems. The Journal of Neuroscience, 7, 3915–34. Gibb, R. G., Luthert, P. J. & Marsden, C. D. (1989). Corticobasal degeneration. Brain, 112, 1171–92. Goldman-Rakic, P. S. (1982). Cytoarchitectonic heterogeneity of the primate neostriatum: subdivision into island and matrix cellular compartments. The Journal of Comparative Neurology, 205, 398–413. Gómez-Tortosa, E., del Barrio, A., Garcia Ruiz, P. J. et al. (1998). Severity of cognitive impairment in juvenile and lateonset Huntington disease. Archives of Neurology, 55, 835– 843. Gómez-Tortosa, E., MacDonald, M. E., Friend, J. C. et al. (2001). Quantitative neuropathological changes in presymptomatic Huntington’s disease. Annals of Neurology, 49, 29–34. Gourfinkel-An, I., Cancel, G., Duyckaerts, C. et al. (1998). Neuronal distribution of intranuclear inclusions in Huntington’s disease with adult onset. NeuroReport, 9, 1823–6. Graveland, G. A., Williams, R. S. & DiFiglia, M. (1985a). Evidence for degenerative and regenerative changes in neostriatal spiny neurons in Huntington’s disease. Science, 227, 770–3. Graveland, G. A., Williams, R. S. & DiFiglia, M. (1985b). A Golgi study of the human neostriatum: neurons and afferent fibers. The Journal of Comparative Neurology, 234, 317–33. Graybiel, A. M. (1990). Neurotransmitters and neuromodulators in the basal ganglia. Trends in Neurosciences, 13, 244–54. Graybiel, A. M. & Ragsdale, C. W., Jr. (1983). Biochemical anatomy of the striatum. In Chemical Neuroanatomy, pp. 427–504. New York: Raven. Graybiel, A. M., Aosaki, T., Flaherty, A. W. & Kimura, M. (1994). The basal ganglia and adaptive motor control. Science, 265, 1826– 31. Group, T. H. s. D. C. R., MacDonald, M. E., Ambrose, C. M. et al. (1993). A novel gene containing a trinucleotide repeat that is

expanded and unstable on Huntington’s disease chromosomes. Cell, 72, 971–83. Growdon, J. H. & Vonsattel, J. P. (1993). Case Records of the Massachusetts General Hospital. Case 17–1993. A 53-year-old woman who died after several years of a dementing illness with intermittent generalized seizures and abnormal movements of the extremities and head. The New England Journal of Medicine, 328, 1259–66. Gu, M., Gash, M. T., Mann, V. M., Javoy-Agid, F., Cooper, J. M. & Schapira, A. H. V. (1996). Mitochondrial defect in Huntington’s disease caudate nucleus. Annals of Neurology, 39, 385–9. Gutekunst, C.-A., Levey, A. I., Heilman, G. J. et al. (1995). Identification and localization of huntingtin in brain and human lymphoblastoid cell lines with anti-fusion protein antibodies. Proceedings of the National Academy of Sciences of the United States of America, 92, 8710–14. Gutekunst, C.-A., Li, S.-H., Mulroy, J. S. et al. (1999). Nuclear and neuropil aggregates in Huntington’s disease: relationship to neuropathology. The Journal of Neuroscience, 19, 2522–34. Hallervorden, J. (1957). Huntingtonsche Chorea (Chorea chronica progressiva hereditaria). In Handbuch der speziellen pathologischen Anatomie und Histologie (XIII/1 Bandteil A), pp. 793– 822. Berlin: Göttingen, Heidelberg: Springer Verlag. Halliday, G. M., McRitchie, D. A., Macdonald, V., Double, K. L., Trent, R. J. & McCusker, E. (1998). Regional specificity of brain atrophy in Huntington’s disease. Experimental Neurology, 154, 663–72. Hardie, R. J., Pullon, H. W. H., Harding, A. E. et al. (1991). Neuroacanthocytosis. A clinical, haematological and pathological study of 19 cases. Brain, 114, 13–49. Harper, P. S. (1996). Huntington’s disease. Major Problems in Neurology (31), 2nd edn London, W. B. Saunders Company Ltd. Harper, P. S., Morris, M. R., Quarrell, O. W. J., Shaw, D. J., Tyler, A. & Youngman, S. (1991). The clinical neurology of Huntington’s disease. In Huntington’s Disease. Major Problems in Neurology, 22. pp. 37–80. London, Philadelphia, Toronto, Sydney, Tokyo: W. B. Saunders Company Ltd. Hayden, M. R. (1981). Huntington’s Chorea. Berlin, Heidelberg, New York: Springer-Verlag. Hedreen, J. C. & Folstein, S. E. (1995). Early loss of neostriatal striosome neurons in Huntington’s disease. Journal of Neuropathology and Experimental Neurology, 54, 105–20. Hedreen, J. C., Peyser, C. E., Folstein, S. E. & Ross, C. A. (1991). Neuronal loss in layers V and VI of cerebral cortex in Huntington’s disease. NeurosciLett. 133, 257–61. Hemachandra, P., Williams, M., Charles, V. et al. (1998). Behavioural abnormalities and selective neuronal loss in HD transgenic mice expressing mutated full-length HD cDNA. Nature Genetics 20, 198–202. Hersch, S., Jones, R., Koroshetz, W. & Quaid, K. (1994). The neurogenetics genie: testing for the Huntington’s disease mutation. Neurology, 44, 1369–1373. Ho, M., Chelly, J., Carter, N., Danek, A., Crocker, P. & Monaco, A. P. (1994). Isolation of the gene for McLoed syndrome that encodes a novel membrane transport protein. Cell, 77, 869–80.

397

398


Hodgson, J. G., Agopyan, N., Gutekunst, C.-A. et al. (1999). A YAC mouse model for Huntington’s disease with full-length mutant huntingtin, cytoplasmic toxicity, and selective striatal neurodegeneration. Neuron, 23, 182–92. Holt, D. J., Graybiel, A. M. & Saper, C. B. (1997). Neurochemical architecture of the human striatum. The Journal of Comparative Neurology, 384, 1–25. Huang, C. C., Faber, P. W., Persichetti, F. et al. (1998). Amyloid formation by mutant huntingtin: Threshold, progressivity and recruitment of normal polyglutamine proteins. Somatic Cell and Molecular Genetics, 24, 217–33. Iizuka, R. & Hirayama, K. (1986). Dentato-rubro-pallido-luysian atrophy. In Handbook of Clinical Neurology, Vol. 5 (49) Extrapyramidal Disorders, pp. 437–43. Amsterdam: Elsevier Science Publishers B.V. Ikeuchi, T., Koide, R., Tanaka, H. et al. (1995). Dentatorubralpallidoluysian atrophy: clinical features are closely related to unstable expansions of trinucleotide (CAG) repeat. Annals of Neurology, 37, 769–75. International Huntington Association (IHA) and the World Federation of Neurology (WFN) Research Group on Huntington’s Chorea. (1994). Guidelines for the molecular genetics predictive test in Huntington’s disease. Neurology, 44, 1533–6. Iwata, M., Fuse, S., Sakuta, M. & Toyokura, Y. (1984). Neuropathological study of chorea-acanthocytosis. Japanese Journal of Medicine, 23, 118–22. Jelgersma, G. (1908). Neue anatomische Befunde bei Paralysis agitans und bei chronischer Chorea. Neurologisches Centralblatt, 27, 995–6. Jéquier, M. (1947). Remarques sur la chorée de Huntington: le rôle des lésions médullaires. Schweizer Archiv fur ¨ Neurologie und Psychiatrie, 60, 405–7. Jervis, G. A. (1963). Huntington’s chorea in childhood. Archives of Neurology, 9, 244–57. Jeste, D. V., Barban, L. & Parisi, J. (1984). Reduced Purkinje cell density in Huntington’s disease. Experimental Neurology, 85, 78–86. Karluk, D., Myers, R. H., DiFiglia, M. et al. (1999). Nonagenarians with Huntington disease (HD) have low CAG repeats. Journal of Neuropathology and Experimental Neurology, 58, 549. Kiesselbach, G. (1914). Anatomischer Befund eines Falles von Huntingtonscher Chorea. Monatsschr Psychiat Neurol. 35, 525–43. Kim, M., Lee, H.-S., LaForet, G. et al. (1999). Mutant huntingtin expression in clonal striatal cells: Dissociation of inclusion formation and neuronal survival by caspase inhibition. The Journal of Neuroscience, 19, 964–73. Koh, J.-Y., Peters, S. & Choi, D. W. (1986). Neurons containing NADPH-Diaphorase are selectively resistant to quinolinate toxicity. Science, 234, 73–6. Koide, R., Ikeuchi, T., Onodera, O. et al. (1994). Unstable expansion of CAG repeat in hereditary dentatorubral pallidoluysian atrophy (DRPLA). Nature Genetics, 6, 9–13. Kremer, B., Goldberg, P., Andrews, S. E. & et al. (1994). A worldwide study of the Huntington’s disease mutation. The sensitivity and

specificity of measuring CAG repeats. The New England Journal of Medicine, 330, 1401–6. Kremer, H. P. H., Roos, R. A. C., Dingjan, G. M. Bots, G. T. A. M., Bruyn, G. W. & Hofman, M. A. (1991). The hypothalamic lateral tuberal nucleus and the characteristics of neuronal loss in Huntington’s disease. Neurosci Letters, 132, 101–4. Kuemmerle, S., Gutekunst, C.-A., Klein, A. M. et al. (1999). Huntingtin aggregates may not predict neuronal death in Huntington’s disease. Annals of Neurology, 46, 842–9. ¨ Kunzle, H. (1975). Bilateral projections from precentral motor cortex to the putamen and other parts of the basal ganglia. An autoradiographic study in Macaca fascicularis. Brain Research, 88, 195–209. (1977). Projections from the primary somatosensory cortex to basal ganglia and thalamus in the monkey. Experimental Brain Research, 30, 481–92. Landwehrmeyer, G. B., McNeil, S. M., Dure, L. S. et al. (1995). Huntington’s disease gene: regional and cellular expression in brain of normal and affected individuals. Annals of Neurology, 37, 218–30. Lange, H., Thörner, G., Hopf, A. & Schröder, K. F. (1976). Morphometric studies of the neuropathological changes in choreatic diseases. Journal of Neurological Science, 28, 401–25. Lange, H. W. (1981). Quantitative changes of telencephalon, diencephalon, and mesencephalon in Huntington’s chorea, postencephalitic, and idiopathic parkinsonism. Verh Anat Ges, 75, 923–5. Lewy, F. H. (1923). Die Histopathologie der choreatischen Erkrankungen. Zeitschrift fur ¨ die gesamte Neurologie und Psychiatrie (Berlin), 85, 622–58. Li, X.-J., Li, S.-H., Sharp, A. H. et al. (1995). A huntingtin-associated protein enriched in brain with implications for pathology. Nature, 378, 398–402. ¨ Luers, T. & Spatz, H. (1957). Picksche Krankheit (Progressive umschriebene Grosshirnatrophie). In Handbuch der speziellen pathologischen Anatomie und Histologie (XIII/1 Bandteil A), pp. 614–715. Berlin, Göttingen, Heidelberg: Springer Verlag. Maat-Schieman, M. L. C., Dorsman, J. C., Smoor, M. A. et al. (1999). Distribution of inclusions in neuronal nuclei and dystrophic neurites in Huntington disease brain. Journal of Neuropathology and Experimental Neurology, 58, 129–37. McCaughey, W. T. E. (1961). The pathologic spectrum of Huntington’s chorea. Journal of Nervous and Mental Diseases, 133, 91– 103. Macdonald, V., Halliday, G. M., Trent, R. J. & McCusker, E. A. (1997). Significant loss of pyramidal neurons in the angular gyrus of patients with Huntington’s disease. Neuropathology and Applied Neurobiology, 23, 492–5. McIntosh, G. C., Jameson, H. D. & Markesbery, W. R. (1978). Huntington disease associated with Alzheimer disease. Annals of Neurology, 3, 545–8. MacMillan, J. C., Snell, R. G., Tyler, A. et al. (1993). Molecular analysis and clinical correlations of the Huntington’s disease mutation. Lancet, 342, 954–8.


Mangiarini, L., Sathasivam, K., Seller, M. et al. (1996). Exon 1 of the HD gene with an expanded CAG repeat is sufficient to cause a progressive neurological phenotype in transgenic mice. Cell, 87, 493–506. Margolis, R. L., O’Hearn, E., Rosenblatt, A. et al. (2001). A disorder similar to Huntington’s disease is associated with a novel CAG repeat expansion. Annals of Neurology, 50, 373–80. Markham, C. H. & Knox, J. W. (1965). Observations on Huntington’s chorea in childhood. The Journal of Pediatrics, 67, 46–57. Matsumura, M., Tremblay, L., Richard, H. & Filion, M. (1995). Activity of pallidal neurons in the monkey during dyskinesia induced by injection of bicuculline in the external pallidum. Neuroscience, 65, 59–70. Mazurek, M. F., Garside, S. & Beal, M. F. (1997). Cortical peptide changes in Huntington’s disease may be independent of striatal degeneration. Annals of Neurology, 41, 540–7. Metzler, M., Chen, N., Helgason, C. D. et al. (1999). Life without huntingtin: normal differentiation into functional neurons. Journal of Neurochemistry, 72, 1009–18. Moss, R. J., Mastri, A. R. & Schut, L. J. (1988). The coexistence and differentiation of late onset Huntington’s disease and Alzheimer’s disease. A case report and review of the literature. Journal of the American Geriatrics Society, 36, 237–41. Myers, R. H., Sax, D. S., Schoenfeld, M. et al. (1985). Late onset of Huntington’s disease. Journal of Neurology, Neurosurgery, and Psychiatry, 48, 530–4. Myers, R. H., Leavitt, J., Farrer, L. A. et al. (1989). Homozygote for Huntington disease. American Journal of Human Genetics, 45, 615–8. Myers, R. H., Vonsattel, J. P., Paskevich, P. A. et al. (1991). Decreased neuronal and increased oligodendroglial densities in Huntington’s disease caudate nucleus. Journal of Neuropathology and Experimental Neurology, 50, 729–42. Myers, R. H., MacDonald, M. E., Koroshetz, W. J. et al. (1993). De novo expansion of a (CAG)n repeat in sporadic Huntington’s disease. Nature Genetics, 5, 168–73. Nagafuchi, S., Yanagisawa, H., Sato, K. et al. (1994). Dentatorubral and pallidoluysian atrophy expansion of an unstable CAG trinucleotide on chromosome 12p. Nature Genetics, 6, 14–8. Nance, M. A., Mathias-Hagen, V., Breningstall, G., Wick, M. J. & McGlennen, R. C. (1999). Analysis of a very large trinucleotide repeat in a patient with juvenile Huntington’s disease. Neurology, 52, 392–4. Neustaedter, M. (1933). Concerning the striatal localization in chronic progressive chorea. With a report of three cases, two of the Huntington type in siblings and one senile arteriosclerotic, with necropsies. Journal of Nervous and Mental Disease, 78, 470–91. Nicoletti, F., Bruno, V., Copani, A., Casabona, G. & Knöpfel, T. (1996). Metabotropic glutamate receptors: a new target for the therapy of neurodegenerative disorders? Trends in Neurosciences, 19, 267–71. Ona, V. O., Li, M., Vonsattel, J. P. G. et al. (1999). Inhibition of caspase1 slows disease progression in a mouse model of Huntington’s disease. Nature, 399, 263–7.

Oyanagi, K. & Ikuta, F. (1987). A morphometric reevaluation of Huntington’s chorea with special reference to the large neurons in the neostriatum. Clinical Neuropathology, 6, 71–9. Passani, L. A., Bedford, M. T., Faber, P. W. et al. (2000). Huntingtin’s WW domain partners in Huntington’s disease post-mortem brain fulfill genetic criteria for direct involvement in Huntington’s disease pathogenesis. Human Molecular Genetics, 9, 2175–82. Patzig, B. (1935). Vererbung von Bewegungsstörungen. Zeitschrift fur ¨ Induktive Abstammungs- und Vererbungslehre, 70, 476–84. Paulson, H. L. (1999). Protein fate in neurodegenerative proteinopathies: polyglutamine diseases join the (mis)fold. Human Genetics, 64, 339–45. Penney, J. B., Vonsattel, J.-P., MacDonald, M. E., Gusella, J. F. & Myers, R. H. (1997). CAG repeat number governs development rate of pathology in Huntington’s disease. Annals of Neurology, 41, 689–92. Peppard, R. F., Lu, C. S., Chu, N.-S., Teal, P., Martin, W. R. W. & Calne, D. B. (1990). Parkinsonism with neuroacanthocytosis. The Canadian Journal of Neurological Sciences, 17, 298–301. Persichetti, F., Srinidhi, J., Kanaley, L. et al. (1994). Huntington’s disease CAG trinucleotide repeats in pathologically confirmed post-mortem brains. Neurobiology of Disease, 1, 159–66. Persichetti, F., Trettel, F., Huang, C. C. et al. (1999). Mutant huntingtin forms in vivo complexes with distinct contextdependent conformations of the polyglutamine segment. Neurobiology of Disease, 6, 364–75. Perutz, M. F., Johnson, T., Suzuki, M. & Finch, T. (1994). Glutamine repeats as polar zippers: Their possible role in inherited neurodegenerative diseases. Proceedings of the National Academy of Sciences of the United States of America, 91, 5355–8. Portera-Cailliau, C., Hedreen, J. C., Price, D. L. & Koliatsos, V. E. (1995). Evidence for apoptotic cell death in Huntington disease and excitotoxic animal models. The Journal of Neuroscience, 15, 3775–87. Potter, N. T., Meyer, M. A., Zimmerman, A. W., Eisenstadt, M. L. & Anderson, I. J. (1995). Molecular and clinical findings in a family with dentatorubral–pallidoluysian atrophy. Annals of Neurology, 37, 273–7. Rajkowska, G., Selemon, L. D. & Goldman-Rakic, P. S. (1998). Neuronal and glial somal size in the prefrontal cortex. A postmortem morphometric study of schizophrenia and Huntington disease. Archives of General Psychiatry, 55, 215–24. Rebeiz, J. J., Kolodny, E. H. & Richardson, E. P., Jr. (1968). Corticodentatonigral degeneration with neuronal achromasia. Archives of Neurology, 18, 20–33. Reddy, P. H., Williams, M., Charles, V. et al. (1998). Behavioural abnormalities and selective neuronal loss in HD transgenic mice expressing mutated full-length HD cDNA. Nature Genetics, 20 October, 198–202. Reiner, A., Albin, R. L., Anderson, K. D., D’Amato, C. J., Penney, J. B. & Young, A. B. (1988). Differential loss of striatal projection neurons in Huntington disease. Proceedings of the National Academy of Sciences of the United States of America, 85, 5733–7.

399

400


Reyes, M. G. & Gibbons, S. (1985). Dementia of the Alzheimer’s type and Huntington’s disease. Neurology, 35, 273–7. Richardson, E. P., Jr. (1990). Huntington’s disease: some recent neuropathological studies. Neuropathology and Applied Neurobiology, 16, 451–60. Richfield, E. K., Maguire-Zeiss, K. A., Cox, C., Gilmore, J. & Voorn, P. (1995). Reduced expression of preproenkephalin in striatal neurons from Huntington’s disease patients. Annals of Neurology, 37, 335–43. Riley, D. E., Lang, A. E., Lewis, A. et al. (1990). Cortical-basal ganglionic degeneration. Neurology, 40, 1203–12. Rinne, J. O., Daniel, S. E., Scaravilli, F., Harding, A. E. & Marsden, C. D. (1994a). Nigral degeneration in neuroacanthocytosis. Neurology, 44, 1629–32. Rinne, J. O., Daniel, S. E., Scaravilli, F., Pires, M., Harding, A. E. & Marsden, C. D. (1994b). The neuropathological features of neuroacanthocytosis. Movement Disorders, 9, 297–304. Robitaille, Y., Lopes-Cendes, I., Becher, M., Rouleau, G. & Clark, A. W. (1997). The neuropathology of CAG repeat diseases: review and update of genetic and molecular features. Brain Pathology, 7, 901–26. Rodda, R. A. (1981). Cerebellar atrophy in Huntington’s disease. Journal of Neurological Sciences, 50, 147–57. Roos, R. A. C., Pruyt, J. F. M., de Vries, J. & Bots, G. T. A. M. (1985). Neuronal distribution in the putamen in Huntington’s disease. Journal of Neurology, Neurosurgery, and Psychiatry, 48, 422–5. Rosas, H. D., Goodman, J., Chen, Y. I. et al. (2001). Striatal volume loss in HD as measured by MRI and the influence of CAG repeat. Neurology, 57, 1025–8. Rosas, H. D., Liu, A. K., Hersch, S. et al. (2002). Regional and progressive thinning of the cortical ribbon in Huntington’s disease. Neurology, 58, 695–701. Rosenblatt, A., Ranen, N. G., Rubinsztein, D. C. et al. (1998). Patients with features similar to Huntington’s disease, without CAG expansion in huntingtin. Neurology, 51, 215–20. Rubinsztein, D. C., Leggo, J., Coles, R. et al. (1996). Phenotypic characterization of individuals with 30–40 CAG repeats in the Huntington disease (HD) gene reveals HD cases with 36 repeats and apparently normal elderly individuals with 36–39 repeats. The American Journal of Human Genetics, 59, 16–22. Rubio, J. P., Danek, A., Stone, C. et al. (1997). Chorea-acanthocytosis: Genetic linkage to chromosome 9q21. The American Journal of Human Genetics, 61, 899–908. Sakai, T., Iwashita, H. & Kakugawa, M. (1985). Neuroacanthocytosis syndrome and choreoacanthocytosis (Levine–Critchley syndrome). Neurology, 35, 1679. Sapp, E., Ge, P., Aizawa, H. et al. (1995). Evidence for a preferential loss of enkephalin immunoreactivity in the external globus pallidus in low grade Huntington’s disease using high resolution image analysis. Neuroscience, 64, 397–404. Sapp, E., Schwarz, C., Chase, K. et al. (1997). Huntingtin localization in brains of normal and Huntington’s disease patients. Annals of Neurology, 42, 604–12. Sapp, E., Kegel, K. B., Aronin, N. et al. (1999a). Microglia accumulate in the HD striatum and cortex. Society of Neuroscience, 25, 829.

Sapp, E., Penney, J., Young, A. B., Aronin, N., Vonsattel, J.-P. & DiFiglia, M. (1999b). Axonal transport of N-terminal huntingtin suggests early pathology of corticostriatal projections in Huntington disease. Journal of Neuropathology and Experimental Neurology, 58, 165–73. Sathasivam, K., Hobbs, C., Turmaine, M. et al. (1999). Formation of polyglutamine inclusions in non-CNS tissue. Human Molecular Genetics, 8, 813–22. Saudou, F., Finkbeiner, S., Devys, D. & Greenberg, M. E. (1998). Huntingtin acts in the nucleus to induce apoptosis but death does not correlate with the formation of intranuclear inclusions. Cell, 95, 55–66. Sax, D. S. & Vonsattel, J.-P. (1992). Case Records of the Massachusetts General Hospital. Case 2–1992. Chorea and progressive dementia in an 88-year-old woman. The New England Journal of Medicine, 326, 117–25. Scherzinger, E., Lurz, R., Turmaine, M. et al. (1997). Huntingtinencoded polyglutamine expansions form amyloid-like protein aggregates in vitro and in vivo. Cell, 90, 549–58. Schroeder, K. (1931). Zur Klinik und Pathologie der Huntingtonschen Krankheit. Journal fur ¨ Psychologie und Neurologie, 43, 183–201. Schulz, J. B. & Dichgans, J. (1999). Molecular pathogenesis of movement disorders: are protein aggregates a common link in neuronal degeneration? Current Opinion in Neurology, 12, 433–9. Selemon, L. D., Rajkowska, G. & Goldman-Rakic, P. S. (1995). Abnormally high neuronal density in the schizophrenic cortex. Archives of General Psychiatry, 52, 805–18. (1998). Elevated neuronal density in prefrontal area 46 in brains from schizophrenic patients: Application of a threedimensional, stereologic counting method. The Journal of Comparative Neurology, 392, 402–12. Singer, C. (1992). ‘Haw River Syndrome’ or dentato-rubro-pallidoLuysian atrophy? Archives of Neurology, 49, 13–14. Sotrel, A., Paskevich, P. A., Kiely, D. K., Bird, E. D., Williams, R. S. & Myers, R. H. (1991). Morphometric analysis of the prefrontal cortex in Huntington’s disease. Neurology, 41, 1117– 23. Spatz, H. (1952). La maladie de Pick, les atrophies systématisées progressives et la sénescence cérébrale prématurée localisée. In The Proceedings of the First International Congress of Neuropathology, vol. 2. (Torino, Rosenberg and Sellier), pp. 375– 406. Spielmeyer, W. (1926). Die anatomische Krankheitsforschung am Beispiel einer Huntingtonschen Chorea mit Wilsonschem Symptomenbild. Zeitschrift fur ¨ die gesamte Neurologie und Psychiatrie (Berlin), 101, 701–28. Steffan, J. S., Bodai, L., Pallos, J. et al. (2001). Histone deacetylase inhibitors arrest polyglutamine-dependent neurodegeneratin in Drosophila. Nature, 413, 739–43. Telenius, H., Almqvist, E., Kremer, B. et al. (1995). Somatic mosaicism in sperm is associated with intergenerational (CAG)n changes in Huntington disease. Human Molecular Genetics, 4, 189–95.


Terplan, K. (1924). Zur pathologischen Anatomie der chronischen progressiven Chorea. Virchow’s Archive of Pathology and Anatomy (Berl), 252, 146–76. Tissot, R., Constantinidis, J. & Richard, J. (1975). La Maladie de Pick. Paris: Masson. ¨ Tokay, L. (1930). Studien uber die Chorea chronica und die Beziehung des Striatum zu dieser. Arbeiten aus dem Institut fur ¨ Anatomie und Physiologie des Centralnervensystems an der Wiener Universität, 32, 209–30. Tsuchiya, K., Oyanagi, S., Arima, K. et al. (1998). Dentatorubropallidoluysian atrophy: clinicopathological study of dementia and involvement of the nucleus basalis of Meynert in seven autopsy cases. Acta Neuropathologica, 96, 502–8. Ueno, S., Maruki, Y., Nakamura, M. et al. (2001). The gene encoding a newly discovered protein, chorein, is mutated in choreaacanthosis. Nature Genetics, 28, 121–2. Van Broeckhoven, C. L. (1995). Molecular genetics of Alzheimer disease: identification of genes and gene mutations. European Neurology, 35, 8–19. Vonsattel, J.-P. G. & DiFiglia, M. (1998). Huntington disease. Journal of Neuropathology and Experimental Neurology, 57, 369–84. Vonsattel, J.-P., Myers, R. H., Stevens, T. J., Ferrante, R. J., Bird, E. D. & Richardson, E. P., Jr. (1985). Neuropathological classification of Huntington’s disease. Journal of Neuropathology and Experimental Neurology, 44, 559–77. Vonsattel, J.-P., Myers, R. H., Bird, E. D., Ge, P. & Richardson, E. P., Jr. (1992). Maladie de Huntington: sept cas avec ˆılots néostriataux relativement préservés. Revue Neurologique, 148, 107–16. Wanker, E. E. (2000). Protein aggregation and pathogenesis of Huntington’s disease: mechanisms and correlations. Biological Chemistry, 381, 937–42. Weeks, R. A., Harding, A. E. & Brooks, D. J. (1996). Striatal D1 and D2 dopamine receptor loss in asymptomatic mutation carriers of Huntington’s disease. Annals of Neurology, 40, 49–54. ¨ Weisschedel, E. (1938). Uber eine systematische Atrophie der Oberen Olive. Archiv fur ¨ Psychiatrie und Nervenkrankheiten, 108, 219–27. ¨ (1939). Uber histopathologische Befunde an der oberen Olive und deren Beziehung zur Hörfunktion. Zeitschrift fur ¨ die gesamte Neurologie und Psychiatrie (Berlin), 165, 248–56.

Wexler, N. S., Young, A. B., Tanzi, R. E. et al. (1987). Homozygotes for Huntington’s disease. Nature, 326, 194–7. Wheeler, V. C., White, J. K., Gutekunst, C.-A. et al. (2000). Long glutamine tracts cause nuclear localization of a novel form of huntingtin in medium spiny striatal neurons in HdhQ92 and HdhQ111 knock-in mice. Human Molecular Genetics, 9, 503– 13. White, J. K., Auerbach, W., Duyao, M. P. et al. (1997). Huntingtin is required for neurogenesis and is not impaired by the Huntington’s disease CAG expansion. Nature Genetics, 17, 404–10. Witt, t. N., Danek, A., Reiter, M., Heim, M. U., Disrschinger, J. & Olsen, E. G. J. (1992). McLeod syndrome: a distinct form of neuroacanthocytosis. Report of two cases and literature review with emphasis on neuromuscular manifestations. Journal of Neurology, 239, 302–6. Yamada, M., Wood, J. D., Shimohata, T. et al. (2001). Widespread occurrence of intranuclear atrophin-1 accumulation in the central nervous system neurons of patients with dentatorubral-pallidoluisian atrophy. Annals of Neurology, 49, 14–23. Yamamoto, T., Hirose, G., Shimazaki, K., Takado, S., Kosoegawa, H. & Saeki, M. (1982). Movement disorders of familial neuroacanthocytosis syndrome. Archives of Neurology, 39, 298–301. Yeterian, E. & Van Hoesen, G. W. (1978). Cortico-striate projections in the rhesus monkey: the organization of certain corticocaudate connections. Brain Research, 139, 43–63. Young, A. B., Shoulson, I., Penney, J. B. et al. (1986). Huntington’s disease in Venezuela: neurologic features and functional decline. Neurology, 36, 244–9. Zalneraitis, E. L., Landis, D. M. D., Richardson, E. P., Jr. & Selkoe, D. J. (1981). A comparison of astrocytic structure in cerebral cortex and striatum in Huntington’s disease. Neurology, 31, 151. Zweig, R. M., Koven, S. J., Hedreen, J. C., Maestri, N. E., Kazazian, H. H., Jr. & Folstein, S. E. (1989). Linkage to the Huntington’s disease locus in a family with unusual clinical and pathological features. Annals of Neurology, 26, 78–84. Zweig, R. M., Ross, C. A., Hedreen, J. C. et al. (1992). Locus coeruleus involvement in Huntington’s disease. Archives of Neurology, 49, 152–6.

401

17 Human prion diseases James W. Ironside and Mark W. Head National Creutzfeldt–Jakob Disease Surveillance Unit, University of Edinburgh, Western General Hospital, UK

Introduction Prion diseases (also known as transmissible spongiform encephalopathies) are fatal neurodegenerative diseases occurring in a wide range of mammals, including humans (for review see Prusiner, 1998). The first of these diseases shown to be transmissible under experimental conditions was scrapie, a disease of sheep and goats, which is endemic in the UK, North America and many countries in Europe. Since then, much of the work to identify and characterize the transmissible agents responsible for this unique group of disorders has been performed on the scrapie agent. Prion diseases are unique amongst human neurodegenerative disorders in that they occur in sporadic, familial and acquired forms (Ironside, 1996; Prusiner, 1998). The clinical manifestations of human prion diseases frequently include dementia, but a wide range of other neurological abnormalities may also occur (Will et al., 1999). Accordingly, the range of neuropathological abnormalities in these diseases is broad, but four cardinal features occur: spongiform change (vacuolation of the grey matter and occasionally neurons), neuronal loss, gliosis and the accumulation of an abnormal isoform of a host encoded protein, the prion protein (Ironside, 1998). This abnormal protein can accumulate as amyloid plaques in some cases, but most examples of human prion diseases do not contain amyloid plaques and the accumulation of the abnormal form of the prion protein can only be detected by immunocytochemistry or Western blot techniques (see Appendices). The transmissible agent responsible for prion diseases is different from other conventional organisms associated with infectious diseases in humans and other species. Although the precise nature of the transmissible agent is not

fully understood, there is a large and increasing body of evidence to favour the prion hypothesis (Prusiner, 1982). In this hypothesis, the infectious agents are defined as prions, which are proteinaceous infectious particles that are devoid of nucleic acid and appear to be composed exclusively of a modified isoform of the prion protein, designated PrPSc . The normal cellular isoform of the prion protein (PrPC ) which is thought to be the precursor of PrPSc is encoded by a single gene termed PRNP (Prusiner, 1998).

The prion protein gene (PRNP) The human PRNP gene is located on the short arm of chromosome 20. The entire open reading frame is located within a single exon, which encodes 253 amino acid residues (Prusiner, 1998). Nascent PrPC undergoes cleavage to form a molecule comprising residues 23–231. Over 20 pathogenic mutations have been identified in PRNP, including point mutations (which mostly occur in the Cterminal region of the PRNP) and insertional mutations, comprising 24 base pair (bp) repeats which are inserted in the octapeptide repeat region of the protein between residues 51 and 91 (Fig. 17.1). In addition, almost 20 nonpathogenic polymorphic sites have been identified including nonsense, missense, and silent mutations. The polymorphisms at codon 129 (methionine/valine) and codon 219 (glutamic acid/lysine) are important in determining disease susceptibility (Windl et al., 1996); there is a considerable body of evidence for the role codon 129 plays in sporadic and variant CJD (see Table 17.1) (Will et al., 1996; Alperovitch et al., 1999). Codons 129 and 219 may also influence the clinical and pathological phenotype of human


402

Human prion diseases

Fig. 17.1. Schematic representation of the PRNP coding sequence showing the location of non-pathogenic polymorphisms above (green boxes) and disease-associated mutations below (red boxes). Codon positions identified with both types are shown with labels above and below, and red boxes. The clinical disorders occurring in patients with disease-associated mutations are summarized in Table 17.7.

Table 17.1. PRNP codon 129 genotypes in sporadic and variant CJD PRNP codon 129 genotype (%)

MM

MV

VV

Normal population Sporadic CJD Variant CJD

37 74 100

51 15 0

12 11 0

prion diseases (see below). A number of silent mutations have been identified including codons 117 and 124. The locations of the polymorphic sites in the PRNP are depicted in Fig. 17.1.

The cellular prion protein, PrPC The cellular prion protein (PrPC ) is a GPI-anchored cell surface glycoprotein (for review see Prusiner, 1998). It is expressed in a wide variety of tissues but it is expressed at particularly high levels in CNS neurons, where it is found at the pre-synaptic terminal (Herms et al., 1999). While the function of PrPC is not known with certainty it has been implicated in copper-related oxidative stress resistance (see, for example, Herms et al., 1999; Brown et al., 2001; Klampt et al., 2001). PrPC has a structured C-terminus (comprising three regions of -helix and two short anti-parallel -pleated sheet regions) and a largely unstructured Nterminus. This N-terminal sequence contains the copperbinding octarepeat region and a further high affinity copper binding site has also been recently described downstream. PrP can exist in three possible glycoforms (di-, mono-, and

non-glycosylated PrP) resulting from variable occupancy of the two asparagine-linked C-terminal glycosylation sites. The methionine/valine polymorphism at codon 129 of the PRNP lies close to the boundary between the structured and unstructured regions (for review see Prusiner, 1998).

The disease-associated prion protein, PrPSc PrPC does not appear to be essential for normal neurological development and function in PRNP knock-out mice, although there is some evidence of occult oxidative stress, though no frank disease (Klampt et al., 2001). The absence of PrPC in PRNP knock-out mice does, however, render them resistant to infection with transmissible spongiform encephalopathies (for review see Telling, 2000) indicating a key role for PrPC in TSE pathogenesis. Accordingly the transmissible spongiform encephalopathies including the human prion diseases are uniquely characterised by the presence of an altered form of the prion protein, termed PrPSc , which accumulates in the brain during the degenerative process. While PrPC is a soluble, protease-sensitive protein dominated by -helical structure, PrPSc has a much-increased -sheet content, is partially resistant to proteolytic digestion and exists in highly aggregated (insoluble) forms. The prion hypothesis equates PrPSc with the infectious agent in TSE and proposes that PrPSc , once present, results in a self-propagating wave of conformational conversion of endogenous PrPC to PrPSc (for review see Prusiner, 1998). How this process might result in the phenotypic diversity of human prion diseases is a subject of intensive debate. One possible explanation is that the

403

404

J. W. Ironside and M. W. Head

phenotypic aspects of human prion diseases are determined at least in part by different physico-chemical isoforms of PrPSc that can replicate with a degree of fidelity. Alternatively PrPSc formation may be viewed as a host response to an as yet uncharacterized agent. In either event the presence of PrPSc and the range of biochemical forms in which it occurs have proved diagnostically useful.

Structural variation in PrPSc Potential sources of structural variation in PrPSc that have been proposed to correlate with disease phenotype include: (i) The possible presence of mutations in the coding region of the PRNP gene (for review see Prusiner, 1998). (ii) The presence of either methionine (M) or valine (V) at the polymorphic residue 129 (Gambetti et al., 1995; Collinge et al., 1996; Parchi et al., 1996). (iii) The conformation, most commonly determined by the extent of experimental proteolytic N-terminal truncation in the region spanning residues 74–102 (Collinge et al., 1996; Parchi et al., 1996; Telling et al., 1996). (iv) The occupancy of the two potential glycosylation sites at asparagine 181 and 197 (Collinge et al., 1996; Parchi et al., 1997, 1999b). (v) The composition and structural diversity of the attached glycans (Pan et al., 2001). (vi) The presence and identity of bound divalent cations at two different sites (Wadsworth et al., 1999; Jackson et al., 2001; Wong et al., 2001).

glycine 82 whereas the N-terminus of type 2 is predominantly at serine 97. However, these two types are actually populations of PrP molecules with ragged ends and the dispersion of these ragged ends around glycine 82 and serine 97 is accentuated by the presence of a valine residue at codon 129. Hence codon 129 and PrPRes isotype are not fully independent variables. A further distinction has been made on the basis of the occupancy of the two potential glycosylation sites in PrPRes with two broad categories: those where the diglycosylated form predominates being designated B, and those where the diglycosylated form is not the most abundant being designated A.

PrPRes isotypes

Human prion diseases – classification

In practice only the first four of these are currently used to classify PrPSc in a diagnostic setting and, even within these, points three and four remain controversial. What is becoming clear is that human PrPSc occurs in at least two major variants that are distinguished by their electrophoretic mobility on Western blot analysis after limited proteolysis with proteinase K (Fig. 17.2). The term PrPRes is reserved for PrPSc defined operationally by this resistance to proteolytic degradation. The nonglycosylated fragment of type 1 PrPRes has a molecular weight of approximately 21 kDa, whereas the non-glycosylated fragment of Type 2 PrP Res is smaller having a molecular weight of approximately 19k Da (Parchi et al., 1997). These mobilities on Western analysis have recently been shown to correspond to differing degrees of Nterminal truncation (Parchi et al., 2000). After proteolytic degradation the N-terminus of type 1 is predominantly at

The current classification of human prion diseases is summarized in Table 17.2. Creutzfeldt–Jakob disease (CJD) was first described in the 1920s (Creutzfeldt, 1920; Jakob, 1921) but review of Creutzfeldt’s original case has indicated that it was probably not an example of what we would now recognize as a human prion disease and may represent a form of metabolic encephalopathy (Richardson & Masters, 1995). Among Jakob’s patients was a family with an inherited form of this disease, which was subsequently shown to be due to a pathogenic mutation at codon 178 in the (PRNP) (Brown et al., 1994a). In the next decade, Gerstmann, Sträussler and Scheinker described another family with an inherited form of spinocerebellar ataxia (Gerstmann et al., 1936), which was subsequently shown to be due to a proline to leucine mutation at codon 102 in the PRNP (Hsiao et al., 1989). Since then, an ever-increasing number of pathogenic

Fig. 17.2. Western blot analysis of the PrPRes isotype of post-mortem cerebral cortex from cases of sporadic (s) and variant (v) Creutzfeldt–Jakob disease (CJD). The codon 129 genotype is shown as methionine (M) or valine (V). The non-glycosylated (lowest) band is classified as type 1 (∼21 kDa) or type 2 (∼19 kDa). The diglycosylated (top) band predominates (*) in cases of vCJD and this pattern is classified as type 2B.


Table 17.2. Classification of human prion diseases Idiopathic:

Sporadic CJD Sporadic fatal insomnia

Inherited:

Familial CJD GSS (classical and variant forms) Fatal familial insomnia

Acquired:

Human source: Bovine source:

Kuru Iatrogenic CJD Variant CJD

1994b; Will et al., 1998). However, it has been observed that individuals who are homozygotes at codon 129 in the PRNP are over-represented in patients with sCJD (Palmer et al., 1991), iatrogenic CJD (Collinge et al., 1991) and vCJD (Will et al., 1996) when compared with the relevant normal population (see Table 17.1). The mechanism for this genetic predisposition to human prion disease is uncertain, but it may relate to the relative efficiency at which conversion from PrPC to PrPSc can occur.

Human prion diseases – neuropathology mutations and insertions in the prion protein gene have been reported (for review see Gambetti et al., 1999); these are associated with a wide range of clinical disease phenotypes which are described below. In 1957, Gajdusek and Zigas described kuru, an unusual neurodegenerative disorder that was one of the commonest causes of death in the Fore tribe in Papua New Guinea. The histological similarity between kuru and scrapie was noted by William Hadlow, who suggested that experiments should be performed to see if kuru was a transmissible disorder (Hadlow, 1959). Experimental transmission of kuru was achieved in 1966 (Gajdusek et al., 1966), followed by experimental transmission of sporadic CJD (sCJD) in 1968 (Gibbs et al., 1968) and the Gerstmann–Sträussler–Scheinker syndrome in 1981 (Masters et al., 1981). In 1974, the first case of iatrogenic CJD was reported in a corneal transplant recipient in whom a cornea from a patient who had subsequently died from CJD had been implanted (Duffy et al., 1974). Since then, increasing numbers of patients with iatrogenic forms of CJD have been recorded, most of which are associated with dura mater grafts and treatment with human pituitary hormones (Brown et al., 2000). In 1996, a new variant form of CJD (vCJD) was reported in the UK by the National CJD Surveillance Unit (Will et al., 1996). This novel prion disease had a unique set of clinical and pathological features, and it was suggested on the basis of epidemiological evidence that this disorder was likely to be linked to the epidemic of bovine spongiform encephalopathy (BSE) that had occurred in the UK since 1986. Subsequent biochemical (Collinge et al., 1996) and experimental transmission studies (Bruce et al., 1997; Scott et al., 1999) have indicated that the transmissible agents responsible for vCJD and BSE are identical, indicating that vCJD is the first example of a zoonotic form of prion disease in humans. The commonest human prion disease is sCJD, for which the cause is unknown. Detailed epidemiological studies over several decades in many countries have failed to identify any consistent environmental risk factors (Brown et al.,

All forms of human prion disease share a similar spectrum of neuropathological abnormalities, which have for many years been the mainstay of diagnosis. Of these, the most consistent abnormality is spongiform change (Masters & Richardson, 1978), defined as diffuse or focally clustered (morula-type) small discrete round or oval vacuoles in the neuropil of the whole depth or deep layers of the cerebral cortex (Budka et al., 1995) (Fig. 17.3), in the cerebellar cortex (predominantly in the molecular layer) or subcortical grey matter. The vacuoles may become confluent to form larger cyst-like cavities. A range of other unrelated disorders can produce vacuolation in the brain, which may be confused with ‘true’ spongiform change (see Table 17.3 and Fig. 17.3). With experience, these can be distinguished by their distribution within the brain and the absence of the small rounded neuropil vacuoles that are characteristic of human prion diseases. The pathogenetic mechanisms behind spongiform change are uncertain, but ultrastructural studies have revealed that the spongiform vacuolation within the grey matter is derived from dilatation of cell processes (probably dendrites) with swelling of organelles, including the endoplasmic reticulum and Golgi apparatus. This suggests an underlying abnormality of the plasma membrane, resulting in abnormal permeability to both water and electrolytes (DeArmond & Ironside, 1999). Since PrPC is linked to the plasma membrane by a GPI anchor, it is perhaps not difficult to imagine that conversion to PrPSc might disturb the structure and function of the plasma membrane, thereby facilitating the development of spongiform change. Similarly, the mechanisms of neuronal death in prion diseases remain poorly understood, and experimental models have suggested a number of potential mechanisms for neuronal toxicity. These include a direct toxic effect of PrPSc (or a fragment of PrPSc ) on neurons, an indirect action via microglial cells, causing cytokine release and subsequent neuronal death, or interference with a normal cytoprotective mechanism which deals with oxidative stress (DeArmond & Ironside 1999; Giese & Kretzschmar 2001).

405

(a)

(b)

(c)

(d)

(e)

(f )

(g)

(h)

Fig. 17.3. (a) Non-specific artefactual vacuolation in ischaemic cerebral cortex. The neurones are shrunken and there is marked retraction of the grey matter around these shrunken cells and blood vessels. Haematoxylin and eosin. (b) Status spongiosus in sporadic CJD. Widespread neuronal loss and extensive gliosis have resulted in collapse of the cortical cytoarchitecture, with coarse vacuolation in this case of long standing sporadic CJD. Haematoxylin and eosin. (c) Sporadic CJD MM1. A focus of confluent spongiform change is present within the occipital cortex (centre). The vacuoles are rounded and in the centre of the lesion appear to coalesce to form cyst-like spaces. Haematoxylin and eosin. (d ) Sporadic CJD MM1. Immunocytochemistry shows marked perivacuolar deposition of PrPRes , with smaller deposits present elsewhere in the neuropil. (e) Sporadic CJD MM2. Widespread microvacuolar spongiform change is present within the occipital cortex, with neuronal loss and gliosis. Haematoxylin and eosin. ( f ) Sporadic CJD MM2. Immunocytochemistry in the occipital cortex shows a synaptic (reticular) pattern of PrPRes accumulation. (g) Sporadic CJD VV1. The frontal cortex shows spongiform change involving all cortical layers with neuronal loss and gliosis, but no confluent spongiform change is present. Haematoxylin and eosin. (h) Sporadic CJD VV1. Immunocytochemistry shows a synaptic (reticular) pattern of PrP accumulation with focal random accentuation.


Table 17.3. Conditions resembling spongiform change: apart from status spongiosis in CJD, none are associated with the accumulation of PrPRes in the brain Disorder Status spongiosis

Cerebral oedema Ischaemia Alzheimer’s disease, Dementia with Lewy bodies Pick’s disease, dementia lacking specific histological features Metabolic encephalopathies Toxic encephalopathies Fixation/processing artefact

Differentiation from ‘true’ spongiform change Cystic cavities of irregular size associated with severe neuronal loss, gliosis and collapse of the cortical cytoarchitecture. This can occur in CJD and a wide range of other neurodegenerative disorders Vacuolation also occurs in the white matter Vacuolation usually restricted to the margins of infarcts, possibly due to oedema Very similar to spongiform change in prion diseases, but restricted to the temporal cortex Superficial coarse vacuolation, often in layer 2 of the cerebral cortex

Vacuolation may be exclusively intraneuronal due to accumulation of storage products Vacuolation may occur only in the white matter, e.g. with cuprizone or triethyl tin Coarse vacuolation which is usually present around neurons and blood vessels; fine vacuolation within the neuropil of the grey matter is absent

One of the most striking aspects of the pathology of prion diseases is the distribution of the pathological changes within the central nervous system. The pathological abnormalities do not uniformly occur within the brain, but instead are concentrated at various sites, a concept known as neuropathological targeting (Bell & Ironside, 1993). The basis of this precise targeting is also uncertain, but it appears to relate to interactions between the infectious agent and the host, and may in part be explained by differences in the configuration or glycosylation of PrPC within different neuronal subpopulations (DeArmond et al., 1999). Amyloid plaques formed from PrPSc occur in a variety of human prion diseases including sporadic, familial and acquired forms (Figs. 17.4–17.7) (Ironside, 1998; Gambetti et al., 1999). The presence of these striking lesions also appears to reflect an interaction between the infectious agent and the host, where host genetics appear particularly important (Gambetti et al., 1999). The neuropathological features of the major forms of human prion disease will be discussed below, accompanied by a summary of the principal clinical and genetic features and the biochemical pathology.

usually include electroencephalography (EEG), which is particularly useful in the diagnosis of sCJD, magnetic resonance imaging (MRI) of the brain, and analysis of various brain-associated proteins in the cerebrospinal fluid (CSF) (Will et al., 1998, 1999) (see Table 17.4). Analysis of the PRNP is usually performed in patients with suspected prion disorders in order to investigate the possibility of a diseaseassociated mutation or insertion, and to characterize the polymorphic alleles at codon 129 and 219. Neuropathological examination of the brain is required to confirm a clinical diagnosis of human prion disease (Budka et al., 1995). This is usually performed after autopsy, but biopsy of the brain in patients with suspected prion disorders is occasionally performed. However, this investigation is carried out less frequently nowadays, largely as a result of better-defined clinical diagnostic criteria for human prion diseases. Brain biopsy is now reserved for patients in whom the possibility of an underlying treatable condition is still being considered. Since there are clear health and safety implications for those performing the biopsy, as well as for the handling of the neurosurgical instruments subsequently if the patient is confirmed to have a human prion disease (see Chapter 5), this investigation is not a routine procedure in most countries.

Investigation of human prion diseases Human prion diseases present as progressive neurodegenerative disorders with a wide range of neurological signs and symptoms throughout the course of the illness. Accordingly, most patients are referred to clinical neurologists for investigation and diagnosis. The clinical investigations

Sporadic CJD: clinical features Sporadic Creutzfeldt–Jakob disease (sCJD) is the commonest form of human prion disease, occurring in a worldwide distribution with an incidence of around 1 per million of the

407

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

Fig. 17.4. (a) Sporadic CJD VV2. The temporal cortex shows widespread spongiform change which is predominantly microvacuolar, but some larger cyst-like spaces are also present. Haematoxylin and eosin. (b) Sporadic CJD VV2. Immunocytochemistry shows a characteristic perineuronal pattern of PrPRes accumulation in large neurons in layer five of the cortex, with decoration of apical ascending dendrites. (c) Sporadic CJD MV1. The molecular layer in the cerebellum shows patchy spongiform change, with focal loss of Purkinje cells and granule cells and accompanying astrocytosis. No amyloid plaques are present. Haematoxylin and eosin. (d ) Sporadic CJD MV1. Immunocytochemistry shows synaptic (reticular) accumulation of PrP in the molecular layer with a coarser pattern of deposition in the granular layer. Purkinje cells are unstained and no plaques are identified. (e) Sporadic CJD MV2. A kuru-type plaque is present in the granular layer, composed of a central eosinophilic core and a rounded fibrillary periphery. There is little evidence of spongiform change. Haematoxylin and eosin. ( f ) Sporadic CJD MV2. Immunocytochemistry shows strong staining of the kuru-type plaques in the granular layer, and also reveals smaller plaque-like deposits that are not evident on routine stains. (g) Sporadic fatal insomnia (thalamic variant of sporadic CJD). Immunocytochemistry for glial fibrillary acidic protein shows increased numbers of hypertrophied astrocytes within the dorsomedial thalamic nucleus. (h) Fatal familial insomnia. Immunocytochemistry for GFAP shows increased numbers of reactive astrocytes in the anteromedial thalamic nucleus. There is no spongiform change present.

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

Fig. 17.5. (a) Familial CJD E200K. There is widespread spongiform change in the molecular layer of the cerebellum with a marked loss of granule cells and Purkinje cells and accompanying astrocytosis. Haematoxylin and eosin. (b) Familial CJD E200K. Immunocytochemistry shows a composite synaptic (reticular) pattern of PrPRes accumulation with small plaque-like deposits present in both the molecular and granular layer. (c) GSS P102L. Large multicentric amyloid plaques are present in both the molecular and granular layers of the cerebellum, with mild spongiform change in the molecular layer. Haematoxylin and eosin. (d ) Immunocytochemistry shows strong staining for PrPRes in the large multicentric plaques, and also demonstrates multiple smaller plaque-like deposits in the cerebellum. (e) GSS A117V. Several pale-staining amyloid plaques are present in the molecular layer of the cerebellum, associated with mild spongiform change and astrocytosis. Haematoxylin and eosin. ( f ) GSS A117V. Immunocytochemistry for PrPRes shows strong staining of the plaques in the molecular layer, but no deposits are identified in the granular layer of the cerebellum. (g) GSS D202N. Numerous large rounded amyloid plaques are present in the cerebellar molecular layer. There is extensive loss of Purkinje cells and accompanying astrocytosis. Haematoxylin and eosin. (h) GSS D202N. Immunocytochemistry for PrPRes shows strong staining of the rounded amyloid plaques in the molecular layer, with intense labelling at the plaque periphery. The granular layer is unstained.

(a)

(b)

(c)

(d)

(e)

(f )

(g)

(h)

Fig. 17.6. (a) Familial CJD 144 bp insertion. There is moderate spongiform change in the molecular layer of the cerebellum with marked astrocytosis and focal loss of Purkinje cells and granule cells. Haematoxylin and eosin. (b) Familial CJD 144 bp insertion. Immunocytochemistry for PrPRes shows unusual linear deposits in the molecular layer, with faint synaptic positivity elsewhere in the molecular layer and the granular layer. (c) Kuru. Several small rounded amyloid plaques are present in the granular layer of the cerebellum. There is moderate spongiform change in the molecular layer and focal loss of Purkinje cells and granule cells, with accompanying astrocytosis. Haematoxylin and eosin. (d ) Kuru. Immunocytochemistry for PrPRes shows strong staining of the plaques in the granular layer and also reveals a plaque-like deposit in the molecular layer. (e) Iatrogenic CJD, dura mater graft recipient. The cerebellum shows mild spongiform change and astrocytosis in the molecular layer, but no amyloid plaques are identified. Haematoxylin and eosin. ( f ) Iatrogenic CJD, dura mater graft recipient. Immunocytochemistry for PrPRes shows widespread synaptic (reticular) positivity in the molecular layer with coarser positivity in the granular layer and occasional rounded plaque-like deposits. (g) Iatrogenic CJD, growth hormone recipient. The cerebellum exhibits mild spongiform change in the molecular layer, but occasional small kuru-type plaques are present in the granular layer, with focal loss of Purkinje cells. Haematoxylin and eosin. (h) Iatrogenic CJD, growth hormone recipient. Immunocytochemistry for PrPRes shows strong staining of the amyloid plaques and in addition reveals a coarse pattern of plaque-like positivity in the granular layer with synaptic (reticular) positivity in the molecular layer.

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

Fig. 17.7. (a) Variant CJD. A large florid plaque is present in the occipital cortex, comprising a central eosinophilic core with a pale fibrillary periphery and surrounding spongiform change. Haematoxylin and eosin. (b) Variant CJD. Immunocytochemistry for PrPRes in the cerebral cortex shows strong staining of the large florid plaques, and also demonstrates a smaller cluster of plaques and amorphous PrPRes deposits around small neurons and astrocytes. (c) Variant CJD. Numerous amyloid plaques are present in both the granular layer and the molecular layer, with patchy spongiform change in the surrounding grey matter of the molecular layer. There is marked loss of Purkinje cells. Haematoxylin and eosin. (d ) Variant CJD. Immunocytochemistry for PrPRes in the cerebellum shows strong staining of the amyloid plaques in the granular and molecular layer, with extensive amorphous accumulation of PrPRes around small neurons in the molecular layer. (e) Variant CJD. Extensive confluent spongiform change is present within the caudate nucleus. No amyloid plaques are identified. Haematoxylin and eosin. ( f ) Variant CJD. Immunocytochemistry for PrPRes shows perineuronal labelling and decoration of dendritic processes in a linear pattern. No florid plaques are identified. (g) Variant CJD. Immunocytochemistry for PrPRes in the tonsil shows strong staining of follicular dendritic cells and macrophages within a germinal centre. (h) Variant CJD. Immunocytochemistry for PrPRes in a dorsal root ganglion shows varying positivity in ganglion cells with intense positivity in the surrounding satellite cells.

412


Table 17.4. Clinical diagnostic criteria for sporadic and iatrogenic CJD and familial prion diseases (from www.cjd.ed.ac.uk) 1.

SPORADIC I Rapidly progressive dementia II A Myoclonus B Visual or cerebellar problems C Pyramidal or extrapyramidal features Akinetic mutism III Typical EEG 1.1 Definite: Neuropathological/ immunocytochemically confirmed 1.2 Probable: I + 2 of II + III or Possible + positive 14-3-3 1.3 Possible: I + 2 of II + duration < 2 years

2.

IATROGENIC CJD 2.1 Definite: Definite CJD with a recognized risk 2.2 Probable: 2.2.1 Progressive cerebellar syndrome in human pituitary hormone recipients 2.2.2 Probable CJD with recognized risk

3.

FAMILIAL PRION DISEASESa 3.1 Definite Definite TSE plus definite or probable TSE in a first-degree relative 3.2 Probable 3.2.1 Probable TSE plus definite or probable TSE in a first-degree relative 3.2.2 Progressive neuropsychiatric disorder plus disease-specific mutation

a

Including GSS and FFI

population per annum (Will et al., 1998). As its name implies, the cause of this disorder is unknown, and extensive epidemiological studies over several decades in different countries have yielded no consistent findings. Analysis of the PRNP has shown a significant excess of homozygotes at codon 129 in comparison with the normal population (see Table 17.1) (Alperovitch et al., 1999). The clinical and pathological features of sCJD are variable, but clinical and pathological diagnostic criteria have been established and evaluated in prospective surveillance projects in Europe (Tables 17.4 and 17.5) (Will et al., 1998). Most cases of sCJD occur in the elderly, with a peak incidence in the seventh decade of life; most major series show no significant differences in the incidence of the disease in males and females (Brown et al., 1994b; Will et al., 1998). A wide age range has been reported in sCJD ranging from the (very rare) cases identified in teenagers to cases in

the tenth decade of life (see Will et al., 1999 for review). As the diagnostic clinical criteria imply, the commonest clinical presentation for sCJD is rapidly progressive dementia with movement disorders (particularly myoclonus, pyramidal and extrapyramidal signs and visual abnormalities). However there is a wide spectrum of clinical features and the historical literature on sCJD is full of eponymous syndromes used to describe particular clinical variants, e.g. the Heidenhain variant with marked visual disturbances, or the Brownell–Oppenheimer variant with predominant cerebellar features (Will et al., 1999). Reviews of large series of cases have indicated that there is a significant minority of sCJD cases which present with features other than rapidly progressive dementia, particularly cerebellar signs and symptoms, (Brown et al., 1994b) and unusual subacute presentations resembling stroke have also been described (McNaughton & Will, 1997). There appears to be a relationship between the age of onset of sCJD and the codon 129 genotype, with valine homozygotes tending to have a significantly younger age of onset than other genotypes (Alperovitch et al., 1999). The relationship between PRNP genotype and clinical and pathological phenotype is discussed more fully below. The clinical investigation of sCJD cases usually includes EEG, which in the majority of cases will show a characteristic pattern of triphasic synchronous periodic discharges (Bortone et al., 1994). MRI scans have shown signal abnormalities in the basal ganglia (particularly in the caudate nucleus and putamen) in most cases of sCJD (Finkenstaedt et al., 1996), which may be of diagnostic significance (Schroter et al., 2000). A number of brain-associated Table 17.5. Pathological diagnostic criteria for CJD, GSS, FFI and Kuru (Budka et al., 1995) 1 CREUTZFELDT–JAKOB DISEASE: sporadic, iatrogenic (recognized risk) or familial (with the same disease in first-degree relative or a pathogenic PRNP mutation): A Spongiform encephalopathy in cerebral and/or cerebellar cortex and/or subcortical grey matter; and/or: B Encephalopathy with PrPRes immunoreactivity (plaque and/or diffuse synaptic and/or patchy/perivacuolar types) 2 GERSTMANN–STRAUSSLER–SCHEINKER SYNDROME (GSS) (in family with dominantly inherited progressive ataxia and/or dementia): Encephalo(myelo)pathy with muticentric PRPRes plaques 3 FATAL FAMILIAL INSOMNIA (FFI) (in family with PRNP178 mutation): Thalamic degeneration, with focal cerebral spongiform change 4 KURU (in the Fore population)


Table 17.6. CSF brain-associated proteins used in the diagnosis of human prion diseases Disorder Sporadic CJD

Variant CJD

CSF Protein

Sensitivity

Specificity

14-3-3 Neuron-specific enolase S-100b tau 14-3-3 Neuron-specific enolase S-100b tau

90–97% 80% 84%, 94% 100% 50% 52% 78% 80%

87–100% 92% 85%, 91% 95% 91% 86% 76% 94%

proteins have been reported to be increased in the cerebrospinal fluid (CSF) of patients with sCJD (see Table 17.6). Of these, only 14–3-3 has reliably been shown to have a high degree of both sensitivity and specificity for the diagnosis of sCJD (Hsich et al., 1996; Zerr et al., 1998). 14-3-3 belongs to a family of low molecular weight proteins that are found in high concentration within neuronal cells and are involved in signal transduction, cell cycle regulation, and apoptosis (Aitken et al., 1995). It is rarely detected in other forms of dementia such as Alzheimer’s disease or Lewy body disease, which makes it a useful test for discriminating sCJD from these conditions. Neuron-specific enolase, another neuronal protein, is also elevated in the CSF of patients with sCJD, but is less sensitive than 14-3-3 (Zerr et al., 1995).The astrocytic protein S-100b and the microtubular protein, tau are both sensitive markers for sCJD(Otto et al., 1997 a,b) but are also elevated in Alzheimer’s disease and other forms of dementia which limits their diagnostic potential (Green et al., 1997; Andreasen et al., 1998).

Sporadic CJD: neuropathology In many cases of sCJD there are no significant macroscopic abnormalities in the brain (Ironside, 1996). However, a significant minority of cases exhibit cerebral atrophy which may be either global or local, e.g. involving particularly the occipital lobe in some cases; or cerebellar atrophy may also be a prominent feature (Budka et al., 1995). In general, the longer the clinical illness, the greater the chance of cerebral atrophy being noted on macroscopic examination. There is a rare variant of sCJD named the panencephalopathic variant of CJD which has been reported predominantly in the Japanese literature (for review see Budka et al., 1995). In this rare entity, in addition to severe cerebral cortical and cerebellar atrophy there is marked shrinkage and degeneration of the white matter throughout the central nervous system. This is evident on the cut surfaces

of the cerebral hemispheres, but it remains uncertain as to whether this is a primary phenomenon or whether it is related to secondary axonal degeneration in patients with prolonged clinical illness. The microscopic features of sCJD comprise spongiform change, neuronal loss, gliosis and amyloid plaque formation (Ironside, 1996). However, these features are markedly variable both from one area of the brain to another within an individual case, and from one case to another (Bell & Ironside, 1993). Furthermore, these changes are accentuated when the distribution of the abnormal form of PrP is studied by immunocytochemistry (Bell et al., 1997; Ironside, 1998). The pathological phenotype in sCJD is profoundly influenced by the PRNP codon 129 genotype and by the PrPRes isotype detected on Western blotting. Recent studies of large series of sCJD cases have identified at least six variants which appeared to be consistent and share both clinical and pathological features, but are readily distinguishable on the basis of these findings (Parchi et al., 1999b). The major clinical and pathological features of these subgroups are summarized in Table 17.11, where the PRNP subgroups (MM, MV and VV) and the PrPRes isotype (Type 1 and Type 2) seem to be major determinants of disease phenotype (Parchi et al., 1996, 1999b). In MM and MV1 cases, there is widespread spongiform change, gliosis and neuronal loss in the cerebral cortex, basal ganglia, thalamus and cerebellum with sparing of the hippocampus and brainstem. Spongiform change tends to be most severe in the occipital cortex and is associated with synaptic and perivacuolar patterns of PrPRes deposition (Fig. 17.3) (Parchi et al., 1999b). The latter is particularly evident in areas of confluent spongiform change. In occasional cases, the spongiform change is so severe as to result in collapse of the cortical cytoarchitecture with severe neuronal loss and coarse vacuolation, which is termed status spongiosis (Fig. 17.3). Similar features are present in the MV1 subgroup, but severe occipital involvement and perivacuolar PrPRes accumulation are usually less pronounced (Fig. 17.4) (Parchi et al., 1999b). PrPRes amyloid plaques do not occur in these subgroups. The pathology in the MM2 subgroup is somewhat similar, but involvement of the cerebellum is less pronounced and there is usually perivacuolar and synaptic accumulation of PrPRes in the cerebral cortex and basal ganglia (Fig. 17.3) (Parchi et al., 1999b). A subtype of MM2 sCJD has a quite different pattern of pathology centred on the medial thalamus which exhibits severe neuronal loss and gliosis (Fig. 17.4). Similar changes are present in the inferior olivary nuclei, but spongiform change is limited to small foci in the cerebral cortex, entorhinal cortex, and occasionally the cerebellum (Parchi et al., 1999a). The pathological phenotype

413

414


of this ‘thalamic variant’ of sCJD is somewhat similar to fatal familial insomnia. Recent descriptions of this entity have suggested that the term Creutzfeldt–Jakob disease is no longer used for this disorder, for which the term sporadic fatal insomnia has been proposed on the basis of both the clinical and pathological features (see below) (Parchi et al., 1999a). In contrast, VV2 and MV2 subtypes show significant pathology in the limbic structures, the basal ganglia and brainstem nuclei. The cerebral cortical pathology is more restricted, and often exhibits a laminar distribution of spongiform change which is most evident in layers 5/6 (Fig. 17.4). Immunocytochemistry for PrPRes shows a strong synaptic pattern in a laminar distribution with decoration of large neurons in layer five and punctate staining around the apical ascending dendrites from these cells (Fig. 17.4). Small plaque-like structures are often identified in layer three, but amyloid plaques are not visible on routinely stained preparations in the VV2 subgroup (Parchi et al., 1999b). PrPRes amyloid plaques (kuru-type plaques) are a pathological hallmark of the MV2 subgroup, comprising the unicentric rounded fibrillary plaques which are most readily identified within the cerebellum (Fig. 17.4), but which can occur in other brain regions including the thalamus, basal ganglia and less frequently the cerebral cortex. These stain intensely on immunocytochemistry for PrPRes (Fig. 17.4) and can occasionally be identified occurring as linear deposits within the cerebellar granular layer. Previous studies had suggested that the presence of fibrillary PrPRes amyloid plaques in sCJD was related to the length of the clinical illness (Brown et al., 1994b), but this observation has not been borne out in more recent series and there is no doubt that the PRNP codon 129 genotype and PrPRes isotype are far more important in influencing disease phenotypes. The rarest subgroup of sCJD is the VV1 subgroup, in which there is predominant involvement of the cerebral cortex and basal ganglia, with relative sparing of the thalamus, cerebellum, hippocampus and brainstem (Parchi et al., 1999b). Within the cerebral cortex the occipital lobe is often relatively spared and there is widespread synaptic positivity for PrPRes in all cortical regions (Fig. 17.3). Spongiform change is often most evident in the deeper layers of the cerebral cortex (Fig. 17.3) and occasional ballooned neurons have been identified particularly on immunocytochemistry for neurofilament protein and B-crystallin. Although this approach to the sub-classification of sCJD has much to commend it, there have been a number of cases reported which emphasize the limits of this approach in our understanding of the disease. In particular, the existence of more than one PrPRes isotype has been demonstrated in several cases of sCJD, with one recent

series suggesting that as many as 33% of cases can contain more than one PrPRes isotype (Puoti et al., 1999). The significance of this in influencing the clinical and pathological phenotype is uncertain, although it has been claimed that the pattern of PrPRes distribution as detected on immunocytochemistry is profoundly influenced by changes in PrPRes isotype. Other isolated case reports have demonstrated findings which appear not to conform to this general classification, and one recent intriguing case has been reported where there was a change in the PrPRes isotype at autopsy compared with brain biopsy (Head et al., 2001), again emphasizing the limits on our understanding of the role of PrP in determining disease progression and regional heterogeneity. Further understanding of these complexities requires detailed neuropathological investigations in all cases where all major brain regions are examined and immunocytochemistry for PrPRes is performed. Western blot analysis for PrPRes and PRNP analysis are clearly also essential for this approach to classification and these needs should be borne in mind when a case of suspected CJD is encountered at autopsy (see Chapter 5).

Inherited prion diseases: clinical features and neuropathology The inherited prion diseases as a group account for around 10–15% of all human prion disease. These are inherited as autosomal dominant conditions, most of which appear to have full penetrance. An ever-increasing number of inherited prion diseases have been identified as the use of molecular genetic techniques to detect mutations and insertions in the PRNP become more widespread (for review see Gambetti et al., 1999). The clinical features of these disorders are remarkably variable, even within families in which the affected members all carry the identical PRNP mutation or insertion. One of the important factors that may modify substantially the clinical and pathological features is the effect of the polymorphisms at codon 129 and codon 219 (Gambetti et al., 1995; Barbanti et al., 1996; Hainfellner et al., 1999). For this reason, many investigators prefer to specify these polymorphisms in addition to the pathogenic mutation or insertion in order to more fully characterize the genetic basis of the clinical and pathological phenotype. Not all affected individuals appear to have a family history of prion disease, and it is conceivable that occasional patients in whom a pathogenic mutation is detected may represent an apparently new mutation. Far more commonly, affected individuals may give a family history of neurological disease which may have been mislabelled in the past as other disorders, for example Alzheimer’s


Table 17.7. Classification of familial prion diseases by clinical and genetic features 1

CJD phenotype:

Rapidly progressive dementia, myoclonus, ataxia, pyramidal/extrapyramidal signs, visual disturbances, characteristic EEG Associated mutations: D178N:129V, V180I, T183A, E200K, R208H, V210I, M232R 2 GSS phenotype: Progressive ataxia, incoordination, dysarthria, pyramidal/extrapyramidal signs, late dementia, no characteristic EEG Associated mutations: P102L, P105L, A117V, Y145STOP, F198S, D202N, Q217R 3 FFI phenotype: Reduced total sleep time, dysautonomia, enacted dreams, myoclonus, ataxia, late or absent dementia, no characteristic EEG Associated mutation: D178N:129M 4 Variable phenotype: Progressive dementia in most cases, myoclonus, cerebellar and extrapyramidal signs, no characteristic EEG Octapeptide repeat region inserts: 24, 48, 96, 120, 144, 168, 192, 216 bp

disease or Huntington’s disease. Review of large kindreds has indicated that these individuals are likely to be suffering from the same form of inherited prion disease as the index patient (Collinge et al., 1990). Despite the marked variability in the disease phenotype in inherited prion diseases, there are four main groups that can be recognized; the genetic and clinical features of these groups are summarized in Table 17.7. One of the main clinical subgroups exhibits a sCJD-type phenotype in which the patients suffer from rapidly progressive dementia with myoclonus, movement abnormalities and visual abnormalities, often with triphasic periodic complexes in the EEG resembling those in sCJD. The disease onset can be from the third to the ninth decade of life, with duration of illness ranging from several weeks to over 3 years. In patients with a sCJD phenotype, histological examination of the brain usually shows a spongiform encephalopathy which is similar to that occurring in MM1 or MV1 form of sCJD (Fig. 17.5). However, in the patients with the D178N-129V genotype, the neuropathology resembles the VV2 form of sCJD (Gambetti et al., 1995) although cerebral cortical gliosis tends to be pronounced, sometimes resulting in status spongiosis. No amyloid plaques are detectable in these cases. Another common group is the Gerstmann–Sträussler– Scheinker disease (GSS) phenotype, in which the disease

onset occurs from the third to the seventh decade of life with a prolonged duration of illness of around five years (Piccardo et al., 1998). The typical clinical presentation is a slowly progressive cerebellar syndrome with the subsequent development of other movement disorders (particularly pyramidal and extrapyramidal signs), and dementia usually occurring late in the illness. Occasional patients within this group also have suffered from seizures and sensory abnormalities, and occasional cases of Parkinsonism or spastic paraparesis have been described. In patients with the GSS phenotype, histological examination of the brain shows amyloid plaques with a characteristic multicentric structure, first described in the original GSS kindred (which were subsequently found to be a P102L mutation) (Gerstmann et al., 1936). These plaques are most readily identified in the cerebellar molecular layer (Fig. 17.5), but also occur in a more random distribution within the cerebral cortex and basal ganglia. Spongiform change in this disorder is variable and often appears to be related to the duration of the clinical illness. Similar histological features have been described in families with GSS A117V (Fig. 17.5). Certain PRNP mutations with the GSS phenotype are associated with other pathological features; in a Japanese family, the P102L-129M-219K genotype is not associated with amyloid plaque formation (Furukawa et al., 1995). In individuals with the F198S mutation (also known as the Indiana kindred) the neuropathology is characterized by severe amyloid plaque formation accompanied by neurofibrillary tangles in the cerebral grey matter (Ghetti et al., 1995). The PrPRes amyloid deposits can be unicentric or multicentric and can occur in the hippocampus, basal ganglia, thalamus and brainstem in addition to the cerebral and cerebellar cortex. These plaques are associated with abnormal neurites in the neocortex, which stain positively with antibodies to tau, ubiquitin and APP. The neurofibrillary tangles are similar to those found in Alzheimer’s disease and react with antibodies to phosphorylated tau protein. Spongiform change is not a prominent feature of this disorder. Similar neuropathological features have been identified in patients with a Q217R mutation with the clinical features of progressive dementia, ataxia and Parkinsonism, and in patients with the D202N mutation (Fig. 17.5) (Ghetti et al., 1995; Piccardo et al., 1998). A unique Japanese patient has been reported with a nonsense (stop) mutation at codon 145 in the PRNP gene (Kitamoto et al., 1993). This patient presented clinically with memory disturbance and progressive severe dementia in the fourth decade of life with a very long duration of illness of over two decades. Histology of the brain showed amyloid deposits in the walls of the small and medium sized blood vessels within the meninges, which reacted with antibodies to PrP. Neurofibrillary tangles similar to those

415

416


occurring in Alzheimer’s disease were also identified, but no spongiform change was present. In 1986, fatal familial insomnia (FFI) was described (Lugaresi et al., 1986), but the underlying genetic basis of this disease was not identified until 1992, when the PRNP was sequenced and a D178N mutation was identified in association with methionine at codon 129 on the affected allele (Medori et al., 1992). It is of considerable interest that the D178N mutation had previously been identified in some families with an inherited form of CJD, including the family described by Jakob in the 1920s. However, this form of familial CJD has valine at codon 129 on the mutated allele (Gambetti et al., 1995). In fatal familial insomnia the disease onset can occur between the third and eighth decade of life with a clinical history ranging between 6 and 48 months in duration. Patients with FFI usually present with insomnia, motor signs and dysautonomia, which progressively worsen and can be accompanied subsequently by dysarthria and ataxia. Severe dementia is not a characteristic feature. Heterogeneity within the FFI phenotype has been identified and in patients who are homozygous for methionine at codon 129, there tends to be an earlier onset of disease and a more rapid clinical course, whilst in heterozygotes (with valine at codon 129 in the non-mutated allele) there tends to be a later age of onset and a longer duration of the clinical illness (Parchi et al., 1995). The major histological feature of fatal familial insomnia is thalamic gliosis and neuronal loss which is most marked in the anterior and dorsomedial thalamic nuclei (Fig. 17.4), but is also present in other nuclei in the thalamus (particularly the anterior nuclei) to a lesser degree (Gambetti et al., 1995). Neuronal loss and gliosis are also present in the inferior olivary nuclei within the medulla, but there is little evidence of spongiform change in the central nervous system. Spongiform change has been identified in a restricted distribution within the entorhinal cortex, the severity of which often relates to the duration of the clinical illness. Focal spongiform change can be identified in patients with a prolonged clinical illness in the frontal, temporal and parietal lobes. The distribution of the pathology is somewhat similar to the ‘thalamic variant’ of sCJD (Parchi et al., 1999b) and in cases where the thalamic pathology is pronounced, with clinical features similar to those described above, the term sporadic fatal insomnia has been proposed (see above) (Parchi et al., 1999a).

Insertional mutations The clinical and pathological phenotype associated with these mutations is markedly variable, but some relatively

distinctive features can be identified. In patients with 24–96 bp insertions the phenotype closely resembles sCJD (MM1 or MV1) with a rapidly progressive dementia associated with ataxia, myoclonus, visual disturbances and triphasic periodic complexes on the EEG. Pathologically, these cases exhibit spongiform change in the cerebral cortex which may be severe with status spongiosis, but no PrPRes amyloid plaques are identified (Capellari et al., 1997). In patients with 120–216 bp insertions the clinical phenotype is more variable and slowly progressive, only rarely resembling that seen in sCJD. The clinical features may include psychiatric features, ataxia, movement disorders and occasionally myoclonus (Collinge et al., 1990). Pathologically, these cases also exhibit spongiform change of varying severity in the cerebral cortex and in the cerebellum there is a characteristic pattern of PrPRes accumulation with multicentric amyloid plaques often present, sometimes with a linear arrangement (Cochran et al., 1996). This unusual feature is not present in other regions of the brain and is best revealed on immunocytochemistry for PrPRes (Fig. 17.6).

Acquired human prion diseases Acquired human prion diseases are fortunately very rare, but interest in this group of disorders has been rekindled by the identification of vCJD as a zoonotic disorder caused by the bovine spongiform encephalopathy (BSE) agent. However, studies of other acquired human prion diseases have yielded much important information on the pathogenesis, genetic susceptibility and incubation times for these disorders and much of this information has informed the current predictions concerning likely future numbers of vCJD cases. Experimental studies have shown that prion diseases can be transmitted by a variety of routes, and the acquired human diseases show a corresponding diversity in the route of transmission. With the exception of vCJD, the neuropathological features of the other forms of the acquired human prion diseases are not in themselves sufficiently distinct to suggest an infectious aetiology, and this reinforces the importance of a full medical history when assessing an individual case.

Kuru: clinical features and neuropathology Kuru was described in the 1950s as an unusual progressive neurodegenerative disorder affecting the Fore tribe in Papua New Guinea, who lived in the Northern Highlands (Gajdusek & Zigas 1957). The aetiology of this disease was initially unknown, and it was not until transmission


of kuru was achieved experimentally in 1966 that it was recognized to be a form of transmissible spongiform encephalopathy (Gajdusek et al., 1966). The epidemiology of kuru is complex, but it was largely confined to the Fore tribe. This tribe practised ritualistic endocannibalism, and it appears that disease transmission occurred either by consumption of the brain or by handling the brain tissue by cutaneous contact in which numerous superficial lesions were likely to be present (for review see Will et al., 1999). Kuru was a common cause of death in adult females, (who were more likely than the males to have eaten the brain tissue) and the disease was recorded between the ages of approximately 5 years and into late adulthood. Kuru occurred as an epidemic, which gradually declined when the ritualistic practices were abandoned, although it is not yet completely extinct and very occasional recent cases have been reported, which presumably represent an incubation period of several decades (Cervenakova et al., 1998). The word kuru means to shake or to tremble and this describes the characteristic clinical features of tremor, followed by ataxia and dysarthria, with dementia occurring later in the illness. EEG examinations have indicated in a relatively small number of cases that the characteristic triphasic synchronous discharges of sCJD are absent. Genetic studies in kuru have indicated that all three genotypes at codon 129 at PRNP are susceptible to the disease. Interestingly, it has been suggested that methionine homozygotes were affected early in the epidemic, with the other genotypes affected afterwards and in particular the heterozygotes developing the illness after a relatively prolonged incubation period (Cervenakova et al., 1998). Pathologically, kuru exhibits a spectrum of changes which are similar to those found in sCJD, particularly in the VV2 and the MV2 subtypes. The most characteristic histological features in kuru are the rounded fibrillary amyloid plaques which are present in up to 50% of cases, particularly in the granular layer of the cerebellum (Fig. 17.6), but also occasionally present in smaller numbers in the molecular layer of the cerebellar cortex and the basal ganglia and thalamus (Hainfellner et al., 1999). Spongiform change has been identified in kuru (Fig. 17.6), but was not included in the original descriptions. PrP immunocytochemistry in kuru shows features similar to those in VV2 and MV2 subtypes of sCJD, with numerous kuru-type plaques staining positively in the cerebellum (Fig. 17.6), whilst in individuals who are valine homozygotes at codon 129 in the PRNP there is a characteristic laminar distribution of cerebral cortical pathology with perineuronal accumulation of PrPRes (McLean et al., 1998).

Iatrogenic CJD: clinical features and neuropathology The first case of iatrogenic CJD was reported in a corneal transplant recipient in 1974 (Duffy et al., 1974). Since then, several hundred cases of iatrogenic CJD have been reported in association with a range of surgical and medical procedures; these are summarized in Table 17.8. The routes of transmission can be divided into three groups: neurosurgical (involving contaminated neurosurgical instruments and intracerebral electrodes), which have the shortest incubation periods; tissue graft-associated (involving corneal and dura mater grafts), which have intermediate incubation periods, and human pituitary-associated cases which represent transmission by the intramuscular or subcutaneous route and are associated with prolonged incubation periods which may reach several decades (Brown et al., 2000). The clinical features of iatrogenic CJD are correspondingly variable, and appear to be associated with the route of transmission. In neurosurgical transmissions, the clinical features resemble those of sporadic Creutzfeldt–Jakob disease and appear to be associated with a rapidly progressive dementia with movement disorder. In the tissue graft cases there is a wider variation in the clinical phenotype, some cases resembling that of sCJD, but others appear to be associated with a cerebellar syndrome which is only followed by cognitive decline in the later stage of the illness. In both the neurosurgical and tissue graft associated cases there are periodic triphasic synchronous discharges on the EEG, similar to that occurring in most cases of sCJD (Will et al., 1999; Brown et al., 2000). Histologically, these cases also resemble sCJD (Fig. 17.6) (Billette de Villemeur et al., 1994), reinforcing the need for adequate clinical information in every case of suspected CJD. However, one interesting subgroup of dura mater graft recipients in Japan

Table 17.8. Summary of cases of iatrogenic CJD Type of exposure

Route of exposure

Surgical

Neurosurgical

instruments

instruments

Number of cases

5

Incubation period

12–28 months

Intracerebral Tissue grafts

electrodes

2

16–20 months

Corneal graft

3

16 months–26 years

117

18 months–18 years

141

5–30 years

Dura mater graft: Hormone group:

Human growth hormone Human gonadotraphin

4

12–16 years

417

418


exhibits florid plaques in small numbers in the cerebral and occasionally the cerebellar cortex; these are rounded fibrillary PrP amyloid plaques which are visible on routine microscopy and surrounded by spongiform change (Shimizu et al., 1999). These resemble the florid plaques occurring in vCJD (see below) but are far less numerous and occur in a more restricted distribution within the brain. PRNP analysis in these forms of iatrogenic CJD have shown that there is a preponderance of homozygotes at codon 129, with methionine homozygotes accounting for the majority of cases in France and the USA (Deslys et al., 1994; Brown et al., 2000) and valine homozygotes in the majority of UK cases (Collinge et al., 1991). The forms of iatrogenic CJD occurring in human pituitary hormone recipients are different in their clinical and pathological features. Most cases present as a progressive cerebellar syndrome with accompanying movement disorders and eventually dementia. Examination of the brain may show selective cerebellar atrophy, particularly in cases with a lengthy clinical duration of illness (15 months on average). On histological examination the cerebellum is usually severely affected with severe neuronal loss and gliosis and extensive confluent spongiform change in the molecular layer. PrP amyloid plaques have been described in several cases (Billette de Villemeur et al., 1994). The neuropathology in the cerebral hemispheres tends to resemble that of sCJD in VV or MV subjects, often with a laminar pattern of pathology in the cerebral cortex with PrPRes accumulation around large neurons in layer 5, small plaque-like deposits in layers 2 and 3 and widespread synaptic positivity for PrPRes (Fig. 17.6). However, other cases show more pronounced spongiform change with perivacuolar accumulation of PrPRes and fewer amyloid plaques. It is of interest that the admittedly restricted studies of lymphoid tissues in these forms of acquired human prion disease have not indicated any accumulation of PrPRes either by immunocytochemistry or by Western blot analysis (Hill et al., 1999).

Variant CJD: clinical features In 1990, surveillance of Creutzfeldt–Jakob disease was reinstituted in the UK because of the epidemic of BSE occurring in the cattle population and concerns that this agent might be a pathogen for humans, as it had subsequently affected a wide range of other species including antelopes, domestic cats and large wild cats in zoos. In 1996 the National CJD Surveillance Unit described ten patients with a new variant form of CJD (Will et al., 1996). This disorder occurred in young individuals (mean age 28 years), with a prolonged duration of illness (mean 13 months) characterized by psychiatric features at onset, followed by

sensory abnormalities (dysaesthesia involving the face, arms, back or legs) and cerebellar ataxia (Will et al., 2000). Subsequently, the patients developed movement disorders, including myoclonus and chorea, and eventually exhibited evidence of cognitive decline and dementia, which led to a terminal state of akinetic mutism. Clinical investigations have shown that the characteristic EEG appearances of sCJD are absent in these patients although the EEG is abnormal with non-specific changes in a widespread distribution. MRI scans have shown a characteristic area of high signal intensity in T2 and proton-weighted images in the posterior thalamus, predominantly involving the pulvinar but also extending in some cases to involve the dorsomedial nucleus (Zeidler et al., 2000). Similar changes have been identified in occasional patients in the periaqueductal grey matter of the midbrain. The value of CSF brain-associated proteins in the investigation of vCJD is less promising than in sCJD (see Table 17.6). CSF 14-3-3 is only elevated in 50% of vCJD patients; however, it does have a high degree of specificity (Green et al., 2001). As a result the positive predictive value of 14-3-3 in vCJD is high. Of all the other markers evaluated, tau protein has the highest degree of sensitivity and specificity (Green et al., 2001). Further studies are under way to assess whether tau protein may be useful in a clinical setting. The current clinical diagnostic features for vCJD are summarized in Table 17.9.

Variant CJD: neuropathology The pathological diagnostic features for vCJD are summarized in Table 17.10. Neuropathological studies have shown that the brain in most patients was of normal size and shape, but in patients with a prolonged duration of illness (around two years or more) there was evidence of both cerebral and cerebellar atrophy (Ironside et al., 2000). The cerebral atrophy is often most prominent in the occipital cortex. Microscopic examination of the brain shows a widespread spongiform encephalopathy characterized by the appearance of numerous large florid plaques. These lesions comprise rounded fibrillary amyloid plaques (measuring up to 100m) surrounded by a zone or halo of spongiform change (Fig. 17.7). These lesions occur both in isolation and in clusters, and in patients with a prolonged clinical illness they can be surrounded by more extensive spongiform change and gliosis (Ironside et al., 2001). Neuronal loss is often evident in the occipital cortex and in the cerebellum (predominantly involving the granular neurons). However, there is severe neuronal loss in the posterior thalamic nuclei and in the dorsomedial nucleus, accompanied by severe astrocytosis, often in the absence of spongiform change and amyloid plaque formation. The distribution of these


Table 17.9. Clinical diagnostic criteria for variant CJD (from www.cjd.ed.ac.uk)

Table 17.10. Pathological diagnostic criteria for variant CJD (Ironside et al., 2000).

I A B C

1

D II A B C D E III A

B

Progressive neuropsychiatric disorder. Duration of illness > 6 months. Routine investigations do not suggest an alternative diagnosis. No history of potential iatrogenic exposure. Early psychiatric symptomsa Persistent painful sensory symptomsb Ataxia. Myoclonus or chorea or dystonia. Dementia. EEG does not show the typical appearance of classical CJD (after review by CJDSU staff) c OR no EEG performed Posterior thalamic high signal on MRI scan (after review by CJDSU staff).

IV A Positive tonsil biopsy. DEFINITE VARIANT CJD PROBABLE VARIANT CJD PROBABLE VARIANT CJD POSSIBLE VARIANT CJD

IA and neuropathological confirmation of vCJDd I and 4/5 of II and IIIA and IIIB OR 1 and IV A I and 4/5 of II and IIIA

a

depression, anxiety, apathy, withdrawal, delusions. including both frank pain and/or unpleasant dysaesthesia. c generalized triphasic periodic complexes at approximately one per second. d spongiform change and extensive PrPRes deposition with florid plaques, throughout the cerebrum and cerebellum. b

thalamic lesions corresponds to the areas of high signal noted on MRI examination in the thalamus (Ironside et al., 2000; Zeidler et al., 2000). Spongiform change is most evident in the caudate nucleus and putamen, often without numerous amyloid plaques (Fig. 17.7). In the brainstem, neuronal loss and gliosis are evident particularly in the periaqueductal grey matter and in the colliculi with spongiform change occasionally present in the pontine nuclei. Immunocytochemistry for PrPRes shows a striking pattern of accumulation which is quite distinct from all other forms of human prion disease (Ironside, 1998). There is intense staining of the large florid plaques, but numerous smaller plaques (which cannot be identified on haematoxylin and eosin stained preparations) usually present in small clusters in the cerebral and cerebellar cortex (Fig. 17.7). There is widespread amorphous deposition of PrPRes around capillary walls and around small neurons

2

3 4 5

Multiple florid plaques in haematoxylin and eosin-stained sections of the cerebral and cerebellar cortex, with numerous cluster plaques on PrP immunocytochemistry and amorphous pericellular and perivascular PrPRes accumulation Severe spongiform change in the caudate nucleus and putamen with perineuronal and periaxonal PrPRes accumulation Marked astrocytosis and neuronal loss in the posterior thalamic nuclei and midbrain PrPRes accumulation in lymphoid tissue throughout the body Predominance of diglycosylated PrPRes in the central nervous system and lymphoid tissues

and astrocytes in the cerebral and cerebellar cortex. These non-amyloid deposits can also be visualized on PAS and Alcian blue stains, and more successfully with a Gallyas silver stain. In the basal ganglia and thalamus there is a striking perineuronal and linear dendritic accumulation of PrPRes (Fig. 17.7). Synaptic accumulation of PrPRes is also evident in the thalamus and also in both regions occasional amyloid plaques are identified. Synaptic staining for PrPRes and perineuronal staining are also evident in the hypothalamus and brainstem and within the grey matter of the spinal cord. In the cerebellum, the molecular layer shows a similar pattern of PrPRes accumulation to the cerebral cortex. Large numbers of plaques are also identified in the granular layer and occasionally in the cerebellar white matter (Fig. 17.7). The dentate nucleus in the cerebellum shows intense deposition of PrPRes in both a synaptic and plaque-like pattern. In the hippocampus there is little evidence of spongiform change and no florid plaques are visible but there is both synaptic and plaque-type deposition involving the pyramidal cells and the dentate neurons. Western blot analysis of PrPRes from vCJD brain shows a highly consistent pattern characterized by a 19 kDa nonglycosylated fragment and a predominance of the diglycosylated form (Collinge et al., 1996; Parchi et al., 1997; Ironside et al., 2000). An example of the pattern found in vCJD is shown in Fig. 17.2 where it is contrasted with the typical appearance of the PrPRes found in the two most common sCJD subtypes (MM1 and VV2). Unlike sCJD there is no evidence for regional variation PrPRes isotype within the brain in vCJD. While this pattern is characteristic of vCJD highly glycosylated PrPRes is also characteristic of some familial prion diseases (see Table 17.11). Studies of non-CNS tissues in vCJD have revealed no characteristic pathological features on routine histological

419

420


Table 17.11. Summary of genotypes and phenotypes of human prion diseases PRNP mutation

PRNP Codon 129a

PrPRes isotype

Proposed histological correlate

Reference

P102L

MM, MV

Type 1B and 8 kDa

Spongiform change, synaptic PrP, amyloid plaques

Parchi et al., 1998a Piccardo et al., 1998

P102L, A117V, D202N, Q212P, Q217R F198S

MM, MV MV, VV VV MM MV MV, VV

8 kDa

Multicentric amyloid plaques. Spongiform change and synaptic PrP not prominent

Parchi et al., 1998a Piccardo et al., 1998

Multicentric amyloid plaques and neurofibrillary tangles Thalamic atrophy

Piccardo et al., 1996

MM, MV

Novel type and 8kDa Type 2B

VV, MV MM MV

Type 1B Type 1B Type 2B

VV

Type 2B

None

MM, MV

Type 1A

Cortical spongiform degeneration Diffuse PrP staining Diffuse PrP staining in cortex, focal deposits in the cerebellum Plaque-like deposits in the cerebellum Synaptic and coarse granular PrP staining in cortex

None

VV

Type 2A

None

MV

Type 2A

None

MM

Type 2A

None

MM

Sporadic CJD Iatrogenic CJD (growth hormone)

None None

VV MM

Type 2A (Basic glycans) Type 1A Type 1A

None

VV MM

Type 2A

Kuru

None

MV VV MM

Type 2A Type 2A Type 2B

Disease GerstmannSträusslerScheinker disease

Fatal familial insomnia Familial CJD Familial CJD

Sporadic CJD (Myoclonic, Heidenhain variants) Sporadic CJD (Ataxic variant) Sporadic CJD (Kuru-plaque variant) Sporadic CJD (Sporadic fatal insomnia) Sporadic CJD (Cortical variant)

Variant CJD a

D178N D178N E200K

Plaque-like, focal and perineuronal staining Amyloid plaques in the cerebellum

Parchi et al., 1998b Parchi et al., 1998b Chardone et al., 1999 Puoti et al., 2000 Hainfellner et al., 1999 Parchi et al., 1999b

Parchi et al., 1999b Parchi et al., 1999b

Thalamic atrophy. PrP staining faint and variable Cortical perivacuolar staining

Parchi et al., 1999a,b Pan et al., 2001 Parchi et al., 1999a,b Pan et al., 2001

Faint synaptic staining Kuru plaques in the cerebral and cerebellar cortex

Parchi et al., 1999b Billette de Villemeur et al., 1994 Parchi et al., 1997

Kuru plaques in the cerebellum

Parchi et al., 1997 McLean et al., 1998 Cervenakova et al., 1998

Florid and cluster plaques

Ironside et al., 2000

M or V indicates the polymorphic allele coupled with the causative mutation

sections. However, immunocytochemistry for PrPRes and Western blot studies have revealed an accumulation of the abnormal PrPRes isoform in lymphoid tissues, particularly within follicular dendritic cells and macrophages within germinal centres (Hill et al., 1999; Ironside et al., 2000). This occurs in a widespread distribution involving the

tonsils, lymph nodes, spleen, thymus and gut-associated lymphoid tissue including Peyer’s patches and in the wall of the appendix (Fig. 17.7). There is also accumulation of PrPRes in the peripheral nervous system, particularly in the spinal dorsal root ganglia and in the trigeminal ganglia (Fig. 17.7). Recent studies have also confirmed the


presence of PrPRes in the retina and optic nerve (Wadsworth et al., 2001). The cellular context in which PrPRes accumulates appears to have an effect on the glycoform ratio. Western blot analysis of PrPRes in vCJD lymphoreticular system tissues has a further accentuation of the highly glycosylated type found in the brain. In contrast, the PrPRes found in the vCJD retina is less glycosylated than that found in corresponding brain samples (Head et al., 2003).

Variant CJD: relationship with BSE In the initial description of vCJD it was suggested on the basis of epidemiological evidence that this disorder was most likely to be caused by human exposure to the BSE agent (Will et al., 1996). Subsequent neuropathological examination of macaques experimentally infected with BSE by the intracerebral route showed neuropathological features which were closely similar to those of vCJD (Lasmézas et al., 1996). In addition, biochemical analysis of the PrPRes accumulating in the brain in BSE, vCJD and other BSE-related conditions in other species has shown a similar pattern of glycosylation (Collinge et al., 1996). The most compelling evidence of a relationship with BSE has come from the results of experimental transmissions into mice, and in studies using both transgenic and nontransgenic mice it has been shown that the agents responsible for BSE and vCJD exhibit identical transmission characteristics, indicating a single strain of origin (Bruce et al., 1997; Hill et al., 1997; Scott et al., 1999). Similar studies have shown markedly different characteristics for sCJD. Although it now seems clear that the BSE agent is responsible for vCJD, the precise mode of transmission to human is uncertain (although the dietary route appears the most likely). At present, it is not possible to state with certainty the numbers of future vCJD cases in the UK because of uncertainties over the incubation period, population exposure and genetic susceptibility. All patients with vCJD identified so far (143 definite and probable cases at the time of writing in the UK, six in France and single cases in Canada, Ireland, Italy and USA) when tested have shown methionine homozygosity at codon 129 in the PRNP (see Table 17.1). It is uncertain as to whether other PRNP genetic subgroups at codon 129 will be susceptible to the BSE agent. If so, individuals who are heterozygotes at codon 129 may have a longer incubation period than the methionine homozygotes (as in kuru and iatrogenic CJD in growth hormone recipients): this would significantly increase the future numbers of patients affected by vCJD.

The identification of PrPRes outside the central nervous system in vCJD has been matched by demonstration of infectivity in spleen and tonsil (Bruce et al., 2001). Concerns have also been raised about the potential infectivity of blood, although no direct evidence exists for this so far. Accordingly, it is possible that vCJD may be transmitted accidentally by medical or surgical procedures, since infectivity is likely to be present in the lymphoid tissues during the incubation period of the disease for some time before neurological symptoms appear. In one patient, study of an appendectomy specimen showed accumulation of PrPRes 8 months before the onset of neurological disease (Hilton et al., 1989). The impact that potential iatrogenic transmission of vCJD will have on future numbers of cases in the UK is uncertain.

Conclusions: PRNP genotype, PrPRes isotype and pathological phenotype Table 17.11 summarizes the literature of human prion diseases in which the authors attempt to correlate PRNP genotype, PrPRes isotype and pathological phenotype. The survey is incomplete and in some instances based on very few cases; nevertheless, a number of important points can be made from the comparative data:

Idiopathic forms (i) sCJD is not associated with any known mutations in the PRNP coding sequence, nor are affected individuals known to have been exposed to exogenous sources of infectious prions. (ii) sCJD phenotype is known to be influenced by the M/V polymorphism at codon 129 and this in combination with the PrPRes isotype has been proposed to determine six definable clinico-pathological variants (Parchi et al., 1999b).

Familial forms (i) The familial prion diseases GSS, familial CJD and FFI are associated with mutations in the coding sequence of PRNP, the gene that encodes PrP. (ii) The methionine (M)/valine (V) polymorphism profoundly affects disease phenotype. The D178N mutation coupled with 129V results in a CJD-like phenotype (familial CJD) characterized by cortical spongiform change whereas the same mutation coupled with 129M results in the thalamic atrophy characteristic

421

422


of fatal familial insomnia (Parchi et al., 1998b). The PrPRes isotypes produced differ between the two diseases, with type 1 present in familial CJDD178N and type 2 present in FFI. PrPRes in familial CJD D178N and FFI derive solely from the mutant allele (Chen et al., 1997) and this difference in PrPRes isotype is stable on transmission to transgenic mice (Telling et al., 1996). The presence of M or V at codon 129 in combination with the familial CJD E200K mutation also appears to influence PrPRes isotype and histopathology, but here the presence of V is associated with type 2 PrPRes . (iii) Amyloid plaques in GSS are associated with the presence of small PrPRes fragments (∼8 kDa) (Parchi et al., 1998a; Piccardo et al., 1998).

Acquired forms (i) Iatrogenic CJD in human growth hormone recipients is associated with type 1 PrPRes in methionine homozygotes and type 2 PrPRes in valine homozygotes without dramatically affecting the histopathological phenotype. (ii) The pathological phenotype of Kuru, which was acquired by ritual cannibalism in the Fore people of New Guinea, resembles iatrogenic CJD resulting from exposure to human prions in cadaveric hormone preparations. (iii) Variant CJD, which results from presumed oral exposure to bovine prions has a neuropathological profile distinct from that of Kuru, which results from oral exposure to human prions (McLean et al., 1998). Taken together these data suggest that there is no simple direct correlation between PrPRes isotype (as determined by current assays) and neuropathological profile, but that the human prion disease phenotype depends upon a more subtle and complex interplay between host genotype, prion strain and route of exposure. The prion diseases or transmissible spongiform encephalopathies are a unique group of disorders with a diverse spectrum of clinical and neuropathological features. Careful study of the neuropathology of these disorders has yielded much important information, and the use of techniques to identify the abnormal form of PrPRes in the brain and other tissues has added immensely to the value of pathological examination in these diseases. This places a high priority on autopsy examination for the confirmation of the clinical diagnosis, and reinforces the need for high standards of autopsy and laboratory practice when handling brain and other potential infectious tissues from

patients who died from these diseases (see Chapter 5). Although human prion diseases are fatal, trials of experimental treatments have now been established (Korth et al., 2001), and it is clear that neuropathology will continue to play an important role in the investigation of these devastating disorders, not only for diagnostic purposes, but to study the potential effects of these experimental treatments and to expand frontiers of research in collaboration with clinicians, biochemists, geneticists, and molecular and cell biologists.

Acknowledgements The National CJD Surveillance Unit is funded by the Department of Health and the Scottish Executive. We thank our colleagues Dr A Green and Mr M Bishop for helpful discussion, and would like to thank Ms B. A. Mackenzie for assistance in preparing the manuscript for publication and Ms D Ritchie for assistance with photography.

REFERENCES Aitken, A., Jones, D., Soneji, Y. & Howell, S. (1995). 14-3-3 proteins: biological function and domain structure, Biochem. Soc. Trans., 23, 605–11. Alperovitch, A., Zerr, I., Pocchiari, M. et al. (1999). Codon 129 genotype and sporadic Creutzfeldt–Jakob disease. Lancet, 353, 1673–4. Andreasen, N., Vanmechelen, E., Van de Voorde, A. et al. (1998). Cerebrospinal fluid tau protein as a biochemical marker for Alzheimer’s disease: a community based follow up study. J. Neurol. Neurosurg., Psychiatry, 64, 298–305. Barbanti, P., Fabbrini, G., Salvatore, M. et al. (1996). Polymorphism at codon 129 or codon 219 of PRNP and clinical heterogeneity in a previously unreported family with Gerstmann–Sträussler– Scheinker disease (PrP-P102L mutation). Neurology, 47,734– 41. Bell, J. E. & Ironside, J. W. (1993). Neuropathology of spongiform encephalopathies in humans. Br. Med. Bull., 43,738–77. Bell, J. E., Gentleman, S. M., Ironside, J. W. et al. (1997). Prion protein immunocytochemistry – UK five centre consensus report. Neuropathol. Appl. Neurobiol., 23, 26–35. Billette de Villemeur, T., Gelot, A., Deslys, J. P. et al. (1994). Iatrogenic Creutzfeldt–Jakob disease in three growth hormone recipients: a neuropathological study. Neuropathol. Appl. Neurobiol., 20,111–17. Bortone, E., Bettoni, L., Giori, C., Terzano, M. G., Trabattoni, G. R. & Mancia D. (1994). Reliability of EEG in the diagnosis of Creutzfeldt–Jakob disease. Electroencephalogr. Clin. Neurophysiol., 90, 323–30.


Brown, D. R., Clive, C. & Haswell, S. J. (2001). Antioxidant activity related to copper binding of native prion protein. J. Neurochem., 76, 69–76. Brown, P., Cervenakova, L., Boellaard, J. W., Stavrou, D., Goldfarb, L. G. & Gajdusek, D. C. (1994a). Identification of a PRNP gene mutation in Jakob’s original Creutzfeldt–Jakob disease family. Lancet, 344, 130–1. Brown, P., Gibbs, C. J., Rodgers-Johnson, P. et al. (1994b). Human spongiform encephalopathy: The National Institutes of Health series of 300 cases of experimentally transmitted disease. Ann. Neurol., 35, 513–29. Brown, P., Preece, M., Brandel, J.-P. et al. (2000). Iatrogenic Creutzfeldt–Jakob disease at the millennium. Neurology, 55, 1075–81. Bruce, M. E., Will, R. G., Ironside, J. W. et al. (1997). Transmissions to mice indicate that ‘new variant’ CJD is caused by the BSE agent. Nature, 389, 498–501. Bruce, M. E., McConnell, I., Will, R. G. & Ironside, J. W. (2001). Detection of variant Creutzfeldt–Jakob disease infectivity in extraneural tissues. Lancet, 358, 208–9. Budka, H., Aguzzi, A., Brown, P. et al. (1995). Neuropathological diagnostic criteria for Creutzfeldt–Jakob disease (CJD) and other human transmissible spongiform encephalopathies (prion diseases). Brain Pathol., 5, 459–66. Capellari, S., Vital, C., Parchi, P. et al. (1997). Familial prion disease with a 144 bp insertion in the prion protein gene in a Basque family. Neurology, 49, 133–41. Cervenakova, L., Goldfarb, L. G., Garruto, R., Lee, H.-S., Gajdusek, D. C. & Brown, P. (1998). Phenotype-genotype studies in kuru: implications for new variant Creutzfledt–Jakob disease. Proc. Natl Acad. Sci. USA, 95, 13239–41. Chardone, F., Liu, Q. G., Petraroli, R. et al. (1999). Prion protein glycotype analysis in familial and sporadic Creutzfeldt–Jakob disease patients. Brain Res. Bull., 49, 429–33. Cochran, E. J., Bennett, D. A., Cervenakova, L. et al. (1996). Familial Creutzfeldt–Jakob disease with a five-repeat octapeptide insert mutation. Neurology, 47, 727–33. Collinge, J., Owen, F., Poulter, M. et al. (1990). Prion dementia without characteristic neuropathology. Lancet, 336, 7–9. Collinge, J., Palmer, M. S. & Dryden, A. J. (1991). Genetic predisposition to iatrogenic Creutzfeldt–Jakob disease. Lancet, 337, 1441–2. Collinge, J., Sidle, K. C. L., Meads, J., Ironside, J. W. & Hill, A. F. (1996). Molecular analysis of prion strain variation and the aetiology of ‘new variant’ CJD. Nature, 383, 685–90. ¨ Creutzfeldt, H. G. (1920). Uber eine eigenartige herdförmige Erkrankung des Zentralnervensystems. Z. Neurol. Psychiat., 57, 1–18. DeArmond, S. J. & Ironside, J. W. (1999). Neuropathology of Prion Diseases. In Prion Biology and Diseases, ed. S. B. Prusiner, pp. 585–652. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. DeArmond, S. J., Qiu, Y., Sanchez, H. et al. (1999). PrPC glycoform heterogeneity as a function of brain region: implications for

selective targeting of neurons by prion strains. Neuropathol. Appl. Exp. Neurol., 58, 1000–9. Deslys, J. P., Marce, D. & Dormont, D. (1994). Similar genetic susceptibility in iatrogenic and sporadic Creutzfeldt–Jakob disease. J. Gen. Virol., 75, 23–7. Dickson, D. W. & Brown, P. (1999). Multiple prion types in the same brain: is a molecular diagnosis of CJD possible? Neurology, 53, 1903–4. Duffy, P., Wolf, J., Collins, G., DeVoe, A. G., Streeten, B. & Cowen, D. (1974). Possible person-to-person transmission of Creutzfeldt–Jakob disease. New Eng. J. Med., 290, 692–3. Finkenstaedt, M., Szudra, A., Zerr, I. et al. (1996). MR imaging of Creutzfeldt–Jakob disease. Radiology, 199, 793–8. Furukawa, H., Kitamoto, T., Anaka, Y. & Tateishi, J. (1995). New variant prion protein in a family with Gerstmann–Straussler syndrome. Brain Res. Mol. Biol., 30, 385–8. Gajdusek, D. C. & Zigas, V. (1957). Degenerative disease of the central nervous system in New Guinea. The endemic occurrence of ‘Kuru’ in the native population. N. Engl. J. Med., 257, 974–8. Gajdusek, D. C., Gibbs, C. J. & Alpers, M. P. (1966). Experimental transmission of a kuru-like syndrome to chimpanzees. Nature, 209, 794–6. Gambetti, P., Parchi, P., Peterson, R. B., Chen, S. G. & Lugaresi, E. (1995). Fatal familial insomnia and Creutzfeldt–Jakob disease: clinical, pathological and molecular features. Brain Pathol., 5, 43–51. Gambetti, P., Petersen, R. B., Parchi, P. et al. (1999). Inherited Prion Diseases. In Prion Biology and Diseases, ed. S. B. Prusiner, pp. 509–84. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. ¨ Gerstmann, J., Sträussler, E. & Scheinker, I. (1936). Uber eine eigenartige hereditär-familiäre Erkrankung des Zentralnervensystems. Zugleich ein Beitrag zur Frage des vorzeitigen lokalen Alterns. Z. Neurol. Psychiat., 154, 736–62. Ghetti, B., Dloughy, S. R., Giaccone, G. et al. (1995). Gerstmann– Sträussler–Scheinker disease and the Indiana kindred. Brain Pathol., 5, 61–75. Gibbs, C. J., Gajdusek, D. C., Asher, D. M. & Alpers, M. P. (1968). Creutzfeldt–Jakob disease (spongiform encephalopathy): transmission to the champanzee. Science, 161, 388–9. Giese, A. & Kretzschmar, H. A. (2001). Prion-induced neuronal damage: the mechanism of neuronal destruction in the subacute spongiform encephalopathies. Curr. Top. Microbiol. Immunol., 253, 203–17. Green, A. J. E., Harvey, R. J., Thompson, E. J. & Rossor, M. N. (1997). Increased S-100b in the cerebrospinal fluid of patients with frontotemporal dementia Neurosci Lett., 235, 5–8. Green, A. J. E., Thompson, E. J., Stewart, G. E. et al. (2001). Use of 14-3-3 and other brain-specific proteins in the diagnosis of variant Creutzfeldt–Jakob disease. J. Neurol. Neurosurg. Psych., 70, 744–8. Hadlow, W. J. (1959). Scrapie and kuru. Lancet, 2, 289–90. Hainfellner, J. A., Liberski P. P., Guiroy, D. C. et al. (1997). Pathology and immunocytochemistry of a kuru brain. Brain Pathol., 7, 547–53.

423

424


Hainfellner, J. A., Parchi, P., Kitamoto, T., Jarius, C., Gambetti, P. & Budka, H. (1999). A novel phenotype in familial Creutzfeldt– Jakob disease: Prion protein gene E200K mutation coupled with valine at codon 129 and type 2 protease-resistant prion protein. Ann. Neurol., 45, 812–16. Head, M. W., Tissingh, G., Uitdehaag, B. M. et al. (2001). Sporadic Creutzfeldt–Jakob disease in a young Dutch valine homozygote: atypical molecular phenotype. Ann. Neurol., 50, 258–61. Head M. W., Northcott V., Rennison, K. et al. (2003). Prion protein accumulation in eyes of patients with sporadic and variant Creutzfeldt–Jakob disease. Invest. Ophthalmol. Vis. Sci., 44, 342–6. Herms, J., Tings, T., Gall, S. et al. (1999). Evidence of presynaptic location and function of the prion protein. J. Neurosci., 19, 8866–75. Hill, A. F., Desbruslais, M., Joiner, S. et al. (1997). The same prion strain causes vCJD and BSE. Nature, 389, 448–50. Hill, A. F., Butterworth, R. J., Joiner, S. et al. (1999). Investigation of variant Creutzfeldt–Jakob disease and other human prion diseases with tonsil biopsy samples Lancet, 353, 183–9. Hilton, D. A., Fathers, E., Edwards, P., Ironside, J. W. & Zajicek, J. (1998). Prion immunoreactivity in appendix before clinical onset of variant Creutzfeldt–Jakob disease. Lancet, 352, 703–4. Hsiao, K., Baker, H., Crow, T. J. et al. (1989). Linkage of prion protein missense variant to Gerstmann–Sträussler syndrome. Nature, 338, 342–5. Hsich, G., Kenney, K., Gibbs, C. J., Lee, K. H. & Garrubgtib, M. G. (1996). The 14-3-3 brain protein in cerebrospinal fluid as a marker for transmissible spongiform encephalopathies. N. Engl. J. Med., 335, 924–30. Ironside, J. W. (1996). Review: Creutzfeldt–Jakob disease. Brain Pathol., 6, 379–88. Ironside, J. W. (1998). Prion diseases in man. J. Pathol., 186, 227–34. Ironside, J. W., Head, M. W., Bell, J. E., McCardle, L. & Will, R. G. (2000). Laboratory diagnosis of variant Creutzfeldt–Jakob disease. Histopathology, 37, 1–9. Jackson, G. S., Murray, I., Hosszu, L. P. et al. (2001). Location and properties of metal-binding sites on the human prion protein. Proc. Natl Acad. Sci. USA, 98, 8531–5. ¨ Jakob, A. (1921). Uber eine eigenartige Erkrankungen des Zentralnervensystems mit bemerkenswertem anatomichen Befunde (Spastische Pseudosklerose-Encephalomyelopathie mit anatomishen disseminierten Degenerationsherden). Z. Gesamte Neurol. Psychiat., 64, 147–228. Kitamoto, T., Iizuka, R. & Tateishi, J. (1993). An amber mutation of prion protein in Gerstmann–Straussler syndrome with mutant PrP plaques. Biochem. Biophys. Res. Commun., 192, 525–31. Klampt, F., Dal-Pizzol, F., Conte da Frota, M. L. J. R. et al. (2001). Imbalance of antioxidant defense in mice lacking cellular prion protein. Free Radic. Biol. Med., 30, 1137–44. Korth, C., May, B. C., Cohen, F. E. & Prusiner, S. B. (2001). Acridine and phenothiazine derivatives as pharmacotherapeutics for prion diseases. Proc. Natl Acad. Sci. USA, 98, 9836–41. Lasmézas, C. I., Ironside, J. W., Chiach, C. et al. (1996). BSE transmission to Macaques. Nature, 381, 743–4.

Lugaresi, E., Medori, R., Montagna, P. et al. (1986). Fatal familial insomnia and dysautonomia with selective degeneration of thalamic nuclei. N. Engl. J. Med., 315, 997–1003. McLean, C. A., Ironside, J. W., Alpers, M. P. et al. (1998). Comparative neuropathology of Kuru and the new variant of Creutzfeldt– Jakob disease: evidence for the strain of agent over genotype of host. Brain Pathol., 8, 429–37. McNaughton, H. K. & Will, R. G. (1997). Creutzfeldt–Jakob disease presenting acutely as stroke; an analysis of 30 cases. Neurol. Infect. Epidemiol., 2, 19–24. Masters, C. L. & Richardson, E. P. (1978). Subacute spongiform encephalopathy (Creutzfeldt–Jakob disease). The nature and progression of spongiform change. Brain, 101, 333–44. Masters, C., Gajdusek, D. C. & Gibbs, Jnr C. J. (1981). Creutzfeldt– Jakob disease virus isolations from the Gerstmann–Strässler– Scheinker syndrome with an analysis of the various forms of amyloid plaque deposition in the virus-induced spongiform encephalopathies. Brain, 104, 559–88. Medori, R., Tritschler, H.-J., LeBlanc, A. et al. (1992). Fatal familial insomnia, a prion disease with a mutation at codon 178 of the prion protein gene. N. Engl. J. Med., 326, 444–9. Otto, M., Stein, H., Szudra, A. et al. (1997a). S-100 protein concentration in the cerebrospinal fluid of patients with Creutzfeldt– Jakob disease. J. Neurol., 244, 566–70. Otto, M., Wiltfang, J., Tumani, H. et al. (1997b). Elevated levels of tau-protein in cerebrospinal fluid of patients with Creutzfeldt– Jakob disease. Neurosci. Lett., 225, 210–12. Palmer, M. S., Dryden, A. J., Hughes, J. T. & Collinge J. (1991). Homozygous prion protein genotype predisposes to sporadic Creutzfeldt–Jakob disease. Nature, 352, 340–2. Pan, T., Colucci, M., Wong, B.-S. et al. (2001). Novel differences between two human prion strains revealed by two-dimensional gel electrophoresis. J. Biol. Chem., 276, 37284–8. Parchi, P., Castellani, R., Cortelli, P. et al. (1995). Regional distribution of protease-resistant prion protein in fatal familial insomnia. Ann. Neurol., 38, 21–9. Parchi, P., Castellani, R., Capellari, S. et al. (1996). Molecular basis of phenotypic variability in sporadic Creutzfeldt–Jakob disease. Ann. Neurol., 39, 767–78. Parchi, P., Capellari, S., Chen, S. G. et al. (1997). Typing prion isoforms. Nature, 386, 232–3. Parchi, P., Chen, S. G., Brown, P. et al. (1998a). Different patterns of truncated prion protein fragments correlate with distinct phenotypes in P102L Gerstmann-Sträussler–Scheinker disease. Proc. Natl Acad. Sci. USA, 95, 8322–7. Parchi, P., Petersen, R. B., Chen, S. G. et al. (1998b). Molecular pathology of fatal familial insomnia. Brain Pathol., 8, 539–48. Parchi, P., Capellari, S., Chin, S. et al. (1999a). A subtype of sporadic prion disease mimicking fatal familial insomnia. Neurology, 52, 1757–63. Parchi, P., Giese, A., Capellari, S. et al. (1999b). Classification of sporadic Creutzfeldt–Jakob disease based on molecular and phenotypic analysis of 300 subjects. Ann. Neurol., 46, 224–33.


Parchi, P., Zou, W., Wang, W. et al. (2000). Genetic influence on the structural variation of the abnormal prion protein. Proc. Natl Acad. Sci. USA, 97, 10168–72. Piccardo, P., Seiler, C., Dlouhy, S. R. et al. (1996). ProteinaseK-resistant prion protein isoforms in Gerstmann–Straussler– Scheinker disease (Indiana kindred). J. Neuropathol. Exp. Neurol., 55, 1157–63. Piccardo, P., Dlouhy, S. R., Lievens, P. M. J. et al. (1998). Phenotypic variability of Gerstmann–Straussler–Scheinker disease is associated with prion protein heterogeneity. J. Neuropathol. Exp. Neurol., 57, 979–88. Prusiner, S. B. (1982). Novel proteinaceous particles cause scrapie. Science, 216, 136–44. (1998). Prions. Proc. Natl Acad. Sci. USA, 95, 13363–83. Puoti, G., Giaconne, G., Rossi, G., Canciani, B., Bugiani, O. & Tagliavini, F. (1999). Sporadic Creutzfelt–JaKob disease: cooccurrence of different types of PrP (Sc) in the same brain. Neurology, 53, 2173–6. Puoti, G., Rossi, G., Giaccone, G. et al. (2000). Polymorphism at codon 129 of PRNP affects the phenotypic expression of Creutzfeldt–Jakob disease linked to E200K mutation. Ann. Neurol., 48, 269–70. Richardson, E. P. & Masters, C. (1995). The nosology of Creutzfeldt– Jakob disease and conditions related to the accumulation of PrPCJD in the nervous system. Brain Pathol., 5, 33–41. Schroter, A., Zerr, I., Henkel, K. Tschampa, H. J., Finkenstaedt, M. & Poser, S. (2000). Magnetic resonance imaging in the clinical diagnosis of Creutzfeldt–Jakob disease. Arch. Neurol., 57, 1751–7. Scott, M. R., Will, R., Ironside, J. et al. (1999). Compelling evidence for transmission of bovine spongiform encephalopathy prions to humans. Proc. Natl Acad. Sci. USA, 96, 15137–42. Shimizu, S., Hoshi, K., Muramoto, M. et al. (1999). Creutzfeldt–Jakob disease with florid-type plaques after cadaveric dura mater grafting. Arch. Neurol., 56, 357–62. Telling, G. C. (2000). Prion protein genes and prion diseases: studies in transgenic mice. Neuropathol. Appl. Neurobiol., 26, 209–20. Telling, G. C., Parchi, P., De Armond, S. J. et al. (1996). Evidence for the conformation of the pathologic isoform of the prion protein enciphering and propagating prion diversity. Science, 274, 2079–82. Wadsworth, J. D. F., Hill, A. F., Joiner, S., Jackson, G. S., Clarke, A. R. & Collinge, J. (1999). Strain-specific prion-protein conformation determined by metal ions. Nat. Cell Biol., 1, 55–9. Wadsworth, J. D. F., Joiner, S., Hill, A. F. et al. (2001). Tissue distribution of protease resistant prion protein in variant Creutzfeldt– Jakob disease using a highly sensitive immunoblotting assay. Lancet, 358, 171–80. Will, R. G., Ironside, J. W., Zeidler, M. et al. (1996). A new variant of Creutzfeldt–Jakob disease in the UK. Lancet, 347, 921–5. Will, R. G., Alperovitch, A., Poser, S. et al. (1998). Descriptive epidemiology of Creutzfeldt–Jakob disease in six European countries, 1993–1995. Ann. Neurol., 43, 763–7. Will, R. G., Alpers, M. P., Dormont, D., Schonberger, L. B. & Tateishi, J. (1999). Infectious and Sporadic Prion Diseases. In Prion

Biology and Diseases, ed. S. B. Prusiner, pp. 465–508. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Will, R. G., Zeidler, M., Stewart G. E. et al. (2000). Diagnosis of new variant Creutzfeldt–Jakob disease. Ann. Neurol., 47, 575–82. Windl, O., Dempster, M., Estibeiro, P. et al. (1996). Genetic basis of Creutzfeldt–Jakob disease in the United Kingdom: a systematic analysis of predisposing mutations and allelic variation in the PRNP gene. Hum. Genet., 98, 259–64. Wong, B. S., Chen, S. G., Colucci, M. et al. (2001). Aberrant metal binding by prion protein disease. J. Neurochem., 78, 1400–8. Zeidler, M., Sellar, R. J., Collie, D. A. A. et al. (2000). The pulvinar sign on magnetic resonance imaging in variant Creutzfeldt–Jakob disease. Lancet, 355, 1412–18. Zerr, I., Bodemer, M., Racker, S., et al. (1995). Cerebrospinal fluid concentration of neuron-specific enolase in diagnosis of Creutzfeldt–Jakob disease. Lancet, 345, 1609–10. Zerr, I. Bodemer, M., Gefeller, O. et al. (1998). Detection of 14-33 protein in the cerebrospinal fluid supports the diagnosis of Creutzfeldt–Jakob disease. Ann. Neurol., 43, 32–40.

Appendix 17.1 Technique for PrP immunocytochemistry on paraffin-embedded sections. 1. Take sections to water. 2. Remove formalin pigment with saturated picric acid – 15 min. 3. Rinse well with water to remove picric. 4. Wash in water. 5. Block with 3% hydrogen peroxide/methanol – 30 min 6. Wash in water – 5 min 7. Autoclave at 121 ◦ C in distilled water – 10 min 8. Remove from autoclave when temperature falls to about 60 ◦ C. Allow to cool further by adding small amounts of cold distilled water – 5min 9. 96% formic acid – 5 min 10. Wash well with tap water (three changes) 11. Immerse in Proteinase K at 10 g/ml – 5min at room temp (150 l BDH stock/300 ml PBS) 12. Rinse in PBS 13. Immerse in TBS – 5 min 14. Diluted normal blocking serum (3 drops of yellow stock to 10 ml of TBS) – 20 min. 15. Drain. 16. Primary antiserum (diluted in TBS) – overnight. 3F4 1:50 (Dako, UK) 17. Rinse in TBS ×2. 18. Diluted biotinylated secondary antibody solution (3 drops of Elite yellow stock and 1 drop of Elite blue stock to 10 ml of TBS) – 30 min. 19. Rinse in TBS × 2.

425

426


20. Vectastain Elite ABC reagent (Vector laboratories, UK)(2 drops of reagent A and B to 5 ml TBS) – 30 mins. 21. Rinse in TBS × 2. 22. DAB (visualize microscopically). 23. Wash well in water. 24. Counterstain lightly in haematoxylin. 25. Dehydrate, clear and mount.

Appendix 17.2 Western blot analysis of CJD brain (Ironside et al., 2000) 1. Sample ∼100 mg frozen cerebral (usually frontal) cortex. 2. Homogenise in 9 volumes (w/v) of non-denaturing detergent-containing buffer (0.5% NP40, 0.5% sodium deoxycholate, TBS). 3. Clear homogenate by low speed centrifugation.

4. Digest supernatant with proteinase K (50 g/ml, 1 hour, 37◦ C). 5. Separate 10 l of SDS-denatured sample by minigel SDS-PAGE. 6. Transfer to PVDF membrane. 7. Immuno-detection with monoclonal antibody 3F4 (Dako) or 6H4 (Prionics) and an appropriate antimouse IgG horseradish peroxidase-conjugated secondary antibody. 8. Enhanced chemiluminescent (ECL or ECL Plus) detection. 9. Estimation of PrPRes mobility by reference to (ECL) molecular weight markers or standard samples. 10. Analysis of glycoform ratio by densitometry. Note: Steps 1–8 should be carried out using appropriate personal protection in a category 3∗ laboratory with steps 1– 4 being performed in a class 1 microbiological containment hood.

18 Alcoholism and dementia Clive Harper1 and Richard A. Scolyer2 2

Introduction There is little doubt that excessive consumption of alcohol over a considerable period of time leads to an impairment of cognitive function. Specific alcohol-related disorders such as the Wernicke–Korsakoff syndrome, hepatic encephalopathy, and pellagra cause clinical dementia syndromes but when these have been excluded there is still a considerable number of alcoholics who can be classified as demented. The Liverpool Longitudinal Study of mental health of the elderly residing in a community dwelling found that dementia was 4.6 times more likely to occur in men aged 65 and older who had a lifetime history of heavy drinking (Saunders et al., 1991). A more recent epidemiological study of older African-American men found that increasing alcohol consumption was associated with a worsening performance on dementia screening scales (Hendrie et al., 1996). In a sample of 130 cognitively impaired residents of long-term care facilities, alcohol-related dementia comprised 24% of this population compared with Alzheimer’s disease (35%), vascular dementia (19%), and other causes (22%) (Carlen et al., 1994). The most commonly used clinical definition of alcohol-related dementia is given in the Diagnostic and Statistical Manual Version IV (DSM IV) (Frances, 1994). This definition requires the presence of dementia that, in the opinion of the clinician, is intrinsically linked to the abuse of alcohol. The diagnostic criteria are vague and subjective and there have been no published validation or reliability reports using these criteria. Two recent papers dealing with operational diagnostic criteria in alcohol-related disorders might help clarify the issue. Oslin and colleagues have proposed diagnostic criteria

1 Neuropathology Unit, University of Sydney, Australia Anatomical Pathology Department, Royal Prince Alfred Hospital, Camperdown, Australia

for alcohol-related dementia (Oslin et al., 1998) and Caine and colleagues have proposed criteria for the Wernicke– Korsakoff syndrome (Caine et al., 1997). In 1985 Victor and Adams (1985) reviewed the clinical, neuropsychological, neuropathological and neuroradiological evidence concerning alcohol specific neurotoxicity and concluded that there was no need to invoke a separate entity due to the toxic effect of alcohol on the brain as they could practically always account for the clinical state of their patients by one or a combination of the Wernicke–Korsakoff syndrome, acute and chronic hepatic encephalopathy, communicating hydrocephalus, Alzheimer’s disease, Marchiafava Bignami disease, ischaemic infarction or anoxic encephalopathy. They emphasized that no one had established a discrete pathological basis for the syndrome of alcoholic dementia but suggested that morphometric and other quantitative techniques might disclose the lesions. Nevertheless, it must be remembered that not all alcoholics have impairment of cerebral function. Butters and his colleagues (Butters et al., 1987) have shown that 30–50% of alcoholics will perform a range of neuropsychological tests within the normal range for controls. However, as Tuck and Jackson (1991) report, careful study of people who drink excessively (median daily intake of 180 grams per day) and are not overtly demented frequently reveals a cognitive impairment, which takes the form of frontal lobe dysfunction and may be relatively subtle. These signs may precede those of other alcohol related neurological disorders such as cerebellar degeneration, peripheral neuropathy and Wernicke–Korsakoff syndrome by more than 10 years. Thus, there seems to be a wide spectrum of the effects of alcohol on the brain with considerable individual variability in susceptibility which, to some extent, may be


427

428

C. Harper and R. A. Scolyer

determined genetically (Parsons, 1987). Other factors such as the pattern of alcohol consumption might be important but are very difficult to study. The increasing incidence of polydrug abuse has also raised concerns regarding the extent of brain damage due to interactive effects of alcohol and other licit and illicit drugs (Miller & Gold, 1990).

Primary alcoholic dementia The existence of specific neurotoxic effects of alcohol on the central nervous system (primary alcoholic dementia) can be approached from a number of points of view. While there has been some controversy, clinical and neuropsychological data point towards such an entity. Moreover, as new neuroimaging techniques such as MRI diffusion tensor imaging are introduced, abnormalities at the structural and functional level are being identified in uncomplicated alcoholics who are impaired cognitively (Pfefferbaum et al., 2000). Nevertheless, the ultimate proof of the existence of this entity rests with the identification of the pathological substrate of alcohol specific neurotoxicity. There is a significant literature on animal models of alcohol neurotoxicity. Arendt (Arendt et al., 1988) has shown that rats that are well nourished and given alcohol develop memory deficits. Pathologically, both hippocampal neurons (Walker et al., 1980) and cerebellar Purkinje cells (Pentney, 1982; Cragg & Phillips 1983) appear to be damaged by prolonged exposure to alcohol. McMullen and her colleagues (McMullen et al., 1984) made an important observation in this regard when they showed that 5 months of exposure to alcohol in rats caused a significant reduction in the branching of the dendritic arbor of hippocampal neurons but that the arbur returned towards normality after two months of abstinence. Human neuropathological studies are far more difficult. Ideally, cases should have been tested clinically and neuropsychologically before death and cognitive deficits documented. Other causes of dementia such as Alzheimer’s disease, strokes, Wernicke–Korsakoff syndrome, etc. must be excluded in order to address the question of alcohol specific neurotoxicity. There is only limited data on such uncomplicated alcoholic cases. Brain weight studies show that a group of uncomplicated alcoholics (drinking more than 80 grams of alcohol per day for more than 15 years) had a significantly reduced brain weight compared to controls (Harper & Kril, 1993). The loss of brain tissue can be determined more accurately from measurements of brain volume (BV) and intracranial volume (ICV). The ratio of ICV minus BV to ICV has been termed the pericerebral space (PICS) and in uncomplicated alcoholics the PICS is 11.3% compared to 8.3% in controls (Harper & Kril 1985).

The loss of brain tissue is largely accounted for by a reduction in the volume of the white matter rather than loss of cortical tissue (Harper & Kril 1993). These changes are more severe in those alcoholics with either Wernicke– Korsakoff syndrome or cirrhosis (Harper & Kril, 1985). The explanation for this white matter change has yet to be elucidated but simple explanations such as changes in hydration (Harper et al., 1988) or chemical composition (Olsson et al., 1996) have been excluded. Some of the white matter loss appears to be reversible in that, after prolonged abstinence, the brain shrinkage reverts towards normality (Shear et al., 1994; Liu et al., 2000). Nevertheless there is almost certainly permanent damage with axonal degeneration following neuronal loss. MRI diffusion tensor imaging permits in vivo quantification of the directionality and coherence of white matter fibre tracts and abnormalities have been noted in alcoholism (Pfefferbaum et al., 2000). These authors suggest that the disruption contributes to disturbances in attention and working memory. The apparent atrophy of the cerebral cortex that is seen commonly in alcoholic cases is partly explained by the white matter loss, however, there is also generally a slight reduction in the volume of the cerebral cortex. This has been demonstrated both pathologically (De la Monte, 1988) and using MRI (Jernigan et al., 1991) although not all the alcoholic cases have reduced cortical grey matter volumes (Harper & Kril, 1985). Alcohol-related neuronal loss has been documented in specific regions of the cerebral cortex (superior frontal association cortex, Brodmanns area 8) (Harper et al., 1987), hypothalamus (supraoptic and paraventricular nuclei) (Harding et al., 1996), and cerebellum (Baker et al., 1999). There was no significant change in the primary motor (area 4), frontal cingulate (area 32) or inferior temporal (areas 20 & 36) cortices (Kril & Harper 1989), basal ganglia, nucleus basalis (Cullen et al., 1997), or serotonergic raphe nuclei (Halliday et al., 1995). The data in the literature is conflicting for several other regions: the hippocampus (Harding et al., 1997), amygdala and locus ceruleus (Halliday & Baker, 1996). The severe damage to the frontal cortex in alcoholics is consistent with clinical (OscarBerman & Hutner, 1993) and neuroradiological studies (Jernigan et al., 1991) which suggest that the frontal lobe is more susceptible to alcoholic-related brain damage than other cortical regions. Not all studies have shown neuronal loss in alcoholics. Jensen and Pakkenberg (1993) estimated the total number of neurons in the neocortex in alcoholic and control patients and found no difference in the two groups, although it should be noted that selective neuronal loss from particular gyri could be missed with the technique used in this study. There may be particular

Alcoholism and dementia

groups of neurons which are more likely to be damaged. An analysis of the pattern of neuronal loss from the superior frontal cortex in alcoholics revealed that large pyramidal neurons, with a somal area greater than 90 were selectively lost (Harper & Kril, 1989). This population of large neurons has been recognized as being more vulnerable in both Alzheimer’s disease (Terry et al., 1981) and the normal ageing process (Terry & Hansen, 1987). The tau protein, a marker for Alzheimer’s disease, has been measured in the cerebrospinal fluid (CSF) of a range of different dementia cases. It was shown that alcohol-induced organic brain disorders are characterized by normal CSF-tau levels, and that the CSF examination for tau in combination with other clinical findings may help in differentiating these cases from Alzheimer’s disease (Morikawa et al., 1999). There does not appear to be any link between alcohol-related brain damage and Alzheimer’s disease although there is some work that suggests a relationship between alcohol and ageing (Harper et al., 1998a). Many of the regions that are normal in uncomplicated alcoholics are damaged in those with the Wernicke–Korsakoff syndrome as discussed below. One other cortical abnormality which has been described in alcoholics is Morels laminar sclerosis (Victor et al., 1989). This is a rare condition characterized by necrosis and degeneration of layers III and IV of the cerebral cortex. It is frequently associated with Marchiafava–Bignami disease (Victor et al., 1989). Dendritic and synaptic changes have been documented in uncomplicated alcoholics (Harper & Corbett, 1990) and these, together with receptor and transmitter changes, may explain functional changes and cognitive deficits that precede the more severe structural neuronal changes (Hakim & Pappius, 1983; Freund & Ballinger, 1988). Pathological changes that have been found to correlate with alcohol intake include white matter loss (Kril et al., 1997) and neuronal loss in the hypothalamus (Harding et al., 1996) and cerebellum (Baker et al., 1999). The specific mechanisms linking the long-term abuse of alcohol (ethanol) with the pathological changes described above are still largely unknown. It has been stated that alcohol or one of its metabolites is directly neurotoxic to the nervous system but most of the speculation is based on experimental animal models and in vitro tissue culture studies. Mechanisms of damage have been well reviewed by Charness and his colleagues (Charness et al., 1989) in their paper entitled ‘Ethanol and the nervous system’. There are many other factors which can play a role in the long-term effects of alcohol on the nervous system. Some of these will be discussed in the latter parts of this chapter but one should not forget that alcoholics are prone to recurrent head injuries, fitting is common, as is polydrug abuse (Miller & Gold, 1990), and there appears to be a link between

alcohol and the sleep-apnoea syndrome (Carlen & Wilkinson, 1987).

The Wernicke–Korsakoff syndrome The Wernicke–Korsakoff syndrome is an easily treated but potentially fatal neurological disorder which is caused by thiamin (vitamin B1) deficiency. In most acute cases there is a dramatic clinical improvement following treatment with thiamin. The disorder is more common than generally realized with a prevalence at autopsy ranging from 0.10 to 4.7%, the latter being in a selected forensic population (Harper, 1983; Harper et al., 1995). Case reports are seen commonly in current world literature. After a public health programme of thiamin enrichment of bread in Australia in 1991, the prevalence of the Wernicke–Korsakoff syndrome dropped significantly from 4.7% to 1.1% in 1997 (Harper et al., 1998). Enrichment of flour with thiamin has been practised in a number of countries for many years (Axford & Williams, 1981). In the UK, Canada, and Denmark the requirement is mandatory. In the USA enrichment is not mandatory however standards are laid down and the bulk of flour marketed is enriched. It is said that thiamin deficiency has virtually disappeared in countries such as Japan and the USA since the introduction of thiamin enrichment of rice and flour respectively (Figueroa et al., 1953; Sebrel, 1966; Axford & Williams, 1981). Although the disorder is seen most commonly in the alcoholic population, there are a number of other ‘at risk’ groups (Harper, 1980). Most of these are self-evident; patients on starvation diets, gastric stapling, haemodialysis (Ihara et al., 1999), prolonged intravenous feeding without vitamin supplementation and other causes of severe malnourishment. It should also be noted that the Wernicke–Korsakoff syndrome can occur as the result of other diencephalic lesions such as tumours and trauma (Von Cramon et al., 1985; Armstrong, 1986). The requirement of thiamin in humans is 1.0–1.5 mg/day (Freeman, 1979) and body stores can be depleted within about 3 weeks. The major food sources of thiamin are cereal products. Thiamin transport across the blood brain barrier occurs at a rate similar to that calculated for thiamin turnover in the brain suggesting that thiamin transport may be just sufficient to meet cerebral requirements (Butterworth, 1989). Thiamin deficiency is often noted in alcoholic populations (Darnton-Hill & Truswell, 1990), probably because alcohol interferes with normal thiamin metabolism (Rindi et al., 1987). Thiamin levels are reduced in the brain in human and animal studies of alcohol toxicity and thiamin deficiency (Abe & Itokawa, 1977; Summers et al., 1991). The main role of thiamin is that of a coenzyme in the form of thiamin pyrophosphate. In the central

429

430


nervous system there are four thiamin-dependent enzymes which largely relate to glucose metabolism; pyruvate dehydrogenase, -ketoglutarate dehydrogenase, transketolase and branched-chain -ketoacid dehydrogenase (Heroux & Butterworth, 1992). Thiamin also plays a role in nerve conduction and probably membrane transport, independent of its function as an enzyme cofactor. It has been shown that acute Wernicke’s encephalopathy (WE) can be precipitated when glucose loads are given to patients with thiamin deficiency (Harper, 1980). Thus, there is an intimate relationship between alcohol use and thiamin deficiency. It is interesting to note that, in published reports of pure thiamin deficiency, unaccompanied by chronic and excessive consumption of alcohol, the rate of progression to Korsakoff’s syndrome is low (Homewood & Bond, 1999). An additional confounding factor is that there are many cases of chronic WE (diagnosed pathologically) who do not have the characteristic clinical syndrome of chronic mental impairment in which memory deficits over-shadow other aspects of cognitive failure (Korsakoff’s psychosis). The clinical diagnosis of Korsakoff’s psychosis in alcoholics has only recently been standardized (Caine et al., 1997; Sullivan et al., 1999). The amnestic syndrome in Korsakoff’s psychosis is characterized by persistent anterograde episodic memory loss and preserved semantic memory, intelligence and learned behaviour (Squire et al., 1993). The acute syndrome is more straightforward. Patients may develop acute confusion, impairment of consciousness, ophthalmoplegia, nystagmus and ataxia and examination of the brain shows characteristic pathological changes in specific regions. It should be noted that the full clinical triad as outlined above is the exception rather than the rule and many patients will present with only one or two of the clinical signs (Harper et al., 1986). Hepatic coma from end-stage alcoholic cirrhosis can also mask the diagnosis. Kril and Butterworth (1997) showed that 30% of alcoholics in hepatic coma had the characteristic pathological lesions of WE but only two of the nine cases had been diagnosed during life. The diagnosis is being made more frequently with the aid of neuroimaging techniques in both acute and chronic cases (Ashikaga et al., 1997; Sullivan et al., 1999). It has been found that proton-density and T2 -weighted images are most useful for the diagnosis of acute Wernicke’s encephalopathy (Mascalchi et al., 1999). Studies have confirmed the usefulness of measuring mamillary body volumes in the diagnosis of chronic WE although there are other causes of shrinkage including Alzheimer’s disease (Charness, 1999; Sheedy et al., 1999). Proposed mechanisms of brain damage in thiamin deficiency include compromised cerebral energy metabolism

(Butterworth, 1989) and focal accumulation of lactate and decreased pH (Hakim & Pappius, 1983). Two other mechanisms which may play a role and which have received a lot of attention include excitotoxicity and free radicals. Several authors have drawn comparisons between the pathological changes seen in experimental WE and excitotoxic neuronal damage (Langlais & Mair, 1990; Lancaster, 1992). The link between thiamin deficiency and the excessive accumulation of excitatory amino acids is unclear but several factors may play a role. First, an important function of astrocytes is the re-uptake of excitatory amino acids. Since these cells appear to be a principle target in the pathology of thiamin deficiency, their damage may permit the accumulation of higher than normal concentrations of excitatory amino acids in the extracellular fluid. Secondly, in the late stages of thiamin deficiency in both animal models and in humans, epileptic fits are quite common and this can result in excessive accumulation of excitatory amino acids. The most important data which suggests that excitotoxicity plays a role is that the pathological changes in experimental models can be attenuated by the use of antagonists to the excitotoxins (Langlais & Zhang, 1993). Langlais and his colleagues (Langlais et al., 1994) suggested that histamine release may also contribute to glutamate-N-methyl-daspartate (NMDA)-mediated excitotoxic neuronal death in thiamin deficiency. The characteristic topography of brain lesions in WE might also be explained on the basis of excitotoxicity. Glutamic acid decarboxylase (GAD) protects most regions of the brain from glutamate that accumulates when the activity of -ketoglutarate dehydrogenase (a thiamin-dependent enzyme complex) is reduced. It has been proposed that during severe thiamin deficiency, glutamate accumulates in GAD-free peripheral tissues and reaches a concentration in blood at which it passes through circumventricular organs into the cerebral ventricles or contiguous brain and finally diffuses into the extracellular space of proximate diencephalic and brain stem tissues causing the characteristic lesions of WE (McEntee, 1997). Currently, the excitotoxic mechanism does not explain the specific susceptibility of the glial and vascular endothelial cells in thiamin deficiency. Cavanagh and his colleagues have found that nicotinamide analogues and nitroheterocyclic compounds can cause neuropathological changes similar to acute WE in rats (Cavanagh, 1988; Romero et al., 1991). The glio-vascular lesions in the brainstem caused by 1, 3-dinitrobenzene are accompanied by regional increases in blood flow similar to those described by Hakim (1986) in his animal models of acute thiamin deficiency. Romero and his colleagues (Romero et al., 1991)


postulate that these changes may be caused by free radical generation from the interaction of the 1, 3-dinitrobenzene with xanthine oxidase, which is found in high concentration in endothelial cells. The evolution of the disease can be documented both clinically and pathologically. In pathological studies of Wernicke–Korsakoff syndrome, approximately 17% of the cases are acute, 66% of the cases are chronic and there is a significant proportion of cases (17%) which show acute on chronic changes (Harper, 1983). These patients may have suffered repeated episodes of thiamin deficiency, some of which may have been subclinical (Lishman, 1981). In a Scandinavian study of 45 cases 53% were acute and 47% were chronic (Torvik, 1987). The topography of the lesions in acute WE were described in great detail by Victor and his colleagues in 1971(Victor et al., 1971). However, quantitative analyses of cortical and subcortical regions such as basal nucleus of Meynert, locus ceruleus and raphe nuclei were not reported. Damage to these latter regions, which project widely throughout the brain, could cause significant memory and cognitive deficits (Wilcock et al., 1988). Studies of raphe nuclei have shown a significant reduction (56%) in numbers of serotonergic neurons in alcoholics with both WE and Korsakoff’s psychosis (Halliday et al., 1995). However, no correlation was shown between the severity of the neuronal loss and memory impairment. Lesions in the noradrenergic locus ceruleus are said to cause impairment of attention and information processing. There may also be links with learning and memory as indicated by experimental models (Squire et al., 1993). Several groups of workers have emphasized the importance of this nucleus and its noradrenergic pathways in alcoholics with the Wernicke–Korsakoff syndrome (Mair et al., 1979; McEntee & Mair, 1990). Victor et al. (1971) noted abnormalities in the locus ceruleus in 19 of the 28 cases studied (67.9%). Various other groups have studied the locus coeruleus in a quantitative fashion but there are contradictory data and further studies will be necessary to determine whether or not neurons in the locus ceruleus are affected in the Wernicke–Korsakoff syndrome (Arango et al., 1994; Halliday & Baker, 1996). Cullen and colleagues (Cullen et al., 1997) found that numbers of magnocellular neurons in the cholinergic nucleus basalis are reduced in thiamin deficient alcoholic patients but cell loss is minor and does not account for the profound amnesia of Korsakoff’s psychosis (although the nonamnesic patient with the greatest cell loss was impaired on attentional tasks). An interesting observation is that the nucleus basalis is an exclusive site of neurofibrillary degeneration in alcoholic patients with WE (Cullen & Halliday, 1995).

Neuropathology The mamillary bodies are almost invariably affected and the structures around the third and fourth ventricles and aqueduct are frequently involved (Harper, 1979; Victor et al., 1989). The diagnosis is not always evident on macroscopic examination of the brain. In studies by two different groups the diagnosis was suspected on gross examination in only 75% of the cases (Harper, 1983; Torvik, 1987). The most easily recognized acute lesions are haemorrhages, often petechial or in some cases only seen on microscopic examination. In acute cases red cells are often present in the perivascular spaces, suggesting that the vessel walls are leaky. The tissues appear oedematous and spongy. Silver impregnation methods (e.g. modified Bielschowsky) show disruption of axons and axonal irregularities. After several days the endothelial cells become plump with active looking nuclei and they proliferate so that the lumina of the vessels become narrowed (Fig. 18.1). There is evidence of

Fig. 18.1. Acute Wernicke’s encephalopathy with endothelial proliferation and luminal narrowing (arrow) in the mamillary body. Haematoxin and eosin × 125.

431

432


Fig. 18.2. Acute Wernicke’s encephalopathy with necrotizing lesions in the medial thalamus (arrows). There is relative sparing of the subependymal tissues. III: third Ventricle, T: thalamus, RN: red nucleus. Nissl ×2.

a breakdown in the blood brain barrier in experimental animal models (Phillips & Cragg, 1984). Capillary budding commences and by about 7–10 days the affected region has a highly vascular appearance. Rosenblum and Feigin (1965) examined the question of haemorrhage in Wernicke’s encephalopathy and concluded that patients who develop severe and extensive haemorrhages are likely to have an additional cause such as a bleeding diathesis due to cirrhosis or uraemia. Such lesions are, in most cases, incompatible with survival and may be the cause of sudden death (Harper, 1979, 1980). It has been said that there is a relative sparing of the neurons in acute WE. In the less severe cases this may be correct but there is sometimes frank necrosis of tissues and destruction of neurons. The most common site for the necrotizing lesion is the medial thalamus as demon-

strated by Victor and his colleagues (Victor et al., 1989) in 6 of 45 cases (Fig. 18.2). There is sparing of the immediate subependymal tissues. Several authors have drawn our attention to the fact that a second pattern of pathology is seen in the thalamic nuclei and olives (Torvik, 1985; Byrne et al., 1993). In these regions the neurons appear to be the principle target and show changes which resemble acute ischaemic cell change with eosinophilia of the cytoplasm and ultimately, neuronal disintegration. The endothelial and capillary changes are much less florid than in the mamillary body. It may be that the mechanism underlying this neuronal change is excitotoxic as discussed above. In the periventricular lesions one rarely sees any form of inflammatory cell infiltrate. The astrocytes are slow to react and are not easily recognized until the vascular changes begin to subside. They are more evident in the chronic stages of the disease. Myelin stains show some generalized pallor but there is no definite demyelination. In chronic Wernicke–Korsakoff syndrome the characteristic macroscopic lesion is shrinkage and brown discolouration of the mamillary bodies. Microscopically the mamillary bodies show spongy degeneration and a relative excess of small blood vessels which can be highlighted by the use of reticulin stains (Fig. 18.3). There is some loss of neurons and gliosis. Occasionally there are haemosiderin-laden macrophages, which are evidence of previous haemorrhages. Neuronal loss, gliosis and pallor of myelin staining are the principal findings in thalamic nuclei and periventricular regions. The thalamic lesions tend to be more severe in mediodorsal, centro-medial and anterior nuclei (Victor et al., 1989). Until recently, pathological studies did not adequately explain the difference between the clinical entities of WE and Korsakoff’s psychosis although lesions in the thalamus were thought to correlate best with the amnesia seen in Korsakoff’s psychosis (Victor et al., 1971; Mair et al., 1979; Mayes et al., 1988). Damage to the mediodorsal nucleus was first correlated with amnesia by Victor et al. (1971). Harding and his colleagues have now shown that degeneration of anterior thalamic nuclei differentiates those alcoholics with amnesia (Harding et al., 2000). The number of neurons in the mamillary nuclei and the anterior and mediodorsal thalamic nuclei were estimated using unbiased stereological techniques. Neurodegeneration of the hypothalamic mamillary nuclei and the mediodorsal thalamic nuclei was substantial in both non-amnesic and amnesic alcoholics with WE. However, neuronal loss in the anterior thalamic nuclei was found consistently only in alcoholic Korsakoff’s psychosis indicating that this is the anatomical substrate


for the amnestic state (Harding et al., 2000). Data from a recent quantitative study of the cerebellum confirm significant changes in alcoholics with the Wernicke–Korsakoff syndrome. In the cerebellar vermis there was a decrease in Purkinje cell density (reduced on average by 43%) (Baker et al., 1999), confirming previous findings (Phillips et al., 1990). There was also a 36% reduction in estimated Purkinje cell numbers in the flocculi (Baker et al., 1999). Baker and colleagues correlated the severity of cerebellar damage and clinical signs – there was a 36% loss of Purkinje cells in the lateral lobes in alcoholics with mental state signs. This correlation is of particular interest given recent data showing the importance of the cerebellum in the organization of higher order cerebral functions (Schmahmann & Sherman 1998). As discussed in the previous section dealing with alcoholic dementia, brain weight and volume studies have shown that there is a significant loss of tissue, particularly in those alcoholics with the Wernicke–Korsakoff syndrome (Harper & Blumbergs 1982; Torvik et al., 1982). The loss of tissue is largely from the white matter of the cerebral hemispheres (Harper & Kril 1985; De la Monte, 1988) and some of this white matter loss appears to be reversible after prolonged abstinence (Carlen & Wilkinson 1987; Shear et al., 1994). Jernigan and her colleagues (Jernigan et al., 1991) have analysed MRI scans of patients with the Wernicke–Korsakoff syndrome and have shown a reduction of cerebral cortical volume. Kril and her colleagues have shown a 22% reduction in the number of neurons in the superior frontal cortex. However, analysis of counts in the different alcoholic groups revealed that there is no significant difference between those alcoholics with the Wernicke–Korsakoff syndrome or cirrhosis of the liver and the uncomplicated alcoholics (Kril & Harper, 1989; Kril et al., 1997). In a recent study of the prevalence of the Wernicke–Korsakoff syndrome in 2212 forensic autopsies there were 130 cases affected with a history of alcoholrelated problems and 12 affected (9.2%) had sclerosis of the hippocampus (Harper et al., 1998a). Other studies of the hippocampus in the Wernicke–Korsakoff syndrome have shown a reduction in the volume of the hippocampus but this change could be explained on the basis of loss of white matter (Harding et al., 1997; Kril et al., 1997) as there was no evidence of neuronal loss in either of these studies. Similarly, there is a significant reduction in the volume of the amygdala complex in alcoholics with Wernicke– Korsakoff syndrome but no neuronal loss (Kril et al., 1997). It should also be noted that sudden unexpected death occurs commonly in alcoholic patients with the Wernicke–

Fig. 18.3. Chronic Wernicke’s encephalopathy. There is an increase in the number of blood vessels (arrows) associated with shrinkage and spongiosis of the mamillary body (MB). Reticulin ×10.

Korsakoff syndrome (Harper, 1979, 1980). In one study, 20% of Wernicke–Korsakoff syndrome cases died suddenly and there was no evident cause of death at autopsy. Eighty per cent of the cases were acute with lesions involving vital cardiorespiratory centres in the brainstem that could have caused their sudden death (Harper, 1980). Alternatively, there may be a link between sudden death and the autonomic dysfunction that is well documented in these cases (Johnson & Robinson, 1988). Thus, as with many other diseases of the central nervous system, the complexity of the pathology of the Wernicke–Korsakoff syndrome increases as we look more closely at brain structures and circuitry with improved technologies. Moreover, we must use a multidisciplinary approach, each discipline providing new data and ideas which give directions for further research.

433

434


Hepatic encephalopathy The clinical manifestations of hepatic encephalopathy or portal systemic encephalopathy range from minimal changes in personality and motor activity, to overt deterioration of intellectual function, decreased consciousness and coma, and appear to reflect primarily a variable imbalance between excitatory and inhibitory neurotransmission. A flapping tremor (asterexis) is often elicitable if looked for carefully (Sherlock, 1981). Deep tendon reflexes are usually increased, as is muscle tone. As the coma deepens, the patients may become flaccid. This clinical syndrome is largely reversible with improvement of liver function and reduction of blood ammonia levels. As may therefore be anticipated, pathological changes are relatively minor and, based on experimental studies, appear to be largely reversible. Patients should be carefully evaluated to ensure that other alcohol-related diagnoses are not missed. Kril and Butterworth showed that seven of thirty patients who died while in hepatic coma also had undiagnosed Wernicke’s encephalopathy (Kril & Butterworth 1997). Macroscopically the brain usually appears normal although in fulminant hepatic failure acute brain swelling is a common cause of death (Tholen, 1971; Blei, 2000a). With the improved management of hepatic failure, this is now seen less commonly but is a feature of failed liver transplant cases and some cases of acute viral hepatitis. While dysfunction of the blood-brain barrier cannot be excluded (Ede & Williams, 1986), astroglial swelling appears to be the factor responsible for brain oedema in fulminant hepatic failure (Cordoba & Blei, 1996). The mechanism for the ammonia-induced astrocytic swelling remains unclear but the generation of excessive osmolytes, chiefly glutamine and possibly glutathione are thought most likely to be responsible (Norenberg, 1996; Albrecht & Dolinska, 2001). Ultrastructural studies show swelling of endothelial cells and perivascular astrocytes and astrocytes have swollen mitochondria and dilation of endoplasmic reticulum (Kato et al., 1992). Portal systemic encephalopathy (PSE) is the most common form of hepatic encephalopathy. It accompanies the development of portal-systemic collaterals arising as a result of portal hypertension secondary to liver cirrhosis. The use of transjugular intraphepatic portal systemic shunts (TIPS) for the prevention of oesophageal variceal bleeding may precipitate PSE in 30% of patients, particularly the more elderly (Conn, 1993). The characteristic pathological change in PSE is the presence of astrocytes with enlarged pale nuclei, marginated chromatin and often prominent nucleoli. These cells were

described first by von Hosslin and Alzheimer, (von Hosslin and Alzheimer, 1912) and are now called Alzheimer type II astrocytes. A morphometric study showed that the astrocytic nuclei have a mean surface area of 60 m2 compared to a normal measurement of 39 m2 (Martin et al., 1987). In patients with long-standing hepatic encephalopathy, the astrocyte nuclei also contain glycogen inclusion bodies. The nuclei are often lobulated or bean shaped (Fig. 18.4). Norenberg (1994), in a review of astrocyte responses to CNS injury, claimed that lobulaton is more prominent in certain brain regions including pallidum, substantia nigra and dentate nucleus. With conventional stains, there is no visible cytoplasm, although an excess of lipofuscin can sometimes be identified. There is a marked diminution of the amount of glial fibrillary acidic protein staining using immunohistochemical techniques (Sobel et al., 1981). Characteristically the astrocyte nuclei occur in pairs, suggesting cellular division, although mitoses are rarely seen (Norenberg et al., 1990). These Alzheimer II astrocytes are seen most easily in specific areas of the central nervous system (pontine nuclei of the basis pontis, Ammon’s horn, lower layers of the cortex just above the cortical–white matter interface and in the basal ganglia and dentate nucleus). Only a proportion of astrocytes are affected, and the change is rarely seen in the white matter. There appears to be a correlation between the severity of the astrocytic change and the severity of the hepatic failure, although this has not been well documented scientifically. In the original description of the pathology by von Hosslin and Alzheimer (1912), a second astrocytic change was noted – the Alzheimer I astrocyte. There has been considerable debate as to whether or not this cell exists but in a recent review Ma has provided evidence of a link between the Alzheimer I and Alzheimer II astrocytes (Ma, 2001). Although ammonia is still thought to be the leading toxin influencing brain function in this condition, endogenous benzodiazepines and cytokines may contribute to astrocyte swelling (Blei, 2000a). The abnormal astrocytes also manifest altered expression of several key proteins and enzymes including monoamine oxidase B, glutamine synthetase, nNOS and peripheral-type benzodiazepine receptors (Hazell and Butterworth, 1999). Alzheimer II astrocytes are also seen in other disease states which result in an hyperammonaemia (e.g. inherited metabolic disorders of the urea cycle) and in uraemia, hypercapnia and in the early stages of anoxia (Norenberg, 1994). There is little evidence to suggest significant neuronal loss in portal systemic encephalopathy. Kril and Harper (1989) found that cirrhosis of the liver does not accentuate the loss of cortical neurons from the superior frontal cortex in alcoholics. Other histopathological studies reveal


Fig. 18.4. Hepatic encephalopathy showing Alzheimer type II astrocytes in pons (arrows). Note the changes are more pronounced in grey matter (GM) (pontine nucleus) than in white matter (WM). Haematoxin and eosin ×125.

no evidence of neuronal damage and the measurement of neuronal marker enzymes (Lavoie et al., 1987) and furthermore, specific binding sites for postsynaptic neuronal ligands (Butterworth et al., 1988) provide no evidence of neuronal loss. The basic effect of ammonia on the neuronal membrane is inhibition of chloride extrusion and neuronal depolarization. This disturbs the interaction between individual neurons and the processing of information in neuronal circuits (Rabbe, 1990). The mechanism of ammonium chloride (NH4 Cl) neurotoxicity involves interruption of oxidative metabolism. This leads to decreased levels of ATP concentration and subsequent degradation of glial fibrillary acidic protein (Haghighat et al., 2000). However, evidence from both biochemical measurements and from noninvasive techniques suggests that neurotransmission failure rather than primary energy failure is the major cause of hepatic encephalopathy (Albrecht & Jones, 1999). Neurotransmitter systems in which abnormalities have been identified include the glutamatergic, monoaminergic and opioid systems. Two major factors play a role in the dysfunction of the glutamatergic system. There is downregulation of Glu receptors following excessive extrasynaptic accumulation of Glu resulting from impaired re-

uptake into nerve endings and astrocytes. Secondly, there is an increase in inhibitory neurotransmission by gammaaminobutyric acid (GABA). There are several recent reviews of the complex neurotransmitter system changes in portal systemic encephalopathy (Albrecht & Jones, 1999; Hazell & Butterworth, 1999; Albrecht & Dolinska, 2001). It has also been suggested that alterations of serotonin transport (Michalak et al., 2001) and the histaminergic system (Lozeva et al., 2001) may be implicated in the pathogenesis of the neuropsychiatric symptoms. Further elucidation of these neurotransmitter changes could provide novel pharmacological approaches in the treatment of hepatic encephalopathy. Brumback and Lapham (1989), in a study of rats injected with methionine sulfoximine (which causes an elevation in brain ammonia levels), noted the development of Alzheimer type II astrocytes as soon as 18 hours after injection. There was also labelling of these astrocytes by tritiated thymidine given to the rats at the same time as the methionine sulfoximine. This suggests that astrocytes undergo DNA replication and division early in the course of hepatic encephalopathy as part of the Alzheimer type II change. Earlier studies found no increase in the total

435

436


number of glial cells (Diemer, 1978) and although a recent study found an increase in astrocyte cell areas in rats administered with a hepatotoxin (Matsushita et al., 1999), whether or not there is a true astrocytosis still remains unclear. There are also changes in brain amino acids. Therrien and Butterworth (1991), in a study of portacaval shunted rats injected with ammonium acetate, observed increases in the cerebro spinal fluid concentrations of glutamine, glutamate, alanine, phenylalanine, and tyrosine. GABA levels were unchanged. Of the amino acids studied, alanine best reflected the progression of neurological dysfunction. Magnetic resonance imaging (MRI) in cirrhotic patients with PSE reveal bilateral signal hyperintensity of T1 weighted images in the globus pallidus (Kulisevsky et al., 1992; Naegele et al., 2000). Direct measurement of tissue removed from the globus pallidus (at autopsy) from patients with chronic liver disease who died in hepatic coma reveal two- to seven-fold increases of pallidal manganese. This suggests that pallidal magnetic resonance signal hyperentensity is the result of manganese deposition (Butterworth et al., 1995). In positron emission tomography (PET) studies of patients with portal systemic encephalopathy, brain ammonia utilization is increased 2–3 fold and the blood–brain barrier appears to be more permeable to ammonia (Lockwood et al., 1991). There is a selective loss of dopaminergic D2 receptors in the globus pallidus of these patients (Mousseau et al., 1993) and it has been suggested that these findings reflect dopaminergic neuronal dysfunction which could contribute to the extrapyramidal signs in these patients. Hamuro and colleagues suggest that proton MR spectroscopy (MRS) may be useful for the early detection of hepatic encephalopathy (Hamuro et al., 2000); others find it useful for understanding pathogenesis but not for clinical diagnosis (Kostler, 1998; Rovira et al., 2001).

Pellagra Pellagra is a disease caused primarily by a deficiency of niacin (nicotinic acid). Deficiencies of the amino acid tryptophan, a niacin precursor, may also produce the syndrome. Pellagra is much less common today when compared with the epidemic prevalence during the early part of the twentieth century. The diminished incidence is in large part related to public health programmes whereby niacin has been added to staple foods such as flour and bread (Park et al., 2000). Currently pellagra is seen most commonly in the alcoholic population and in patients being treated with isoniazid for tuberculosis (Ishii & Nishihara 1981, 1985) or chemotherapeutic agents. Pyridoxine supplementation has been used to reduce the risk of isoniazid-

induced pellagra, however, this is not always effective (Darvay et al., 1999). Deaths attributable to pellagra are said to be two-fold higher in women than men and it seems that oestrogens may be partly responsible for that as they inhibit the synthesis of niacin from tryptophan (Shibata & Toda, 1997). Pellagra is more likely to occur in patients who are vegetarians or vegans. Defective intestinal absorption of niacin is sometimes the cause of secondary pellagra which may occur even though there is adequate niacin in the diet. In standard medical texts the clinical features of pellagra are listed as dermatitis, dementia and diarrhoea (Wilson, 1991), however, clinical papers from as early as the 1940s emphasized that neuropsychiatric features were often the only manifestation of the disorder (Jolliffe et al., 1940; Joliffe, 1941; Gottlieb, 1944). In most cases the evidence for this was based upon the observed clinical improvement following nicotinic acid therapy (most patients were also given thiamine) and pathological confirmation was the exception. In a more recent study by Serdaru et al. (1988) the cases of pellagra were identified pathologically and then the hospital clinical notes were reviewed. The findings echoed warnings that in the more acute cases of nicotinic acid deficiency the skin and gastrointestinal manifestations of the disease are often absent and may only develop after a prolonged deficiency state allowing structural changes to develop (Jolliffe, 1941; Gottlieb, 1944). The characteristic clinical picture in the 22 cases described by Serdaru et al. (1988) were confusion and/or clouding of consciousness (100%), oppositional hypertonia (100%) and myoclonic jerking (55%). Fluctuations in levels of mentation and consciousness are also a feature. The hypertonia was seen days to weeks before the development of coma. The myoclonus tends to affect face and shoulders. Many patients also had evidence of involvement of peripheral nerves with a burning sensation in the hands and feet and numbness. Four of the cases also had Wernicke–Korsakoff syndrome, eight had Marchiafava-Bignami disease and one had both. This finding emphasises the close interrelationship between the vitamin B group deficiency states. As suggested by Lishman (1981), it may well be that treatment using one of the deficient vitamins (e.g. thiamine only) may precipitate say pellagra by increasing the metabolic demand and requirement for nicotinic acid. The brain is macroscopically normal. Microscopically the characteristic feature of pellagra is chromatolytic change in neurons. In particular regions of the central nervous system there is ballooning and enlargement of affected neurons and an apparent clearing of the cytoplasm due to loss of nissl substance (rough endoplasmic reticulum). The nuclei become eccentrically placed and appear smaller than normal. No changes have been noted in the


glia, myelin, blood vessels or meninges. The topography of these changes varies to some extent depending upon whether the pellagra is sporadic, endemic or induced by isoniazid. In the latter two cases the cortical neurons, especially the Betz cells are always heavily affected. Hauw and his colleagues (Hauw et al., 1988) deal with this in great detail. They showed that in the sporadic alcoholic cases the central chromatolysis is seen predominantly in the brainstem, especially the pontine nuclei and the nuclei of cranial nerves (third, sixth, seventh and eighth). The reticular nuclei, arcuate nuclei, posterior horn cells in the spinal cord and cerebellar dentate nuclei were also frequently affected. Other brain stem and spinal cord changes were described in detail by Hauw et al. (1988). One of us (CH) has had the opportunity of reviewing several of Hauw et al’s (1988) cases, and although seemingly straight forward, it is evident that the chromatolytic change in neurons typical of pellagra could be easily overlooked by even experienced neuropathologists. In our own neuropathological studies of many hundreds of alcoholic cases in Australia we have not identified a single case of pellagra. The explanation for this is unclear but may relate to governmental laws on supplementation of foods with nicotinic acid. In France there is no provision for vitamin supplementation of flour, whereas in Australia it is recommended that nicotinic acid and other vitamins be added to flour and bread (Axford & Williams, 1981). Potential false positive diagnoses could be made as a result of several factors. It should be noted that, in some regions (e.g. third and fourth cranial nerve nuclei), normal neurons often have eccentrically placed nuclei. Excessive lipofuscin in neurons (e.g. olivary nuclei) can also mimic chromatolysis and ischaemic cell change can cause confusion.

REFERENCES Abe, T. & Itokawa, Y. (1977). Effect of ethanol administration on thiamin metabolism and transketolase activity in rats. International Journal of Nutrition Research, 47:307–14. Albrecht, J. & Dolinska, M. (2001). Glutamine as a pathogenic factor in hepatic encephalopathy. J Neurosci Res, 65:1–5. Albrecht, J. & Jones, E. A. (1999). Hepatic encephalopathy: molecular mechanisms underlying the clinical syndrome. Journal of Neurological Sciences, 170:138–46. Arango, V., Underwood, M. D. & Mann, J. J. (1994). Fewer pigmented neurons in the locus coeruleus of uncomplicated alcoholics. Brain Research, 650:1–8. Arendt, T., Allen, Y., Sinden, J. et al. (1988). Cholinergic-rich transplants reverse alcohol-induced memory deficits. Nature, 332:448–50.

Armstrong, E. (1986). Enlarged limbic structures in the human brain: the anterior thalamus and medial mamillary body. Brain Research, 362:394–7. Ashikaga, R., Araki, Y., Ono, Y. et al. (1997). FLAIR appearance of Wernicke encephalopathy. Radiation Medicine, 15:251–3. Axford, D. W. E. & Williams, D. A. (1981). Flour enrichment around the world. British Flour Milling and Baking Research Association, 4:156–63. Baker, K. G., Harding, A. J., Halliday, G. M. et al. (1999). Neuronal loss in functional zones of the cerebellum of chronic alcoholics with and without Wernicke’s encephalopathy. Neuroscience, 91:429–38. Blei, A. T. (2000a). Brain edema and portal-systemic encephalopathy. Liver Transplant, 6:14–20. (2000b). Diagnosis and treatment of hepatic encephalopathy. Baillières Best Practice Research Clinical Gastroenterology, 14:959–74. Brumback, R. A. & Lapham, L. W. (1989). DNA synthesis in Alzheimer type II astrocytosis. The question of astrocytic proliferation and mitosis in experimentally induced hepatic encephalopathy. Archives of Neurology, 46:845–8. Butters, N. E., Granholm, E., Salmon, D. P. et al. (1987). Episodic and semantic memory: a comparison of amnesic and demented patients. Journal of Clinical and Experimental Neuropsychology, 9:479–97. Butterworth, R. (1989). Effects of thiamin deficiency on brain metabolism: implications for the pathogenesis of the Wernicke– Korsakoff syndrome. Alcohol and Alcoholism, 24:271–9. Butterworth, R. F. J., Lavoie, J., Giguere, J. F. et al. (1988). Affinities and densities of GABA-A receptors and of central benzodiazepine receptors are unchanged in autopsied brain tissue from cirrhotic patients with hepatic encephalopathy. Hepatology, 8:1084–8. Butterworth, R. F. J., Spahr, L., Fontaine, S. et al. (1995). Manganese toxicity, dopaminergic dysfunction and hepatic encephalopathy. Metabolic Brain Disease, 10:259–67. Byrne C., Halliday G., Ellis J. et al. (1993). Thalamic vacuolation in acute Wernicke’s encephalopathy. Metabolic Brain Disease, 8:107–13. Caine, D., Halliday, G. M., Kril, J. J. et al. (1997). Operational criteria for the classification of chronic alcoholics: identification of Wernicke’s encephalopathy. Journal of Neurology, Neurosurgery and Psychiatry, 62:51–60. Carlen, P. L. & Wilkinson, D. A. (1987). Reversibility of alcoholrelated brain damage: clinical and experimental observations. Acta Medica Scandinavica, 222:19–26. Carlen, P. L., McAndrews, M. P., Weiss, R. T. et al. (1994). Alcoholrelated dementia in the institutionalized elderly. Alcoholism: Clinical and Experimental Research, 18:1330–4. Cavanagh, J. B. (1988). Lesion localisation: implications for the study of functional effects and mechanisms of action. Toxicology, 49:131–6. Charness, M. E. (1999). Intracranial voyeurism: revealing the mammillary bodies in alcoholism. Alcoholism: Clinical and Experimental Research, 23:1941–4.

437

438


Charness M. E., Simon, R. P. & Greenberg, D. A. (1989). Ethanol and the nervous system. New England Journal of Medicine, 321:442–54. Conn, H. O. (1993). Transjugular intrahepatic portal–systemic shunts: the state of the art. Hepatology, 1:148–58. Cordoba, J. & Blei, A. T. (1996). Brain edema and hepatic encephalopathy. Seminars in Liver Disease, 16:271–80. Cragg, B. & Phillips, S. (1983). Toxic effects of alcohol on brain cells and alternative mechanisms of brain damage in alcoholism. Australian Drug and Alcohol Review, 2:64–70. Cullen, K. M. & Halliday, G. M. (1995). Tangles in chronic alcoholics. Neuropathology and Applied Neurobiology, 21:312–18. Cullen, K. M., Halliday, G. M., Caine, D. et al. (1997). The nucleus basalis (Ch4) in the alcoholic Wernicke–Korsakoff syndrome: reduced cell number in both amnesic and non-amnesic patients. Journal of Neurology, Neurosurgery and Psychiatry, 63:315–20. Darnton-Hill, I. & Truswell, A. S. (1990). Thiamin status of a sample of homeless clinic attenders in Sydney. Medical Journal of Australia, 152:5–9. Darvay, A., Basarab, T., McGregor, J. M. et al. (1999). Isoniazid induced pellagra despite pyridoxine supplementation. Clin Exp Dermatol, 24:167–9. De la Monte, S. M. (1988). Disproportionate atrophy of cerebral white matter in chronic alcoholics. Archives of Neurology, 45:990–2. Diemer, N. H. (1978). Glial and neuronal changes in experimental hepatic encephalopathy. Acta Neurologica Scandinavica, 58: 1–144. Ede, R. J. & Williams, R. W. (1986). Hepatic encephalopathy and cerebral edema. Seminars in Liver Disease, 6:107–18. Figueroa, W. G., Sargent, F., Imperiale, L. et al. (1953). Lack of avitaminosis amongst alcoholics. Journal of Clinical Nutrition, 1:179–99. Frances, A. (1994). Diagnosis and Statistical Manual of Mental Disorders, 4th edn. Washington, DC: American Psychiatric Press. Freeman R. M., Ed. (1979). Rational use of vitamins in practice. In Quick Reference to Clinical Nutrition. A Guide for Physicians. Toronto: J. B. Lippincott. Freund, G. & Ballinger, W. E. (1988). Decrease of benzodiazepine receptors in frontal cortex of alcoholics. Alcohol, 5:275–82. Gottlieb, B. (1944). Acute nicotinic acid deficiency (anioacinosis). British Medical Journal, 1:392–3. Haghighat, N., McCandless, D. W., Geraminegad, P. et al. (2000). Responses in primary astrocytes and C6-glioma cells to ammonium chloride and dibutyryl cyclic-AMP. Neurochemistry Research, 25:277–84. Hakim, A. M. (1986). Effect of thiamine deficiency and its reversal on cerebral blood flow in the rat. Observations on the phenomena of hyperfusion, ‘no reflow’, and delayed hypoperfusion. Journal of Cerebral Blood Flow and Metabolism, 6:79–85. Hakim, A. M. & Pappius, H. M. (1983). Sequence of metabolic, clinical, and histological events in experimental thaimine deficiency. Annals of Neurology, 13:365–75.

Halliday, G. & Baker, K. (1996). Noradrenergic locus coeruleus neurons [letter]. Alcoholism: Clinical and Experimental Research, 20:191–2. Halliday, G., Baker, K. & Harper. C. (1995). The role of serotonin in alcohol-related brain damage. Metabolic Brain Disease, 10:25– 30. Hamuro, M., Nakamura, K., Matsuoka, T. et al. (2000). Proton MR spectroscopy of the brain in patients treated with TIPS. Acta Radiologica, 41:567–71. Harding, A., Halliday, G. M., Caine, D. et al. (2000). Degeneration of anterior thalamic nuclei differentiates alcoholics with amnesia. Brain, 123:141–54. Harding, A. J., Halliday, G. M., Ng J. L. et al. (1996). Loss of vasopressin-immunoreactive neurons in alcoholics is doserelated and time-dependent. Neuroscience, 72:699–08. Harding, A. J., Wong, A., Svoboda, M. et al. (1997). Chronic alcoholic consumption does not cause hippocampal neuron loss in humans. Hippocampus, 7:78–87. Harper, C. (1979). Wernicke’s encephalopathy: a more common disease than realised. Journal of Neurology, Neurosurgery and Psychiatry, 42:226–31. Harper, C. & Corbett, D. (1990). A quantitative Golgi study of cortical neurons from alcoholic patients. Journal of Neurology, Neurosurgery and Psychiatry, 53:856–61. Harper, C. & Kril, J. (1985). Brain atrophy in chronic alcoholic patients: a quantitative pathological study. Journal of Neurology, Neurosurgery and Psychiatry, 48:211–17. (1989). Patterns of neuronal loss in the cerebral cortex in chronic alcoholic patients. Journal of Neurological Sciences, 92:81–9. Harper, C., Kril, J. & Daly, J. (1987). Are we drinking our neurones away? British Medical Journal, 294:534–6. Harper, C., Fornes, P., Duyckaerts, C. et al. (1995). An international perspective on the prevalence of the Wernicke–Korsakoff syndrome. Metabolic Brain Disease, 10:17–24. Harper, C. G. (1980). Sudden, unexpected death and Wernicke’s encephalopathy. A complication of prolonged intravenous feeding. Australian and New Zealand Journal of Medicine, 10:230–5. (1983). The incidence of Wernicke’s encephalopathy in Australia – a neuropathological study of 131 cases. Journal of Neurology, Neurosurgery and Psychiatry, 46:593–8. Harper, C. G. & Blumbergs, P. C. (1982). Brain weights in alcoholics. Journal of Neurology, Neurosurgery and Psychiatry, 45:838–40. Harper, C. G & Kril, J. J., eds. (1993). Neuropathological changes in alcoholics. In Alcohol-induced Brain Damage. Washington, DC, US Department of Health and Human Services. Harper, C. G., Giles, M. & Finlay-Jones, R. (1986). Clinical signs in the Wernicke–Korsakoff complex – a retrospective analysis of 131 cases diagnosed at necropsy. Journal of Neurology, Neurosurgery and Psychiatry, 49:341–5. Harper, C. G., Kril, J. J. & Daly, J. M. (1988). Brain shrinkage in alcoholics is not caused by changes in hydration: a pathological study. Journal of Neurology, Neurosurgery and Psychiatry, 51:124–7. Harper, C. G., Kril, J. & Sheedy, D. (1998a). Neuropathological studies: the relationship between alcohol and aging. In Alcohol


Problems and Aging. E. S. L. Gomberg, A. M. Hegedus & R. A. Zucker, pp. 117–34. Bethesda, Maryland: NIAAA Research Monograph. Harper, C. G., Sheedy, D. L., Lara, A. I. et al. (1998b). Prevalence of Wernicke–Korsakoff syndrome in Australia: has thiamine fortification made a difference? [see comments]. Medical Journal of Australia, 168:542–5. Hauw, J.-J., De Baecque, C., Hausser-Hauw, C. et al. (1988). Chromatolysis in alcoholic encephalopathies. Brain, 111:843–57. Hazell, A. S. & Butterworth, R. F. (1999). Hepatic encephalopathy: an update of pathophysiologic mechanisms. Proceedings of the Society for Experimental Biology and Medicine, 222:99–112. Hendrie, H. C., Gao, S., Hall, K. S. et al. (1996). The relationship between alcohol consumption, cognitive performance, and daily functioning in an urban sample of older black Americans. Journal of the American Geriaritrics Society, 44:1158–65. Heroux, M. & Butterworth, R., Eds. (1992). Animal models of neurological disease,II. In Neuromethods, The Humana Press Inc. Homewood, J. & Bond, N. W. (1999). Thiamin deficiency and Korsakoff’s syndrome: failure to find memory impairments following nonalcoholic Wernicke’s encephalopathy. Alcohol, 19:75–84. Ihara, M., Ito, T., Yanagihara, C. et al. (1999). Wernicke’s encephalopathy associated with hemodialysis: report of two cases and review of the literature [In Process Citation]. Clinical Neurology and Neurosurgery, 101:118–21. Ishii, N. & Nishihara, Y. (1981). Pellagra among chronic alcoholics: clinical and pathological study of 20 necropsy cases. Journal of Neurology, Neurosurgery and Psychiatry, 44:209–15. (1985). Pellagra encephalopathy among tuberculous patients: its relation to isoniazid theraphy. Journal of Neurology, Neurosurgery and Psychiatry, 48:628–34. Jensen, G. B. & Pakkenberg, B. (1993). Do alcoholics drink their neurons away? Lancet, 342:1201–4. Jernigan, T. L., Butters, N., DiTraglia, G. et al. (1991). Reduced cerebral grey matter observed in alcoholics using magnetic resonance imaging. Alcoholism: Clinical & Experimental Research, 15:418–27. Johnson, R. H. & Robinson, B. J. (1988). Mortality in alcoholics with autonomic neuropathy. Journal of Neurology, Neurosurgery and Psychiatry, 51:476–80. Jolliffe, N. (1941). Treatment of neuropsychiatric disorders with vitamins. Journal of the American Medical Association, 117:1496–502. Jolliffe, N., Bowman, K., Rosenblum, L. et al. (1940). Nicotinic acid deficiency encephalopathy. Journal of the American Medical Association, 114:307–9. Kato, M., Hughes, R. D., Keays, R. T. et al. (1992). Electron microscopic study of brain capillaries in cerebral edema from fulminant hepatic failure. Hepatology, 15:1060–6. Kostler, H. (1998). Proton magnetic resonance spectroscopy in portal–systemic encephalopathy. Metabolic Brain Disease, 13:291–301. Kril, J. J. & Butterworth, R. F. (1997). Diencephalic and cerebellar pathology in alcoholic and nonalcoholic patients with endstage liver disease. Hepatology, 26:837–41.

Kril, J. J. & Harper, C. G. (1989). Neuronal counts from four cortical regions of alcoholic brains. Acta Neuropathologica (Berl), 79:200–4. Kril, J. J., Halliday, G. M., Svoboda, M. D. et al. (1997). The cerebral cortex is damaged in chronic alcoholics. Neuroscience, 79:983–98. Kulisevsky, J., Pujol, J., Balanzo, J. et al. (1992). Pallidal hyperintensity on magnetic resonance imaging in cirrhotic patients: clinical correlations. Hepatology, 16:1382–8. Lancaster, F. E. (1992). Alcohol, nitric oxide and neurotoxicity: Is there a connection? – a review. Alcoholism: Clinical and Experimental Research, 16:539–41. Langlais, P. J. & Mair, R. G. (1990). Protective effects of the glutamate antagonist MK-801 on pyrithiamine-induced lesions and amino acid changes in rat brain. Journal of Neuroscience, 10:1664–74. Langlais, P. J. & Zhang, S. X. (1993). Extracellular glutamate is increased in thalamus during thiamine deficiency-induced lesions and is blocked by MK-801. Journal of Neurochemistry, 61:2175–82. Langlais, P. J., Zhang, S. X., Weilersbacher, G. et al. (1994). Rapid Communication: Histamine-mediated neuronal death in a rat model of Wernicke’s encephalopathy. Journal of Neuroscience Research, 38:565–74. Lavoie, J., Giguere, J. F., Layrargues, G. et al. (1987). Activities of neuronal and astrocytic marker enzymes in autopsied brain tissue from patients with hepatic encephalopathy. Metabolic Brain Disease, 2:283–90. Lishman, W. A. (1981). Cerebral disorders in alcoholism. Syndromes of impairment. Brain, 104:1–20. Liu, R. S., Lemieux, L., Shorvon, S. D. et al. (2000). Association between brain size and abstinence from alcohol. Lancet, 355:1969–70. Lockwood, A. H., Yap, E. W. H., Wong, W. H. et al. (1991). Cerebral ammonia metabolism in patients with severe liver disease and minimal hepatic encephalopathy. Journal of Cerebral Blood Flow and Metabolism, 11:337–41. Lozeva, V., Tuomisto, L., Sola, D. et al. (2001). Increased density of brain histamine H(1) receptors in rats with portacaval anastomosis and in cirrhotic patients with chronic hepatic encephalopathy. Hepatology, 33:1370–6. Ma, K. C. (2001). Alzheimer-type I astrogliopathy (AIA) and its implications for dynamic plasticity of astroglia: a historical review of the significance of AIA. Journal of Neuropathlogy and Experimental Neurology, 60:121–31. McEntee, W. J. (1997). Wernicke’s encephalopathy: an excitotoxicity hypothesis. Metabolic Brain Disease, 12:183–92. McEntee, W. J. & Mair, R. G. (1990). The Korsakoff syndrome: a neurochemical perspective. Trends in Neurosciences, 13:340–4. McMullen, P. A., Saint-Cyr, J. A., Carlen, P. L. et al. (1984). Morphological alterations in rat CA1 hippocampal pyramidal cell dendrites resulting in chronic ethanol consumption and withdrawal. Journal of Comparative Neurology, 225:111–18. Mair, W. G. P., Warrington, G. K., Weiskrantz, L. et al. (1979). Memory disorder in Korsakoff’s psychosis. Brain, 102:749–83.

439

440


Martin, H., Voss, K., Hufnagl, P. et al. (1987). Morphometric and densitometric investigations of protoplasmic astrocytes and neurons in human hepatic encephalopathy. Experimental Pathology, 32:241–50. Mascalchi, M., Simonelli, P., Tessa, C. et al. (1999). Do acute lesions of Wernicke’s encephalopathy show contrast enhancement? Report of three cases and review of the literature. Neuroradiology, 41:249–54. Matsushita, M., Yamamoto, T., Gemba, H. et al. (1999). The role of astrocytes in the development of hepatic encephalopathy. Neuroscience Research, 34:271–80. Mayes, A. R., Meudell, P. R., Mann, D. et al. (1988). Locations of the lesions in Korsakoff’s syndrome: neuropathological data on two patients. Cortex, 24:367–88. Michalak, A., Chatauret, N., Butterworth, R. F. et al. (2001). Evidence for a serotonin transporter deficit in experimental acute liver failure. Neurochemistry International, 38:163–8. Miller, N. S. & Gold, M. S. (1990). The contemporary alcoholic. New Jersey Medicine, 87:35–9. Morikawa, Y., Arai, H., Matsushita, S. et al. (1999). Cerebrospinal fluid tau protein levels in demented and nondemented alcoholics [see comments]. Alcoholism: Clinical and Experimental Research, 23:575–7. Mousseau, D. D., Perney, P., Layrargues, G. P. et al. (1993). Selective loss of pallidal dopamine D2 receptor density in hepatic encephalopathy. Neuroscience Letters, 162:192–6. Naegele, T., Grodd, W., Viebahn, R. et al. (2000). MR imaging and (1)H spectroscopy of brain metabolites in hepatic encephalopathy: time-course of renormalization after liver transplantation. Radiology, 216:683–91. Norenberg, M. D. (1994). Astrocyte response to CNS injury. Journal of Neuropathology and Experimental Neurology, 53:213–20. (1996). Astrocytic-ammonia interactions in hepatic encephalopathy. Seminars in Liver Disease, 16:245–53. Norenberg, M. D., Neary, J. T., Norenberg, L. O. et al. (1990). Ammonia induced decrease in glial fibrillary acidic protein in cultured astrocytes. Journal of Neuropathology and Experimental Neurology, 49:399–405. Olsson, N. U., Harding, A. J., Harper, C. et al. (1996). Highperformance liquid chromatography method with lightscattering detection for measurements of lipid class composition: analysis of brains from alcoholics. Journal of Chromatography. B, Biomedical Applications, 681:213–18. Oscar-Berman, M. & Hutner, N. (1993). Frontal lobe changes after chronic alcohol ingestion. In Alcohol Induced Brain Damage. W. A. Hunt & S. J. Nixon. pp. 121–56. Rockville: NIH Publications. Oslin, D., Atkinson, R. M., Smith, D. M. et al. (1998). Alcohol related dementia: proposed clinical criteria. International Journal of Geriatric Psychiatry, 13:203–12. Park, Y. K., Sempos, C. T., Barton, C. N. et al. (2000). Effectiveness of food fortification in the United States: the case of pellagra. American Journal of Public Health, 90:727–38. Parsons, O. A. (1987). Intellectual impairment in alcoholics: persistent issues. Acta Medica Scandinavia, 171:33–46.

Pentney, R. J. (1982). Quantitative analysis of ethanol effects on Purkinje cell dendritic tree. Brain Research, 249:397–401. Pfefferbaum, A., Sullivan, E. V. et al. (2000). In vivo detection and functional correlates of white matter microstructural disruption in chronic alcoholism. Alcoholism: Clinical and Experimental Research, 24:1214–21. Phillips, S. C. & Cragg, B. C. (1984). Alcohol withdrawal causes a loss of cerebellar Purkinje cells in mice. Journal of Studies on Alcohol, 45:475–80. Phillips, S. C., Harper, C. G. & Kril, J. J. (1990). The contribution of Wernicke’s encephalopathy to alcohol-related cerebellar damage. Drug and Alcohol Review, 9:53–60. Rabbe, W., ed. (1990). Effects of NH4 on the function of the CNS. In Advances in Experimental Medicine and Biology – Cirrhosis, Hepatic Encephalopathy and Ammonium Toxicity. New York: Plenum Press. Rindi, G., Comincioli, V., Reggiani, C. et al. (1987). Nervous system thiamine metabolism in vivo. III. Influence of ethanol intake on the dynamics of thiamine and its phosphoesters in different brain regions and sciatic nerve of the rat. Brain Research, 413:23–35. Romero, I., Brown, A. W., Cavanagh, J. B. et al. (1991). Vascular factor in the neurotoxic damage caused by 1,3-dinitrobenzene in the rat. Neuropathology and Applied Neurobiology, 17:495–508. Rosenblum, W. I. & Feigin, I. (1965). The hemorrhagic component of Wernicke’s Encephalopathy. Archives of Neurology, 13:627–32. Rovira, A., Grive, E., Pedraza, S. et al. (2001). Magnetization transfer ratio values and proton MR spectroscopy of normal-appearing cerebral white matter in patients with liver cirrhosis. American Journal of Neuroradiology, 22:1137–42. Saunders, P. A., Copeland, J. R., Dewey, M. E. et al. (1991). Heavy drinking as a risk factor for depression and dementia in elderly men. Findings from the Liverpool longitudinal community study. British Journal of Psychiatry, 159:213–16. Schmahmann, J. D. & Sherman, J. C. (1998). The cerebellar cognitive affective syndrome [see comments]. Brain, 121:561–79. Sebrel, W. H. (1966). Enrichment: good gift of yesterday. Cereal Science Today, 11:228–30. Serdaru, M., Hausser-Hauw, C., Laplane, D. et al. (1988). The clinical spectrum of alcoholic pellagra encephalopathy. Brain, 111:829–42. Shear, P. K., Jernigan, T. L. & Butters, N. (1994). Volumetric magnetic resonance imaging quantification of longitudinal brain changes in abstinent alcoholics. Alcoholism: Clinical and Experimental Research, 18:172–6. Sheedy, D., Lara, A., Garrick, T. et al. (1999). Size of mamillary bodies in health and disease: useful measurements in neuroradiological diagnosis of Wernicke’s encephalopathy. Alcoholism: Clinical and Experimental Research, 23:1624–8. Sherlock, S., ed. (1981). Hepatic encephalopathy. Acute (fulminant) hepatic failure. In Diseases of the liver and biliary system. Oxford: Blackwell Scientific Publications. Shibata, K. & Toda, S. (1997). Effects of sex hormones on the metabolism of tryptophan to niacin and to serotonin in male rats. Bioscience, Biotechnology and Biochemistry, 61:1200–2.


Sobel, R. A., DeArmond, S. J., Forno, L. S. et al. (1981). Glial fibrillary acidic protein in hepatic encephalopathy: an immunohistochemical study. Journal of Neuropathology and Experimental Neurology, 40:625–32. Squire, L. R., Knowlton, B., Musen, G. et al. (1993). The structure and organization of memory. Annual Review of Psychology, 44:435–95. Sullivan, E., Lane, B., Deshmukh, A. et al. (1999). In vivo mamillary body volume deficits in amnesic and nonamnesic alcoholics. Alcoholism: Clinical and Experimental Research, 23:1629–36. Summers, J., Pullan, P., Kril, J. J. et al. (1991). Increased central immunoreactive B-endophin content in patients with Wenicke– Korsakoff syndrome and in alcoholics. Journal of Clinical Pathology, 44:126–9. Terry, R. D. & Hansen, L. A. (1987). Neocortical cell counts in normal human adult aging. Annals of Neurology, 21:530–9. Terry, R. D., Peck, A., De Teresa, R. et al. (1981). Some morphological aspects of the brain in senile dementia of the Alzheimer type. Annals of Neurology, 10:184–92. Therrien, G. & Butterworth, R. F. (1991). Cerebrospinal fluid amino acids in relation to neurological status in experimental portal-systemic encephalopathy. Metabolic Brain Disease, 6:65–73. Tholen, H. (1971). Hironeden. Eine todesursache beim endogenen Leberkoma. Klinische Wochenschrift, 50:196–301. Torvik, A. (1985). Two types of brain lesions in Wernicke’s encephalopathy. Neuropathology and Applied Neurobiology, 11:179–90.

(1987). Topographic distribution and severity of brain lesions in Wernicke’s encephalopathy. Clinical Neuropathology, 6:25–9. Torvik, A., Lindboe, C. F. & Rogde, S. (1982). Brain lesions in alcoholics. A neuropathological study with clinical correlations. Journal of the Neurological Sciences, 56:233–48. Tuck, R. R. & Jackson, M. (1991). Social, neurological and cognitive disorders in alcoholics. Medical Journal of Australia, 155:225– 9. Victor, M. & Adams, R. D. (1985). The Alcoholic Dementias. Amsterdam: Elsevier Science Publishers. Victor, M., Adams, R. D. & Collins, G. H. (1971). The Wernicke– Korsakoff Syndrome. Oxford: Blackwell Scientific. (1989). The Wernicke–Korsakoff Syndrome. Philadelphia: Davis. Von Cramon, D. Y., Hebel, N. & Shuri, U. (1985). A contribution to the anatomical basis of thalamic amnesia. Brain, 108:993–1008. von Hosslin, D. & Alzheimer, A. (1912). Ein Beitrag zur Klinik und pathologischen Anatomie der westphal-Strumpellschen Pseudosklerose. Zeitschrift gesellschaft Neurologische Psychiatrie, 8:183–209. Walker, D. W., Barnes, D. E., Zornetzer, S. et al. (1980). Neuronal loss in hippocampus induced by prolonged ethanol consumption in rats. Science, 209:711–13. Wilcock, G., Esiri, M., Bowen, D. M. et al. (1988). The differential involvement of subcortical nuclei in senile dementia of Alzheimer’s type. Journal of Neurology, Neurosurgery and Psychiatry, 51:842–9. Wilson, J. D., ed. (1991). Vitamin deficiency and excess. In Harrison’s Principles of Internal Medicine. New York: McGraw Hill Inc.

441

19 Hydrocephalus and dementia Margaret M. Esiri1 and Gary A. Rosenberg2 1 2

Department of Neuropathology The Radcliffe Infirmary, Oxford, UK Department of Neurology University of New Mexico, USA

Introduction The ancient term hydrocephalus refers to an excessive accumulation of fluid (water) inside the head. It is most readily detected in infants and children with congenital hydrocephalus since its existence before the skull sutures are closed causes the formation of an enlarged head. Such children have long been recognized commonly to have mental impairment, manifest as mental retardation, although the extent of this is very variable. Morgagni (1769) first described hydrocephalus due to enlarged cerebral ventricles in an adult without head enlargement. If this condition develops insidiously during adult life, dementia is an almost invariable accompaniment, though not the only one. More familiarly hydrocephalus in adults develops acutely or subacutely and presents with symptoms of raised intracranial pressure – headache, vomiting and drowsiness. However, in this chapter we are concerned with chronic hydrocephalus in adults, a condition in which symptoms and signs of raised intracranial pressure are commonly clinically absent and in which the cerebrospinal fluid (CSF) pressure is often normal when measured randomly – hence the commonly used term – normal pressure hydrocephalus (NPH). The recognition of this condition, and the realization that it is responsible for some cases of progressive dementia in adults are recent (Riddoch, 1936; Foltz & Ward 1956; McHugh, 1964; Hakim & Adams, 1965; Adams et al., 1965). Cases of NPH, some of which respond well to a shunting procedure, are to be distinguished from cases of dementia due to neurodegenerative diseases in which the ventricles dilate as a consequence of cerebral atrophy (hydrocephalus ex vacuo). It should be noted that making this distinction can be difficult, and therefore

neurodegenerative diseases need to be excluded when a case of possible NPH is being examined neuropathologically. As outlined below, the pathophysiology of NPH is still only poorly understood (Cummings & Benson, 1983; Pickard, 1991).

Cerebrospinal fluid physiology The fluids surrounding the brain and within the ventricles, which were thought by early investigators to act as a buffer against injury, play a more important role in brain function, namely they act as the lymph of the brain, delivering nutrients and removing metabolites. Early physiological studies by Weed and Cushing at the beginning of this century established the concept that the interstitial fluid (ISF), which bathes the brain cells, and the CSF, which fills the ventricles and subarachnoid spaces, were in essence one fluid that formed a ‘Third Circulation’ (Cushing, 1925; Weed, 1935). They recognized that the brain lacked lymphatics and that the CSF acted as the lymph of the brain. The third circulation begins as secretions from cerebral capillaries and cellular metabolism, which forms the ISF, and from the choroid plexuses, which forms the CSF. Once formed, the CSF/ISF deliver essential nutrients, such as glucose and amino acids, to the cells, and carry away the products of cellular metabolism. The CSF/ISF circulate out of the ventricles and over the convexities to be transferred to the blood through the one-way valve system of the arachnoid granulations. When excess fluid is formed or the routes of absorption are impaired, increases in CSF pressure occur. Increased tissue pressure, such as is found in traumatic or ischaemic insults, leads to cerebral oedema with decreased


442

Hydrocephalus and dementia

CSF space and small ventricles. Obstruction of CSF outflow results in an increase in the size of the ventricles, which is referred to as hydrocephalus. The discussion in this chapter will focus on the adult-onset form of hydrocephalus, which has been termed normal pressure hydrocephalus (NPH). A more accurate term would be idiopathic adult-onset hydrocephalus or intermittent, elevated pressure hydrocephalus, but NPH has persisted in the literature, and it will be used in this chapter. The basic features of the CSF hydrodynamics have been described in several reviews and monographs (Katzman & Pappius, 1973; Davson, 1987; Rosenberg, 1990; Fishman, 1992).

Blood–brain interfaces Brain tissues are separated from the systemic circulation by tight-junctions (zonae occludens). Cerebral endothelial cells comprise the major interface because of their extent. Two other lesser interfaces include the ependymal cells overlying the choroid plexuses, which are joined together by tight junctions, and the arachnoidal cells that form one of the meninges covering the brain. Collectively, these interfaces function as the blood–brain barrier (BBB). As opposed to the high electrical resistance in the cells joined by tight junctions, gap junctions with low electrical resistance are found in the ependymal and pial layers. The distinction between the tightness of the junctions is important in understanding the ease or difficulty with which substances move from the blood to the brain and within the brain. For example, a highly lipid soluble molecule, such as an anaesthetic gas, easily moves from the lungs to the blood to the brain, crossing the BBB because of its lipid solubility. Whereas the anticancer agent, methotrexate, is lipophobic, and fails to enter the brain after intravenous injection. However, when it is injected into the CSF space it crosses the ependyma and the pia to gain access to brain cells. In the hydrocephalic brain, extravasation of CSF into the tissue surrounding the ventricles is seen on CT and MRI. This fluid collection is referred to as transependymal absorption, implying that it is an alternate route of CSF removal from the obstructed ventricles. Although the fluid collection appears at the onset of hydrocephalus, and disappears after treatment, whether it is a route of absorption or a static region of oedema remains uncertain. Cerebrospinal fluid is formed by the choroid plexuses in the lateral, third and fourth ventricles (Fig. 19.1). Energy is required to create the osmotic gradients used in the secretion of the CSF. The ependymal cells lining the cerebral ventricles, which contain cilia, function to move the fluid through the ventricles, while the ependymal cells that

Fig. 19.1. Diagram of the mechanisms involved in CSF formation by the choroid plexus. Capillaries in the stroma lack tight junctions, allowing proteins to escape. Ependymal cells overlie the stroma, and are joined together by tight junctions (TJ) at the apical surface. Exchange pumps are located on the choroid cell surfaces. Carbonic anhydrase (CA) forms bicarbonate from carbon dioxide. A sodium–potassium ATPase pump on the apical surface provides the osmotic gradients for CSF production.

form the choroid plexuses have a secretory function (Roth et al., 1985). The secretory function of the choroid plexus is derived from two enzyme systems: sodium–potassium adenosine triphosphatase (Na–K ATPase) located on the ventricular surface and carbonic anhydrase within the cell. Na–K ATPase uses the energy of ATP to pump three sodium molecules into the CSF, bringing two of potassium into the blood. The excess ion concentration on the CSF side of the cell creates an osmotic imbalance, pulling water into the ventricles. Carbonic anhydrase forms bicarbonate from carbon dioxide and water (Saito & Wright 1983). Acetazolamide (Diamox), which inhibits the action of carbonic anhydrase and reduces CSF production, is used therapeutically in the treatment of pseudotumour cerebri. Inhibition of Na–K ATPase has only been possible experimentally with the toxic agent, oubain, which acts like digitalis. The cerebral capillaries, by a mechanism similar to that found in the choroid plexuses, form an estimated 30 to 60%

443

444

M. M. Esiri and G. Rosenberg

Fig. 19.2. Diagram to show pathways of flow of cerebrospinal fluid. (1) From lateral ventricles through foramina of Munro to (2) third ventricle via aqueduct to (3) fourth ventricle from whence it escapes into the subarachnoid space. (4) represents circulation around the spinal cord in the subarachnoid space, (5) descent in the central canal of the spinal cord and (6) circulation in the subarachnoid space around the outer surface of the brain.

of the CSF. A Na–K ATPase pump is found on the abluminal surface of the capillaries (Betz et al., 1994). Production of CSF is similar between different species based on weight of the choroid plexus (Cserr, 1971). However, the amount of CSF formed initially as the ISF varies between species. Removal of the choroid plexuses in monkeys lowers CSF production by only 30%, suggesting that as much as 60% is coming from nonchoroidal sources (Milhorat et al., 1971). In the rabbit and the cat the amount of non-choroidally produced CSF is estimated to be 30% (Pollay, 1975; Rosenberg et al., 1980). Circulating CSF/ISF exits from the ventricular system into the subarachnoid space through three openings in the roof of the fourth ventricle, namely, the lateral openings, which are called the foramen of Luschka, and the medial one, which is named after Magendie (Fig. 19.2). Once in the subarachnoid space, the fluid either moves down into the spinal canal, where some absorption can occur around the sleeves of the nerve roots, or up over the convexity of the cerebral hemispheres, where it is absorbed through the arachnoid granulations into the blood. The route of circulation of the CSF has been studied by

cisternography in which a radiolabelled tracer is injected into the CSF. This isotopic method has been used as a diagnostic test of the circulation of the CSF. Normally, the injected radiolabelled tracer flows into the cisterns of the brain and is removed from the brain by 24 hours. In situations where the flow of CSF is obstructed, but the communication between the subarachnoid space and the ventricles remains patent, the tracer enters the ventricles, and can persist in that location for several days. In such pathological situations, a diagnosis of communicating hydrocephalus can be made (Benson et al., 1970). This method was used extensively in the past to aid in the diagnosis of NPH, but has fallen out of favour because of it nonspecificity. Absorption of the CSF occurs mainly through the arachnoid granulations, which form a one-way valve system (Fig. 19.3). This series of channels opens when the CSF pressure exceeds the pressure in the venous sinuses, and closes when the pressure is lower. Large particles, such as red blood cells and white blood cells, interfere with the normal absorption of the CSF (Tripathi & Tripathi, 1974). Because of the direct flow of CSF into the blood across the arachnoid villi the CSF pressure is regulated by that of the venous pressure. Thin-walled veins transmit the venous pressure directly to the CSF/ISF, while the arterial walls with their elastic fibres and smooth muscle cells fail to transmit the arterial pressure to the CSF (Davson, 1987). When blood is released into the CSF by the rupture of a subarachnoid aneurysm or an intracerebral haemorrhage, there is a rise

Fig. 19.3. Diagram to show the manner of cerebrospinal fluid (csf ) reabsorption via the arachnoid granulations (Ag) into the superior sagittal venous sinus (SSvS). DM: dura mater, FC: falx, PM: pia mater, AM: arachnoid mater, SS: subarachnoid space, CC: cerebral cortex, eCs: endothelial cell.


in the CSF pressure that can lead to enlargement of the cerebral ventricles. The enlargement of the cerebral ventricles that is seen with a subarachnoid haemorrhage is generally transient, and only occasionally leads to hydrocephalus that requires the placement of a shunt. Occlusion of the venous sinuses that drain the blood from the brain causes an increase in the CSF also, but rarely results in the enlargement of the ventricles. Rather, it leads to idiopathic intracranial hypertension or pseudotumour cerebri.

Hydrocephalus Hydrocephalus is defined as abnormal accumulation of CSF in the ventricles without loss of brain tissue. Noncommunicating (obstructive) hydrocephalus is used to describe enlargement of the ventricles due to a blockage of the CSF outflow from any of the ventricles. Communicating hydrocephalus implies movement of fluid between the ventricular system and the subarachnoid space with obstruction at the sites of CSF absorption. The pathophysiology of hydrocephalus and the clinical features are different in acute and chronic hydrocephalus and between adults and children. Enlargement of the cerebral ventricles in children less than 2 years of age may be associated with acute symptoms and the enlargement of the head because of the open sutures in the skull. In the adult, acute obstruction of the ventricles occurs with masses compressing the foramen of Monro and the fourth ventricle outflow routes. Patients have ventricular enlargement with headache, vomiting, and gait disturbances. When the onset of symptoms is more gradual, diagnosis may be more difficult at the onset. However, when problems of gait and intellect appear, and full diagnostic work-up is begun, the hydrocephalus is generally diagnosed by neuroimaging. Increases in CSF pressure due to obstruction of CSF circulation are thought to occur at some point in the course of an adult-onset hydrocephalus, but the CSF pressure may be normal by the time that the symptoms have progressed to the point where diagnosis is apparent. In some patients with dementia, a loss of tissue results in a compensatory ventricular enlargement, which is called ‘hydrocephalus ex vacuo’ to distinguish it from excess fluid due to obstruction of CSF outflow. When the ventricles are obstructed at the level of the third or fourth ventricle, acute hydrocephalus may develop rapidly in both children and adults. Ventricular size may reach 80% of maximal enlargement within six hours. This rapid growth is followed by a slower phase of enlargement as compensatory mechanisms develop. Fluid accumulation in the periventricular tissue may be seen on CT and MRI. This is referred to as transependymal absorption or

hydrocephalic oedema, depending on whether it is considered as a drainage route for CSF or a static collection of fluid (Fishman, 1975). With long-standing hydrocephalus tissue changes occur in the tissues around the ventricles. In the chronic form of adult hydrocephalus, stabilisation of the ventricular size occurs, and the CSF pressure returns to normal. With long-standing hydrocephalus atrophy may occur in the periventricular white matter. When the rate of ventricular enlargement stabilizes in patients with incomplete ventricular obstruction, CSF production is balanced by the absorption at alternate sites. Arrested hydrocephalus occurs more commonly in communicating hydrocephalus, which tends to produce an incomplete block to CSF circulation, but is occasionally found in noncommunicating hydrocephalus due to aqueductal stenosis. For reasons that are not well understood, patients that have arrested hydrocephalus can undergo decompensation after many years (Little et al., 1975).

Normal pressure hydrocephalus In the adult, symptoms of acute hydrocephalus include headaches, papilloedema, diplopia, and mental status changes. Sudden death may occur with severe increases in pressure that lead to dysfunction of the vital nuclei that surround the third and fourth ventricles. In rare instances hydrocephalus causes an akinetic mutism, resembling a persistent vegetative state. In aqueductal stenosis of the adult, enlargement of the third ventricle can lead to symptoms which can include temporal lobe seizures, CSF rhinorrhoea, endocrine dysfunction, such as amenorrhoea, polydipsia and polyuria, and obesity (Fukuhara & Luciano, 2001). Gait disturbances are reported in patients with aqueductal stenosis, but hyper-reflexia with Babinski signs is infrequent. Adults are more likely than children to present with an acute blockage of CSF flow by intraventricular masses, such as a colloid cyst of the third ventricle or an ependymoma of the fourth ventricle (Table 19.1). These tumours cause sudden headaches, ataxia, and loss of consciousness with symptoms that may be intermittent due to the ball-valve effect of the masses. Diagnosis can be made by CT or MRI, which reveals the structural lesion in the ventricular system. Cerebellar haemorrhage, or less commonly cerebellar infarction with oedema, cause an acute hydrocephalus by compression of the brainstem, blocking the Sylvian aqueduct and the outflow of CSF from the fourth ventricle (Ott et al., 1974). Patients with cerebellar haemorrhage usually have a history of hypertension or more rarely an arteriovenous malformation. On rare occasions,

445

446


Table 19.1. Causes of obstructive hydrocephalus Non-communicating

Communicating

Aqueduct stenosis Congenital Inflammatory Haemorrhagic

Post-haemorrhagic

Intraventricular obstructive lesion Tumours meningiomas papillomas ependymomas vascular formations carcinomas lymphomas craniopharyngiomas pituitary adenomas

Post-infectious

Compressive lesions Gliomas Other tumours as above Colloid cysts Haemorrhagic obstruction

Post-traumatic

Post-craniotomy Idiopathic Neoplastic sagittal sinus meningioma carcinomatosis of leptomeninges lymphomatosis tentorial tumour Basilar impression Ectatic basilar artery

Obstruction of outlet foramina Posterior fossa neoplasms Posterior fossa vascular malformations Inflammatory processes in basilar cisterns Dandy–Walker malformation Arnold–Chiari malformation Arachnoid cysts Haemorrhagic obstruction Source: After Cummings & Benson (1983).

the cerebellar bleeding is caused by anticoagulation with coumadin. Cerebellar haemorrhage with hydrocephalus causes an acute headache, which is often followed by increasing drowsiness and ataxia of gait. Hemiparesis and brainstem findings evolve after the ataxia, providing a clue that the origin of the problem is in the posterior fossa. The expanding mass in the posterior fossa demands urgent neurosurgical attention, with placement of a ventricular catheter to decompress the lateral and third ventricles, followed by posterior fossa craniectomy to remove the mass effect and pressure on the brainstem. Chronic hydrocephalus in the adult can produce more subtle symptoms that may be confused with other progressive neurological disorders. Prior to CT, patients with adult-onset hydrocephalus were diagnosed and treated for Parkinson’s disease. Lacunar infarcts were another diagnosis mistaken for hydrocephalus because of the gait problems and dementia. Since the routine use of neuroimaging

has become available, neither of these diagnoses need be mistaken for adult-onset hydrocephalus. Adult onset hydrocephalus is diagnosed clinically by the triad of gait disturbance, incontinence, and memory loss (Adams et al. 1965). Causes of chronic hydrocephalus include subarachnoid haemorrhage, chronic meningeal infections, slow growing tumours blocking the CSF pathways, and arrested hydrocephalus. However, about one-third of the patients have no obvious cause and are classified as idiopathic (Koto et al., 1977). Epidemiological data on the relative incidence of various causes of hydrocephalus is extremely limited. One study, which is frequently cited, includes data prior to the use of MRI (Table 19.2) (Katzman, 1977). Treatment of hydrocephalus involves an operation to insert a tube to shunt CSF from the ventricles to another body cavity, such as peritoneum (ventriculo-peritoneal shunt). Hydrocephalic ventricles can usually be entered with a catheter without difficulty due to the increase in their size, making this a relatively simple surgical procedure. However, there is a high surgical complication rate in shunted patients. Infection and subdural haematomas are the two most frequent with intracerebral bleeding being a rare complication. Shunts may dysfunction after long intervals, causing abrupt decompensation. Symptoms of acute increased intracranial pressure from a shunt malfunction are similar to those seen with the onset of the hydrocephalic process. Only a small percentage of the patients with adult-onset hydrocephalus due to subarachnoid bleeding or infection Table 19.2. Causes of NPH in 914 cases assembled by Katzman (1977) Subarachnoid haemorrhage Head injury Aqueduct stenosis Meningitis Post-craniotomy Obstructive lesions of the fourth ventricle Tumours obstructing the third ventricle Basilar artery ectasia or aneurysm Other tumours, including pituitary adenoma Basilar impression with or without Paget’s disease Syringomyelia ‘Encephalopathy’, ‘encephalitis’ Idiopathic (unknown) associated with: Parkinson’s disease Alzheimer’s disease Cerebrovascular disease Huntington’s chorea Chronic epilepsy

315 102 34 34 43 21 26 11 7 4 1 2 314 15 7 4 2 2


require insertion of a shunt. Many patients with subarachnoid haemorrhage have transient hydrocephalus, but most resolve spontaneously. Syphilis, tuberculosis, and fungal infections can lead to chronic obstruction of the subarachnoid pathways, which can cause enlargement of the ventricles. Because of the relationship of infection to the ventricular obstruction, cultures of the CSF are indicated in the elderly patient with enlarged ventricles, and searching for other sources of infection in the lungs and other organs may be helpful in establishing the type of infection.

Diagnosis of normal pressure hydrocephalus Although there was initially optimism that many patients with dementia would be cured by the drainage of CSF, diagnosis of NPH remains a challenge 35 years after its initial description. The clinical examination is used by many as the selection criteria. When the typical triad of mental impairment, a gait disturbance, and incontinence begins with problems of gait, the prognosis is the best. The presence of enlarged ventricles on CT or MRI is not useful as a diagnostic test, although the CT findings of transependymal flow, enlarged temporal poles, and reduced cortical margins are suggestive (Fig. 19.4). Those with a defined aetiology, such as trauma, infection, or subarachnoid haemorrhage, respond better to treatment than the one-third in whom no aetiology is found. NPH causes a gait disturbance

that is referred to by some authors as an apraxic gait, which is an inability to lift the legs as if they were stuck to the floor. This resembles the gait disturbance found in patients with Parkinson’s disease. The motor strength is intact, reflexes are usually normal, and Babinski signs are absent. It has been recognized for many years that patients with cerebrovascular disease respond poorly to shunting. Lacunar state was described in several patients at autopsy prior to CT and MRI (Koto et al., 1977; Earnest et al., 1974). More recent studies have confirmed that cerebrovascular disease is common in patients with NPH (Graff-Radford & Godersky, 1987; Boon et al., 1998). Since many of these patients also have hypertension, and some small or large strokes, such patients may have other neurological findings, including spasticity and hyper-reflexia with Babinski signs. The combination of cerebrovascular disease and hydrocephalus is a poor prognostic sign for treatment with shunts. NPH leads to a reduction in intellect, which at times may be subtle. The dementia involves slowing of the verbal and motor responses with preservation of cortical functions, such as language and spatial resolution. Neuropsychological testing quantifies the fall in intellect and the degree of dementia. Patients are apathetic, and may appear depressed. Incontinence of urine may occur early in the course, particularly in patients with prominent gait disturbance. In the early stages of the illness, presumably as

Fig. 19.4. Computed tomographic (CT) scan from a patient with NPH. (a) Enlarged lateral ventricles are seen without concomitant enlargement of the cortical sulci. (b) In this lower section in the same patient are seen the enlarged lateral ventricles in the temporal poles (arrow) and a normal fourth ventricle.

447

448


the ventricles are undergoing enlargement, patients can experience drop attacks or brief loss of consciousness. Headache and papilloedema are not a part of the syndrome. Diagnosis of adult onset NPH and selection of patients for placement of a ventriculoperitoneal shunt has been difficult. In spite of many studies attempting to find a definitive diagnostic test to predict those that will benefit from treatment, the goal remains elusive. Confounding the treatment is the fact that many of these patients have hypertensive vascular disease with lacunar infarcts. These patients continue to progress in spite of temporary improvement after the shunt is placed because of the progression of the vascular disease. Features of Parkinson’s disease were noted in earlier reports of the syndrome, and it is now recommended that all patients with Parkinson’s disease have scans to rule out hydrocephalus. Separating Parkinson’s disease, lacunar state and NPH has been aided by CT and MRI, though NPH may occasionally coexist with these diseases. An increasing number of patients with NPH have vascular diseases, such as lacunar state or subcortical arteriosclerotic encephalopathy (Binswanger’s disease). Selection of patients for shunting requires a combination of clinical findings and diagnostic test results, since no test can totally predict the patient likely to benefit from an operation. The use of cisternography has fallen out of favour because of the overlap with enlarged ventricles due to atrophy. The Dutch Normal-Pressure Hydrocephalus study, which was done at multiple centres in Holland, has revived interest in the Katzman infusion test (Boon et al., 1999, 2000). In the infusion test, artificial CSF is infused into the lumbar sac at a constant rate. Normally the absorption proceeds without an increase in the CSF pressure, which is measured by a pressure monitor. In those with NPH, however, the pressure continues to rise due to the obstruction of the CSF drainage. A series of measures was used by the investigators, and the one that was the best predictor of improvement after one year was the finding of an outflow resistance that was >18 mm Hg/ml/minute. Several patients with CT evidence of NPH and low outflow resistance also responded to shunt, leading the authors to equivocate on the use of a single test. Another problem with the study was that the follow-up was only 1 year, and those with progressive vascular problems may not have been detected. But it could be argued that a year of improvement was worthwhile even in those destined to progress for other reasons. The Dutch study also confirmed the importance of cerebrovascular disease in this illness. Two early pathological studies showed the connection between cerebrovascular disease and NPH (Koto et al., 1977; Earnest et al., 1974). More recently, series of patients with NPH have revealed a high incidence of hypertension, which was found in 28

to 83% of patients in reported series. Strokes seen on CT predicted poor outcome better than the finding of cerebrovascular disease risk factors. Some authors have argued that the NPH leads to vascular disease, and that as the population ages, differentiation between cerebrovascular disease and NPH may be increasingly difficult (Gallassi et al., 1991). The pathophysiology of NPH is controversial. Pulse pressure from the formation of CSF has been implicated. Weakened ependymal walls are another postulated cause. Transependymal flow of CSF has been found experimentally in animals with increased intracranial pressure (Rosenberg et al., 1983). MRI suggests that transependymal clearance of fluid is occurring. Whether this is a drainage route or merely evidence of so-called hydrocephalic oedema is unresolved (Fishman, 1975). Most likely many factors are involved. In those with subarachnoid bleeding or infection with a normal vasculature otherwise, the obstruction of the arachnoid granulations is the cause of the hydrocephalus. But in the elderly individual with long-standing hypertension and lacunar infarctions other mechanisms must be acting. Some increase in resistance to outflow of CSF occurs with normal aging as shown by cisternography (Larsson et al., 1994). One possibility is that the remodelling of the cerebral vessels results in the release of fragments of fibrin, which over time impede the outflow of CSF. Increased proteolytic activity of the extravascular coagulation system, involving urokinase plasminogen activator and matrix metalloproteinases, is found in animals with experimental stroke and in humans with vascular dementia (Mun-Bryce & Rosenberg, 1998). There may be a correlation between the improvement in the gait after the removal of CSF and improvement after shunting. Cisternography is a useful procedure that involves the injection of a radiolabelled tracer into the CSF with monitoring of its absorption for three days (Fig. 19.5). Normally, the labelled material fails to enter the ventricles, moving over the convexity of the brain, and leaving the CSF space by 12 to 24 hours (Larsson et al., 1994). In patients with large ventricles due to atrophy, there may be a delay in circulation time, with some isotope being seen in the ventricles during the first 24 hr. Communicating hydrocephalus with abnormal CSF circulation shows persistent ventricular filling for more than 48 hours. In patients with NPH there is reflux of the tracer into the cerebral ventricles by 24 hours and retention in the ventricles for 48 to 72 hours. This suggests that transependymal absorption is occurring, and the periventricular white matter has become an alternate route of CSF absorption. Unfortunately, a positive cisternogram is seen in some patients with hypertensive cerebrovascular disease and Binswanger’s encephalopathy.


Computed tomography and MRI in NPH show that the temporal horns of the lateral ventricles are enlarged, and that cortical atrophy is less than anticipated for age. This is in contrast to patients with hydrocephalus ex vacuo due to a degenerative disease, such as Alzheimer’s disease where there is atrophy of the cerebral gyri and enlargement of the ventricles. Another useful finding on proton density or diffusion-weighted MRI is the presence of presumed transependymal fluid in the frontal and occipital periventricular regions (Fig. 19.6(a)). Quantitative cisternography using single photon emission computed tomography (SPECT) has been successfully used to predict the results of a shunt (Larsson et al., 1994). Other proposed diagnostic methods, including the measurement of rate of absorption of CSF by the infusion of saline or artificial CSF into the thecal sac, clinical improvement following the removal of CSF, or the prolonged monitoring of intracranial pressure, have been used with some success to select patients for surgery (Chen et al., 1994; Raftopoulos et al., 1994; Sand et al., 1994). Decreased cerebral blood flow has been reported in NPH; regional cerebral blood flow is reduced in both cortical and subcortical regions (Kristensen et al., 1996). Those patients showing a clinical improvement to shunting had a concomitant increase in cerebral blood flow (Kimura et al., 1992; Waldemar et al., 1993). Removal of CSF may result in an increase in cerebral blood flow in those patients where NPH is likely to respond to shunt therapy. The number of patients undergoing shunt operations at most centres has fallen as the initial enthusiasm, which resulted in many shunts and a low success rate, has waned. None of the currently available tests by themselves identifies the patients that will benefit from shunting. Most helpful is a combination of clinical signs, and judiciously chosen laboratory tests. Success rates vary between investigators with some reports describing improvement in around 80% of the treated patients, while others have lower rates. One study of 166 patients receiving shunts for presumed NPH found that 36% had a poor response and 28% had major complications, including a high rate of infection and subdural haematomas after shunt placement (Fig. 19.6(b)) (Vanneste, 2000). Clearly, more information is needed to aid the management of patients with this uncommon but treatable syndrome. Most clinicians rely on a series of tests along with the clinical impression to decide whether to insert a ventricular drain. Rarely is the ideal patient with a known aetiology, primary gait disturbance, and typical neuroimaging encountered. More commonly, the patients have hypertensive vascular disease and the idiopathic form with equivocal diagnostic test results. This situation is likely to

Fig. 19.5. A cisternogram of a patient with NPH is shown. (a) 24 hours after injection of the radiolabelled tracer into the lumbar sac, the label is seen accumulating in the ventricles in this lateral view. (b) An anterior view shows the outline of the ventricles at the same time. In a normal subject the tracer would not have entered the ventricles from the subarachnoid space, and the majority of the tracer would have exited the brain by 24 hours.

continue because of the increase in the numbers of elderly people and the rise of cerebrovascular disease. Future studies should be directed at understanding the relationship of the vascular changes to the development of the syndrome with the goal of reducing the factors related to the development of hydrocephalus before the insertion of a shunt is necessary.

Neuropathological findings in NPH There have been rather few neuropathological studies of NPH published. Few centres see large numbers of such cases, and some of those cases that are diagnosed post mortem have inadequate clinical documentation. There is also a problem in defining the idiopathic condition neuropathologically other than by exclusion of other neuropathology associated with dementia in a case with well-preserved cerebral cortex and marked ventricular dilatation. Moreover, what constitutes ‘marked’ ventricular dilatation is generally not precisely defined, since measurement of ventricular volume post-mortem is not readily accomplished. Ventricular dilatation is, in any event, a not uncommon incidental finding in clinically undemented elderly subjects, so that without a good clinical history its significance may be dismissed easily (Messert et al., 1972). Some clinically diagnosed cases have clear-cut evidence of communicating hydrocephalus with marked leptomeningeal thickening. Others, including some that have been demonstrated to benefit from a shunt, have been shown after death to have changes of Alzheimer’s disease or multiple small infarcts with apparently normal leptomeninges or leptomeningeal thickening (De Land et al., 1972; Vessal et al., 1974; Earnest et al., 1974; Koto et al., 1977; Di Rocco et al., 1977; Katzman, 1977; Akai et al., 1987;

449

450


Fig. 19.6. Magnetic resonance images (MRI) from a patient with idiopathic NPH. (a) Intermediate-weighted (proton density) scan showing the enlargement of the ventricles seen on CT, and the extravasation of fluid from the ventricles into the periventricular white matter. The intermediate-weighted image is best to show the transependymal flow because the CSF has a greyish colour while the water in the surrounding tissues appears white. The arrowhead shows the prominent fluid accumulation in the frontal white matter. (b) The patient had a lumboperitoneal shunt, which is seen as a dark tract (arrow). Post-shunt a subdural haematoma developed (arrowhead) with fluid-level seen for the blood.

Derouesne et al., 1987; Newton et al., 1989). A few cases of AD are known to have severe ventricular dilatation (Tomlinson et al., 1970) and these may have been mistaken for cases of NPH. However, a few cases of AD with not necessarily severe cortical pathology may show a marked degree of amyloid deposition in leptomeningeal blood vessel walls. This predisposes the vessels to leak and cause multiple small, and occasionally larger, subarachnoid haemorrhages which could, at least potentially, result in a noncommunicating hydrocephalus, although such cases do not appear to have been reported. It would be of interest to know if ventricular size in AD is larger in those cases with severe amyloid angiopathy than in other cases. Some cases with multiple deep infarcts may likewise have been mistakenly diagnosed as NPH, since the infarcts might be expected to have interrupted the same long fibres that are affected in NPH, thus giving rise to similar symptomology, while the cerebral atrophy that results gives rise to the ventricular dilatation. However, cases with multiple deep infarcts that have responded to shunting may be true cases of NPH precipitated by blood from partly haemorrhagic infarcts lying close to the lateral or third ventricles reaching the CSF and clogging up the subarachnoid space intermittently.

Our own experience of the neuropathology of NPH amounts to 16 cases so diagnosed at post mortem examination over a 29-year period in which a total number of approximately 700 cases of dementia were examined neuropathologically (Table 19.3). Cases of NPH therefore constituted 2.3% of the total. Most cases were not clinically diagnosed as NPH, but in all of them there was a history of dementia. There was a predominance of males and the average age was 72 years. In most cases a history of gait disturbance, albeit ill defined, was present. The neuropathological findings, apart from severe hydrocephalus, were variable. One had a non-communicating hydrocephalus caused by obstruction by a pituitary adenoma. All the other cases had a communicating hydrocephalus. This was associated with pronounced leptomeningeal thickening in the subarachnoid space over the cerebral convexities, but without any inflammation or evidence of old haemorrhage, in three cases. One of these cases also had an old infarct in the caudate nucleus and another moderate numbers of argyrophilic plaques insufficient to diagnose Alzheimer’s disease. In five cases there were deep vascular lesions, the majority old, small, white or deep grey matter infarcts. However, almost all these lesions contained


451

Table 19.3. Details of 16 cases of normal pressure hydrocephalus examined necropathologically at Oxford, 1971–2000 Case

Sex/Age

Clinical features

Pathological findings

Brain wt

1

M 66

History of two brainstem strokes several years before death. Then had become ‘forgetful’ with an unsteady gait.

Gross hydrocephalus, thickened leptomeninges, old infarcts at right occipital pole, adjacent to right thalamus, in left occipital white matter and in the pons. No Alzheimer type pathology or other dementing illness pathology.

1471 g

2

M 76

Loss of consciousness after head injury 30 years before death following which he had slurred speech and unsteady gait. Developed mild dementia during the last few years of his life.

Gross hydrocephalus with a brain that appeared swollen externally. Arachnoid cyst in posterior fossa containing clear cerebrospinal fluid on right with brain stem slightly shifted to the left. Syringobulbia.

1767 g

3

M 79

History of dementia, gait disturbance and Paget’s disease. Details not available.

Basilar impression of skull with Paget’s disease. External appearance of the brain unremarkable. Marked hydrocephalus. No Alzheimer type pathology or other dementing illness pathology.

N/A

4

M 53

History of progressive dementia, clinically diagnosed as Alzheimer’s disease.

N/A

5

M 77

Five year history of dementia and 3 year history of temporal lobe epilepsy. Increased limb tone and spastic gait.

Moderately thickened leptomeninges. Gross hydrocephalus. Moderate numbers of plaques in many areas of cortex and rare neurofibrillary tangles, limited to the hippocampus and para-hippocampal gyrus (features not sufficient to diagnose Alzheimer’s disease). No other dementing illness pathology. Gross enlargement of lateral ventricles, and compression of third ventricle by suprasellar non-functioning pituitary adenoma. No Alzheimer’s disease or other dementing illness pathology.

6

M 65

One year history of dementia and abnormal behaviour.

1490 g

7

M 75

Two year history of progressive dementia and incontinence. Previous history of high alcohol intake.

Gross hydrocephalus, normal leptomeninges. No Alzheimer’s disease or other dementing illness pathology. Gross hydrocephalus. Mild generalized cerebral atrophy. Few widespread plaques and no neurofibrillary tangles interpreted as insufficient to diagnose Alzheimer’s disease. No other dementing illness pathology.

8

F 67

Several months’ history of progressive dementia and unsteadiness.

Slight generalized cerebral atrophy and slight leptomeningeal thickening. Gross hydrocephalus. No Alzheimer’s disease or other dementing illness pathology.

1200 g

9

M 81

Severe head injury 12 years before death, with subdural and extradural haemorrhage, From 9 years before death, occasional incontinence and 4 years before death onset of difficulty walking and episodes of confusion and aggression, progressing to frank dementia. Able to walk only with two assistants, gait broad-based.

Five burr holes, taut subdural membrane over right frontal lobe with calcification. A few old contusional scars and an old right frontal pole infarct. Gross hydrocephalus. No Alzheimer’s disease pathology or other dementing illness pathology.

1300 g

10

M 80

Three year history of hesitant gait and frequent falls, 18 month history of incontinence, weak legs and some ‘blackouts’. One year history of mild dementia. Spastic, bilateral weakness on examination, oral dyskinesia and tremor of right hand. Found to be anaemic with subacute myelo-monocyte leukaemia.

Thickened leptomeninges and gross hydrocephalus with no cerebral atrophy. Small old infarct in head of left caudate nucleus. No Alzheimer’s disease pathology or other dementing illness pathology.

1350 g

11

M 55

Many years’ history of treated hypertension. 1 year’s history of being withdrawn, slovenly confused and incontinent. On examination, lower limbs spastic. First encephalogram showed communicating hydrocephalus

Three right-sided burr holes. Massive right intracerebral haemorrhage in white matter and lentiform nucleus. Acute posterior cerebral artery territory infarction secondary to uncal herniation. Gross dilatation of left

1490 g

1251 g

1230 g

452


Table 19.3. (cont.) Case

Sex/Age

Clinical features

Pathological findings

and a shunt was inserted in the right lateral ventricle. Post-operative intracerebral haemorrhage occurred followed by death on the second post-operative day.

lateral ventricle. Old softening with residual haemosiderin deposition in left caudate nucleus and dorsal thalamus.

Brain wt

12

M 79

Five year history of increasing immobility and double incontinence. Previous excessive alcohol intake. Eighteen months’ history of progressive dementia and aggressive behaviour.

Gross hydrocephalus. Old small, left cerebellar infarct. Rare neurofibrillary tangles confined to the hippocampus and a few plaques in all cortical areas, insufficient to diagnose Alzheimer’s disease. No other dementing illness pathology.

1330 g

13

M 81

Six months’ history of difficulty walking and ‘drop attacks’, followed by incontinence and mild confusion. On examination he had a staggering gait, brisk reflexes but downgoing plantar responses. Died shortly after.

Gyri at vertex clearly flattened, but leptomeninges normal. Severely stenosed left vertebral artery but no infarcts seen. Gross hydrocephalus. No Alzheimer’s disease or other dementing illness pathology.

1510 g

14

F 70

History extending back several years of schizophrenia, falls and incontinence. Diagnosis of NPH made elsewhere. Four months before death examination showed dementia and Parkinsonian features, the latter considered to be drug-induced.

Slight generalised atrophy, leptomeninges normal. Gross hydrocephalus, multiple small old infarcts in deep grey and cerebral white matter. Severe arteriolar sclerosis. Fresh almond-sized haemorrhage in left thalamus.

1151 g

15

M 87

4 year history of dementia, difficulty finding way about and falls.

Marked ventricular dilation with fenestration of septum. No cortical atrophy, but many diffuse cortical argyrophilic plaques. Neurofibrillary tangles few and confined to hippocampus and adjacent medial temporal lobe cortex.

1415 g

16

M 72

About 18 months history of dementia giddiness, unsteadiness, urinary incontinence and falls.

Marked ventricular dilation without cortical atrophy. Mild cribriform change in white matter; mild loss of pigmented neurons in substantia nigra without Lewy bodies; rare Lewy bodies in locus ceruleus, with moderate neuron loss. Occasional diffuse argyrophilic plaques in cortex. Neurofibrillary tangles rare and confined to entorhinal cortex/hippocampus.

1547 g

some haemosiderin, indicating previous haemorrhage and in one case a small thalamic haemorrhage was present extending to the wall of the third ventricle in addition to old deep grey and white matter softenings (Fig. 19.7). An association between NPH and deep white matter ischaemic lesions has been increasingly recognized in the imaging and neuropathology literature (Bech et al., 1997; Krauss et al., 1997; Hahnel et al., 2000). Such cases tend to have a poor outcome when shunted (Boon et al., 1999; Tullberg et al., 2000). Some cases show pathological changes of Alzheimer’s disease in biopsies taken at the time of shunt insertion or post mortem (Bech et al., 1999; Savolainen et al., 1999; Del Bigio et al., 1997). Some studies record improvement in clinical state after shunting even if Alzheimer’s disease has been demonstrated in biopsies (Golomb et al., 2000; Bech et al., 1999). At post-mortem examination the brain weights in our cases of NPH averaged 1379 grams, with a range of

1151–1767 grams that is well within the normal range. Externally the cerebral cortex usually appeared normal or slightly flattened. In only three cases was there mild atrophy. The leptomeninges were normal, minimally or more markedly thickened. After fixation coronal slicing revealed brains with marked dilatation of the lateral ventricles and third ventricle, sometimes with fenestration of the septum. The corpus callosum appeared of normal or only slightly reduced width. (The stretching that the corpus callosum undergoes makes it appear slightly narrowed). (Fig. 19.8). Obstructive lesions or vascular softenings were present in some cases (Figs. 19.9 and 19.10) and basilar impression in one case (Fig. 19.11). In microscopic sections of the deep cerebral white matter the myelin staining usually appeared normal, but in some cases it appeared diffusely pale and fibre densities in silver stains, attenuated. The ependymal lining of the lateral ventricles was fragmented and the subependymal tissue variably gliotic. In cases with


Fig. 19.8. Normal pressure hydrocephalus (case 13). Coronal slice across frontal and anterior temporal lobes. Frontal horns of the ventricles are markedly enlarged, the corpus callosum appears slightly narrow because of stretching and the cerebral cortex appears relatively well preserved and sulci narrow (compare with Fig. 3.6(e)) in which the degree of ventricular dilatation is comparable but the cortex is much more atrophic).

Fig. 19.7. Case of normal pressure hydrocephalus (case 14) with small organizing right thalamic haemorrhage lying close to the ventricular system.

thickened meninges there was dense acellular collagen deposition, without inflammatory change. In some cases there was haemosiderin deposition. Arachnoid granulations invaginating the sagittal sinus appeared normal or fibrotic. In the hindbrain the fourth ventricle may or may not appear enlarged (even in communicating hydrocephalus, the fourth ventricle is not invariably enlarged because more pressure is required to dilate it than to dilate the already larger lateral ventricles and third ventricle with which they communicate).

Recommended procedure for diagnosis of NPH The diagnosis of NPH should be considered if the clinical features are consistent (and in any case of dementia if clinical details are scanty or non-existent) if the brain

Fig. 19.9. Normal pressure hydrocephalus (case 2) due to an arachnoid cyst in the posterior fossa preventing normal outlet of cerebrospinal fluid from the fourth ventricle. Note enlarged third and lateral ventricles, relatively well-preserved corpus callosum and cerebral cortex and absence of a septum.

at autopsy appears normal or slightly swollen externally and the ventricles are markedly enlarged in the absence of significant atrophy. Marked leptomeningeal thickening in the presence of the above features further strengthens the case, as does the finding of gross obstructive lesions along cerebrospinal fluid pathways. Arteries at the base of the brain should be examined carefully for aneurysms and a search made for rusty discoloration of the leptomeninges which would suggest previous subarachnoid haemorrhage.

453

454


Fig. 19.10. Obstructive hydrocephalus due to non-functioning pituitary tumour in the third ventricle. The inferior horns of the lateral ventricles are particularly enlarged in this case (case 5).

(a)

A careful search should be made for old or fresh vascular lesions in the brain particularly in deep white matter and basal ganglia. Sections should be examined of any focal lesions found and of periventricular white matter and leptomeninges. Because NPH and Alzheimer’s disease may co-exist it is also necessary to examine silver- and immuno-stained sections of hippocampus and neocortex for features of Alzheimer’s disease (Chapter 9). A congo red stain will reveal the extent of amyloid deposition in leptomeningeal vessels. In addition, since the features of gait disturbance and dementia occur in other neurodegenerative diseases, these should also be excluded by examining sections of the substantia nigra, which will show cell loss and Lewy bodies if Parkinson’s disease is present (Chapter 15) and cell loss and neurofibrillary tangles if progressive supranuclear palsy is

(b)

(c)

Fig. 19.11. Paget’s disease with dementia due to hydrocephalus (case 3). (a) View of the inside of the skull during removal of the brain showing greatly thickened skull and compressed hindbrain. (b) Flattened cerebellum and slightly enlarged fourth ventricle. (c) Coronal slice through the cerebrum showing enlarged third and lateral ventricles and dorso-ventral compression of the hemispheres by excessive growth of the base of the skull.


present (Chapter 11). The pons and dentate nuclei of the cerebellum, should also be examined, and will show loss of cells and Lewy bodies in the locus ceruleus in Parkinson’s disease and cell loss and neurofibrillary tangles in pontine tegmentum with cell loss in the dentate nuclei in progressive supranuclear palsy. Old head injury, multiple sclerosis and Pick’s disease, which also enter into the differential diagnosis clinically, should have become apparent from the naked eye appearance of the brain.

REFERENCES Adams, R. D., Fisher, C. M., Hakim, S., Ojemann, R. & Sweet, W. (1965). Symptomatic occult hydrocephalus with ‘normal’ cerebrospinal fluid pressure: a treatable syndrome. New Engl J Med, 273: 117–26. Akai, K., Uchigasaki, S., Tanaka, U. & Komatsu, A. (1987). Normal pressure hydrocephalus. A neuropathological study. Acta Pathol Jpn, 37: 97–110. Bech, R. A., Juhler, M., Waldemar, G., Klinken, L. & Gjerris, F. (1997). Frontal brain and leptomeningeal biopsy specimens correlated with cerebrospinal fluid outflow resistance and B-wave activity in patients suspected of normal-pressure hydrocephalus. Neurosurgery, 40: 497–502. Bech, R. A., Waldemar, G., Gjerris, F., Klinken, L. & Juhler, M. (1999). Shunting effects in patients with idiopathic normal pressure hydrocephalus; correlation with cerebral and leptomeningeal biopsy findings. Acta Neurochir, 141: 633–9. Benson, D. F., LeMay, M., Patten, D. H. & Rubens, A. B. (1970). Diagnosis of normal-pressure hydrocephalus. N Engl J Med, 283: 609–15. Betz, A. L., Goldstein, G. W. & Katzman, R. (1994). Blood–braincerebrospinal fluid barriers. In Siegel, G. J., Agranoff, B. W., Albers, R. W., Molinoff, P. B. (eds) Basic Neurochemistry: Molecular, Cellular, and Medical Aspects, pp. 681–702. New York: Raven Press. Boon, A. J., Tans, J. T., Delwel, E. J. et al. (1998). Does CSF outflow resistance predict the response to shunting in patients with normal pressure hydrocephalus? Acta Neurochir Suppl (Wien), 71: 331–3. Boon, A. J., Tans, J. T., Delwel, E. J. et al. (1999). Dutch NormalPressure Hydrocephalus Study: the role of cerebrovascular disease. J. Neurosurg, 90(2): 221–6. (2000). The Dutch normal-pressure hydrocephalus study. How to select patients for shunting? An analysis of four diagnostic criteria. Surg Neurol, 53(3): 201–7. Chen, I. H., Huang, C. I., Liu, H. C. & Chen, K. K. (1994). Effectiveness of shunting in patients with normal pressure hydrocephalus predicted by temporary, controlled-resistance, continuous lumbar drainage: a pilot study. J Neurol Neurosurg Psychiat, 57(11): 1430–2.

Cserr, H. F. (1971). Physiology of the choroid plexus. Physiol Rev, 51: 273–311. Cummings, J. L. & Benson, D. F. (1983). Hydrocephalic dementia. Chapter 7 in Dementia: A Clinical Approach, pp. 213–236. Boston: Butterworths. Cushing, H. (1925). The Third Circulation. London: Oxford University Press. Davson, H., Welch, H. K., Segal, M. B. (eds) (1987). Physiology and Pathophysiology of the Cerebrospinal Fluid. Edinburgh: Churchill Livingstone. De Land, F. H., James, A. E. Jr, Ladd, D. J. & Konigsmark, B. W. (1972). Normal pressure hydrocephalus: a histologic study. Am J Clin Pathol, 58: 58–63. Del Bigio, M. R., Cardoso, E. R. & Halliday, W. C. (1997). Neuropathological changes in chronic adult hydrocephalus: cortical biopsies and autopsy findings. Can J Neurol Sci, 24: 121–6. Derouesne, C., Gray, F., Escourolle, R. & Castaigne, P. (1987). ‘Expanding cerebral lacunae’ in a hypertensive patient with normal pressure hydrocephalus. Neuropathol Appl Neurobiol, 13: 309–20. Di Rocco, C., Di Trapani, G., Maira, G. et al. (1977). Anatomo-clinical correlations in normotensive hydrocephalus. J Neurol Sci, 33: 437–52. Earnest, M. P., Fahn, S., Karp, J. H. & Rowland, L. P. (1974). Normal pressure hydrocephalus and hypertensive cerebrovascular disease. Arch Neurol, 31: 262–6. Fishman, R. A. (1975). Brain edema N Engl J Med, 293: 706–11. (1992). Cerebrospinal Fluid in Diseases of the Nervous System 2nd edn. Philadelphia: W. B. Saunders Co. Foltz, E. L. & Ward, A. A. (1956). Communicating hydrocephalus from subarachnoid bleeding. J Neurosurg, 13: 546–66. Fukuhara, T. & Luciano, M. G. (2001). Clinical features of late-onset idiopathic aqueductal stenosis. Surg Neurol, 55(3): 132–6. Gallassi, R., Morreale, A., Montagna, P. et al. (1991). Binswanger’s disease and normal-pressure hydrocephalus. Clinical and neuropsychological comparison. Arch Neurol, 48(11): 1156–9. Golomb, J., Wisoff, J., Miller, D. C. et al. (2000). Alzheimer’s disease comorbidity in normal pressure hydrocephalus: prevalence and shunt response. J Neurol Neurosurg Psychiatry, 68: 778–81. Graff-Radford, N. R. & Godersky, J. C. (1987). Idiopathic normal pressure hydrocephalus and systemic hypertension. Neurology, 37(5): 868–71. Hahnel, S., Freund, M., Munkel, K. et al. (2000). Magnetisation transfer ratio is low in normal-appearing cerebral white matter in patients with normal pressure hydrocephalus. Neuroradiology, 42: 174–9. Hakim, S., Adams, R. D. (1965). The special clinical problem of symptomatic hydrocephalus with normal cerebrospinal fluid pressure. J Neurol Sci, 2: 307–27. Katzman, R. (1977). Normal pressure hydrocephalus Chapter 4 In Wells CE (ed) Dementia 2nd edn, pp. 69–92. Philadelphia: F. A. Davis. Katzman, R. & Pappius, H. M. (1973). Brain Electrolytes and Fluid Metabolism. Baltimore: Williams & Wilkins Co.

455

456


Kimura, M., Tanaka, A. & Yoshinaga, S. (1992). Significance of periventricular hemodynamics in normal pressure hydrocephalus. Neurosurgery, 30(5): 701–4; discussion 704–5. Koto, A., Rosenberg, G., Zingesser, I. H., Horoupian, D. & Katzman, R. (1977). Syndrome of normal pressure hydrocephalus: possible relation to hypertensive and arteriosclerotic vasculopathy. J Neurol Neurosurg Psychiatry, 40: 73–9. Krauss, J. K., Regel, J. P., Vach, W. et al. (1997). White matter lesions in patients with idiopathic normal pressure hydrocephalus and in an age-matched control group: a comparative study. Neurosurgery, 40: 491–5; discussion 495–6. Kristensen, B., Malm, J., Fagerland, M. et al. (1996). Regional cerebral blood flow, white matter abnormalities, and cerebrospinal fluid hydrodynamics in patients with idiopathic adult hydrocephalus syndrome. J Neurol Neurosurg Psychiatry, 60(3): 282–8. Larsson, A., Arlig, A., Bergh, A. C. et al. (1994). Quantitative SPECT cisternography in normal pressure hydrocephalus. Acta Neurol Scand, 90(3): 190–6. Little, J. R., Houser, O. W. & MacCarty, C. S. (1975). Clinical manifestations of aqueductal stenosis in adults. J Neurosurg, 43: 546–52. McHugh, P. R. (1964). Occult hydrocephalus. Q J Med, 33: 297–308. Messert, B., Wannamaker, B. & Dudley, A. (1972). Re-evaluation of the size of the lateral ventricles of the brain. Neurology, 224: 941–51. Milhorat, T. H., Hammock, M. K., Fenstermacher, J. D. & Levin, V. A. (1971). Cerebrospinal fluid production by the choroid plexus and brain. Science, 173: 330–2. Morgagni, G. B. (1769). The Seats and Causes of Diseases Investigated by Anatomy. Alexander B. Trans. London: Miller and Caddell. Mun-Bryce, S. & Rosenberg, G. A. (1998). Matrix metalloproteinases in cerebrovascular disease. J Cereb Blood Flow Metab, 18(11): 1163–72. Newton, H., Pickard, J. D. & Weller, R. O. (1989). Normal pressure hydrocephalus and cerebrovascular disease: findings of post mortem. J Neurol Neurosurg Psychiatry, 52: 804. Ott, KH., Kase, C. S., Ojemann, R. G. & Mohr, J. P. (1974). Cerebellar haemorrhage: diagnosis and treatment. A review of 56 cases. Arch Neurol, 31: 160–7. Pickard, J. D. (1991). Normal pressure hydrocephalus. In Swash M., Oxbury, J. O. (eds) Clinical Neurology, pp. 151–64. Edinburgh: Churchill Livingstone. Pollay, M. (1975). Formation of cerebrospinal fluid. J Neurosurg, 42: 665–73. Raftopoulos, C., Deleval, J., Chaskis, C. et al. (1994). Cognitive recovery in idiopathic normal pressure hydrocephalus: a prospective study. Neurosurgery, 35(3): 397–404; discussion 404–5.

Riddoch, G. (1936). Progressive dementia without headache or changes in the optic disc, due to tumours of the third ventricle. Brain, 59: 225–33. Rosenberg, G. A. (1990). Brain Fluids and Metabolism. New York: Oxford University Press. Rosenberg, G. A., Kyner, W. T. & Estrada, E. (1980). Bulk flow of brain interstitial fluid under normal and hyperosmolar conditions. Am J Physiol, 238: 42–9. Rosenberg, G. A., Saland, L. & Kyner, W. T. (1983). Pathophysiology of periventricular tissue changes with raised CSF pressure in cats. J Neurosurg, 59: 606–11. Roth, Y., Kimhi, Y., Edery, H., Aharonson, E. & Priel, Z. (1985). Ciliary motility in brain ventricular system and trachea of hamsters. Brain Res, 330: 291–7. Saito, Y. & Wright, E. M. (1983). Bicarbonate transport across the frog choroid plexus and its control by cyclic nucleotides. J Physiol, 336: 635–48. Sand, T., Bovim, G., Grimse, R., Myhr, G., Helde, G. & Cappelen, J. (1994). Idiopathic normal pressure hydrocephalus: the CSF tap-test may predict the clinical response to shunting. Acta Neurol Scand, 89(5): 311–16. Savolainen, S., Paljarvi, L. & Vapalahti, M. (1999). Prevalence of Alzheimer’s disease in patients investigated for presumed normal pressure hydrocephalus: a clinical and neuropathological study. Acta Neurochir, 141: 849–53. Tomlinson, B. E., Blessed, G. & Roth, M. (1970). Observations on the brains of demented old people. J Neurol Sci, 11: 205– 42. Tripathi, B. J. & Tripathi, R. C. (1974). Vacuolar transcellular channels as a drainage pathway for cerebrospinal fluid. J Physiol, 239: 195–206. Tullberg, M., Mansson, J. E., Fredman, P. et al. (2000). CSF sulfatide distinguishes between normal pressure hydrocephalus and subcortical arteriosclerotic encephalopathy. J Neurol Neurosurg Psychiatry, 69: 74–81. Vanneste, J. A. (2000). Diagnosis and management of normalpressure hydrocephalus. J Neurol, 247(1): 5–14. Vessal, K., Sperber, E. E. & James, A. E. Jr. (1974). Chronic communicating hydrocephalus with normal CSF pressures: a cisternographic-pathologic correlation. Ann Radiol, 17: 785– 93. Waldemar, G., Schmidt, J. F., Delecluse, F., Andersen, A. R., Gjerris, F. & Paulson, O. B. (1993). High resolution SPECT with [99mTc]-d,l-HMPAO in normal pressure hydrocephalus before and after shunt operation. J Neurol Neurosurg Psychiatry, 56: 655–64. Weed, L. H. (1935). Certain anatomical and physiological aspects of the meninges and cerebrospinal fluid. Brain, 58: 383.

20 Head injury and dementia Colin Smith1,2 , James A. R. Nicoll1,3 and David I. Graham1 1

Department of Neuropathology, University of Glasgow, UK 2 Department of Pathology, University of Edinburgh, UK 3 Division of Clinical Neurosciences, University of Southampton, General Hospital, UK

Introduction Traumatic brain injury remains a significant cause of morbidity and mortality throughout the world. In the United Kingdom more than 150 000 patients are admitted to hospital each year with a head injury. Of this group more than 80% are classified as having a mild head injury, as defined by the Glasgow Coma Scale (GCS). The GCS (Teasdale & Jennett 1974, 1976) provides a means of quantifying the level of consciousness after traumatic brain injury based on the clinical features of verbal performance, eye opening and motor response. Using this scale three levels of severity of head injury are defined; mild (score 13–15), moderate (score 9–12), and severe (score 3–8). Approximately 1–2% of patients admitted to hospital after traumatic brain injury die as a consequence of their injuries. The majority of fatalities are within the severe head injury group, with 40% of the cases resulting in death at 6 months (Murray et al., 1999). Among survivors of traumatic brain injury of all grades chronic disability may have a physical component although it is predominantly the cognitive and behavioural problems which provide the greatest challenge (Jennett et al., 1981). Outcome may be assessed by the Glasgow Outcome Scale (GOS) (Jennett & Bond 1975) which defines four outcome states; death/vegetative state, severe disability, moderate disability, and good recovery. The GOS is based predominantly on assessment of social reintegration after traumatic brain injury involving a structured questionnaire-based interview. This has recently been modified as the extended GOS (Teasdale et al., 1998). Predictors of neurobehavioural outcome in adults include age (greater than 50 years is a poor prognostic factor), the

acute GCS, abnormal brain stem reflexes, subacute ventricular enlargement, neurological deficit, and the duration of post-traumatic amnesia (Capruso & Levin, 2000). Somewhat surprisingly, recent studies have indicated that the incidence of moderate and severe disability in young people and adults one year after mild head injury is similar to that seen in survivors of moderate and severe head injury (Thornhill et al., 2000). In addition there is evidence (discussed below) that traumatic brain injury may be associated with continuing cognitive decline in later years and with an increased incidence of Alzheimer’s disease (AD). The mechanisms underlying the association between head injury and AD are unknown as yet, although, as we shall discuss in this chapter, the response to traumatic brain injury and the pathology of AD have some features in common not only in terms of a cellular and protein response but also striking parallels in the genetic influences. In order to attempt to clarify the mechanisms underlying post-traumatic cognitive deficit a basic understanding of the pathology of traumatic brain injury is helpful. Blunt force head injury results in both focal and diffuse pathologies involving the skull and the underlying brain and its coverings. Focal lesions can take the form of skull fractures, cerebral contusions (Fig. 20.1), focal ischaemic lesions secondary to raised intracranial pressure, and intracranial haematomas. Diffuse lesions may take the form of cerebral ischaemia or cerebral swelling, or may develop as a consequence of rotational forces (diffuse traumatic axonal injury) (Graham et al., 1995b). The primary injury is related to mechanical damage, and can be focal or diffuse, or a combination of both. It is related to the effects of both the impact and inertial forces on the skull and brain. Delayed secondary events such as diffuse traumatic axonal


457

458

C. Smith, J. A. R. Nicoll and D. I. Graham

Fig. 20.1. Healed cerebral contusions. There are orange-brown scars on the under aspect of the frontal and temporal lobes.

injury and cerebral ischaemia develop over a period of hours, days, or weeks after the traumatic episode. The secondary events may be related to neurochemical alterations and the associated cellular and molecular alterations induced by trauma (Graham et al., 2000).

Effects of head injury on cognitive function Concussion refers to an immediate, usually reversible episode of brain dysfunction after traumatic brain injury. A clinical spectrum is recognized ranging from mild concussion, in which consciousness is often preserved, to severe diffuse axonal injury resulting in the vegetative state (Gennarelli, 1993). The anatomical basis of concussion syndromes is currently considered to be diffuse axonal pathology and, in particular, axonal disruption resulting in disconnection between areas involved in consciousness; cerebral cortex, brainstem reticular activating areas, thalamus

and hypothalamus (Gennarelli, 1993). The vegetative state refers to a group of patients who have loss of meaningful cognitive function and awareness, but retain spontaneous breathing and periods of wakefulness. The neuropathological basis of the vegetative state has been defined in a study that examined 49 patients in the vegetative state, 35 of whom had experienced traumatic brain injury (Adams et al., 2000) (Fig. 20.2). In the trauma-related cases diffuse traumatic axonal injury of grade 2 or 3 was found in 71% of cases, and thalamic pathology in 80% of cases. In cases with minimal brainstem and cerebral cortical pathology, thalamic pathology was always present. Therefore, damage to the thalamic nuclei and/or the afferent and efferent white matter pathways of the thalamus appear to play a major role in the genesis of the vegetative state after head injury. White matter (Wallerian) degeneration is a consequence of severe diffuse traumatic axonal injury (Fig. 20.3). The axonal loss results in gliosis and macrophage activation (Fig. 20.4), which may be under genetic control as discussed

Head injury and dementia

Fig. 20.2. Coronal slice-vegetative. 21-month survival post RTA. Note the thinning of the corpus callosum, ventricular enlargement and small rather granular thalami due to diffuse traumatic axonal injury.

later. In contrast, the structural basis of moderate disability after traumatic brain injury is more likely to be a focal lesion rather than diffuse brain pathology, usually an evacuated intracranial haematoma (Adams et al., 2001). In a study of 30 severely disabled patients 50% had focal brain pathology only. Some severely disabled patients did show diffuse brain pathology similar to vegetative state patients, and it may be that there is a greater quantitative amount of damage in the vegetative cases (Jennett et al., 2001).

Long-term outcome from head injury and chronic neurodegeneration Clinical studies Mild head injury (acute GCS 13–15) is associated with a higher than expected incidence of disability (GOS moderate or severe disability) at 1 year post-injury (Thornhill et al., 2000). Of major interest in the context of this discussion are the longer-term effects on cognition many years after the

injury, and the relationship between traumatic brain injury and AD. There is a considerable epidemiological literature examining the relationship between traumatic brain injury and the development of AD in later life. Many of these studies take the form of retrospective case-control studies and are therefore subject to recall bias. A number of these studies have reported an association between traumatic brain injury and AD (French et al., 1985; Graves et al., 1990; Mortimer et al., 1991; van Duijn et al., 1992; Mayeux et al., 1993; Rasmussen et al., 1995; Salib & Hillier, 1997; O’Meara et al., 1997; Nemetz et al., 1999; Guo et al., 2000), although some do not reach statistical significance (Chandra et al., 1987, Amaducci et al., 1986). In particular, the study by Mayeux et al. in 1993 reported an almost four-fold increased risk of developing AD after traumatic brain injury when compared to age-matched controls. Guo et al. (2000) studied 2233 individuals who met the criteria for probable or definite AD, and compared them with 14 668 controls (first-degree relatives or spouses) as part of the MIRAGE (Multi-Institutional Research in Alzheimer Genetic

459

460


(a)

(b) Fig. 20.3. Wallerian degeneration. Same case as Fig. 20.1. (a) Normal corticospinal tract. (b) Pallor of staining with increased cellularity. A&B Luxol fast blue/Cresyl violet (x 180).

Epidemiology) project. They reported that traumatic brain injury was a risk factor for AD and that the risk was proportional to the severity of the injury. For example, comparison of probands with unaffected spouses yielded odds ratios for AD of 9.9 for head injury with loss of consciousness and 3.1 for head injury without loss of consciousness. Comparison of probands with their parents and sibs were 4.0 for head injury with loss of consciousness and 2.0 for head injury without loss of consciousness. At age 93 years the lifetime risk of developing AD was 77.2% for those with and 40.1% for those without a history of head injury. Other retrospective case-control studies, however, have not confirmed that there is an association between traumatic brain injury and AD (Broe et al., 1990; Ferini-Strambi et al., 1990; Li et al., 1992; Mendez et al., 1992; Fratiglioni et al., 1993).

To try and address the problems of recall bias inherent in case-control studies a number of prospective studies have been designed. Again, however, there is conflicting data. Corkin et al. (1989) performed neuropsychological assessment of 57 World War 2 veterans with a penetrating head injury at two time points 30 years apart, and compared their performance with 27 veterans who suffered a peripheral nerve injury only and who were assessed over the same 30-year period. They found that a penetrating head injury exacerbated the decline in cognitive performance over time when compared with the peripheral injury group. Schofield et al. (1997) reported a communitybased longitudinal study of ageing in north Manhattan. 271 participants without significant cognitive impairment at the time of enrolment were interviewed in relation to previous head injury and associated loss of consciousness. Patients were then followed up for 5 years with annual evaluations. They reported that previous traumatic brain injury was a risk factor for AD with a three-fold increased risk. Plassman et al. (2000) examined 1776 World War 2 navy and marine veterans, with military medical records. 548 had a history of non-penetrating traumatic brain injury, 1228 did not. All individuals were assessed for AD. They found that in this group moderate head injury (Frankowski scale, Frankowski et al., 1985) resulted in 2.3 × increased risk of AD, while severe head injury resulted in a four-fold increased risk. Against this data, however, there are a number of prospective studies which have failed to demonstrate an association between traumatic brain injury and AD (Katzman et al., 1989; Aronson et al., 1990; Williams et al., 1991; Breteler et al., 1995). Launer et al. (1999), as part of the European Studies of Dementia (EURODEM), analysed four European population-based prospective studies, with individuals aged 65 years or older at time of recruitment. This large study did not find an association between traumatic brain injury and AD. Mehta et al. (1999) reported the prospective population-based Rotterdam study, which looked at 6645 individuals aged 55 years or older and who did not have dementia when recruited. This study found that mild traumatic brain injury was not associated with an increased risk of AD, although the follow-up period was short being on average 2.1 years after initial assessment. There are many difficulties in assessing the relationship between traumatic brain injury and AD as the conflicting results presented above clearly illustrate. Retrospective case-control studies have both recall and selection bias. The prospective studies retain a lesser degree of recall bias and do not rely on the recollections of cognitively impaired individuals. However, some of the prospective studies have only a short follow up period, 5 years in many cases, and


this may bias the outcome. Comparisons between studies are difficult due to differences in definitions of severity of brain injury, post injury outcome status, and clinical definitions of AD. Also the age at the time of the injury and the age at the time of assessment are likely to be important variables.

Boxers and dementia pugilistica While the data relating to long term associations between traumatic brain injury and AD is currently conflicting, the association between relatively mild repetitive head injury and cognitive impairment has been established in the literature for many years. The ‘punch drunk’ state was first described by Martland in 1928 and was renamed dementia pugilistica by Millspaugh in 1937. Dementia pugilistica was fully reviewed in the previous edition of this book (Bruton 1997) and a summarized account will be presented here. This condition is described in boxers who have competed in many bouts over a long period of time. Clinically, they develop a degree of intellectual deterioration often with an associated movement disorder, usually parkinsonism but in some cases predominantly ataxia. The largest study of this disorder clinically (Roberts, 1969) examined 224 ex-boxers using neurological examination, electroencephalogram, and simple psychometric testing. He found that 17% had varying degrees of movement disorder involving the cerebellar, pyramidal and extrapyramidal systems. Minor degrees of intellectual function were seen in several of the ex-boxers, although only two required long term care as a result of their cognitive impairment. Roberts concluded that the occurrence of encephalopathy increased significantly with the number of bouts and the length of the boxer’s career. He also concluded, however, that the rate of cognitive decline was not greater than that associated with ageing alone. More recent studies (Casson et al., 1984; McLatchie et al., 1987; Brooks et al., 1987; Murelius & Haglund, 1991; Heilbronner et al., 1991) suggest that fullblown dementia pugilistica is now rarely seen, although mild cognitive and movement disorders are still associated with boxing. While dementia pugilistica was initially described in relation to boxing, cases have been described in National Hunt jockeys (Foster et al., 1976). In addition, there is a considerable literature relating to the risks of repetitive mild traumatic head injury and other sports such as soccer (Matser et al., 1999, Kirkendall et al., 2001), rugby union and Australian rules football (McIntosh et al., 2000), American football (Maroon et al., 2000), and ice hockey (Biasca et al., 1993). In the absence of large prospective studies the risk of cognitive impairment and movement

(a)

(b) Fig. 20.4. Wallerian degeneration. Same case as Fig. 20.1. (a) Corticospinal tract in pons (bottom right) showing an astrocytosis compared with normal transverse fibres of pons (top left). GFAP (x180). (b) Corticospinal tract (right) in pons. There are many macrophages compared with normal transverse fibres of pons (left). CD 68 (x 180).

disorders secondary to repetitive mild traumatic brain injury in relation to these contact sports, remains uncertain. The largest pathological assessment of dementia pugilistica was the examination of the brains of 15 boxers, 11 of whom were diagnosed with dementia pugilistica in life (Corsellis et al., 1973). This followed on from previous case reports (Brandenburg & Hallervorden 1954; Grahmann & Ule, 1957; Constantinidis & Tissot, 1967) and the

461

462


Fig. 20.5. Macroscopic view of a boxer’s septum. The septal leaves are widely separated forming a large cavum. Only a few strands of tissue remain (x 2.5).

descriptions by Mawdsley and Ferguson (Mawdsley & Ferguson, 1963; Ferguson & Mawdsley, 1965) of the brains of four ex-boxers. Corsellis et al. (1973) reported four principal features of the brain in dementia pugilistica: (i) abnormalities of the septum pellucidum, (ii) cerebellar damage, (iii) degeneration of the substantia nigra, and (iv) cerebral cortex pathology. (i) A fenestrated cavum septum pellucidum was seen in 77% of ex-boxers but in only 3% of non-boxers. One third of the non-boxers who had a fenestrated cavum septum pellucidum had evidence of a previous head injury (Fig. 20.5). The degree of separation of the two leaflets of the septum pellucidum may be related to repetitive injury being most pronounced in the ex-boxers. (ii) Ataxia may be a feature of dementia pugilistica. Corsellis et al. (1973) described cortical scarring of the inferior aspects of the lateral cerebellar hemispheres adjacent to the tonsils in 10 of the 15 ex-boxers brains studied. Histologically there was gliosis and loss of both Purkinje cells and granule cells (Fig. 20.6). (iii) Parkinsonism is a common feature of dementia pugilistica and substantia nigra pathology appears to be the underlying cause. Pigmented cell loss is often marked, both within the substantia nigra and the locus ceruleus, and neurofibrillary tangles can be seen in some of the

remaining neurons (Fig. 20.7). Lewy bodies are not a feature (Corsellis et al., 1973). (iv) Gross cortical pathology, a common feature of acute traumatic brain injury in the form of contusions, does not appear to be a significant feature of dementia pugilistica (Corsellis et al., 1973; Adams & Bruton, 1989). However, diffuse microscopic cortical pathology is a feature of dementia pugilistica (see below).

Pathological mechanisms which may underlie long term neurodegeneration after head injury Over-representation of late cognitive decline in survivors of traumatic brain injury may simply reflect the additive effects of the acute damage and later age-related functional compromise. From this viewpoint the acute injury acts to decrease the ‘functional reserve’ of the brain and subsequent age-related neurodegeneration is more likely at an earlier age to result in traverse of the threshold of impairment required to manifest as dementia. However, there are remarkable parallels in the pathological processes involved both in the response to traumatic brain injury and AD (see below). It is possible that a component of the acute response to traumatic brain injury acts as a ‘trigger’ to initiate a positive feedback loop that smoulders away to become


Fig. 20.6. Low power microscopic view of cerebellar folia in the tonsillar region of a boxer’s cerebellum. The molecular layer of the scarred cortex is narrowed and intensely gliosed.

manifest in later years as frank neurodegenerative pathology and dementia (Nicoll et al., 1995; Griffin et al., 1998) (Fig. 20.8). Detailed classical neuropathological descriptions of cohorts of long-term survivors of traumatic brain injury including immunohistochemical studies are lacking. However, some of the processes that are believed to be involved in chronic neurodegeneration, including AD, have been explored in the context of both acute injury and long-term survival after trauma.

(a) Cytoskeletal neurodegenerative pathology Cytoskeletal pathology after diffuse traumatic brain injury has been examined experimentally using a pig model with injury induced via controlled head rotational acceleration (Smith et al., 1999). Head-injured pigs were examined at days 1, 3, 7 and 10 post-injury, and compared to control animals without head injury. Within the experimental group tau and neurofilament accumulations were identified immunohistochemically within the white matter, co-localised with damaged axons (APP immunoreactive), and within neuronal perikarya in the cerebral cortex. To date a similar observation has not been made in humans following a single episode of traumatic brain injury although cleaved forms of tau protein are elevated markedly in the CSF of brain-injured patients (Zemlan et al., 1999). However, in

Fig. 20.7. (a) Transverse cut through the midbrain of an elderly male non-boxer to show the normal pigmentation of the substantia nigra (x 0.75). (b) (c) and (d). The substantia nigra of three punch drunk boxers. Some pigment is still visible in (b) but (c) and (d) are almost totally devoid of pigmentation (x 0.75).

cases of repetitive mild head injury cytoskeletal pathology is observed. Neurofibrillary tangles were reported in exboxers by Corsellis et al. (1973) scattered throughout the cerebral cortex and the brainstem, being most prominent in the medial temporal cortex (Fig. 20.9). Recently, Geddes et al. (1999) examined the brains of four young individuals

463

464


Cognitive reserve

Loss of functional reserve

Acute effects of TBI

Recovery

Threshold for dementia

Pre-injury

Post-injury

Age

Fig. 20.8. Graphical illustration of potential mechanisms operating after head injury. See text for further explanation.

(a)

(b) Fig. 20.9. Neurofibrillary tangles affecting most neurons in the parahippocampal gyrus in a case of dementia pugilistica. Note the ¨ absence of senile plaques. von Braunmuhl’s silver stain (×100).

(age range from 23–28 years) with a history of repetitive head injury (two boxers, one footballer, and one mentally subnormal patient with a history of self inflicted head banging) and a frontal lobectomy specimen from an individual with intractable complex partial seizures with recurrent minor head injury. They identified widespread neocortical neurofibrillary tangles and neuropil threads not seen in age matched controls which in areas showed a perivascular distribution (Fig. 20.10). Schmidt et al. (2001) have compared the molecular profiles of the neurofibrillary tangles in dementia pugilistica and AD. They found that dementia pugilistica and AD had a common tau isoform and phosphorylation profile. They concluded that the mechanisms underlying both these conditions might be similar. (b) Amyloid deposition Diffuse A plaques have been identified in approximately 30% of individuals who die shortly after a single episode of severe traumatic brain injury (Roberts et al., 1991, 1994; Graham et al., 1995a). This is a higher proportion than in non-head injury controls. Other groups (Adle-Biassette et al., 1996), however, have not confirmed this observation. Most of this is in the form of A42 (Gentleman et al., 1997; Horsburgh et al., 2000), which is believed to be of pathological significance in AD. The distribution of the plaques does


plaque-like structures were seen in most cases of dementia pugilistica (Roberts et al., 1990) although the neuritic plaques characteristic of AD were absent. In the study by Geddes et al. (1999) A plaques were not seen despite using both a modified Bielschowsky silver stain and A immunohistochemistry with formic acid pre-treatment. They concluded that neurofibrillary tangle formation in the absence of A deposition is an early consequence of repetitive head injury and that, because of their striking perivascular distribution, neurofibrillary tangle formation may be related to damage to blood vessels. The relationship between A deposition and genetic polymorphisms is discussed in a later section.

(a)

(b) Fig. 20.10. Neurofibrillary tangles in the neocortex from a case of repetitive head injury. (a) Neurofibrillary tangles showing a perivascular distribution. Tau (x 100). (b) Intraneuronal tau immunoreactivity in a neocortical neuron. Tau (x 180).

not correlate with focal traumatic lesions such as contusions but may be an expression of a diffuse acute phase response (e.g. hypoxia, acidosis, oedema, reduced cerebral blood flow) (Graham et al., 1995a). For example, the A deposits may be the result of increased production or altered distribution of APP, increased cleavage of APP in a proteolytic environment to produce A, an alteration in the balance of production of A40 :A42, extracellular conditions which favour the precipitation of amyloid fibrils, or decreased removal or drainage of A. In the study by Corsellis et al. (1973) neurofibrillary tangles were found in the almost complete absence of senile plaques when examined using silver (Bielschowsky) and Congo red stains. However, when this was re-examined using immunohistochemistry with formic acid pre-treatment for the -amyloid protein (A), extensive immunoreactive

(c) Neuronal loss Neuronal loss after traumatic brain injury has been reported in the neocortex, the hippocampus, the cerebellum and the thalamus (Adams et al., 1985; Kotapka et al., 1992; Ross et al., 1993). In the acute phase, neuronal loss is related to contusions or as a consequence of cerebral hypoxia/ischaemia, and bilateral hippocampal neuronal loss has been documented in 85% of cases in one study (Adams et al., 1985). The mechanisms of cell death have been extensively studied and the processes of necrosis and programmed cell death have been considered to be separate mechanisms, although this view is being increasingly challenged and shared molecular pathways have been identified in both processes. The role of programmed cell death after traumatic brain injury has been reviewed by Raghupathi et al. (2000). Cell death has been identified in situ after traumatic brain injury in both animal models and in human material using the terminal deoxynucleotidyl transferase mediated dUTP nick end-labelling (TUNEL) technique (Rink et al., 1995, Smith et al., 1997). This technique identifies DNA fragmentation, a feature common to both necrosis and programmed cell death. Differentiation between necrosis and programmed cell death is possible by assessing other mechanisms seen in programmed cell death such as caspase activation, and identification of the morphological expression of programmed cell death, apoptosis. TUNEL positive neurons and oligodendroglial cells have been reported in human traumatic brain injury; Clark et al. (1999) demonstrated elevated levels of bcl-2 and caspase 3, increased cleavage of both caspases 1 and 3, and cells with the morphological appearances of apoptosis in 8 patients who had contusions removed surgically between 1 and 9 days after an episode of traumatic brain injury. Smith et al. (2000) and Shaw et al. (2001) studied a number of brain areas in human post-mortem tissue of 18 patients who survived between 6 hours and 10 days after traumatic brain

465

466


injury. TUNEL-positive cells were seen in both grey and white matter, peaking between 25 and 48 hours although still identifiable at 10 days post injury. There was a mixture of both apoptotic and necrotic morphology in neurons, although white matter TUNEL-positive cells more consistently showed an apoptotic morphology. They concluded that in human frontal lobe contusions both apoptosis and necrosis contributed to post-traumatic pathology, and that multiple cell types, including neurons, were involved. Recent experimental studies suggest that the cellular pathology initiated by an episode of acute traumatic brain injury may indeed be progressive. Rats subjected to severe lateral fluid-percussion brain injury were studied for up to 12 months and showed long term cognitive and neurological motor dysfunction (Pierce et al., 1998) accompanied by continuing cell loss (Smith et al., 1997). Recent studies in human cases have demonstrated TUNEL-positive cells up to 12 months after traumatic brain injury (Williams et al., 2001). Again, the majority of the cells were present in the white matter and were considered to be closely associated with Wallerian degeneration. Long-term DNA fragmentation therefore appears to be a feature of traumatic brain injury in humans. (d)Cholinergic brain pathways The nucleus basalis of Meynert within the basal forebrain provides cholinergic innervation of the cerebral cortex and the hippocampus, and damage to this pathway can result in attention, memory and emotional dysfunction (Everitt & Robbins, 1997). Abnormalities within the cholinergic projection system have been postulated to contribute both to the altered mental state in AD (Geula & Mesulam, 1994) and to the neurobehavioural sequelae which persist after a head injury (Cardenas et al., 1994). In rats there is a reduction in the number of choline acetyltransferase [ChAT] positive neurons after experimentally induced traumatic brain injury (Schmidt & Grady, 1995), and alterations of cholinergic innervation of the cerebral cortex and hippocampus have been detected (Dixon et al., 1995, 1997). Patients who die acutely as a consequence of traumatic brain injury have reduced levels of cortical ChAT when compared to age-matched controls (Dewar & Graham, 1996; Murdoch et al., 1998). Recently, neuronal damage has been demonstrated within the nucleus basalis of Meynert in eight of twelve fatally head injured patients, with a median survival time of 27 hours (Murdoch et al., 2001). Neuronal damage was a result of both mechanical distortion (tissue herniation) and focal ischaemia. The authors concluded that damage to cholinergic neurons may contribute to the dysfunction of memory and cognition in survivors

of traumatic brain injury, although studies of the nucleus basalis of Meynert in long-term survivors has not been undertaken. (e) Neuroinflammation Recent studies have focussed attention on ‘neuroinflammation’ as a potential culprit both in AD and in the response to brain injury (Engel et al., 2000; Nicoll et al., 2000; Griffin et al., 1998). The principal mediator of inflammatory processes in the central nervous system is the microglial cell. Microglia have a variety of functions including antigen presentation, synthesis and secretion of cytokines and phagocytosis. These cells are a source of several of the proteins upregulated both in AD and after traumatic brain injury, including apoE, and pro-inflammatory cytokines such as interleukin 1 (IL-1). This raises the question that patients who sustain a head injury may have a microglial response which plays a role both in influencing their outcome following injury and their increased susceptibility to AD later in life. IL-1 is thought to orchestrate the inflammatory responses within the brain after an insult, resulting in a number of responses including: (a) microglial proliferation (Ganter et al., 1992), (b) induction of neuronal production of APP (Goldgaber et al., 1989), and (c) astrocytic activation with upregulation of astrocytic-derived proteins (Das & Potter, 1995). IL-1 is expressed in increased quantities in the cerebral cortex within hours of traumatic brain injury (Griffin et al., 1994), and chronic overexpression of IL-1 is found in AD (Griffin et al., 1989). Griffin et al. (1998) have proposed a ‘Cytokine Cycle’ in which traumatic brain injury, or other brain insults, can, in susceptible individuals, initiate an overexuberant sustained inflammatory response which can result in neurodegeneration. APP and the astrocyteproduced molecule S-100 are upregulated in response to increased IL-1 levels, and are known to be upregulated in AD (Griffin et al., 1989; Mrak et al., 1996). APP is not only upregulated in acute traumatic brain injury, but there is increased intraneuronal processing of the molecule (Buxbaum et al., 1992) potentially resulting in A production and deposition. The relation between IL-1 and APP in the acute phase is uncertain, but increased levels of IL-1 may result in sustained APP, and therefore A, production. Positive feedback of this interaction may be provided by soluble fragments of APP (sAPP) which are produced by the processing of APP. sAPP promotes microglial activation by a mechanism that is modulated by apoE in an isoform specific fashion (Barger & Harmon, 1997). IL-1 positive microglial cells lie in close relation to APP positive neurons and dystrophic neurites in the


brains of head-injured patients (Griffin et al., 1994) and are also found in close apposition to neurofibrillary tanglecontaining neurons in AD (Sheng et al., 1997). IL-1 is known to be trophic to neurons in low concentrations, but at higher concentrations IL-1 has a neurotoxic effect, inducing overexpression and phosphorylation of both neurofilaments and tau (Sheng et al., 2000). There is a considerable experimental literature relating to neuroinflammation and neuronal death. A recent study of mixed cultures of activated glia and neurons suggested that inflammatory neurodegeneration may be mediated by glial nitric oxide (NO) (Bal-Price & Brown, 2001). They proposed that NO produced by activated microglia or astrocytes inhibits the mitochondrial function of surrounding neurons, causing glutamate release from neurons (and possibly from astrocytes). Activation of NMDA receptors by glutamate triggers massive influx of Ca2+ into neurons, leading to cell death. Other potential mechanisms of NO mediated neuronal death include mitochondrial damage (Heales et al., 1999) and poly(ADP-ribose) synthetase activation (Zhang et al., 1994).

Evidence for genetic influences on outcome after head injury The response to brain injury and AD have in common not only a cellular and protein response but there are striking parallels in the genetic influences. There is a polymorphism of the apolipoprotein E gene (APOE, gene; apoE, protein) of which there are 3 common alleles (ε2, ε3 and ε4). Possession of APOE ε4 allele is the major genetic susceptibility factor for sporadic AD (Saunders et al., 1993). In addition, the APOE polymorphism influences neuropathological findings in patients who die from head injuries (Nicoll et al., 1995). This study examined the brains of 90 individuals who died within 2 weeks of a head injury. A deposits were found in 23 cases and the frequency of the APOE ε4 allele within this group was significantly greater than that seen in either control populations without neurological disease or in AD. In addition all individuals homozygous for the APOE ε4 allele had A deposition. Furthermore, the density of these plaques is related to APOE genotype, with greater numbers of plaques being associated with the APOE ε4 allele in an allele dose dependant manner (i.e. homozygotes having greater numbers of plaques than heterozygotes) (Horsburgh et al., 2000). The initial interpretation of these findings by Nicoll et al. was that in genetically susceptible individuals (i.e. those with an APOE ε4 allele) traumatic brain injury appears to act as a trigger for A deposition. However, there are alternative explanations

for these observations (Roses & Saunders, 1995); A deposits may pre-date the injury, and patients with ε4, who are more likely to have age-related deposits, may have a higher mortality from traumatic brain injury and therefore be selected for an autopsy-based study. Until it is possible to image A plaques during life it may not be possible to resolve this uncertainty. Subsequent clinical studies have indeed shown that head-injured patients (and patients with spontaneous intracerebral haemorrhage) who possess APOE ε4 have poorer outcome than noncarriers of APOE ε4 (Alberts et al., 1995; Teasdale et al., 1997; McCarron et al., 1998). Jordan et al., 1997 studied 30 professional boxers and assessed their cognitive status in relation to APOE genotype and number of professional bouts. They concluded that possession of an APOE ε4 allele is associated with increased severity of chronic neurological deficits in high-exposure boxers. A recent neuropathological study compared A deposits in long term survivors of traumatic brain injury (survival time up to 20 years) with age-matched and APOE genotype-matched controls (MacFarlane et al., 1999). They found A deposits were more common in ε4 patients in both the long term survivors and the control groups, but were not more common among long term survivors than controls. A further genetic polymorphism has recently been suggested to confer susceptibility to AD and this further implicates neuroinflammatory processes. Interleukin 1 (IL1) exists in two distinct forms (IL-1 and IL-1 with IL-1A and IL-1B genes respectively). Polymorphisms have been identified in each of these genes (both have an allele 1 and an allele 2) and an association has been demonstrated between the IL-1A 2,2 genotype and AD (Nicoll et al., 2000; Grimaldi et al., 2000). Nicoll et al., studied 232 pathologically confirmed cases of AD and found the IL-1A 2,2 genotype in almost 13% of cases as compared to 6.6% of agematched and APOE-matched controls. In addition they found that homozygosity for allele 2 of both IL-1A and IL1B conferred an even greater risk, although homozygosity for allele 2 of IL-1B alone was not significant. In AD there is evidence that patients with APOE ε4 have increased microglial activity compared to patients without APOE ε4 (Egensperger et al., 1998). Although there is currently a lack of definitive information relating to APOE genotype, IL-1 genotype and microglial activation in traumatic brain injury these observations raise the possibiltiy that microglial activation (‘neuroinflammation’) may be under genetic influence. Specifically, they raise the question that individuals with the relevant alleles (APOE ε4 or IL-1A allele 2) who sustain a head injury may have a relatively overexuberant microglial response which is associated both with a poorer outcome from injury and greater

467

468


susceptibility to later AD. If this were the case it is possible to speculate that simple anti-inflammatory medication may have a role in the long-term management of the headinjured patient in much the same way that non-steroidal anti-inflammatory drugs (NSAIDs) may be useful in protecting against AD (McGeer et al., 1990, 1996).

REFERENCES Adams, C. W. M. & Bruton, C. J. (1989). The cerebral vasculature in dementia pugilistica. J Neurol Neurosurg Psychiatry, 52: 600–4. Adams, J. H., Doyle, D., Graham, D. I. et al. (1985). The contusion index: A reappraisal in human and experimental non-missile head injury. Neuropath Appl Neurobiol, 11: 299–308. Adams, J. H., Graham, D. I. & Jennett, B. (2000). The neuropathology of the vegetative state after an acute brain insult. Brain, 123: 1327–38. (2001). The structural basis of moderate disability after traumatic brain damage. J Neurol Neurosurg Psychiatry, 71: 521–4. Adle-Biassette, H., Duyckaerts, C., Wasowicz, M. et al. (1996). Beta AP deposition and head trauma. Neurobiol Aging, 17: 415–19. Alberts, M. J., Graffagnino, C., McClenny, C. et al. (1995). ApoE genotype and survival from intracerebral haemorrhage. Lancet, 346: 575. Amaducci, L. A., Fratiglioni, L., Rocca, W. A. et al. (1986). Risk factors for clinically diagnosed Alzheimer’s disease: a case-control study of an Italian population. Neurology, 36: 922–31. Aronson, M. K., Ooi, W. L., Morgenstern, H. et al. (1990). Women, myocardial infarction, and dementia in the very old. Neurology, 40: 1102–6. Bal-Price, A. & Brown, G. C. (2001). Inflammatory neurodegeneration mediated by nitric oxide from activated glia-inhibiting neuronal respiration, causing glutamate release and excitotoxicity. J Neurosci, 21: 6480–91. Barger, S. W. & Harmon, A. D. (1997). Microglial activation by Alzheimer amyloid precursor protein and modulation by apolipoprotein E. Nature, 28: 878–81. Biasca, N., Simmen, H. P. & Trentz, O. (1993). Head injuries in ice hockey exemplified by the National Hockey League ‘Hockey Canada’ and European teams. Unfallchirurg., 96: 259–64. Brandenburg, W. & Hallervorden, J. (1954). Dementia pugilistica mit anatomischem Befund. Virchow’s Arch Path Anat Physiol Klin Med, 325: 680–709. Breteler, M. M., de Groot, R. R., van Romunde, L. K. & Hofman, A. (1995). Risk of dementia in patients with Parkinson’s disease, epilepsy and severe head trauma: a register-based follow-up study. Am J Epidemiol, 142: 1300–5. Broe, G. A., Henderson, A. S., Creasey, H. et al. (1990). A case-control study of Alzheimer’s disease in Australia. Neurology, 40: 1698– 707. Brooks, N., Kupshik, G., Wilson, L., Galbraith, S. & Ward, R. (1987). A neuropsychological study of active amateur boxers. J Neurol Neurosurg Psychiatry, 50: 997–1000.

Bruton, C. J. (1997). Head injury and dementia. In Esiri MM, Morris JH eds. The Neuropathology of Dementia. 1st edn., pp. 344–56. Cambridge: Cambridge University Press. Buxbaum, J. D., Oishi, M., Chen, H. I. et al. (1992). Cholinergic agonists and interleukin 1 regulate processing and secretion of the Alzheimer A4 amyloid protein precursor. Proc Natl Acad Sci USA, 89: 10075–8. Capruso, D. X. & Levin, H. S. (2000). Neurobehavioural sequelae of head injury. In P. R. Cooper, J. G. Golfinos, eds. Head Injury, 4th edn. pp. 525–53. New York: McGraw-Hill. Cardenas, D. D., McLean, A. Jr., Farrell-Roberts, L., Baker, L., Brooke, M. & Haselkorn, J. (1994). Oral physostigmine and impaired memory in adults with brain injury. Brain Inj, 8: 579–87. Casson, I. R., Siegel, O., Sham, R., Campbell, E. A., Tarlan, M. & DiDomenico, A. (1984). Brain damage in modern boxers. J Am Med Ass, 251: 2663–7. Chandra, V., Philipose, V., Bell, P. A., Lazaroff, A. & Schoenberg, B. S. (1987). Case-control study of late onset ‘probable Alzheimer’s disease’. Neurology, 37: 1295–300. Clark, R. S. B., Kochanek, P. M., Chen, M. et al. (1999). Increases in Bcl-2 and cleavage of caspase-1 and caspase-3 in human brain after head injury. FASEB J, 13: 813–21. Constantinidis, J. & Tissot, R. (1967). Lesions neurofibrillaires D’Alzheimer generalisees sans plaques seniles. Arch Suis Neurol Nurochir Psychiatry, 100: 117–30. Corkin, S., Rosen, T. J., Sullivan, E. V. & Clegg, R. A. (1989). Penetrating head injury in young adulthood exacerbates cognitive decline in later years. J Neurosci, 9: 3876–83. Corsellis, J. A. N., Bruton, C. J. & Freeman-Browne, D. (1973). The aftermath of boxing. Psychol Med, 3: 270–303. Das, S. & Potter, H. (1995). Expression of the Alzheimer amyloidpromoting factors 1-antichymotrypsin and apolipoprotein E is induced in astrocytes by IL-1. Neuron, 14: 447–56. Dewar, D. & Graham, D. I. (1996). Depletion of choline acetyltransferase activity but preservation of M1and M2 muscarinic receptor binding sites in temporal cortex following head injury: a preliminary human postmortem study. J Neurotrauma, 13: 181–7. Dixon, C. E., Bao, J., Johnson, K. M. et al. (1995). Basal and scopolamine-evoked release of hippocampal acetylcholine following traumatic brain injury in rats. Neurosci Lett, 198: 111–14. Dixon, C. E., Ma, X. & Marion, D. W. (1997). Reduced evoked release of acetylcholine in the rodent neocortex following traumatic brain injury. Brain Res, 749: 127–30. Egensperger, R., Kosel, S., von Eitzen, U. & Graeber, M. B. (1998). Microglial activation in Alzheimer disease: Association with APOE genotype. Brain Pathol, 8: 439–47. Engel, S., Schluesener, H., Mittelbronn, M. et al. (2000). Dynamics of microglial activation after human traumatic brain injury are revealed by delayed expression of macrophage-related proteins MRP8 and MRP14. Acta Neuropathologica, 100: 313–22. Everitt, B. J. & Robbins, T. W. (1997). Central cholinergic systems and cognition. Annu Rev Psychol, 48: 649–84.


Ferguson, F. R. & Mawdsley, C. (1965). Chronic encephalopathy in boxers. 8th International Congress of Neurology, Vienna. Wiener Medizinische Akademie, Vienna, Vol. 1: 81–4. Ferini-Strambi, L., Smirne, S., Garancicni, P., Pinto, P. & Franceschi, M. (1990). Clinical and epidemiological aspects of Alzheimer’s disease with presenile onset: a case-control study. Neuroepidemiology, 9: 39–49. Foster, J. B., Leiguarda, R. & Tilley, P. J. (1976). Brain damage in National Hunt jockeys. Lancet, 1: 981–3. Frankowski, R. F., Annegers, J. F. & Whitman, S. (1985). Epidemiological and descriptive studies. Part 1. The descriptive epidemiology of head trauma in the United States. In D. P. Becker, J. T. Povlishock, (eds.) Central Nervous System Trauma Status Report – 1985. Bethesda, MD: National Institute of Neurological and Communicative Disorders and Stroke, pp. 33–43. Fratiglioni, L., Ahlbom, A., Viitanen, M. & Winblad, B. (1993). Risk factors for late-onset Alzheimer’s disease: a population-based, case-control study. Ann Neurol, 33: 258–66. French, L. R., Schuman, L. M., Mortimer, J. A., Hutton, J. T., Boatman, R. A. & Christians, B. (1985). A case-control study of dementia of the Alzheimer type. Am J Epidemiol, 121: 414–21. Ganter, S., Northoff, H., Mannel, D. & Gebicke-Harter, P. J. (1992). Growth control of cultured microglia. J Neurosci Res, 33: 218–30. Geddes, J. F., Vowles, G. H., Nicoll, J. A. R. & Revesz, T. (1999). Neuronal cytoskeletal changes are an early consequence of repetitive head injury. Acta Neuropathol, 98: 171–8. Gennarelli, T. A. (1993). Cerebral concussion and diffuse brain injuries. In P. R. Cooper, (ed) Head Injury, 3rd edn. pp. 137–58. Baltimore, USA: Williams and Wilkins. Gentleman, S. M., Greenberg, B. D., Savage, M. J. et al. (1997). A beta 42 is the predominant form of amyloid beta-protein in the brains of short-term survivors of head injury. Neuroreport, 8: 1519–22. Geula, C. & Mesulam, M. M. (1994). Cholinergic systems and related neuropathological predilection patterns in Alzheimer’s disease. In R. D. Terry, R. Katzman, Bick, K. L. eds. Alzheimer’s Disease, pp. 263–94. New York: Raven Press. Goldgaber, D., Harris, H. W., Hla, T. et al. (1989). Interleukin-1 regulates synthesis of amyloid beta-protein precursor mRNA in human endothelial cells. Proc Natl Acad Sci USA, 86: 7606–10. Graham, D. I., Gentleman, S. M., Lynch, A. & Roberts, G. W. (1995a). Distribution of beta-amyloid protein in the brain following severe head injury. Neuropathol Appl Neurobiol, 21: 27–34. Graham, D. I., Adams, J. H., Nicoll, J. A. R., Maxwell, W. L. & Gennarelli, T. A., (1995b). The nature, distribution and cause of traumatic brain injury. Brain Pathology, 5: 397–406. Graham, D. I., McIntosh, T. K., Maxwell, W. L. & Nicoll, J. A. R. (2000). Recent advances in neurotrauma. J Neuropath Exp Neurol, 59: 641–51. Grahmann, H. & Ule, G. (1957). Beitrag zur Kenntnis der chronischen cerebralen Krankheitsbilder bei Boxern. Psychiat Neurol, 134: 261–83.

Graves, A. B., White, E., Koepsell, T. D. et al. (1990). The association between head trauma and Alzheimer’s disease. Am J Epidemiol, 131: 491–501. Griffin, W. S. T., Stanley, L. C., Ling, C. et al. (1989). Brain interleukin 1 and S-100 immunoreactivity are elevated in Down syndrome and Alzheimer’s disease. Proc Natl Acad Sci USA, 86: 7611–15. Griffin, W. S. T., Sheng, J. G., Gentleman, S. M., Graham, D. I., Mrak, R. E. & Roberts, G. W. (1994). Microglial interleukin-1 expression in human head injury: correlations with neuronal and neuritic -amyloid precursor protein expression. Neurosci Lett, 176: 133–6. Griffin, W. S. T., Sheng, J. G., Royston, M. C. et al. (1998). Glial– neuronal interactions in Alzheimer’s disease: the potential role of a ‘cytokine cycle’ in disease progression. Brain Pathol, 8: 65– 72. Grimaldi, L. M., Casadei, V. M., Ferri, C. et al. (2000). Association of early-onset Alzheimer’s disease with an interleukin-1alpha gene polymorphism. Ann Neurol, 47: 361–5. Guo, Z., Cupples, L. A., Kurz, A. et al. (2000). Head injury and the risk of Alzheimer’s disease in the MIRAGE study. Neurology, 54: 1316–23. Heales, S. J., Bolanos, J. P., Stewart, V. C., Brookes, P. S., Land, J. M. & Clark, J. B. (1999). Nitric oxide, mitochondria and neurological disease. Biochim Biophys Acta, 1410: 215–28. Heilbronner, R. L., Henry, G. K. & Carson-Brewer, M. (1991). Neuropsychologic test performance in amateur boxers. Am J Sports Med, 19: 376–80. Horsburgh, K., Cole, G. M., Yang, F. et al. (2000). Beta-amyloid (Abeta)42(43), abeta42, abeta40 and apoE immunostaining of plaques in fatal head injury. Neuropathol Appl Neurobiol, 26: 124–32. Jennett, B. & Bond, M. (1975). Assessment of outcome after severe brain damage. Lancet, 1: 480–7. Jennett, B., Snoek, J. & Bond, M. R. (1981). Disability after severe head injury: Observations on the use of the Glasgow Outcome Scale. J Neurol Neurosurg Psychiatry, 44: 285–93. Jennett, B., Adams, J. H., Murray, L. S. & Graham, D. I. (2001). Neuropathology in vegetative and severely disabled patients after head injury. Neurology, 56: 486–90. Jordan, B. D., Relkin, N. R., Ravdin, L. D., Jacobs, A. R., Bennett, A. & Gandy, S. (1997). Apolipoprotein E epsilon4 associated with chronic traumatic brain injury in boxing. J Am Med Assoc, 278: 136–40. Katzman, R., Aronson, M., Fuld, P. et al. (1989). Development of dementing illnesses in an 80 year old volunteer cohort. Ann Neurol, 25: 317–24. Kirkendall, D. T., Jordan, S. E. & Garrett, W. E. (2001). Heading and head injuries in soccer. Sports Med, 31: 369–86. Kotapka, M. J., Graham, D. I., Adams, J. H. & Gennarelli, T. A. (1992). Hippocampal pathology in fatal non-missile human head injury. Acta Neuropathol, 83: 530–4. Launer, L. J., Anderson, K., Dewey, M. E. et al. (1999). Rates and risk factors for dementia and Alzheimer’s disease: results from EURODEM incidence research group and work groups. European studies of dementia. Neurology, 52: 78–84.

469

470


Li, G., Shen, Y. C., Li, Y. T., Chen, C. H., Zhau, Y. W. & Silverman, J. M. (1992). A case-control study of Alzheimer’s disease in China. Neurology, 42: 1481–8. McCarron, M. O., Muir, K. W., Weir, C. J. et al. (1998). The apolipoprotein E ε4 allele and outcome in cerebral vascular disease. Stroke, 29: 1882–7. MacFarlane, D. P., Nicoll, J. A. R., Smith, C. & Graham, D. I. (1999). APOE epsilon 4 allele and amyloid beta-protein deposition in long term survivors of head injury. Neuroreport, 10: 3945–8. McGeer, P. L., McGeer, E., Rogers, J. & Sibley, J. (1990). Antiinflammatory drugs and Alzheimer disease. Lancet, 335: 1037. McGeer, P. L., Schulzer, M. & McGeer, E. G. (1996). Arthritis and anti-inflammatory agents as possible protective factors for Alzheimer’s disease: a review of 17 epidemiologic studies. Neurology, 47: 425–32. McIntosh, A. S., McCrory, P. & Comerford, J. (2000). The dynamics of concussive head impacts in rugby and Australian rules football. Med Sci Sports Exerc, 32: 1980–4. McLatchie, G., Brooks, N. & Galbraith, S. et al. (1987). Clinical neurological examination, neuropsychology, electroencephalography and computed tomographic head scanning in active amateur boxers. J Neurol Neurosurg Psychiatry, 50: 96–9. Maroon, J. C., Lovell, M. R., Norwig, J., Podell, K., Powell, J. W. & Hartl, R. (2000). Cerebral concussion in athletes: evaluation and neuropsychological testing. Neurosurgery, 47: 659–69. Martland, H. S. (1928). Punch drunk. J Am Med Ass, 91: 1103–7. Matser, E. J., Kessels, A. G., Lezak, M. D., Jordan, B. D. & Troost, J. (1999). Neuropsychological impairment in amateur soccer players. JAMA, 282: 971–3. Mawdsley, C. & Ferguson, F. R. (1963). Neurological disease in boxers. Lancet, 2: 795–801. Mayeux, R., Ottman, R., Tang, M. X. et al. (1993). Genetic susceptability and head injury as risk factors for Alzheimer’s disease among community-dwelling elderly persons and their first degree relatives. Ann Neurol, 33: 494–501. Mehta, K. M., Ott, A., Kalmijn, S. et al. (1999). Head trauma and the risk of dementia and Alzheimer’s disease: The Rotterdam study. Neurology, 53: 1959–62. Mendez, M. F., Underwood, K. L., Zander, B. A., Mastri, A. R., Sung, J. H. & Frey, W. H. (1992). Risk factors in Alzheimer’s disease: a clinicopathological study. Neurology, 42: 770–5. Millspaugh, J. A. (1937). Dementia pugilistica. US Nav Med Bull, 35: 297–303. Mortimer, J. A., van Duijn, C. M., Chandra, V. et al. (1991). Head trauma as a risk factor for Alzheimer’s disease: a collaborative re-analysis of case-control studies. EURODEM risk factors research group. Int J Epidemiol 20 Suppl 2, S28–35. Acta Neurol Scand, 83: 9–13. Mrak, R. E., Sheng, J. G. & Griffin, W. S. T. (1996). Correlation of astrocytic S100 expression with dystrophic neurites in amyloid plaques of Alzheimer’s disease. J Neuropathol Exp Neurol, 55: 273–9. Murdoch, I., Perry, E. K., Court, J. A., Graham, D. I. & Dewar, D. (1998). Cortical cholinergic dysfunction after human head injury. J Neurotrauma, 15: 295–305.

Murdoch, I., Nicoll, J. A. R., Graham, D. I. & Dewar, D. (2001). Nucleus basalis of Meynert pathology in the human brain after fatal head injury. J Neurotrauma, 19: 279–84. Murelius, O. & Haglund, Y. (1991). Does Swedish amateur boxing lead to chronic brain damage? A retrospective neuropsychological study. Acta Neurol Scand, 83: 9–13. Murray, G. D., Teasdale, G. M., Braakman, R. et al. (1999). The European Brain Injury Consortium Survey of Head Injuries. Acta Neurochirg, 141: 223–36. Nemetz, P. N., Leibson, C., Naessens, J. M. et al. (1999). Traumatic brain injury and time to onset of Alzheimer’s disease: a population-based study. Am J Epidemiol, 149: 32–40. Nicoll, J. A., Roberts, G. W. & Graham, D. I. (1995). Apolipoprotein E epsilon 4 allele is associated with deposition of amyloid betaprotein following head injury. Nat Med, 1: 135–7. Nicoll, J. A., Mrak, R. E., Graham, D. I. et al. (2000). Associations of Interleukin-1 gene polymorphisms with Alzheimer‘s disease. Ann Neurol, 47: 365–8. O’Meara, E. S., Kukull, W. A., Sheppard, L. et al. (1997). Head injury and the risk of Alzheimer’s disease by apolipoprotein E genotype. Am J Epidemiol, 146: 373–84. Pierce, J. E., Smith, D. H., Trojanowski, J. Q. & McIntosh, T. K. (1998). Enduring cognitive, neurobehavioral and histopathological changes persist for up to one year following severe experimental brain injury in rats. Neuroscience, 87: 359–69. Plassman, B. L., Havlik, R. J., Steffens, D. C. et al. (2000). Documented head injury in early adulthood and risk of Alzheimer’s disease and other dementias. Neurology, 55: 1158–66. Raghupathi, R., Graham, D. I. & McIntosh, T. K. (2000). Apoptosis after traumatic brain injury. J Neurotrauma, 17: 927–38. Rasmussen, D. X., Brandt, J., Martin, D. B. & Folstein, M. F. (1995). Head injury as a risk factor in Alzheimer’s disease. Brain Inj, 9: 213–19. Rink, A. D., Fung, K. M., Trojanowski, J. Q., Lee, V. M.-Y., Neugebauer, E. & McIntosh, T. K. (1995). Evidence of apoptotic cell death after experimental traumatic brain injury in the rat. Am J Path, 147: 1575–83. Roberts, A. H. (1969). Brain Damage in Boxers. A Study of Prevalence of Traumatic Encephalopathy Among Ex-professional Boxers. London: Pitman. Roberts, G. W., Allsop, D. & Bruton, C. (1990). The occult aftermath of boxing. J. Neurol. Neurosurg. Psychiatry, 53: 373–8. Roberts, G. W., Gentleman, S. M., Lynch, A. & Graham, D. I. (1991). -A4 amyloid protein deposition in brain after head trauma. Lancet, 338: 1422–3. Roberts, G. W., Gentleman, S. M., Lynch, A., Murray, L., Landon, M. & Graham, D. I. (1994). -amyloid protein deposition in the brain following severe head injury: implications for the pathogenesis of Alzheimer’s disease. J Neurol Neurosurg Psychiat, 57: 419–25. Roses, A. D. & Saunders, A. (1995). Head injury, amyloid beta and Alzheimer’s disease. Nat Med, 1: 603–4. Ross, D. T., Graham, D. I. & Adams, J. H. (1993). Selective loss of neurons from the thalamic reticular nucleus following severe human head injury. J Neurotrauma, 10: 151–65.


Salib, E. & Hillier, V. (1997). Head injury and the risk of Alzheimer’s disease: a case-control study. Int J Geriatr Psychiatry, 12: 363–8. Saunders, A. M., Strittmatter, W. J., Schmechel, D. et al. (1993). Association of apolipoprotein E allele epsilon 4 with lateonset familial and sporadic Alzheimer’s disease. Neurology, 43: 1467–72. Schmidt, M. L., Zhukareva, V., Newell, K. L., Lee, V. M.-Y. & Trojanowski, J. Q. (2001). Tau isoform profile and phosphorylation state in dementia pugilistica recapitulate Alzheimer’s disease. Acta Neuropathol, 101: 518–24. Schmidt, R. H. & Grady, M. S. (1995). Loss of forebrain cholinergic neurons following fluid-percussion injury: implications for cognitive impairment in closed head injury. J Neurosurg, 83: 496–502. Schofield, P. W., Tang, M., Marder, K. et al. (1997). Alzheimer’s disease after remote head injury: an incidence study. J Neurol Neurosurg Psychiat, 62: 119–24. Shaw, K., MacKinnon, M.-A., Raghupathi, R., Saatman, K. E., McIntosh, T. K. & Graham, D. I. (2001). TUNEL-positive staining in white and grey matter after fatal head injury in man. Clin Neuropathol, 20: 106–12. Sheng, J. G., Mrak, R. E. & Griffin, W. S. T. (1997). Glial–neuronal interactions in Alzheimer’s disease: progressive association of IL-1+ microglia and S100+ astrocytes with neurofibrillary tangle stages. J Neuropathol Exp Neurol, 56: 285–90. Sheng, J. G., Zhu, S. G., Griffin, W. S. T. & Mrak, R. E. (2000). Interleukin-1 promotes expression and phosphorylation of tau protein in vivo. Exp Neurol, 163: 388–91. Smith, D. H., Chen, X. H., Pierce, J. E. et al. (1997). Progressive atrophy and neuron death for one year following brain trauma in the rat. J Neurotrauma, 14: 715–27. Smith, D. H., Chen, X. H., Nonaka, M. et al. (1999). Accumulation of amyloid beta and tau and the formation of neurofilament inclusions following diffuse brain injury in the pig. J Neuropath Exp Neurol, 58: 982–92. Smith, F. M., Raghupathi, R., MacKinnon, M.-A. et al. (2000).

TUNEL-positive staining of surface contusions after fatal head injury in man. Acta Neuropathol, 100: 537–45. Teasdale, G. & Jennett, B. (1974). Assessment of coma and impaired consciousness: a practical scale. Lancet, 2: 81–4. (1976). Assessment and prognosis of coma after head injury. Acta Neurochir, 34: 45–55. Teasdale, G. & Jennett, B. (1976). Assessment and prognosis of coma after head injury. Acta Neurochir. 34: 45–55. Teasdale, G. M., Nicoll, J. A., Murray, G. & Fiddes, M. (1997). Association of apolipoprotein E polymorphism with outcome after head injury. Lancet, 350: 1069–71. Teasdale, G. M., Pettigrew, L. E., Wilson, J. T., Murray, G. & Jennett, B. (1998). Analyzing outcome of treatment of severe head injury: a review and update on advancing the use of the Glasgow Outcome Scale. J Neurotrauma, 15: 587–97. Thornhill, S., Teasdale, G. M., Murray, G. D., McEwen, J., Roy, C. W. & Penny, K. I. (2000). Disability in young people and adults one year after head injury: prospective cohort study. Br Med J, 320: 1631–5. van Duijn, C. M., Tanja, T. A., Haaxma, R. et al. (1992). Head trauma and the risk of Alzheimer’s disease. Am J Epidemiol, 135: 775– 82. Williams, D. B., Annegers, J. F., Kokmen, E., O’Brien, P. C. & Kurland, L. T. (1991). Brain injury and neurologic sequelae: a cohort study of dementia, parkinsonism, and amyotrophic lateral sclerosis. Neurology, 41: 1554–7. Williams, S., Raghupathi, R., MacKinnon, M.-A., McIntosh, T. K., Saatman, K. E. & Graham, D. I. (2001). In-situ DNA fragmentation occurs in white matter up to 12 months after head injury in man. Acta Neuropathol, 102: 581–90. Zemlan, F. P., Rosenberg, W. S., Luebbe, P. A. et al. (1999). Quantification of axonal damage in traumatic brain injury: affinity purification and characterization of cerebrospinal fluid tau proteins. J Neurochem, 73: 437–8. Zhang, J., Dawson, V. L., Dawson, T. M. & Snyder, S. H. (1994). Nitric oxide activation of poly(ADP-ribose) synthetase in neurotoxicity. Science, 263: 687–9.

471

21 Infectious (and inflammatory) diseases causing dementia Harry V. Vinters Brain Research Institute and Neuropsychiatric Institute, UCLA Medical Center and David Geffen School of Medicine at UCLA, Los Angeles, CA, USA

Introduction Dementia secondary to infectious disease, setting aside cases of transmissible spongiform encephalopathy (see Chapter 17) and the cognitive–motor abnormalities frequently associated with acquired immunodeficiency syndrome (AIDS), is distinctly unusual. Infectious disease as the primary cause of a dementing illness is usually a diagnosis of exclusion; lumbar puncture is often carried out in the clinical work-up of a demented patient – especially if the decline in mental status is rapidly progressive – in order to safely rule out a treatable infection (e.g. cryptococcal meningitis), unusual though this is, as the cause for a given individual’s cognitive and neuropsychiatric decline. In a large experience of biopsies (approximately 5– 10 per year) and autopsies (40–50 annually) carried out to establish the cause for dementia, we have rarely encountered – usually as a surprise finding – evidence of a central nervous system (CNS) inflammatory or infectious disease as the primary etiology. Such cases are, however, intriguing insofar as modern diagnostic tools can often establish an aetiological agent for the infection even when the only histopathological evidence for its existence (at autopsy) is widespread chronic inflammation of the brain and its overlying leptomeninges. Throughout the 1980s and 1990s, the range of known viruses and bacteria that target the CNS has widened substantially (Kennedy, 1990; Vinters et al., 1998). The availability of highly sensitive molecular diagnostic tests (especially the polymerase chain reaction, PCR), by which faint traces of molecular ‘footprints’ of bacterial or viral pathogens may be detected within neural tissue, allows the safe prediction that many new associations between micro-organisms and CNS/PNS disease will be

made in the coming years (Darnell, 1993; Tompkins, 1992). In addition, PCR technology – when used appropriately, cautiously and with meticulous attention to controls – has greatly expanded our understanding of the spectrum of viral diseases of the CNS, especially those caused by herpesviruses (Kleinschmidt-DeMasters et al., 2001). Most monographs and review articles on dementia include ‘infective’ causes of cognitive decline in the differential diagnosis. The most commonly cited conditions in such a list include the diseases summarized in Table 21.1 (Mann et al., 1994; Fleming et al., 1995). This chapter will discuss both ‘conventional’ and ‘slow’ CNS infections (especially associated with human immunodeficiency virus, type I, HIV-1) that can present with dementia or have cognitive-motor impairment as a significant component. Purely inflammatory CNS diseases with a variable neurocognitive element but not caused by a known infectious agent (multiple sclerosis, limbic encephalitis, systemic lupus erythematosus, sarcoidosis and Sjogren’s syndrome) will not be considered further. (The contribution of inflammatory mechanisms to AD pathogenesis itself is described elsewhere in this text.) While all of these disorders receive frequent mention, large necropsy series on demented patients whose brain tissue has been accessioned to tissue banks report them only infrequently, even after extremely diligent neuropathologic study of the CNS (Mendez et al., 1992). Nevertheless, evidence of low grade (lymphocytic) inflammation may be found on brain biopsy of patients suspected clinically as having AD or an AD-like encephalopathy, e.g. as reported in two patients with serologically documented Sjogren’s syndrome (Caselli et al., 1993). Another rare entity well documented in careful autopsy studies and not definitively associated with a known


472

Infectious (and inflammatory) diseases

Table 21.1. Infectious and inflammatory conditions that may have dementia/cognitive decline/neurobehavioural abnormality as a significant component Infectious Bacterial meningitis (chronic tuberculous, or partially treated suppurative) Spirochete infection (neurosyphilis, Lyme disease) Fungal (especially cryptococcal) meningitis CNS Whipple’s disease Spongiform encephalopathy (CJD) Viral encephalitis – ‘conventional’ (e.g. Herpes simplex, arbovirus encephalitis) – retroviral (e.g. HIV-1) Viral infection – ‘slow/atypical/unconventional’ (subacute sclerosing panencephalitis, progressive multifocal leukoencephalopathy) Inflammatory Limbic (including paraneoplastic) encephalitis Multiple sclerosis and other forms of demyelination (including viral-induced, e.g. secondary to measles virus, HTLV-1) Systemic lupus erythematosus Sarcoidosis Sjogren’s syndrome Pathogen-free granulomatous disease of the CNS

causal micro-organism, pathogen-free granulomatous disease of the CNS, may present with behavioral disturbances and dementia in addition to focal neurological signs and seizures (Thomas et al., 1998) (Fig. 21.1).

Dementia associated with HIV-1 infection AIDS-related dementia secondary to opportunistic infections (OIs), lymphoma, vascular disease It is now almost universally accepted that AIDS results from systemic infection with the lentivirus human immunodeficiency virus, type 1 (HIV-1). This results in infection of CD4+ T-lymphocytes, causing their death, with subsequent deficiency in T-cell mediated immunity, though many patients can remain HIV-seropositive, without significant immunosuppression, for years. An HIV-infected individual is considered to have AIDS when he/she develops one or more ‘AIDS-defining illnesses’ (ADIs), including opportunistic fungal, viral and/or parasitic infections, nonHodgkin lymphoma, Kaposi sarcoma, and HIV dementia, the latter to be considered in detail below (Vinters & Anders, 1990). Although dementia related to direct HIV-1 infection of the brain has ‘captured the imagination’ of basic neurobiologists, retrovirologists and clinical neuroscientists, it is

Fig. 21.1. Pathogen-free granulomatous disease of the CNS. Note dense granulomatous inflammatory infiltrate adjacent to cerebellum (a). (b) Shows magnified view of a granuloma, containing epithelioid macrophages (at its centre) and lymphocytes (at the periphery). Arrow indicates a multinucleated giant cell. [For details, see G. Thomas et al., 1998; illustrations courtesy of Prof. Michael A. Farrell, Dublin, Ireland]. In this and subsequent figures, photomicrographs are from sections that have been stained with hematoxylin and eosin, unless indicated otherwise.

only one of the protean manifestations of AIDS that may be associated with mental status decline in an HIV-infected individual. A major challenge to a clinician faced with a cognitively impaired AIDS patient is to establish, using clinical examination, neuroimaging and CSF studies, the likely cause(s), among many, of the symptoms. Table 21.2 summarizes the major neurological/neuropathological

473

474

H. V. Vinters

Table 21.2. Neurological/neuropathological manifestations of HIV infection/AIDS 1

Direct effect(s) of HIV-1 on the CNS HIV-associated dementia (HAD, subacute encephalitis of AIDS) Chronic, low grade meningitis Vacuolar myelopathy of AIDS (VMA) 2 Opportunistic infections v i r a l: Cytomegalovirus Papovavirus (JC virus)/ Progressive multifocal leukoencephalopathy Herpes simplex Varicella zoster Others (e.g. adenovirus) f u n g a l: Cryptococcus neoformans Candida albicans Aspergillus spp. Coccidioides immitis Others b a c t e r i a l: Mycobacterium avium-intracellulare (MAI) complex Mycobacterium tuberculosis p a r a s i t i c: Various 3 Neoplasms Primary CNS lymphoma Disseminated lymphoma with CNS (usually meningeal) involvement Kaposi sarcoma (rare?) 4 Ischaemic/haemorrhagic lesions Ischaemic infarcts (e.g. due to HIV vasculopathy, non-bacterial thrombotic endocarditis) Haemorrhages (various causes) Anoxic-ischaemic encephalopathy 5 Miscellaneous/metabolic abnormalities of uncertain significance Alzheimer II astrocytosis

complications of AIDS that must be considered in the differential diagnosis of such a patient. Of the entities listed, almost all can cause dementia or have dementia as one component of a complex neurologic syndrome that may involve both the CNS and PNS. More than one disease process may be present in a given patient. Though a detailed presentation of all neuropathological manifestations of HIV-1 infection associated with immunosuppression (rather than direct HIV infection of brain) is beyond the scope of this chapter, some of the more common relevant clinicopathologic phenomenology merits description. The interested reader is directed to several monographs and review articles that present details and dramatic images of these individual entities (Gray, 1993; Kure et al., 1991; Anders et al., 1986; Artigas et al., 1993; Vinters & Anders, 1990).

In view of aggressive new therapies for HIV infection and AIDS, including nucleoside analogue reverse transcriptase inhibitors and protease inhibitors (highly active anti-retroviral therapy, or HAART) the natural history of HIV infection and AIDS-related illnesses is changing before our eyes (Temesgen & Wright, 1997). An investigation from New York City focusing on autopsy results (in almost 400 patients) obtained over two decades at a large HIV diagnostic and treatment centre found that women died of AIDS at significantly younger ages than men, even after adjustment for risk factors, ethnicity and therapeutic era (Morgello et al., 2002). Caucasian patients had a longer survival than those of other ethnicities. CMV infection and systemic lymphoma appeared to be more common in whites and Hispanic individuals than in African-Americans. Of interest, the prevalence of many disease entities changed over time: specifically, patients autopsied during the years 1996– 2000 had lower prevalences of many viral, parasitic and fungal diseases (especially Pneumocystis carinii pneumonia, PCP), infection with Mycobacterium avium intracellulare (MAI), and Kaposi sarcoma, but higher prevalences of hepatitis, cirrhosis, arteriosclerosis, some bacterial infections, and traumatic lesions (Morgello et al., 2002). The changing disease patterns in HIV-1 infected patients were attributed, at least in part, to potent new therapies directed against HIV infection itself and many opportunistic infections (OIs). Detailed studies from Europe of neuropathological complications of AIDS in the pre-HAART vs. postHAART eras have shown comparatively less frequent involvement of the CNS in the latter group, though this is still the second most likely organ (after lung) to be affected. Extracerebral protozoal, MAI, viral infections and Kaposi sarcoma have decreased significantly over time, while bacterial organ and CNS infections (with the exception of mycobacteriosis), lymphomas, HIV-associated CNS lesions, and non-HIV-associated changes (e.g. vascular, metabolic) have remained largely unchanged (Jellinger et al., 2000). When drug users were considered as a separate group, HIV encephalitis, PML, bacterial infections, and hepatic encephalopathy were more frequently encountered than in non-drug users, who showed an increased incidence of CMV, toxoplasmosis, or other opportunistic CNS infections. Clinical studies performed fairly early in the AIDS epidemic found that, at that time, there was no significant difference in prevalence of neurological disease between HIVinfected parenteral drug users and non-drug users (Malouf et al., 1990), though co-morbidity of potentially neurotoxic drugs (e.g. methamphetamine) and HIV-1 infection of brain is being re-evaluated in both human and animal studies. Neuropathological study of HIV infected haemophiliacs has shown that the predominant cause of death in such


patients was intracranial haemorrhage and cirrhosis of the liver, whereas OIs were less frequently encountered than in non-haemophiliacs, presumably because neuropathological abnormalities associated with HIV infection were at an early developmental stage when patients succumbed to haemorrhagic complications of their disease (Esiri et al., 1989). Given the careful surveillance of the blood supply for HIV contamination, at least in developed countries, HIV infection resulting from blood/blood product transfusion should now be a vanishingly rare event. Mass lesions in the CNS of AIDS patients Space-occupying lesions (SOLs) are frequently found in the diagnostic work-up of HIV-infected patients; diagnostic biopsies of these masses are often presented to the neuropathologist for review, sometimes intraoperatively (Vinters et al., 1996). In many centres, the practice of treating mass lesions in AIDS patients empirically as toxoplasmosis (with anti-toxoplasma therapy), and performing brain biopsy only if this fails, represents the ‘standard of care’ (Andrews & Kenefick, 1993; Martinez et al., 1993). However, stereotactic CT-guided biopsy can easily be performed with high sensitivity and specificity of diagnosis, and comparatively low morbidity and mortality (Pell et al., 1991). The most common diagnoses made in such individuals are toxoplasmosis, primary CNS lymphoma (PCNSL), and progressive multifocal leukoencephalopathy (PML), other types of infection (e.g. cryptococcosis) and infarcts/haemorrhages being encountered occasionally (Ciricillo & Rosenblum, 1990; Andrews & Kenefick, 1993). Even high resolution CT and MRI scanning have limited diagnostic value in the work-up of HIV infected patients with SOLs, though PCNSL more commonly presents as a single mass, toxoplasmosis as multiple lesions. A multimodality diagnostic approach to individuals with SOLs has been suggested, whereby CSF is examined for Epstein–Barr virus (EBV), JC virus, or T. gondii DNA by polymerase chain reaction (PCR) before brain biopsy is performed. Brain biopsy is then carried out in patients with EBV-DNA positive CSF (strongly suggestive of primary CNS lymphoma), and in those whose Toxoplasma serology is negative but in whom scans show a SOL with significant mass effect (Antinori et al., 1997). Cytomegalovirus (CMV) infection CMV is notorious for producing extremely variable neurologic and neuropathologic manifestations in both the CNS and PNS. CMV is a member of the herpesvirus group, an enveloped virus with a diameter of approx. 200 nm, containing a 100 nm icosahedral capsid and double stranded DNA. Almost never a cause of meningoencephalitis in

individuals with an intact immune system, CMV is the commonest opportunistic infection to involve the CNS of patients with AIDS (Vinters et al., 1989; Vinters & Anders, 1990; Morgello et al.,1987), being found at autopsy in 15–20% of all cases, depending upon the care with which it is sought. CMV encephalitis (CMVE) may be difficult to diagnose antemortem because of its protean clinical manifestations, and is usually found in the context of widespread CMV infection throughout the viscera (Holland et al., 1994). In a detailed clinicopathological comparison of individuals with autopsy-confirmed CMVE and those with HIV dementia (to be discussed below), CMVE had a more subacute onset (3.5 vs. 18 weeks), more malignant course (4.6 week survival vs. 28 weeks), and was usually associated with systemic and visceral manifestations of CMV infection, especially CMV adrenalitis (92%) and pneumonitis (42%). CMVE was associated with periventricular enhancement on CT scans and periventricular lesions with meningeal enhancement on MRI scans. Emphasis was placed on the observation that CMVE should be suspected in homosexual men presenting with a subacute encephalopathy who had carried the diagnosis of AIDS for more than 1 year and had a history of systemic CMV infection (Holland et al., 1994). CMV is also a significant pathogen in the PNS of HIV-infected individuals, where it may cause severe inflammatory polyradiculoneuropathy or more subtle neuropathies (Vinters et al., 1989; Cornford et al., 1992; Anders & Goebel, 1998; Robert et al., 1989). On gross inspection of a brain affected by CMVE, findings may be extremely variable. Its most dramatic manifestation is a severe ventriculitis, sometimes including obliteration of the ependymal lining of the cerebral aqueduct. Cut sections of the fixed brain may show an icing-like material on the ependymal linings of the lateral ventricles, though this is rarely present even with fairly advanced CMVE. Histological sections of the CNS show characteristic cytomegalic cells (Fig. 21.2), often in close apposition to an ependymal surface (ependymal cells may themselves be infected and focal ependymal disruption is common), associated with acute or chronic inflammation and microfoci of necrosis. In its more subtle guise, CMV in the brain may simply present as a low-grade microglial nodule encephalitis, with scattered cytomegalic cells present among the nodules. Rarely, CMV can infect capillary endothelial cells, a phenomenon that is especially common in the PNS (Cornford et al., 1992). Studies on autopsy specimens using PCR have found that when applied to analysis of CSF in a search for CMV infection, this is a highly sensitive method to detect evidence for infection; however, it is not predictive for the presence of CMVE (Achim et al., 1994a).

475

476

H. V. Vinters

Fig. 21.2. CMV encephalitis and ependymitis/ventriculitis in an AIDS patient. (a) Brain parenchymal involvement. Note cytomegalic cells containing typical Cowdry type A intranuclear inclusions (e.g. arrow), surrounded by scattered lymphocytes. (b), (c) are low power and magnified views of a region of ependymal/subependymal infection by CMV. Note focal disruption of ependyma (arrow, (b)), with edema and spongy change of underlying neuropil. Magnified view of this region (c) shows numerous cytomegalic cells, including some in or immediately adjacent to the disrupted ependyma. (d ) Vasculitis affecting a small vessel in a patient with CMV encephalomyelitis and necrotizing radiculopathy; arrows indicate cytomegalic cells. Though not shown in this Figure, CMV can rarely be seen in endothelial cells. E. In situ hybidization for CMV shows numerous labelled cells.

Other herpesviruses Other herpesviruses may also be opportunistic pathogens in the CNS of an HIV-infected individual, but are encountered only rarely and usually present with fulminant encephalopathy and encephalitis. Varicella zoster virus (VZV) causes multifocal necrosis with numerous intranuclear Cowdry type A inclusion bodies in multiple cell types in and around lesions. Zoster-induced vasculopathy occurs in some infected individuals, as do other CNS infections and primary CNS lymphoma (KleinschmidtDeMasters & Gilden, 2001a,b; Gray et al., 1992). Herpes simplex virus type 1 (HSV 1) encephalitis, in the context of AIDS, usually does not produce the characteristic necrotizing hemorrhagic lesions involving temporal and frontal lobes that result from this brain infection in nonimmunosuppressed individuals (see below), suggesting that an immune reaction may be required to cause a typical necrotizing encephalitis (Chrétien et al., 1996).

Progressive multifocal leukoencephalopathy (PML) First described in the late 1950s in patients with lymphomas and haematologic malignancies, PML is a subacute demyelinating disorder now seen most commonly in AIDS patients. PML has been estimated to be present in 1–5% of AIDS autopsy cases, and the diagnosis may be made

at the time of brain biopsy in a patient with impaired cell-mediated immunity resulting from HIV infection or any other cause, whose imaging studies show patchy white matter abnormalities (Vinters & Anders, 1990). It results from infection of the brain by the papovavirus JC (JCV, 30 nm particles containing double-stranded DNA), which has a predilection for oligodendroglia; when JCV infects these cells, it often causes cell lysis and demyelination (von Einsiedel et al., 1993; Stohlman & Hinton, 2001). Complete genomic analysis of JCV has shown subtypes of the virus, including JCV Type 2, which may be of varying importance as a cause of PML (Agostini et al., 1998). Virus is probably present in most healthy individuals, possibly in the kidney, where it is kept in check by the host immune response; when this fails, the virus replicates peripherally, including in the CNS. PML is a rare complication of haematologic malignancy (estimated less than 0.1%), vs. as high as 5% among AIDS patients) (Stohlman & Hinton, 2001). Clinicopathological features of PML are, however, identical in the two groups of patients. As the name implies, it frequently presents with multifocal neurologic findings, though patients are often encephalopathic and may appear to have a rapidly progressive dementia – sometimes one that is indistinguishable from HIV-associated dementia (Zunt et al., 1997). Survival from the time of diagnosis of PML is usually brief, measured


Fig. 21.3. Progressive multifocal leukoencephalopathy (PML); gross neuropathological features. (a) Coronal section of fixed brain from an AIDS patient with extensive PML on microscopic examination, but relatively normal appearance of the white matter on gross inspection (a representative slice is shown). (b), (c) Coronal sections of fixed brain from another patient with longstanding symptomatic PML. In both panels, note extensive poorly demarcated regions of cystic collapse and cavitation of the white matter within one cerebral hemisphere.

in weeks or months. It has been suggested that highly active antiretroviral therapy (HAART) can substantially improve the prognosis in HIV-associated PML (Clifford et al., 1999). PML is a diagnosis that may be made at biopsy or autopsy. In autopsy specimens of affected brain, coronal sections can show a range of abnormalities ranging from negligible patchy discoloration of the subcortical white matter to multifocal cavitary lesions, usually in an asymmetric distribution between the cerebral hemispheres, and often involving the cerebellum and brainstem (Fig. 21.3.) Microscopic features typically include destruction of the white matter in a ‘moth eaten’ pattern, i.e. with poorly demarcated regions of cystic cavitation – these foci may contain macrophages and collections of lymphocytes. Defining microscopic features of PML (Fig. 21.4) include bizarre atypical astrocytes, often with hyperchromatic, enlarged and lobulated nuclei, though the astrocytic density is never that of a glial neoplasm despite the glial cytologic atypia. Many oligodendroglia, especially (in the author’s experience) at the edge of an advancing lesion, show effacement of their nuclear chromatin, with presence of a glassy, somewhat amphophilic inclusion that tends to fill the nucleus, by contrast with the situation in herpesvirus inclusions (see below). If an intra-operative consultation (quick section) is requested of the pathologist for a patient undergoing brain biopsy for suspected PML, these inclusions

are often best seen on a smear/squash preparation from the biopsy (Vinters & Anders, 1990). Confirmation of the diagnosis is achieved by either (a) ultrastructural visualization of characteristic 30–50 nm diameter spherical or filamentous intranuclear viral particles, or (b) immunocytochemistry/in situ probe studies using JC virus specific reagents (von Einsiedel et al., 1993). Recently, an in situ PCR technique has been developed to detect extremely small quantities of JC virus genome within cells in brain lesions of patients with PML; currently, this method is useful largely as a research tool (Samorei et al., 2000) (Fig. 21.5). Fungal meningitis Cryptococcus neoformans is the most common fungal OI to infect the brain in patients with AIDS, and is an infection that can also occur in non-immunocompromised individuals (Anders et al., 1986; Vinters & Anders, 1990). In addition to meningitis, cryptococcus may cause pneumonitis or pericarditis (Zuger et al., 1986). In both groups of patients, cryptococcal meningitis can present with encephalopathy. The diagnosis is made clinically by studying the CSF, looking for characteristic fungal organisms on an India ink preparation (Fig. 21.6), or finding cryptococcal antigen in the fluid. The inflammatory response to cryptococcus can be extremely variable in patients with or without AIDS; it may be extremely indolent, almost non-existent

477

478

H. V. Vinters

Fig. 21.4. PML, microscopic features. (a) Note atypical gemistocytic astrocytes in brain parenchyma, some with bizarre nuclei, and scattered lymphocytes. (Mag ×90) (b) PML, though classically confined to the subcortical white matter, may extend into overlying cortex, as in this micrograph (white matter is at right, cortex at left). (Mag ×35) (c) PML may involve the cerebellum, as in this case, causing extensive cavitation of the deep white matter. (Mag ×35) (d ) More preserved regions of cerebellar white matter showed atypical astrocytes. (Mag ×350; Panels (a)–(d ) are from the same patient). (e) Focus of PML (from a different patient) showing less conspicuous atypical astrocytes, but scattered smudged, glassy (oligodendroglial) nuclei with loss of the normal chromatin pattern (arrows), characteristic inclusions seen with this infection. (Mag ×175)


Fig. 21.5. Though PML may be strongly suspected on routine (H & E-stained) histologic sections, confirmation of the diagnosis can be made by immunohistochemistry (a), using primary antibody directed against papovavirus, or in situ hybridization using JC virus-specific probe, either without (b) or with (c) PCR enhancement of the signal (for details of the technique, see Samorei et al., 2000). (d ) Electron microscopy demonstrates a nucleus filled with viral particles (arrow) among myelinated axons. (Panels (b), (c) from material studied by courtesy of Dr. Regina von Einsiedel, Heidelberg, Germany; Mag (a) ×180).

(hence the need to find evidence for the micro-organism in the CSF, even when few inflammatory cells are seen), or there may be a robust, even overwhelming granulomatous meningoencephalitis (von Einsiedel et al., 1992). Determinants of which type of inflammatory response is likely to be seen in a given patient are unknown. AIDS patients with cryptococcal infection must be maintained on long-term anti-fungal therapy, which may include amphotericin and flucytosine (Chuck & Sande, 1989). Other fungal infections occur with lower frequency in AIDS patients; those that produce a granulomatous meningitis (e.g. Coccidiodes immitis, seen quite commonly in Southern California) may present with a dementia-like syndrome or a more fulminant course (Mischel & Vinters, 1995). Primary CNS lymphoma (PCNSL) PCNSL is usually high in the differential diagnosis of a mass lesion, when such is discovered in a patient with HIV

infection or AIDS (see above). Curiously, PCNSL appears to be on the increase, independent of its high frequency in HIV-infected individuals (Miller et al., 1994); this study estimated that PCNSL tripled in frequency between the epochs 1958–77 and 1978–89, to the point where by the early 1990s PCNSL constituted 6–7% of all primary brain neoplasms. This tumour is said to occur in 10% of AIDS patients, and is by far the most common brain tumor in that population (DeAngelis, 1995). Most commonly, patients present clinically with altered mental status and lateralizing signs, though symptom onset is usually much more fulminant than with HIV-associated dementia. Nevertheless, over half of patients with HIV-associated PCNSL present with confusion, lethargy and memory loss. CT or MRI scanning usually shows single or multiple contrast-enhancing masses, ones indistinguishable from toxoplasmosis abscesses or, for that matter, several other types of CNS mass lesion. Definitive diagnosis is by brain biopsy. However, the finding

479

480

H. V. Vinters

reactive T-cells and lymphoplasmacytoid B-lymphocytes may be seen at the periphery of the lesions. AIDS patients with PCNSL also frequently show concurrent CNS and systemic CMV infections. Indeed, OIs are frequently the proximate cause of death in brain lymphoma patients with HIV infection (Goldstein et al., 1991). Epstein–Barr virus genes can frequently be detected within AIDS-related PCNSL, either by in situ hybridization or PCR (MacMahon et al., 1991; Rouah et al., 1990), although the precise role of this herpesvirus in causing these lymphomas is as yet unclear. When encountered at necropsy, PCNSL (as predicted by neuroimaging studies) may be uni- or multifocal, the lesions often manifesting regions of central necrosis (Fig. 21.7). Microscopically, they are angiocentric collections of atypical lymphoid cells, among which mitoses and bizarre, often hyperlobulated nuclei are frequent. Neoplastic lymphoid cells invade brain parenchyma from their vascular adventitial location, the cell density tapering off

Fig. 21.6. Cryptococcal meningitis. Cryptococcal infection of the brain/CSF can be detected (during life) by examining the CSF for cryptococcal antigen, or the ‘old-fashioned’ technique of staining a drop of CSF with India ink – result is shown in (a), which demonstrates characteristic fungal organisms. (b) Cryptococcal meningoencephalitis, when severe, presents (on cut sections of fixed brain) with a cribriform appearance, because of microorganisms that proliferate in Virchow–Robin spaces (around blood vessels), which are often in continuity with the subarachnoid space.

of Epstein–Barr viral DNA in the CSF is strongly supportive of the diagnosis. Whilst many patients respond to cranial irradiation (with doses of 4000–5000 Gy), survival is usually only for a few months. Longer survivals (sometimes up to 1–2 years) are reported in AIDS patients who receive combined chemotherapy and radiotherapy (DeAngelis, 1995). Neuropathological characterization of PCNSLs shows that they are almost uniformly of B-cell phenotype and difficult to subclassify using the working formulation for non-Hodgkin lymphomas (Morgello et al., 1990). Even in otherwise characteristically aggressive B-cell lymphomas,

Fig. 21.7. Primary CNS lymphoma (PCNSL) in a patient with AIDS. (a) Coronal slice of fixed brain (parieto-occipital region) shows a necrotic, poorly demarcated, infiltrative mass (arrow) in the white matter and overlying cortex. (b) PCNSL usually manifests as an angiocentric atypical lymphoid infiltrate (arrows) that extends into surrounding brain parenchyma (Mag ×70).


away from the blood vessel wall (Fig. 21.7). In rare instances, there may be destruction of the vessel wall at the centre of a lymphoid aggregate (angionecrosis), sometimes with thrombosis of the lumen – this entity has been described by the term ‘lymphomatoid granulomatosis’ (Anders et al., 1989). HIV-1 has been shown (by immunohistochemistry) to ‘accumulate’ at foci of PCNSL (Cornford et al., 1991). Meningeal infiltrates of lymphoma may also be seen in HIVinfected patients with disseminated lymphomatosis who do not show parenchymal space-occupying lymphoma lesions (Anders et al., 1986). Cerebrovascular disease Many of the OIs associated with AIDS can produce extensively necrotic lesions within brain, as can primary CNS lymphoma (PCNSL); these may, at some stage and certainly on neuroimaging studies, mimic ischaemic cerebral infarcts. In necropsy studies during which care has been taken to exclude OIs and PCNSL as a possible cause of cerebral infarct, approximately 5% of cases show cerebral infarcts, though none of the affected individuals manifested dementia on the basis of these ischaemic lesions, and indeed very few were symptomatic with stroke or transient ischaemic attacks (Connor et al., 2000). It is suggested that HIV-1 infection can lead to a relatively non-specific microangiopathy characterized by vessel wall thickening, perivascular pigment deposition, with vessel wall mineralization and perivascular inflammatory cell infiltrates, though no definite evidence of vasculitis. Cerebral hemorrhages, when they (rarely) occur in AIDS (Fig. 21.8), may be related to systemic factors or the above described HIV vasculopathy (Vinters & Anders, 1990; Ellison et al., 1998).

HIV-associated dementia Clinical features and neuroimaging HIV-1 infection of the nervous system, recognized within 2–3 years of the beginning of the AIDS epidemic, has taught basic and clinical neuroscientists a great deal about retroviral infection of the brain. The most intriguing questions about HIV in the CNS include: How does the virus enter the brain parenchyma? What are the determinants of the relatively circumscribed brain cell populations infected by HIV? Given the presence of HIV in a comparatively small number and overall percentage of brain cells, how does it produce (in some patients) a devastating neuropsychiatric syndrome and what are the neuropathologic correlates of this symptomatology? What, if any, impact has HAART therapy had on patterns of HIV infection of the CNS, and how

Fig. 21.8. Intraparenchymal haemorrhages in a patient with AIDS. The patient had widespread Kaposi sarcoma (KS) and was thought to have these tumours intracranially. At autopsy, numerous parenchymal hemorrhages (a), (b) were noted in areas with angiocentric inflammation, though conclusive evidence of a vasculitis was lacking. Toxoplasmosis was found elsewhere in the CNS, but no evidence of intracranial KS was seen.

has it affected ‘evolution’ of our understanding of HIVassociated dementia? That HIV infection of the brain can produce a syndrome characterized by cognitive decline has been appreciated since the mid-1980s. The resultant neuropsychiatric entity, an AIDS-defining illness, has been described by several names, including HIV-1 dementia, HIV-1 encephalopathy, HIV-associated cognitive/motor complex, AIDS dementia complex (ADC) and HIV-associated dementia (HAD)– the latter term being the one I shall subsequently employ (McArthur et al., 1992). The early stages of HAD are

481

482

H. V. Vinters

difficult to recognize because they are relatively nonspecific. As well, diagnostic difficulties often arise because of OIs that affect the brain, or co-morbidity that results from non-HIV-associated psychiatric disorders or substance abuse. HAD/ADC is a syndrome characterized by distinctive early and late symptoms (Navia, 1990). Early manifestations include cognitive impairment (slowness, impaired attention and concentration, forgetfulness), motor abnormalities (clumsiness, deterioration in fine motor tasks, tremor, loss of balance), and behavioural changes (reduced spontaneity, apathy, social withdrawal). Late in the disease, an affected patient shows profound mental status changes (psychomotor slowing, impaired word reversal and serial subtraction, blunted affect) and further motor deterioration (impaired rapid movements, tremor, hyperreflexia, impaired tandem gait, limb paresis). Onset of dementia is insidious, steady progression occurring thereafter over months or years; abrupt deterioration may occur in some individuals, often in the presence of systemic illness. The frequency with which HAD occurs among all infected patients is difficult to ascertain, primarily because centres expert in the treatment of the condition are likely to be referred a disproportionate number of HAD patients, especially complicated ones, and thus may report an inflated ‘incidence’ of the disorder. One epidemiologic study (based upon data in the USA) from the early 1990s concluded that HAD (then defined as HIV encephalopathy) had an incidence of 7.3%, the proportion of AIDS patients with encephalopathy being highest at the extremes of age (Janssen et al., 1992). However, a more recent study from the United Kingdom has estimated that HIV encephalitis (HIVE, which many accept – see below – as the neuropathologic correlate of HAD) is the most prevalent neuropathologic diagnosis among patients who had a post-mortem examination, having been discovered in over 25% of such autopsies, especially in patients who died at less than 30 years of age and were abusers of intravenous (IV) drugs (Davies et al., 1997). Of less interest to neuropathologists, but of immense importance as a social welfare and public health issue, has been the question of how early in systemic HIV-1 infection there is evidence of cognitive, memory or motor decline. There has been a lively debate on this issue among neuropsychologists; however, one of the largest studies to address the matter – utilizing participants in a long-term natural history investigation, the ‘Multicenter AIDS Cohort study’ – came to the important conclusion that the frequency of neuropsychological abnormalities in asymptomatic HIV1-infected males is low, indeed not statistically different from that of seronegative controls (Miller et al., 1990). Others have found that neurological and neuropsychological abnormalities are associated with AIDS-related complex

(ARC, a term sometimes used, especially early in the epidemic, to describe symptomatic pre-AIDS) but not with asymptomatic HIV-1 infection (Janssen et al., 1989). An investigation aimed at identifying risk factors for HAD found that depression, executive dysfunction, and the presence of minor cognitive/motor disorder may be among its very early manifestations (Stern et al., 2001). Neuroradiographically demonstrable cerebral atrophy is a common manifestation of HAD (Dal Pan et al., 1992). Studies aimed at establishing the relationship between cognitive impairment and cerebral atrophy have shown that the ventricle-brain ratio (VBR, an overall measure of brain atrophy) and the bicaudate ratio (a more circumscribed measure of atrophy in the region of the caudate nucleus) appear to be significantly associated with dementia. Caudate region atrophy is especially associated with HAD. Pertinent to this investigation, a detailed study examining neuropathologic substrates of ventricular enlargement detected in autopsy brains (using planimetry) had shown that CMV encephalitis is a common cause of ventricular expansion, but ventriculomegaly correlates poorly with molecular evidence of HIV infection in the brain, assessed using PCR (Gelman et al., 1996). Comparisons of quantitative brain MRI measures with neuropathologic indices of HIV infection have shown that there is no correlation between MRI volumes and astrocytosis (a relatively nonspecific marker of CNS injury). However, significant relationships were noted between severity of CNS HIV infection (assayed immunohistochemically using anti-gp41) and MRI volume estimates of both grey and (abnormal) white matter (Heindel et al., 1994). Non-demented but medically symptomatic HIV-seropositive subjects have significant increases in CSF, reduced volume of subcortical white matter, and diminished grey matter volumes, by comparison with ‘low risk’ control subjects (Jernigan et al., 1993). This study also showed, however, a trend toward similar abnormalities in ‘high risk’ control subjects who were not HIV infected. Longitudinal volumetric MRI studies of cerebral volume loss in HIV infection have convincingly demonstrated that this retroviral infection causes both grey and white matter atrophy, changes being most severe in advanced disease but present in medically asymptomatic HIV-seropositive patients (Stout et al., 1998). The caudate nucleus showed progressive volume loss that appeared linked to both rate of decrease of CD4+ cells, and stage of the disease. Proton magnetic resonance spectroscopy has shown that in early stages of HAD (HIV-cognitive motor complex or HIV-CMC, for purposes of this investigation) frontal white matter shows changes suggestive of glial proliferation and cell membrane injury, but no alterations to imply neuronal injury (at least insofar as can be evaluated using


measurement of N-acetyl compounds). Individuals with minor cognitive motor disorder (MCMD) showed abnormalities confined to the frontal white matter, whereas those with HAD had metabolite abnormalities throughout all brain regions investigated (frontal cortex and white matter, and basal ganglia) (Chang et al., 1999). As with many neurodegenerative illnesses, clinicians have long sought ‘surrogate markers’ for HAD, e.g. CSF measures that may be predictive of rate of neuropsychiatric decline or at least correlate with some component of HADrelated neuropathologic abnormalities. Any CSF investigation in an HIV-infected population must take into account the confounding effect of OIs and PCNSL (see above) that are common in AIDS patients, and may be radiographically undetectable in their earliest stages. With this caveat, elevated CSF levels of beta-2-microglobulin (above 3.8 mg/l) have been suggested to be a clinically useful marker for HAD (McArthur et al., 1992). In the area of molecular diagnostics, CSF HIV-1 mRNA levels have been discovered to be elevated in neurocognitively impaired patients with AIDS, possibly reflecting increased productive CNS HIV infection (Ellis et al., 1997). Asymptomatic HIV infection does not appear to cause specific electroencephalographic abnormalities (Nuwer et al., 1992). Clinicopathologic features Since the start of the AIDS epidemic, neuropathologists have been challenged to understand the pathophysiology of HAD. An overwhelming literature exists on its (putative) neuropathological substrates; many early studies have ‘stood the test of time’ quite well i.e. lesions suggested to be the morphological manifestations of HIV-1 in the brain during autopsy investigations of the late 1980s and early 1990s are still accepted as accurate ‘markers’ for this unfortunate occurrence, even as our understanding of HIVassociated neuropathogenesis has evolved (Budka, 1989, 1991; Ho et al., 1989; Kure et al., 1991). Autopsy brain specimens from AIDS patients can be difficult to analyze in trying to understand the neuropathology of HAD, because great care must be taken to exclude the presence of OIs and PCNSL, and the effects of systemic illness (sepsis, hypoxia) before ascribing histopathological changes to HIV-1 infection. The confusing and ever changing terminology that evolved into the diagnosis ‘HAD’ has been mirrored in the varied terms used to describe pathological consequences of HIV-1 infection in the CNS (Budka et al., 1991). These have included ‘HIV encephalitis’, ‘subacute encephalitis of AIDS’, ‘HIV-associated leukoencephalopathy’ and ‘multifocal giant cell encephalitis’. Despite these semantic controversies, multi-centre studies incorporating autopsy data from large cities in the United Kingdom, continental Europe and the

USA report that HIV encephalitis (HIVE) is the most common neuropathological finding among them (Davies et al., 1998). To simplify, HIV-1 infection of the CNS results in (a) low grade chronic (especially microglial) inflammation, (b) multinucleated (sometimes ‘giant’) macrophages, and (c) evidence of chronic injury to white matter (with pallor of the deep white matter on myelin stains), deep central grey matter and superficial cortex (Budka, 1989,1991; Budka et al., 1987; Ho et al., 1989). Each of these alterations may be seen in varying severity and different combinations in a given brain specimen (Fig. 21.9). HIV-1 can be localized to the CNS by immunohistochemistry (Fig. 21.10), in situ hybridization (ISH), or in situ PCR. It is widely accepted that microglia/macrophages are the primary cell type infected; whether astrocytes and capillary endothelium harbour HIV-1 in significant quantities remains somewhat controversial (Wiley et al., 1986; Takahashi et al., 1996). The possibility of neuronal infection by HIV-1 is debated even more hotly (for review, see Power et al., 2002). Of interest, since the advent of HAART many investigators report a decrease in the incidence of many AIDS-defining illnesses (ADIs), e.g. Pneumocystis carinii pneumonia and Kaposi sarcoma. HAD and its putative neuropathological substrate HIVE, however, have remained relatively constant over time (including in the post-HAART era) or have actually increased in relative frequency, as in one Australian study (Ives et al., 2001; Masliah et al., 2000; Dore et al., 1999). Although HIVE is regarded by many as the neuropathologic substrate of HAD (Wiley & Achim,1994,1995), clinicopathologic correlation is poor, i.e. over 50% of adults with AIDS do not show diffuse myelin pallor, or HIV type multinucleated giant cells (HIV-GC), whereas microglial nodules may be present in as many as 80–90% of AIDS patients who undergo autopsy, including many patients with no evidence of a dementing illness (Power et al., 2002). Microglial nodules are well known, as indicated above, to be associated with OIs in the brain, especially CMV encephalitis and toxoplasmosis (Vinters & Anders, 1990). Early HIV-1 brain invasion has been demonstrated in instances of iatrogenic infection, e.g. in a man who accidentally received intravenous inoculation of white blood cells from an HIVinfected individual and died 15 days later, and only 1 day after virus was first recovered from the blood (Davis et al., 1992). At least some in situ PCR studies of brain tissues from patients who are HIV-1 seropositive but asymptomatic suggest that there is early infection of brain microglia, astrocytes and microvascular endothelial cells by the retrovirus (An et al., 1999). Neuropathological studies of rare patients in the early stages of HAD who come to necropsy

483

484

H. V. Vinters

Fig. 21.10. Sections of brain (autopsy specimens) immunostained with an antibody to HIV p24 antigen. Low magnification (a) and high magnification (b) views show numerous HIV-immunoreactive cells with microglial/macrophage morphology. In (a), p24-immuno-reactive cells tend to cluster around small blood vessels. (Immunohistochemical preparation courtesy of Stanley Sung). (Mags (a) ×110, (b) ×220).

Fig. 21.9. HIV encephalitis, frequently (though not always) found at autopsy in the brain of a patient with HAD. (a) Inflammatory (‘microglial’) nodule shows a loosely defined collection of cells (arrow), including some binucleate cells, possibly macrophages. (Mag ×125) (b) Another microglial nodule shows prominent vacuolization. (c) HIV-type multinucleated giant cells (arrows), virtually pathognomonic of HIV-1 infection of the CNS, within subcortical white matter (Mag panels (b), (c), ×155).

indicate that such individuals show only astrocytic gliosis of the white matter and mild pallor on myelin stains, but no significant inflammation, brain atrophy or HIV-GC (McArthur et al, 1989). Investigations that have undertaken meticulous correlation between neuropathologic evidence

for HIVE and severity of HAD find that HIV-1 infection appears to be confined to microglia/macrophages and that both the presence and frequency of infected cells are highly correlated with histological findings of HIV type giant cell encephalitis (Brew et al., 1995). HIV-immunoreactive cells are abundant in the deep central nuclei and white matter. However, HIV infection was found to be more limited than might have been expected from the clinical severity of dementia. In children, evidence for HIVE – more commonly described as HIV-1-related progressive encephalopathy, HIVPE – is encountered more commonly than OIs and CNS lymphoma, by comparison with adults. Whereas HIV-1associated neurological disease is the initial presenting clinical manifestation of AIDS in 3–7% of infected adults, HIV encephalopathy is the first manifestation of the infection in up to 18–20% of children and adolescents, with an


overall prevalence of 30–60% (Mintz, 1994). It is usually the result of HIV infection acquired in the perinatal period. Most infected children with HIV-PE have very low CD4+ T-cell counts (Gavin & Yogev, 1999). It has been suggested that early HIV encephalopathy in infants has a different pathogenetic mechanism than that occurring in older children, which (in turn) has similarity to that noted in adults (Tardieu et al., 2000). Neuropathological features are similar to those described for HIVE, except that global cerebral atrophy and calcification of the basal ganglia are more commonly observed in children than in adults with HAD. The microcephaly frequently seen in infants and children with AIDS is the result of global loss of brain tissue, including proportional loss of cerebral cortex, subcortical grey and white matter (Kozlowski et al., 1997). The parallel issues of (a) how best to quantify HIV protein or genome within brain tissue or CSF, and (b) what represent(s) the best morphological correlate of HIV-mediated brain injury, have become vexing ones, in particular because correlations between clinical syndromes and HIV ‘viral load’/brain damage rely on accurate determinations of both parameters (Achim et al., 1994b). HIV-1 DNA can be measured by quantitative PCR (Pang et al., 1990). When this technique is applied to brain tissue from asymptomatic HIV seropositive subjects, very low levels of provirus have been found, possibly representing ‘blood contamination’ of the CNS (Bell et al., 1993). Sensitive measures of viral load (using quantitative HIV RNA assays) in autopsy brain and CSF show that high HIV-1 levels in the CSF correlate with (and may be good predictors of) high brain levels of virus. There is anatomic predilection of HIV-1 for basal ganglia and hippocampi. Of interest, some studies also find reasonable agreement between molecular assays of viral load and semi-quantitative immunohistochemical evaluations of brain tissue (using anti-HIV primary antibodies, see Fig. 21.10), suggesting that in laboratories which only have access to the latter methodology, reasonable correlations between viral load and other clinicopathological parameters of HAD can be attempted (Achim et al., 1994b; Wiley et al., 1998). Cellular/molecular pathogenesis Brain infection by any virus is defined in terms of its neuroinvasiveness, neurotropism and neurovirulence (Power et al., 2002). Strains of HIV that cause neurologic abnormalities are ‘neurotropic’, though not ‘neuronotropic’, showing a propensity to infect primarily microglial cells/macrophages (Koyanagi et al., 1987; Power et al., 2002). Neurotropism is probably determined by several viral genes. The two competing hypotheses of how HIV gains entry to the protected environment of the CNS are (i) that it

does so when HIV-infected lymphocytes/monocytes cross the blood-brain barrier (cerebral capillary endothelium), or (ii) that it enters through the choroid plexus, with subsequent seeding of the brain. Several cell surface receptors and co-receptors (CD4, CCR5 and CCR3) are important mediators of HIV-1 entry into cells. Given that HIV-1 appears to infect predominantly brain microglia/macrophages, how then does one explain the devastating consequences of this infection for normal CNS structure and function (Sotrel & Dal Canto, 2000; Dubois-Dalcq et al., 1995)? One suggestion is that HIV-infected macrophages can initiate neurotoxicity, which then becomes amplified through cellto-cell interactions with astrocytes (Dickson et al., 1993; Epstein & Gendelman, 1993). These interactions cause the production of cytokines such as tumour necrosis factor (TNF)-alpha and interleukin-1beta, and arachidonate metabolites that may cause neuronal injury and astrocytic proliferation. Glial cytokines, including microglial-derived interleukin-1 alpha (which stimulates proliferation of astrocytes) and astrocyte-derived S100 beta (which can lead to neuritic degeneration), are abundant in the brains of HIV infected individuals; their presence and putative actions in the CNS establish further cell biologic links between pathogenetic mechanisms that may be important in both HAD and Alzheimer disease (Stanley et al., 1994). In tissue culture studies of macrophages, expression of various cytokines is enhanced when they are co-cultured with astrocytes (Fiala et al., 1996). Careful autopsy investigations comparing brain viral burden with macrophage activation factors and degrees of synaptic loss show that viral burden correlates well with brain parenchymal levels of factors such as IL1beta and IL-6, and that these in turn correlate with quantitative evidence of ‘neuritic injury’ (Achim et al., 1996). Dendritic injury is now well established as a morphological substrate for HIV-related cognitive disorders (Masliah et al., 1997). Cortical synaptic density is diminished in mild to moderate HIV-related neurocognitive disorder (Everall et al., 1999). The role of neuronal death, perhaps by apoptosis, in the evolution of this process is less clear though suggested by some investigators (An et al., 1996). A recently developed technique by which to study morpho-anatomical features of cortical neurons, ‘spatial pattern analysis’, shows distinctive patterns of neuron loss in various stages of HIV-related cognitive impairment (Asare et al., 1996). Precisely how synaptic/dendritic/neuronal injury is mediated in HAD is less clear. HIV-1 proteins, including gp120 (especially its CD4 binding and V3 domains) have been shown to be neurotoxic, as have Tat, gp41, and Nef. The final common pathway of irreversible neuronal injury may be

485

486

H. V. Vinters

through abnormal accumulation of intracellular calcium or activation of NMDA receptors (for review, see Power et al., 2002). White matter abnormalities that are a characteristic finding on both neuroimaging and pathological examination of the CNS in AIDS patients are of uncertain aetiology. Several investigators have suggested that ‘leukoencephalopathy’ may be secondary to subtle abnormalities of the blood– brain barrier (BBB), which may lead (in rare instances) to brain oedema and axonal damage in addition to more subtle white matter abnormalities (Smith et al., 1990; Rhodes, 1991; Power et al., 1993; Gray et al., 1998). Heterogeneous lesions predominantly affecting myelinated fibre tracts, including long tract degeneration and more fulminant, widespread forms of leukoencephalopathy, have been described in AIDS patients (Horoupian et al, 1984; Jones Jr. et al., 1988). Widespread axonal injury may be present in HIV-infected individuals, as may peculiar necrotizing lesions of the white matter, multifocal necrotizing leukoencephalopathy – the latter lesion is also encountered in immunosuppressed patients without AIDS (Giangaspero & Foschini, 1988; Anders et al., 1993).

Dementia associated with other viral infections Relatively rare types of viral encephalitis (e.g. secondary to arboviruses) almost always cause fulminant disease that would generally not be confused with a dementia syndrome (Griffin, 1995). Encephalitis resulting from some types of viral infection of the CNS may rarely mimic a rapidly progressive dementia, though viral encephalitis is usually acute or subacute in its progression, often with focal neurological signs and symptoms. As well, a patient with viral encephalitis in general appears ‘septic’, with fever, headache and a confusional state that may mimic dementia but can often be differentiated from it by detailed clinical history and careful physical examination. Herpes simplex virus encephalitis (HSVE) The commonest form of sporadic necrotizing, often haemorrhagic, encephalitis in both adults and children, results from infection of the brain by herpes simplex virus type 1 (HSV-1), a double stranded DNA virus. Reflecting the pathogen’s predilection for the frontal and temporal lobes, focal neurological signs in HSVE may include ataxia, dysphasia and hemiparesis (Ellison et al., 1998). Focal seizures are often a presenting feature of the encephalitis. Hypomanic symptoms may result from HSVE (Fisher, 1996). CSF examination in affected patients shows a moderate lymphocytosis, while EEG and neuroimaging studies point to

abnormalities in the temporal lobes. Diagnostic testing of the CSF for HSV DNA using PCR has been suggested to be the most specific, rapid and sensitive means by which to make an early diagnosis, often eliminating the need for a diagnostic brain biopsy, thus facilitating early therapy (Kleinschmidt-DeMasters et al., 2001; Anderson et al., 1991; Guffond et al., 1994). HSVE is currently eminently treatable with antiviral agents (Whitley et al., 1989). Autopsy study of an affected brain shows abnormalities that vary with the length of time the infection has been present prior to the patient’s death, and how effectively the infection responded to antiviral therapy, assuming such was administered. Subacute cases may show asymmetric haemorrhagic necrosis of frontal and temporal lobes with oedema. Insular cortex and cingulate gyri may be affected to varying degrees (Vinters et al., 1998). More chronic or long term (e.g. clinically unsuspected or undocumented) cases – still sometimes encountered – occasionally show cystic collapse of the brain parenchyma (Fig. 21.11). Microscopic sections of involved brain show, in subacute cases, acute and chronic, often angiocentric inflammation. Macrophages may be prominent in both the injured brain parenchyma and overlying leptomeninges, which also often show abundant lymphocytes (Love, 2001). A characteristic microscopic feature of HSVE is the presence within infected tissue of eosinophilic intranuclear ‘owl’s eye’ (Cowdry type A) inclusions in several cell types (Fig. 21.12). Presence of HSV-1 antigen can be confirmed by immunohistochemistry. Immunohistochemical studies show that HSV antigen is abundant within the brain in patients dying within the first week after disease onset, and this persists for over 2 weeks. Sites at which antigen was most plentiful included medial and inferior temporal lobes, hippocampi, amygdaloid nuclei, olfactory cortex, insula and cingulate gyri; though bihemispheric involvement by the infection was usually seen, its severity was usually asymmetric between the cerebral hemispheres (Esiri, 1982). We have recently examined the brain of an elderly woman who carried the clinical diagnosis of ‘probable spongiform encephalopathy’ but at brain cutting showed relatively unremarkable gross features of the fixed brain. Microscopic examination, however, revealed widespread chronic encephalitis, typical Cowdry A intranuclear inclusions, and (by immunohistochemistry) widespread HSV antigen in regions where chronic inflammation was most prominent (Fig. 21.13). HSVE is thought to represent reactivation of latent HSV infection (lying dormant in, for example, trigeminal ganglia) as the result of expression of ‘supplemental essential genes’ that determine neuroinvasiveness and neurovirulence (Baringer & Pisani, 1994; Liebert, 2001).


Fig. 21.11. Herpes simplex encephalitis. (a) Subacute case – appearance of (unfixed) brain. Note poorly demarcated regions of haemorrhagic necrosis (resembling traumatic contusions) on undersurface of the left temporal lobe. (b) Chronic case. View of base of (fixed) brain shows extensive encephalomalacia of temporal and frontal lobes. (c), (d ). Coronal sections (same case as illustrated in (b) show asymmetric cystic cavitation and encephalomalacia in frontal lobes, extending (d ) into the left insula (arrow). Histological sections showed foci of chronic inflammation surrounded by rims of gliotic brain tissue. Both in situ hybridization and PCR (of extracted DNA) using HSV-specific primers confirmed presence of HSV in the tissue.

An intriguing idea that has been intermittently under investigation in many laboratories for several years is that reactivation of latent HSV-1 in trigeminal ganglia may be involved in the pathogenesis and progression of Alzheimer’s disease lesions within brain; this hypothetical causal association has yet to be conclusively proven (Ball et al., 2001).

Varicella zoster virus (VZV) This virus may produce disease as a rare OI in patients with AIDS; in such patients, it may cause leukoencephalitis and small vessel disease. In immunocompetent hosts, this virus may cause herpes zoster ophthalmicus with large arterial occlusion and resultant stroke (Kleinschmidt-DeMasters & Gilden, 2001a,b).

487

488

H. V. Vinters

Fig. 21.12. Herpes simplex encephalitis, subacute. The patient was thought to have transmissible spongiform encephalopathy; cut sections of the fixed brain showed minimal abnormalities. Histological sections, however, showed a widespread, often angiocentric, chronic inflammation (a) Numerous Cowdry type A intranuclear inclusions were identified (arrows, (b)) within the tissue. (Mag (a) ×110, (b) ×435)

Fig. 21.13. Herpes simplex encephalitis – immunohistochemistry. Tissue sections are from same patient as illustrated in Fig. 21.12. (a) Section immunostained with primary antibody to a macrophage/microglial marker (CD68) shows numerous immunoreactive cells. (b) Section immunostained with primary antibody to HSV shows scattered immunolabelled cells of differing morphology, including some that have the appearance of neurons. (Magnification: both panels ×220).

Subacute sclerosing panencephalitis (SSPE) This is a rare progressive degenerative disorder, usually seen in children and adolescents, associated with measles virus infection; it may present later in life (Singer et al., 1997). Disease onset is usually between 5 and 15 years of age (Dyken, 2001). Epidemiological studies of SSPE in Turkey have shown that its age of onset decreased from 13 years (before 1994) to 7.6 years (after 1995); the latent period between measles virus infection and disease onset declined from 9.9 to 5.9 years (Anlar et al., 2001a). Adult onset SSPE patients become ill in early adulthood (20–35 years), and often have no well-documented history of measles virus exposure. When a history of such exposure exists, it has generally occurred at an earlier (less than 3 years old) or later (after 9 years) age than usual childhood infection, with unusually long time periods (14–22 years) between the initial infection and SSPE onset. As in children, clinical

symptoms of SSPE can include myoclonus, spastic hemiparesis, bradykinesia and rigidity (as motor manifestations), though presentation is often with visual, and less commonly behavioral complaints. Pathogenesis of SSPE is thought to be related to a persistent form of the ‘wild-type’ measles virus which has remained dormant in the CNS (with restricted viral gene expression) for years (Dyken, 2001; Liebert, 2001). The relationship of SSPE to measles virus infection, suspected since the 1960s, now appears beyond question, there having been a drop in the numbers of SSPE cases associated with measles immunization programmes. However, in recent years there has been a modest increase in numbers of SSPE cases. SSPE has been subclassified into acute, subacute and chronic progressive variants, though progression to maximal neurological disability occurs in the vast majority of patients within 2 years of onset.


Fig. 21.14. Subacute sclerosing panencephalitis (SSPE), in a patient with especially severe brainstem involvement. (a), (b) show prominently angiocentric inflammation without, however, evidence of injury to the blood vessel wall. (c), (d ) Inclusions found in this case included characteristic intranuclear amphophilic structures (arrow, (c)), and other nuclei with pronounced clearing (arrow, (d )); note effacement of the normal chromatin pattern in both inclusions.

Neuropathological microscopic features of SSPE include (as the name of the disease suggests), various degrees of neuron loss, lymphocytic infiltration (including perivascular lymphocytic cuffing), and microglial nodules within both grey and white matter (Fig. 21.14). Demyelination of the deep white matter often results, but may be inconspicuous on inspection of slices of the fixed brain (Anlar et al., 2001b). On histopathological examination of brain sections, typical inclusion bodies may be seen (Fig. 21.14), though measles virus antigen can be demonstrated within endothelial cells, neurons and lymphocytes even in their absence. The inflammatory infiltrate is often composed of both CD4+ (perivascular) and CD8+ (parenchymal) lymphocytes. MHC antigens, interferon-gamma, and tumour necrosis factor-alpha (TNF-alpha) may be expressed in endothelial and glial cells, and a variety of other cytokines may be present within SSPE lesions (Nagano et al., 1994). Patterns of inflammatory cell infiltration and cytokine expression seem to correlate with clinical course of the disease in

some affected patients (Anlar et al., 2001b). Recent studies suggest that hypermutated M protein of SSPE measles virus may actively contribute to chronic progressive CNS disease (Patterson et al., 2001).

Dementia associated with neurosyphilis and Lyme disease In the modern era, neurosyphilis has become a relatively rare cause of dementia, though syphilis itself may be on the rise as a result of the AIDS epidemic. Syphilis results from infection by the spirochete Treponema pallidum, usually acquired by cutaneous or mucous membrane inoculation during sexual intercourse. The three stages of the disease include primary syphilis, secondary to proliferation of micro-organisms at the site of inoculation (usually producing a chancre); secondary syphilis (which develops approximately 6 weeks later in untreated patients) resulting

489

490

H. V. Vinters

Fig. 21.15. Meningovascular syphilis. (a) Coronal sections through the frontal lobes show bilateral cystic infarcts (arrows) in the subcortical white matter. (b) Microscopic section from left frontal lobe shows typical features of a small cystic infarct with a glial rim. (Mag ×11) (c), (d ) Sections of leptomeninges in the interpeduncular cistern show moderately dense lymphoplasmacytic infiltrate, which was noted in all areas (Mag (c) ×55, (d ) ×215). (e), ( f ). Arteritis affecting a leptomeningeal artery. Note pronounced intimal hyperplasia (e) and a modest lymphoplasmacytic infiltrate throughout the vessel wall. ( f ) Shows detail of chronic inflammatory cells adjacent to internal elastic lamina (indicated by arrowheads). (Mag (e) ×55, ( f ) ×215).

from hematogenous dissemination of spirochetes – clinical manifestations include lymphadenopathy and a maculopapular rash – lasts for weeks or months and is followed by a latent phase; and tertiary syphilis, occurring months to years after the lesions of secondary syphilis have subsided in roughly 30–50% of untreated patients with latent syphilis. Manifestations of tertiary syphilis may include vascular, ocular, CNS parenchymal and gummatous lesions. Neurosyphilis is said to be more common in patients with concomitant HIV infection (Ellison et al., 1998).

Meningovascular syphilis This results in a combined multifocal angiitis and chronic meningitis, most commonly occurring 6–7 years after initial infection. We have encountered one elderly patient with a dementing syndrome of mysterious origins who, at necropsy, showed classic manifestations of meningovascular syphilis (Fig. 21.15), even though he lacked an established ante-mortem diagnosis of primary or secondary syphilis. Brain parenchymal infarcts may result from the arteritis (as they apparently did in this patient), and the meninges often contain – in addition to a variably dense lympho-plasmacytic infiltrate – miliary gummas. The angiitis commonly involves media and adventitia of arteries of varying size, with resultant intimal hyperplasia in larger vessels (Fig. 21.15).

General paresis of the insane (GPI) Caused by chronic meningoencephalitis, this usually follows primary infection by 10–20 years, presenting with subtle impairments of attention and cognition, followed by psychiatric symptoms, intellectual decline, loss of motor control, seizures and incontinence. Neuropathological features include grossly apparent cerebral (cortical) atrophy, relatively inconspicuous meningeal and parenchymal clusters of plasma cells and lymphocytes usually centred on blood vessels, modest neuronal loss and astrocytic gliosis, and a dramatic proliferation of rod-like microglia, easily demonstrable using immunohistochemistry incorporating primary antibodies directed against microglial epitopes. Gummas, a late manifestation of tertiary syphilis, may present in the CNS as space occupying lesions. In this location they may resemble tuberculomas but differ from the latter (microscopically) because reticulin is usually preserved (Ellison et al., 1998).

Lyme disease This results from CNS infection by Borrelia burgdorferi, a loosely coiled spirochete (9–30 m in length) transmitted by the bite of ticks. It must be considered an extraordinarily rare cause of dementia, though ‘Stage 2’ infection may be characterized by lymphocytic meningitis, polyradiculitis and cranial nerve palsies while Stage 3 infection


(accepting the fact that the staging of Lyme disease is itself somewhat controversial) is characterized by neurological manifestations that include low grade encephalopathy, encephalomyelitis and cerebral infarcts secondary to ‘angiopathy’ – none of which have been well characterized in rigorous neuropathologic studies, given that Lyme disease is rarely fatal and brain biopsy is only carried out infrequently in affected individuals.

Dementia associated with other bacterial infections and meningitides Most forms of bacterial meningitis have a fulminant course and do not cause a clinical condition that would present with slowly or rapidly progressive dementia. Neurotuberculosis, usually secondary to extracranial (usually pulmonary) infection, may progress with a more indolent course (Ellison et al., 1998). It usually manifests as either (a) a severe basal granulomatous meningitis, with caseating granulomas and ‘endarteritis’ on the intimal aspect of vessels that are entrapped by the meningeal exudate, frequently with resultant cerebral parenchymal infarcts, or (b) tuberculoma, which may mimic many types of primary brain tumour, e.g. a frontal lobe neoplasm that might present with dementia (Dastur et al., 1995; Elisevich & Arpin, 1982). We have recently encountered a rare example of MAI infection in a (non-AIDS) immunosuppressed patient, which presented as a meningioma-like mass lesion (Di Patre et al., 2000). Idiopathic hypertrophic (craniocervical) pachymeningitis This is a very rare disorder that manifests as chronic inflammatory hypertrophy of the dura (Fig. 21.16). It usually affects the spinal dura mater, but may involve cranial dura, in which case it can cause signs and symptoms of raised intracranial pressure with resultant encephalopathy (Botella et al., 1994; Rosenfeld et al., 1987).

Acknowledgements The author acknowledges the generous funding provided through the UCLA Multi-Center AIDS Cohort contract (AI/CA 35040) and the UCLA Alzheimer Disease Research Center grant (P50 AG16570). Carol Appleton assisted with preparation of figures, and Stanley Sung, Alex Brooks and Beth Johnson prepared many of the specimens from which micrographs were made.

Fig. 21.16. Idiopathic hypertrophic pachymeningitis. Dura was markedly thickened and infiltrated by polymorphous chronic inflammatory cells (a), including lymphoid and plasma cell elements (b) and macrophages (c). (Mags (a), (c) ×80, (b) ×160).

491

492

H. V. Vinters

REFERENCES Achim, C. L., Nagra, R. M., Wang, R., Nelson, J. A. & Wiley, C. A. (1994a). Detection of cytomegalovirus in cerebrospinal fluid autopsy specimens from AIDS patients. The Journal of Infectious Diseases, 169, 623–7. Achim, C. L., Wang, R., Miners, D. K & Wiley, C. A. (1994b). Brain viral burden in HIV infection. Journal of Neuropathology and Experimental Neurology, 53, 284–94. Achim, C. L., Masliah, E., Heyes, M. P. et al. (1996). Macrophage activation factors in the brains of AIDS patients. Journal of Neuro-AIDS, 1, 1–16. Agostini, H. T., Shishido-Hara, Y., Baumhefner, R. W., Singer, E. J., Ryschkewitsch, C. F. & Stoner, G. L. (1998). JC virus type 2: definition of subtypes based on DNA sequence analysis of ten complete genomes. Journal of General Virology, 79, 1143–51. An, S. F., Giometto, B., Scaravilli, T., Tavolato, B., Gray, F. & Scaravilli, F. (1996). Programmed cell death in brains of HIV-1-positive AIDS and pre-AIDS patients. Acta Neuropathologica, 91, 169– 73. An, S. F., Groves, M., Gray, F. & Scaravilli, F. (1999). Early entry and widespread cellular involvement of HIV-1 DNA in brains of HIV-1 positive asymptomatic individuals. Journal of Neuropathology and Experimental Neurology, 58, 1156–62. Anders, H.-J. & Goebel, F.-D. (1998). Cytomegalovirus polyradiculopathy in patients with AIDS. Clinical Infectious Diseases, 27, 345–52. Anders, K. H., Guerra, W. F., Tomiyasu, U., Verity, M. A. & Vinters, H. V. (1986). The neuropathology of AIDS. UCLA experience and review. American Journal of Pathology, 124, 537–58. Anders, K. H., Latta, H., Chang, B. S., Tomiyasu, U., Quddusi, A. & Vinters, H. V. (1989). Lymphomatoid granulomatosis and malignant lymphoma of the central nervous system in the acquired immunodeficiency syndrome. Human Pathology, 20, 326–34. Anders, K. H., Becker, P. S., Holden, J. K. et al. (1993). Multifocal necrotizing leukoencephalopathy with pontine predilection in immunosuppressed patients: a clinicopathologic review of 16 cases. Human Pathology, 24, 897–904. Anderson, N. E., Willoughby, E. W., Synek, B. J. L., Croxson, M. C. & Glasgow, G. L. (1991). Brain biopsy in the management of focal encephalitis. Journal of Neurology, Neurosurgery, and Psychiatry, 54, 1001–3. Andrews, B. T. & Kenefick, T. P. (1993). Neurosurgical management of the acquired immunodeficiency syndrome. An update. Western Journal of Medicine, 158, 249–53. ¨ Anlar, B., Köse, G., Gurer, Y., Altunbs¸ak, S¸., Haspolat, S. & Okan, M. (2001a). Changing epidemiological features of subacute sclerosing panencephalitis. Infection, 29, 192–5. Anlar, B., Söylemezo˘glu, F., Aysun, S., Köse, G., Belen, D. & Yalaz, K. (2001b). Tissue inflammatory response in subacute sclerosing panencephalitis (SSPE). Journal of Child Neurology, 16, 895– 900. Antinori, A., Ammassari, A., De Luca, A. et al. (1997). Diagnosis of AIDS-related focal brain lesions: a decision-making analysis

based on clinical and neuroradiologic characteristics combined with polymerase chain reaction assays in CSF. Neurology, 48, 687–94. Artigas, J., Grosse, G. & Niedobitek, F. (Eds.) (1993). The Central Nervous System in AIDS. Berlin-Heidelberg-New York: SpringerVerlag. Asare, E., Dunn, G., Glass, J. et al.(1996). Neuronal pattern correlates with the severity of human immunodeficiency virusassociated dementia complex. Usefulness of spatial pattern analysis in clinicopathological studies. American Journal of Pathology, 148, 31–8. Ball, M. J., Mathews, R., Steiner, I. et al. (2001). Latent HSV 1 virus in trigeminal ganglia: the optimal site for linking prevention of Alzheimer’s disease to vaccination. Neurobiology of Aging, 22, 705–9. Baringer, J. R. & Pisani, P. (1994). Herpes simplex virus genomes in human nervous system tissue analyzed by polymerase chain reaction. Annals of Neurology, 36, 823–9. Bell, J. E., Busuttil, A., Ironside, J. W. et al. (1993). Human immunodeficiency virus and the brain: investigation of virus load and neuropathologic changes in pre-AIDS subjects. The Journal of Infectious Diseases, 168, 818–24. Botella, C., Orozco, M., Navarro, J. & Riesgo, P. (1994). Idiopathic chronic hypertrophic craniocervical pachymeningitis: case report. Neurosurgery, 35, 1144–9. Brew, B. J., Rosenblum, M., Cronin, K. & Price, R. W. (1995). AIDS dementia complex and HIV-1 brain infection: Clinical-virological correlations. Annals of Neurology, 38, 563–70. Budka, H. (1989). Human immunodeficiency virus (HIV)-induced disease of the central nervous system: pathology and implications for pathogenesis. Acta Neuropathologica, 77, 225–36. Budka, H. (1991). Neuropathology of human immunodeficiency virus infection. Brain Pathology, 1, 163–75. Budka, H., Costanzi, G., Cristina, S. et al. (1987). Brain pathology induced by infection with the human immunodeficiency virus (HIV). A histological, immunocytochemical, and electron microscopical study of 100 autopsy cases. Acta Neuropathologica, 75, 185–98. Budka, H., Wiley, C. A., Kleihues, P. et al. (1991). HIV-associated disease of the nervous system: Review of nomenclature and proposal for neuropathology-based terminology. Brain Pathology, 1, 143–52. Caselli, R. J., Scheithauer, B. W., O’Duffy, J. D., Peterson, G. C., Westmoreland, B. F. & Davenport, P. A. (1993). Chronic inflammatory meningoencephalitis should not be mistaken for Alzheimer’s disease. Mayo Clinic Proceedings, 68, 846–53. Chang, L., Ernst, T., Leonido-Yee, M., Walot, I. & Singer, E. (1999). Cerebral metabolite abnormalities correlate with clinical severity of HIV-1 cognitive motor complex. Neurology, 52, 100–8. Chrétien, F., Bélec, L, Hilton, D. A. et al. (1996). Herpes simplex virus type 1 encephalitis in acquired immunodeficiency syndrome. Neuropathology and Applied Neurobiology, 22, 394–404.


Chuck, S. L. & Sande, M. A. (1989). Infections with Cryptococcus neoformans in the acquired immunodeficiency syndrome. The New England Journal of Medicine, 321, 794–9. Ciricillo, S. F. & Rosenblum, M. L. (1990). Use of CT and MR imaging to distinguish intracranial lesions and to define the need for biopsy in AIDS patients. Journal of Neurosurgery, 73, 720–4. Clifford, D. B., Yiannoutsos, C., Glicksman, M. et al. (1999). HAART improves prognosis in HIV-associated progressive multifocal leukoencephalopathy. Neurology, 52, 623–5. Connor, M. D., Lammie, G. A., Bell, J. E., Warlow, C. P., Simmonds, P. & Brettle, R. D. (2000). Cerebral infarction in adults AIDS patients. Observations from the Edinburgh HIV autopsy cohort. Stroke, 31, 2117–26. Cornford, M. E., Said, J. W. & Vinters, H. V. (1991). Immunohistochemical localization of human immunodeficiency virus (HIV) in central nervous system lymphoproliferative disorders of patients with AIDS. Modern Pathology, 4, 232–8. Cornford, M. E., Ho, H. W. & Vinters, H. V. (1992). Correlation of neuromuscular pathology in acquired immune deficiency syndrome patients with cytomegalovirus infection and zidovudine treatment. Acta Neuropathologica, 84, 516–29. Dal Pan, G. J., McArthur, J. H., Aylward, E. et al. (1992). Patterns of cerebral atrophy in HIV-1-infected individuals: Results of a quantitative MRI analysis. Neurology, 42, 2125–30. Darnell, R. B. (1993). The polymerase chain reaction: application to nervous system disease. Annals of Neurology, 34, 513–23. Dastur, D. K., Manghani, D. K. & Udani, P. M. (1995). Pathology and pathogenetic mechanisms in neurotuberculosis. Radiologic Clinics of North America, 33, 733–52. Davies, J., Everall, I. P., Weich, S., McLaughlin, J., Scaravilli, F. & Lantos, P. L. (1997). HIV-associated brain pathology in the United Kingdom: an epidemiological study. AIDS, 11, 1145– 50. Davies, J., Everall, I. P., Weich, S. et al. (1998). HIV-associated brain pathology: a comparative international study. Neuropathology and Applied Neurobiology, 24, 118–24. Davis, L. E., Hjelle, B. L., Miller, V. E. et al. (1992). Early viral brain invasion in iatrogenic human immunodeficiency virus infection. Neurology, 42, 1736–9. DeAngelis, L. M. (1995). Primary brain tumors in the acquired immunodeficiency syndrome. Current Opinion in Neurology, 8, 419–23. Di Patre, P. L., Radziszewski, W., Martin, N. A., Brooks, A. & Vinters, H. V. (2000). A meningioma-mimicking tumor caused by Mycobacterium avium complex in an immunocompromised patient. American Journal of Surgical Pathology, 24, 136–9. Dickson, D. W., Lee, S. C., Mattiace, L. A., Yen, S.-H. C. & Brosnan, C. (1993). Microglia and cytokines in neurological disease, with special reference to AIDS and Alzheimer’s disease. Glia, 7, 75– 83. Dore, G. J., Correll, P. K., Li, Y., Kaldor, J. M., Cooper, D. A. & Brew, B. J. (1999). Changes to AIDS dementia complex in the era of highly active antiretroviral therapy. AIDS, 13, 1249–53.

Dubois-Dalcq, M., Altmeyer, R., Chiron, M. & Wilt, S. (1995). HIV interactions with cells of the nervous system. Current Opinion in Neurology, 5, 647–55. Dyken, P. R. (2001). Neuroprogressive disease of post-infectious origin: A review of a resurging subacute sclerosing panencephalitis (SSPE). Mental Retardation and Developmental Disabilities Research Reviews, 7, 217–25. Elisevich, K. & Arpin, E. J. (1982). Tuberculoma masquerading as a meningioma. Case report. Journal of Neurosurgery, 56, 435–8. Ellis, R. J., Hsia, K., Spector, S. A. et al. (1997). Cerebrospinal fluid human immunodeficiency virus type 1 RNA levels are elevated in neurocognitively impaired individuals with acquired immunodeficiency syndrome. Annals of Neurology, 42, 679–88. Ellison, D., Love, S., Chimelli, L. et al. (1998). Neuropathology. A Reference Text of CNS Pathology. London–Chicago– Philadelphia: Mosby. Epstein, L. G. & Gendelman, H. E. (1993). Human immunodeficiency virus type 1 infection of the nervous system: Pathogenetic mechanisms. Annals of Neurology, 33, 429–36. Esiri, M. M. (1982). Herpes simplex encephalitis. An immunohistological study of the distribution of viral antigen within the brain. Journal of the Neurological Sciences, 54, 209–26. Esiri, M., Scaravilli, F., Millard, P. R. & Harcourt-Webster, J. N. (1989). Neuropathology of HIV infection in haemophiliacs: comparative necropsy study. British Medical Journal, 299, 1312–15. Everall, I. P., Heaton, R. K., Marcott, T. D. et al. (1999). Cortical synaptic density is reduced in mild to moderate human immunodeficiency virus neurocognitive disorder. Brain Pathology, 9, 209–17. Fiala, M., Rhodes, R. H., Shapshak, P. et al. (1996). Regulation of HIV1 infection in astrocytes: expression of Nef, TNF- alpha and IL-6 is enhanced in coculture of astrocytes with macrophages. Journal of Neuro-Virology, 2, 158–66. Fisher, C. M. (1996). Hypomanic symptoms caused by herpes simplex encephalitis. Neurology, 47, 1374–8. Fleming, K. C., Adams, A. C. & Petersen, R. C. (1995). Dementia: Diagnosis and evaluation. Mayo Clinic Proceedings, 70, 1093– 107. Gavin, P. & Yogev, R. (1999). Central nervous system abnormalities in pediatric human immunodeficiency virus infection. Pediatric Neurosurgery, 31, 115–23. Gelman, B. B., Dholakia, S., Casper, K., Kent, T. A. Cloyd, M. W. & Freeman, D. Jr. (1996). Expansion of the cerebral ventricles and correlation with acquired immunodeficiency syndrome neuropathology in 232 patients. Archives of Pathology and Laboratory Medicine, 120, 866–71. Giangaspero, F. & Foschini, M. P. (1988). Diffuse axonal swellings in a case of acquired immunodeficiency syndrome. Archives of Pathology and Laboratory Medicine, 112, 1259–62. Goldstein, J. D., Dickson, D. W., Moser, F. G. et al. (1991). Primary central nervous system lymphoma in acquired immune deficiency syndrome. A clinical and pathologic study with results of treatment with radiation. Cancer, 67, 2756–65. Gray, F. (ed.). (1993). Atlas of the Neuropathology of HIV Infection. Oxford-New York: Oxford University Press.

493

494

H. V. Vinters

Gray, F., Mohr, M., Rozenberg, F. et al. (1992). Varicella-zoster virus encephalitis in acquired immunodeficiency syndrome: report of four cases. Neuropathology and Applied Neurobiology, 18, 502–14. Gray, F., Bélec, L., Chrétien, F. et al. (1998). Acute, relapsing brain oedema with diffuse blood–brain barrier alteration and axonal damage in the acquired immunodeficiency syndrome. Neuropathology and Applied Neurobiology, 24, 209–16. Griffin, D. E. (1995). Arboviruses and the central nervous system. Springer Seminars in Immunopathology, 17, 121–32. Guffond, T., Dewilde, A., Lebert, P-E., Caparros-Lefebre, D., Hober, D. & Wattre, P. (1994). Significance and clinical relevance of the detection of Herpes simplex virus DNA by the polymerase chain reaction in cerebrospinal fluid from patients with presumed encephalitis. Clinical Infectious Diseases, 18, 744–9. Heindel, W. C., Jernigan, T. L., Archibald, S. L., Achim, C. L, Masliah, E. & Wiley, C. A. (1994). The relationship of quantitative brain magnetic resonance imaging measures to neuropathologic indexes of human immunodeficiency virus infection. Archives of Neurology, 51, 1129–35. Ho, D. D., Bredesen, D. E., Vinters, H. V. & Daar, E. S. (1989). The acquired immunodeficiency syndrome (AIDS) dementia complex. Annals of Internal Medicine, 111, 400–10. Holland, N. R., Power, C., Mathews, V. P., Glass, J. D., Forman, M. & McArthur, J. C. (1994). Cytomegalovirus encephalitis in acquired immunodeficiency syndrome (AIDS). Neurology, 44, 507–14. Horoupian, D. S., Pick, P., Spigland, I., Smith, P., Portenoy, R., Katzman, R. & Cho, S. (1984). Acquired immune deficiency syndrome and multiple tract degeneration in a homosexual man. Annals of Neurology, 15, 502–5. Ives, N. J., Gazzard, B. G. & Easterbrook, P. J. (2001). The changing pattern of AIDS-defining illnesses with the introduction of highly active antiretroviral therapy (HAART) in a London clinic. Journal of Infection, 42, 134–9. Janssen, R. S., Saykin, A. J., Cannon, L. et al. (1989). Neurological and neuropsychological manifestations of HIV-1 infection: Association with AIDS-related complex but not asymptomatic HIV-1 infection. Annals of Neurology, 26, 592–600. Janssen, R. S., Nwanyanwu, O. C., Selik, R. M. & Stehr-Green, J. K. (1992). Epidemiology of human immunodeficiency virus encephalopathy in the United States. Neurology, 42, 1472–6. Jellinger, K. A., Setinek, U., Drlicek, M., Bohm, G., Steurer, A. & Lintner, F. (2000). Neuropathology and general autopsy findings in AIDS during the last 15 years. Acta Neuropathologica, 100, 213–20. Jernigan, T. L., Archibald, S., Hesselink, J. R. et al. (1993). Magnetic resonance imaging morphometric analysis of cerebral volume loss in human immunodeficiency virus infection. Archives of Neurology, 50, 250–5. Jones, H. R. Jr., Ho, D. D., Forgacs, P. et al. (1988). Acute fulminating fatal leukoencephalopathy as the only manifestation of human immunodeficiency virus infection. Annals of Neurology, 23, 519–22.

Kennedy, P. G. E. (1990). The widening spectrum of infectious neurological disease. Journal of Neurology, Neurosurgery, and Psychiatry, 53, 629–32. Kleinschmidt-DeMasters, B. K. & Gilden, D. H. (2001a). Varicellazoster virus infections of the nervous system. Clinical and pathologic correlates. Archives of Pathology and Laboratory Medicine, 125, 770–80. Kleinschmidt-DeMasters, B. K. & Gilden, D. H. (2001b). The expanding spectrum of herpesvirus infections of the nervous system. Brain Pathology, 11, 440–51. Kleinschmidt-DeMasters, B. K., DeBiasi, R. L. & Tyler, K. L. (2001). Polymerase chain reaction as a diagnostic adjunct in herpesvirus infections of the nervous system. Brain Pathology, 11, 452–64. Koyanagi, Y., Miles, S., Mitsuyasu, R. T., Merrill, J. E., Vinters, H. V. & Chen, I. S. Y. (1987). Dual infection of the central nervous system by AIDS viruses with distinct cellular tropisms. Science, 236, 819–22. Kozlowski, P. B., Brudkowska, J., Kraszpulski, M. et al. (1997). Microcephaly in children congenitally infected with human immunodeficiency virus – a gross-anatomical morphometric study. Acta Neuropathologica, 93, 136–45. Kure, K., Llena, J. F., Lyman, W. D. et al. (1991). Human immunodeficiency virus-1 infection of the nervous system: An autopsy study of 268 adult, pediatric, and fetal brains. Human Pathology, 22, 700–10. Liebert, U. G. (2001). Slow and persistent virus infections of neurones–a compromise for neuronal survival. Current Topics in Microbiology and Immunology, 253, 35–60. Love, S. (2001). Autopsy approach to infections of the CNS. In Neuropathology. A Guide for Practising Pathologists (Current Topics in Pathology 95), ed. S. Love, pp. 1–50. Berlin-Heidelberg: Springer-Verlag. McArthur, J. C., Becker, P. S., Parisi, J. E. et al. (1989). Neuropathological changes in early HIV-1 dementia. Annals of Neurology, 26, 681–4. McArthur, J. C., Nance-Sproson, T. E., Griffin, D. E. et al. (1992). The diagnostic utility of elevation in cerebrospinal fluid beta2-microglobulin in HIV-1 dementia. Neurology, 42, 1707–12. MacMahon, E. M. E., Glass, J. D., Hayward, S. D. et al. (1991). Epstein–Barr virus in AIDS-related primary central nervous system lymphoma. The Lancet, 338, 969–73. Malouf, R., Jacquette, G., Dobkin, J. & Brust, J. C. M. (1990). Neurologic disease in human immunodeficiency virus-infected drug abusers. Archives of Neurology, 47, 1002–7. Mann, D. M. A., Neary, D. & Testa, H. (1994). Color Atlas and Text of Adult Dementias, London-Baltimore: Mosby-Wolfe. Martinez, E., Marcos, A., Martinez-Chamorro, E., MartinezChamorro, C. & Domingo, P. (1993). Brain biopsy for intracranial mass lesions in AIDS. The Lancet, 341, 242. Masliah, E., Heaton, R. K., Marcotte, T. D. et al. (1997). Dendritic injury is a pathological substrate for human immunodeficiency virus-related cognitive disorders. Annals of Neurology, 42, 963– 72.


Masliah, E., DeTeresa, R. M., Mallory, M. E. & Hansen, L. A. (2000). Changes in pathological findings at autopsy in AIDS cases for the last 15 years. AIDS, 14, 69–74. Mendez, M. F., Mastri, A. R., Sung, J. H. & Frey W. H. II. (1992). Clinically diagnosed Alzheimer disease: Neuropathologic findings in 650 cases. Alzheimer Disease and Associated Disorders, 6, 35–43. Miller, D. C., Hochberg, F. H., Harris, N. L., Gruber, M. L., Louis, D. N. & Cohen, H. (1994). Pathology with clinical correlations of primary central nervous system non-Hodgkin’s lymphoma. The Massachusetts General Hospital Experience 1958–1989. Cancer, 74, 1383–97. Miller, E. N., Selnes, O. A., McArthur, J. C. et al. (1990). Neuropsychological performance in HIV-1-infected homosexual men: The Multicenter AIDS Cohort Study (MACS). Neurology, 40, 197–203. Mintz, M. (1994). Clinical comparison of adult and pediatric NeuroAIDS. Advances in Neuroimmunology, 4, 207–21. Mischel, P. S. & Vinters, H. V. (1995). Coccidioidomycosis of the central nervous system: neuropathological and vasculopathic manifestations and clinical correlates. Clinical Infectious Diseases, 20, 400–5. Morgello, S., Cho, E.-S., Nielsen, S., Devinsky, O. & Petito, C. K. (1987). Cytomegalovirus encephalitis in patients with acquired immunodeficiency syndrome: An autopsy study of 30 cases and a review of the literature. Human Pathology, 18, 289–97. Morgello, S., Petito, C. K. & Mouradian, J. A. (1990). Central nervous system lymphoma in the acquired immunodeficiency syndrome. Clinical Neuropathology, 9, 205–15. Morgello, S., Mahboob, R., Yakoushina, T., Khan, S. & Hague, K. (2002). Autopsy findings in a human immunodeficiency virusinfected population over 2 decades. Influences of gender, ethnicity, risk factors, and time. Archives of Pathology and Laboratory Medicine, 126, 182–90. Nagano, I., Nakamura, S., Yoshioka, M., Onodera, J., Kogure, K. & Itoyama, Y. (1994). Expression of cytokines in brain lesions in subacute sclerosing panencephalitis. Neurology, 44, 710–15. Navia, B. A. (1990). The AIDS dementia complex. In Subcortical Dementia, ed. J. L. Cummings, pp. 181–98. New York-Oxford: Oxford University Press. Nuwer, M. R., Miller, E. N., Visscher, B. R. et al. (1992). Asymptomatic HIV infection does not cause EEG abnormalities: results from the Multicenter AIDS Cohort Study (MACS). Neurology, 42, 1214–19. Pang, S., Koyanagi, Y., Miles, S., Wiley, C., Vinters, H. V. & Chen, I. S. Y. (1990). High levels of unintegrated HIV-1 DNA in brain tissue of AIDS dementia patients. Nature, 343, 85–9. Patterson, J. B., Cornu, T. I., Redwine, J. et al. (2001). Evidence that the hypermutated M protein of a subacute sclerosing panencephalitis measles virus actively contributes to the chronic progressive CNS disease. Virology, 291, 215–25. Pell, M. F., Thomas, D. G. T. & Whittle, I. R. (1991). Stereotactic biopsy of cerebral lesions in patients with AIDS. British Journal of Neurosurgery, 5, 585–9.

Power, C., Kong, P-A., Crawford, T. O. et al. (1993). Cerebral white matter changes in acquired immunodeficiency syndrome dementia: Alterations of the blood–brain barrier. Annals of Neurology, 34, 339–50. Power, C., Gill, M. J. & Johnson, R. T. (2002). The neuropathogenesis of HIV infection: Host-virus interaction and the impact of therapy. Canadian Journal of Neurological Sciences, [in press]. Rhodes, R. H. (1991). Evidence of serum-protein leakage across the blood-brain barrier in the acquired immunodeficiency syndrome. Journal of Neuropathology and Experimental Neurology, 50, 171–83. Robert, M. E., Geraghty, J. J. III, Miles, S. A., Cornford, M. E. & Vinters, H. V. (1989). Severe neuropathy in a patient with acquired immune deficiency syndrome (AIDS). Evidence for widespread cytomegalovirus infection of peripheral nerve and human immunodeficiency virus-like immunoreactivity of anterior horn cells. Acta Neuropathologica, 79, 255–61. Rosenfeld, J. V., Kaye, A. H., Davis, S. & Gonzales, M. (1987). Pachymeningitis cervicalis hypertrophica. Case report. Journal of Neurosurgery, 66, 137–9. Rouah, E., Rogers, B. B., Wilson, D. R., Kirkpatrick, J. B. & Buffone, G. J. (1990). Demonstration of Epstein–Barr virus in primary central nervous system lymphomas by the polymerase chain reaction and in situ hybridization. Human Pathology, 21, 545– 50. Samorei, I. W., Schmid, M., Pawlita, M. et al. (2000). High sensitivity detection of JC-virus DNA in postmortem brain tissue by in situ PCR. Journal of NeuroVirology, 6, 61–74. Singer, C., Lang, A. E. & Suchowersky, O. (1997). Adult-onset subacute sclerosing panencephalitis: Case reports and review of the literature. Movement Disorders, 12, 342–53. Smith, T. W., DeGirolami, U., Hénin, D., Bolgert, F. & Hauw, J.J. (1990). Human immunodeficiency virus (HIV) leukoencephalopathy and the microcirculation. Journal of Neuropathology and Experimental Neurology, 49, 357–70. Sotrel, A. & Dal Canto, M. C. (2000). HIV-1 and its causal relationship to immunosuppression and nervous system disease in AIDS: a review. Human Pathology, 31, 1274–98. Stanley, L. C., Mrak, R. E., Woody, R. C. et al. (1994). Glial cytokines as neuropathogenic factors in HIV infection: pathogenic similarities to Alzheimer’s disease. Journal of Neuropathology and Experimental Neurology, 53, 231–8. Stern, Y., McDermott, M. P., Albert, S. et al. (2001). Factors associated with incident human immunodeficiency virus-dementia. Archives of Neurology, 58, 473–9. Stohlman, S. A. & Hinton, D. R. (2001). Viral induced demyelination. Brain Pathology, 11, 92–106. Stout, J. C., Ellis, R. J., Jernigan, T. L. et al. (1998). Progressive cerebral volume loss in human immunodeficiency virus infection. A longitudinal volumetric magnetic resonance imaging study. Archives of Neurology, 55, 161–8. Takahashi, K., Wesselingh, S. L., Griffin, D. E., McArthur, J. C., Johnson, R. T. & Glass, J. D. (1996). Localization of HIV1 in human brain using polymerase chain reaction/in situ

495

496

H. V. Vinters

hybridization and immunocytochemistry. Annals of Neurology, 39, 705–711. Tardieu, M., Le Chenadec, J., Persoz, A. et al. (2000). HIV-1-related encephalopathy in infants compared with children and adults. Neurology, 54, 1089–95. Temesgen, Z. & Wright, A. J. (1997). Recent advances in the management of human immunodeficiency virus infection. Mayo Clinic Proceedings, 72, 854–9. Thomas, G., Murphy, S., Staunton, H., O’Neill, S., Farrell, M. A. & Brett, F. M. (1998). Pathogen-free granulomatous diseases of the central nervous system. Human Pathology, 29, 110–15. Tompkins, L. S. (1992). The use of molecular methods in infectious diseases. The New England Journal of Medicine, 327, 1290–7. Vinters, H. V. & Anders, K. H. (1990). Neuropathology of AIDS, Boca Raton, Florida: CRC Press. Vinters, H. V., Kwok, M. K., Ho, H. W. et al. (1989). Cytomegalovirus in the nervous system of patients with the acquired immune deficiency syndrome. Brain, 112, 245–68. Vinters, H. V., Mischel, P. S. & Ranchod, M. (1996). Central nervous system. Pathology: State of the Art Reviews, 3(2), 513–34. Vinters, H. V., Farrell, M. A., Mischel, P. S. & Anders, K. H. (1998). Diagnostic Neuropathology, New York: Marcel Dekker. von Einsiedel, R. G., Fiala, M. & Vinters, H. V. (1992). Severe granulomatous meningoencephalitis with CNS cryptococcosis. Journal of Neuropathology and Experimental Neurology, 51, 375 [abstract]. von Einsiedel, R. W., Fife, T. D., Aksamit, A. J. et al. (1993). Progressive multifocal leukoencephalopathy in AIDS: a clinicopathologic

study and review of the literature. Journal of Neurology, 240, 391–406. Whitley, R. J., Cobbs, C. G., Alford, C. A. Jr. et al. (1989). Diseases that mimic herpes simplex encephalitis. Diagnosis, presentation, and outcome. Journal of the American Medical Association, 262, 234–9. Wiley, C. A. & Achim, C. L. (1994). Human immunodeficiency virus encephalitis is the pathological correlate of dementia in acquired immunodeficiency syndrome. Annals of Neurology, 36, 673–6. Wiley, C. A. & Achim, C. L. (1995). Human immunodeficiency virus encephalitis and dementia. Annals of Neurology, 38, 559–60. Wiley, C. A., Schrier, R. D., Nelson, J. A., Lampert, P. W. & Oldstone, M. B. A. (1986). Cellular localization of human immunodeficiency virus infection within the brains of acquired immune deficiency syndrome patients. Proceedings of the National Academy of Sciences USA, 83, 7089–93. Wiley, C. A., Soontornniyomkij, V., Radhakrishnan, L. et al. (1998). Distribution of brain HIV load in AIDS. Brain Pathology, 8, 277–84. Zuger, A., Louie, E., Holzman, R. S., Simberkoff, M. S. & Rahal, J. J. (1986). Cryptococcal disease in patients with the acquired immunodeficiency syndrome. Diagnostic features and outcome of treatment. Annals of Internal Medicine, 104, 234–40. Zunt, J. R., Tu, R. K., Anderson, D. M., Copass, M. C. & Marra, C. M. (1997). Progressive multifocal leukoencephalopathy presenting as human immunodeficiency virus type 1 (HIV)-associated dementia. Neurology, 49, 263–5.

22 Schizophrenia and its dementia Paul J. Harrison University Department of Psychiatry, Warneford Hospital, Oxford, UK

Some readers may be surprised to find a chapter with this title in this book, since schizophrenia is not renowned either for its dementia or for its neuropathology. However, recent studies have shown that cognitive impairment is an important feature of schizophrenia (Aleman et al., 1999; Sharma & Harvey, 2000), and that there are neuropathological correlates of the disorder, albeit still poorly characterized (Harrison, 1999a; Harrison & Roberts, 2000).

Cognitive impairment in schizophrenia Kraepelin, who at the turn of the last century first described the syndrome now called schizophrenia, named it ‘dementia praecox’. However, for many years the cognitive aspects of the disorder were neglected (Johnstone et al., 1978). In recent years, matters have changed, and neuropsychological deficits are now considered an important, perhaps a cardinal, feature of the syndrome. They have clinical relevance, in that they are better predictors of functional outcome than are the psychotic symptoms (Green, 1996), and they are potential therapeutic targets for novel neuroleptic drugs (Harvey & Keefe, 2001). Recent models also postulate that the neuropsychological features are closer to the key pathophysiological and pathogenic processes of the disorder (Goldman-Rakic & Selemon, 1997; Andreasen, 1999). The neuropathological findings to be described may likewise be related more directly to the cognitive deficits than to the psychotic symptoms; this is plausible based upon their characteristics, and given the fact that the cognitive abnormalities are by and large stable (Goldberg et al., 1993),

whereas psychotic symptoms wax and wane, often being absent for many years before death. The neuropsychological impairments of schizophrenia are complex and are only outlined briefly here (for review, see David & Cutting, 1994; Kuperberg & Heckers, 2000; Sharma & Harvey, 2000). There is a generalized intellectual impairment, averaging approximately one standard deviation across most domains (Blanchard & Neale, 1994). The degree of impairment varies markedly between patients, and some perform entirely normally or above average. Deficits are seen at the first presentation of the illness, and premorbidly, and are therefore not merely the consequence of the illness or its treatment (Mohamed et al., 1999; Bilder et al., 2000). Whether they progress during the illness is unclear; there is evidence both for short term improvements associated with amelioration of initial psychotic symptoms (Gold et al., 1999; Hoff et al., 1999), but also evidence from chronic, elderly patients of a decline in later years (Harvey et al., 1999; Fucetola et al., 2000) to the extent that a significant minority are unequivocally demented (Arnold et al., 1995a; Davidson et al., 1995; de Vries et al., 2001). As well as generalized neuropsychological impairment, detailed analyses suggest that there is a greater involvement of language, semantic memory (McKenna, 1994), and executive functioning (Barch et al., 2001). The latter is present even in patients who are otherwise intellectually normal (Weickert et al., 2000; Kremen et al., 2001). These differential deficits implicate prefrontal and medial temporal (Saykin et al., 1994; Heckers et al., 1998) regions, and possibly a preferential left hemisphere involvement (Ragland et al., 1999). Interestingly, as discussed below, similar inferences have been drawn independently from the neuropathological findings.


497

498

P. J. Harrison

Prevalence of Alzheimer’s disease in schizophrenia The belief that Alzheimer’s disease is more frequent in people with schizophrenia seems to have arisen, without good data, in the 1930s (Corsellis, 1962). It was restated in, and thereby given new impetus by, the article which included the famous aphorism about schizophrenia being the neuropathologists’ graveyard (Plum, 1972). The belief may have been encouraged by an assumption that the cognitive deficits experienced by patients were due to Alzheimer-type neuropathology, perhaps complemented by the high prevalence of psychotic symptoms in patients with Alzheimer’s disease (Wragg & Jeste, 1989) and their putative association with greater neurofibrillary pathology (see Sweet et al., 2000). Some empirical support came from three studies which reported Alzheimer-type changes in a high proportion of subjects with schizophrenia (Buhl & Bojsen-Møller, 1988; Soustek, 1989; Prohovnik et al., 1993; Table 22.1). However, these studies had major limitations, either not using adequate control groups, or lacking diagnostic criteria for schizophrenia or Alzheimer’s disease; indeed, reanalysis of a subset of the cases in the study of Prohovnik et al. (1993) confirmed the original diagnosis of Alzheimer’s disease in less than half the subjects. In contrast to these studies, several more powerful investigations in the past decade have shown categorically that Alzheimer’s disease is not more common in schizophrenia. The studies are summarized in Table 22.1. In a meta-analysis of 1293 patients and 479 comparison subjects, Baldessarini et al. (1997) showed an odds ratio for Alzheimer’s disease in schizophrenia of 0.86 (95% confidence interval 0.65 to 1.15). Studies published subsequently uniformly support their negative conclusion (Arnold et al., 1998; Murphy et al., 1998; Niizato et al., 1998; Purohit et al., 1998; Jellinger & Gabriel, 1999). Neither are cholinergic (El-Mallakh et al., 1991; Haroutunian et al., 1994) and immunological (Powchik et al., 1993) markers of Alzheimer’s disease altered in schizophrenia. Of particular relevance here, the lack of association between Alzheimer’s disease and schizophrenia also applies to the subgroup of patients with schizophrenia who are clinically demented. A key study was conducted by Arnold and colleagues (1998), who investigated prospectively assessed demented patients with schizophrenia who had a mean mini-mental state score of 12. Using a battery of quantitative immunocytochemical markers, they found no evidence of Alzheimer’s-type changes (Arnold et al., 1998). Dwork et al. (1998), Purohit et al. (1998) and El-Mallakh et al. (1991) came to a similar conclusion. Other lesions, such as

Lewy bodies, ubiquitinated dystrophic neurites, activated microglia, and aberrant prion protein are also absent in elderly schizophrenics (Arnold et al., 1998, 1999). In summary, there is no evidence for Alzheimer’s disease or any other neurodegenerative process underlying schizophrenia, and the dementia of schizophrenia remains neuropathologically unexplained (Arnold & Trojanowski, 1996a). There are a few minor differences reported in demented compared to non-demented subjects with schizophrenia which may give some clues: an increase in some but not all immunocytochemical indices of astrocytic gliosis (Arnold et al., 1996), a higher apolipoprotein E ε4 allele frequency (Arnold et al., 1997b), more neuritic senile plaques (after exclusion of cases meeting criteria for Alzheimer’s disease; Dwork et al., 1998), and a lower brain weight (Jellinger & Gabriel, 1999). However, there need be no specific pathology waiting to be found: one possibility, raised earlier, is that the neuropathology of schizophrenia itself is related to the cognitive features of the syndrome, in which case demented patients may simply have more extensive, but qualitatively similar, abnormalities than those without. Alternatively, the neuropathology of schizophrenia may render the brain more vulnerable to the effects of a normal amount of age-related neurodegenerative change. These possibilities have yet to be examined.

Gliosis in schizophrenia The question of gliosis has a broader importance in contemporary studies of the neuropathology of schizophrenia. Gliosis has been important for two reasons. First, in a landmark paper, Stevens (1982) reported that periventricular fibrillary gliosis was present in about 70% of her cases of schizophrenia. She proposed that an inflammatory or an infective aetiology might underlie the disorder, and thereby helped to trigger the current generation of neuropathological studies. Secondly, the presence of gliosis has been viewed, simplistically, as evidence for a neurodegenerative process of some kind underlying schizophrenia. (The term ‘gliosis’ has been loosely defined in the field, being taken to mean an increase in one or other measure of astrocytes.) Conversely, the absence of gliosis, in the presence of other macroscopic and microscopic alterations, is taken to imply a neurodevelopmental process, particularly one occurring in the second trimester in utero (Weinberger, 1987, 1995; Roberts, 1991). By extension, this crude dichotomous view also bears upon debates as to the progressive or static nature of the pathology, and the likely underlying molecular processes (e.g. excitotoxicity vs. apoptosis; Olney & Farber, 1995; DeLisi, 1997; Lieberman, 1999).

Table 22.1. Prevalence of Alzheimer’s disease in schizophrenia Cases (number, age)

Clinically demented

Controls

Criteria for scz

Criteria for AD

Prevalence of AD in cases

Comments

Buhl and BojsenMøller, 1988 Soustek, 1989

23

n/k

10

ICD-8

33%

No AD in controls

None

> 5 SP/CX field, and > 5 NFT/HC field > 15 SP/ ×45 field

225; 100 > 60 y

n/k

0

20%

56

St Louis

None stated

–

n/k

0

None

‘Numerous’ SP and NFT in CX or HC

28%

All (Mean CDRs 2.5) n/k

12

DSM-IIIR

Khachaturian

0%

AD in 41% > 60 y, and in 25/41 (61%) > 70 y No difference in SP or NFT densities rated on three-point scale AD in 32% > 65 y, and in 39% > 75 y. NB: On review, AD diagnoses did not meet CERAD/Khachaturian criteria in 42–66% of cases No other dementias found

Bruton et al., 1990

48, mean age 73 y

n/k

Prohovnik et al., 1993

544

Purohit et al., 1993

13, age 54–100 y

Baldessarini et al., 1997a

135, mean age 50 y

201

St Louis

NIA

9%

Arnold et al., 1998

23, mean age 80 y

Most. Mean MMSE 12 68% (on chart review)

14

DSM-IV

Khachaturian

–

Dwork et al., 1998

66, mean age 78 y

16

DSM-IIIR

Khachaturian

8%

Murphy et al., 1998 Niizato et al., 1998

51, mean age 72 y 125, mean age 61 y

n/k No

0 12

DSM-IV DSM-IV

Khachaturian –

2% –

Purohit et al., 1998

100, mean age 78 y

50

DSM-IIIR

CERAD, Khachaturian

9%

Jellinger & Gabriel, 1999

99, mean age 69 y

72% (CDR > 1) 56%

AD cases not stated. No difference in NFT or SP densities in 12 cases > 75 y compared to the 12 age-matched controls Other dementias in 4%

0

DSM-IIIR, ICD-10

NIA, CERAD, Braak

7%

Definite AD in 2%, probable in 5%

Study

Lower AD prevalence than in controls (19%); odds ratio = 0.48 (95% confidence interval 0.23–0.94) NFT and SP densities, also astrocyte, microglial, and ubiquitin densities, same as controls Includes cases from Prohovnik et al. (1993), re-embedded and stained. 9% had other degenerative disorders

Note: AD: Alzheimer’s disease; CDR: clinical dementia rating; CX: cortex; HC: hippocampus; MMSE: mini-mental state examination; NFT: neurofibrillary tangle; n/k: not known; SCZ: schizophrenia; SP: senile plaque. a Unpublished data of Benes and Bird cited in this meta-analysis.

500

P. J. Harrison

The existing studies of gliosis in schizophrenia are summarized in table 22.2. The table is divided into those studies in which brains with neuropathological abnormalities were excluded (such as Alzheimer-type changes, infarcts, trauma), and those which were not. The argument for the former strategy is that such abnormalities are almost certainly coincidental, and inclusion of such brains confounds the search for the intrinsic pathology of schizophrenia (see Bruton et al., 1990; Harrison, 1995; Arnold & Trojanowski, 1996b). The data in Table 22.2 show clearly that gliosis is not a feature of the disorder. However, the possibility that there may be a subgroup in which gliosis is present has not been excluded (Stevens, 1997). Also, there are several important methodological and interpretational caveats to the assumption that a lack of gliosis proves a neurodevelopmental origin of schizophrenia – just as the presence of gliosis would not prove the opposite. These issues have been discussed elsewhere (Roberts & Harrison, 2000), and justify the statement that, in a disorder notable for its neuropathological obscurities, gliosis is ‘the most inscrutable lesion’ (Bruton et al., 1990). Nevertheless, the strength of the evidence for a lack of gliosis, in the context of the positive macroscopic and cytoarchitectural alterations to be described below, does suggest that the pathogenesis of schizophrenia is more likely to be basically developmental rather than degenerative. Epidemiological and other evidence supports this view (Harrison, 1997; Akil & Weinberger, 2000).

Antipsychotic drugs and neurofibrillary pathology Since the 1950s, most patients with schizophrenia have been treated with antipsychotic (neuroleptic) drugs, often at high doses for many years, and it has been suggested that these might contribute to the (alleged) excess of Alzheimertype changes (Wisniewski et al., 1994), and in theory they might also contribute to the other neuropathological findings to be described. Wisniewski’s suggestion was based on the finding of a higher instance of neurofibrillary tangles in cases autopsied after the introduction of these drugs compared to those from an earlier era. However, this association appears to be another myth. First, re-analysis of the data in the Wisniewski et al. (1994) paper concluded there was no significant difference between groups (Baldessarini et al., 1997); in any case, the experimental design precluded firm conclusions being drawn. Secondly, a range of other studies, both human and experimental, do not find any support for a causal relationship between antipsychotic drugs and Alzheimer’s disease (Jellinger, 1977; Harrison, 1999b).

Antipsychotic drugs are associated with some morphometric and ultrastructural changes, particularly in the basal ganglia but there is no evidence that they cause the positive findings reported in schizophrenia (Harrison, 1999b), and they may increase cortical gliosis. Indeed, increased cortical glial density has been reported after long-term administration to monkeys (Selemon et al., 1999), which makes the lack of gliosis in schizophrenia all the more noteworthy.

The cytoarchitectural neuropathology of schizophrenia Meta-analyses of MRI studies show conclusively that there is a neuropathology of schizophrenia, in terms of increased ventricular size and decreased brain volume (Lawrie & Abukmeil, 1998; Wright et al., 2000). Regions of particular involvement include the hippocampal formation (Nelson et al., 1998), cortical grey matter, and perhaps the thalamus (Konick & Friedman, 2001). The neuropathological counterpart of these observations is a brain weight decreased by ∼3–5% (Pakkenberg, 1987; Bruton et al., 1990; Harrison et al., 2003). There is some evidence for an interaction of the pathology with cerebral asymmetry, and for more abnormalities to be seen in the left than the right temporal lobe (Crow et al., 1989; Crow, 1990; Holinger et al., 2000). The alterations seen on MRI are demonstrable at, and before, the onset of symptoms (Lawrie et al., 1999; Velakoulis et al., 1999). There is no evidence that the changes are limited to an ‘organic’ subgroup (Daniel et al., 1991). Longitudinal studies indicate some enhancement of structural changes during the first months of overt illness, but thereafter, as with the cognitive deficits, it is unclear whether further progression occurs; it may be a feature of a subgroup (Woods, 1998). In the absence of evidence for neurodegenerative processes in schizophrenia, or for the changes being attributable to treatment received by the patients, the question arises as to the histological explanation for these macroscopic findings. The answer appears to reside at the level of the cytoarchitecture, with changes in the cellular and synaptic composition of the cortex demonstrated using a range of quantitative morphometric and immunocytochemical approaches. Hippocampal formation. The hippocampal formation has been the most extensively investigated region in contemporary neuropathological studies of schizophrenia, stimulated by striking positive findings from initial studies, notably those by Kovelman and Scheibel (1984) and Jakob and Beckmann (1986). It is also the site of the greatest controversies (Harrison & Eastwood, 2001).

Table 22.2. Studies of gliosis in schizophrenia

Author

Region

(a) Studies without neuropathological purification1 Stevens, 1982 Multiple areas Bogerts et al., 1983 Substantia nigra Benes et al., 1986 Frontal and cingulate cortex (areas 4,10, 24) Hippocampus Falkai and Bogerts,1986a Owen et al., 1987 Frontal and temporal cortex; hippocampus; amygdala; putamen; hypothalamus Falkai et al., 1988a Entorhinal cortex Cortex; periventricular region Bruton et al., 1990b Casanova et al., Dentate gyrus 1990 Pakkenberg, 1990 Mediodorsal thalamus; ventral striatum; basolateral amygdala Karson et al., 1993 Cortical areas; thalamus; cerebellum; pontine tegmentum Arnold et al., 1996 Hippocampus, frontal and visual cortex

Jonsson et al., 1997 Falkai et al., 1999

Hippocampus Entorhinal cortex, subiculum, several white matter regions

(b ) Studies with neuropathological purification 2 Roberts et al., 1986c Multiple areas Roberts et al., 1987 Multiple temporal lobe areas C.D. Stevens et al., Caudate, cingulate white matter, 1988c periventricular nuclei Temporal, frontal and parietal Crow et al., 1989b cortex; hippocampus, amygdala and caudate Benes et al., 1991 Frontal and cingulate cortex (areas 10, 24) Frontal and occipital cortex (areas Selemon et al., 9, 17) 1995d Perrone-Bizzozero Frontal, temporal and occipital et al., 1996 cortex (areas 9, 10, 17, 20) ¨ ur ¨ et al., 1998 Ong Frontal and parietal cortex (areas 24, 3b) Frontal and occipital cortex (areas Rajkowska et al., 9, 17) 1998d Karson et al., 1999 Frontal cortex (area 10) Radewicz et al., 2000 Johnston-Wilson et al., 2000 Benes et al., 2001 Cotter et al., 2001

1

Number of cases/controls

Parameter measured

Technique

25/20 6/6 7–10/5–9

Gliosis Glial number Glial density

Holzer stain Nissl stain Nissl stain

15/9

Glial number

Nissl stain

39/44

MAO-B activity3

Enzyme assay

15/9 48/56 6/7

Glial number Gliosis rating Glial number

Nissl stain Holzer stain Holzer stain

Unchanged Increased Unchanged

12/12

Absolute glial number GFAP abundance

Nissl stain; physical disector Western blot

Unchanged/ decreased Unchanged

GFAP- and vimentin-positive astrocyte density; GFAP IR4 Glial density GFAP-positive astrocyte density

Immunocytochemistry

Unchanged5

Nissl stain Immunocytochemistry

Unchanged Unchanged

Gliosis Gliosis GFAP-positive cell counts Gliosis

GFAP IR4 GFAP IR4 Immunocytochemistry DBI IR6 ; Holzer stain

Unchanged Unchanged Unchanged

Glial density

Nissl stain

Unchanged

Glial density

Nissl stain

Unchanged

GFAP abundance

Western blot

Unchanged

Glial density and number Glial size

Nissl stain

Unchanged

Nissl stain

Unchanged

Western and northern blots Immunocytochemistry 2D-electrophoresis

Unchanged

Nissl stain Nissl stain

Unchanged Decreased in lamina VI

4–12/ 3–10 21/12

4/8 33/26

5/7 18/12 5/7 18/20

15–18/ 9–12 16/19 5–10/ 4–6 11/11 9/10 14/12

Frontal and temporal cortex (areas 9, 22, 24) Frontal cortex (area 10)

8/5–10 24/23

GFAP and GFAP mRNA abundance GFAP-positive cell counts GFAP isoforms

Frontal cortex (area 24) Frontal cortex (area 24)

11/12 15/15

Glial density Glial density

Studies including all schizophrenics, or with clinical criteria for exclusion (e.g. a history of dementia). Studies excluding brains with neuropathological abnormalities (e.g. presence of infarcts, trauma). 3 MAO-B: monoamine oxidase-B. The enzyme is mainly glial. 4 GFAP IR: glial fibrillary acidic protein. IR: immunoreactivity. 5 Schizophrenics with dementia had higher GFAP-positive astrocyte densities than those without (see text). 6 DBI: diazepam-binding inhibitor. DBI is a selective glial marker. a−d Studies sharing the same superscript were carried out on the same brains. 2

Finding in schizophrenia Increased Unchanged Unchanged/ decreased Unchanged/ decreased Unchanged/ decreased

Unchanged

Unchanged Decreased

502

P. J. Harrison

The first influential abnormality to be reported was that of neuronal disarray, with a loss or even reversal of the normal orientation of pyramidal neurons in Ammon’s horn such that the apical dendrite points toward the stratum radiatum (Kovelman & Scheibel, 1984). This disarray was most marked at the boundaries of CA1 with CA2 and with the subiculum. The authors suggested that a developmental migrational disturbance might be responsible, and thereby fostered the emerging neurodevelopmental view of the disorder. Although qualified support for a greater variability of orientation of hippocampal pyramidal neurons in schizophrenia has followed (Conrad et al., 1991; Jonsson et al., 1997), the observation has not been replicated independently using image analysis-based quantitative methods (Christison et al., 1989; Benes et al., 1991b; Arnold et al., 1995b; Zaidel et al.,1997b). Another often stated feature is that of misplaced, smaller, and aberrantly clustered neurons in the entorhinal cortex, particularly the lamina II pre-alpha cells which give rise to the perforant pathway, as reported by Jakob and Beckmann (1986). The finding was interpreted as resulting from aberrant neuronal migration and thence supportive of a prenatal, neurodevelopmental process. Somewhat similar findings were subsequently described by others (Arnold et al., 1991, 1997a). However, several other studies, with larger samples and/or better methodologies have failed to find clear differences between cases and controls (Akil & Lewis 1997; Krimer et al., 1997; Bernstein et al., 1998). Falkai and colleagues (Falkai et al., 2000) revived the controversy by reporting quantitative evidence of abnormally located, and smaller sized clusters, of neurons. It is important to resolve this issue unequivocally, because entorhinal cortex abnormalities of this kind would constitute strong evidence in favour of an early developmental origin for schizophrenia. A loss of hippocampal pyramidal neurons is also sometimes described as being a feature of schizophrenia, although it has only been described in two studies (Jeste & Lohr, 1989; Jonsson et al., 1997) and equivocally in a third (Falkai & Bogerts, 1986), whereas several studies have found no change (Benes et al., 1991a; Arnold et al., 1995b) or a right-sided increase (Zaidel et al., 1997a). Notably, the sole stereological study to date found no differences in neuronal number or density in any hippocampus sub field (Heckers et al., 1991). In contrast to these well-known but relatively uncorroborated findings, a decreased size of hippocampal pyramidal neurons in some or all subfields has been reasonably well replicated (Benes et, al., 1991a; Arnold et al.,1995b; Zaidel et al.,1997a; but see Benes et al., 1998). A change in average neuronal size likely reflects differences in the axodendritic arbourizations of the neurons. Consistent with

this relationship, it is therefore of note that pre-synaptic and dendritic abnormalities have also been reported in the hippocampal formation in schizophrenia. In particular, several studies have found decreases in expression of pre-synaptic proteins, as well as for abnormalities in the structure or molecular characteristics of dendrites. These abnormalities are together suggestive of aberrant neural connectivity (for review see Harrison & Eastwood 2001). There is evidence for involvement of both excitatory (Harrison & Eastwood, 1998) and inhibitory (Benes & Berretta, 2001) circuits. Frontal cortex A separate literature supportive of neuronal, synaptic and dendritic alterations broadly similar to those in the hippocampal formation – including the points of controversy – exists for the prefrontal cortex, particularly dorsalaterally (Brodmann areas 9 and 46). Pyramidal neuron size in lamina III is decreased (Rajkowska et al., 1998; Pierri et al., 2001) and pre-synaptic and dendritic markers are reduced (Glantz & Lewis 1997; Garey et al., 1998; Glantz & Lewis, 2000); compared with the hippocampal formation, the evidence is stronger for a preferential involvement of inhibitory (rather than excitatory) neuronal and synaptic populations (Lewis, 2000; Lewis & Gonzalez-Burgos, 2000; Benes & Berretta, 2001). As in the hippocampus, neuronal density findings in the frontal cortex are equivocal, with increases described in dorsolateral regions (Selemon et al., 1995, 1998), attributed to a loss of neuropil (Selemon & Goldman-Rakic, 1999), but decreased neuronal density in the anterior cingulate cortex (Benes et al., 1991b, 2001). Cortical neuron number appears unaltered (Pakkenberg, 1993). An altered location and number of white matter neurons have been reported in the frontal and temporal cortex (Akbarian et al., 1993, 1996); as with the entorhinal cortex dysplasias these are putatively a remnant of aberrant early neurodevelopmental disturbance (Bloom, 1993), but, equally, require replication before they can be considered robust. Mediodorsal thalamus The mediodorsal nucleus has been found to be smaller and contain fewer neurons in three stereologically based studies (Pakkenberg, 1990; Young et al., 2000; Popken et al., 2000), but not in a fourth study (Cullen et al., 2000). Given that the major anatomical connections of the nucleus are with the prefrontal cortex, the abnormalities have been related to the pathological and other evidence for prefrontal involvement in the disorder (Harrison & Lewis, 2003). Empirical evidence for such a relationship is beginning to emerge (Lewis et al., 2001).

Schizophrenia and dementia

Other regions There are isolated reports for many other brain regions, but for none does sufficient data exist to allow meaningful conclusions to be drawn (for review see Harrison, 1999a). However, a few general comments can be made. Firstly, positive findings are not limited to the above areas, but neither are they seen uniformly across the cortex, or thalamus. For example, pyramidal neuron size is not affected in motor or visual cortex (Arnold et al., 1995b; Rajkowska et al., 1998), but neuronal density may be increased in the latter (Selemon et al., 1995). Secondly, there is emerging evidence that the cerebellum, a hitherto neglected region, may show unexpected pathological as well as functional involvement in the disorder (Andreasen, 1999; Eastwood et al., 2001). Thirdly, there is no consistent evidence for lateralization of the pathology as predicted by some models and macroscopic data (Crow et al., 1989; Crow, 1990), but neither has the issue been well investigated (Holinger et al., 2000).

and dendritic abnormalities, the findings – inconsistencies notwithstanding – suggest that the neuropathology of schizophrenia is likely to be an alteration of the neural circuitry, which in turn contributes to the aberrant functional connectivity which underlies the disorder (Friston, 1998; Harrison, 1999a). At least with the number of studies currently under way, and the increasing technical sophistication afforded by molecular approaches, there is reason to be optimistic that steady progress will be made over the next decade. As part of that process, other crucial issues can be addressed, including identification of the major cross-sectional and longitudinal clinicopathological correlations. For example, what are the similarities and differences from the neuropathology of mood disorder and its dementia, which are also beginning to be investigated (O’Brien et al., 2001; Harrison, 2002)? The specific question as to the neuropathological explanation for the dementia of schizophrenia should also be answerable during this time.

Conclusions In conclusion, no cardinal, robust neuropathological features of schizophrenia have yet been described. Conversely, it is also clear that cytoarchitectural changes, particularly affecting the morphometry of neurons in cortical and hippocampal areas, are the likely histological correlate of the macroscopic brain changes, the existence of which is unequivocal. The current status of the evidence is summarized in Table 22.3. Together with the synaptic Table 22.3. Key positive and negative neuropathological findings in schizophrenia Strong (multiple replications, or meta-analysis) evidence Brain weight decreased No excess of Alzheimer’s disease or other neurodegenerative disorders No gliosis Reasonable (replicated but not incontrovertible) evidence Decreased neuronal size (in several regions) Fewer neurons in mediodorsal thalamus Decreased synaptic markers in hippocampus and prefrontal cortex Fewer dendritic spines on cortical pyramidal neurons Weak (unconfirmed, or inconsistent) evidence Entorhinal cortex heterotopia or dysplasia Hippocampal pyramidal neuron disarray Abnormal dendritic trees on subicular pyramidal neurons Altered cortical neuronal density or number Alterations greater in left than right hemisphere Source: For references, see text and Harrison (1999a).

Acknowledgements Work in the author’s laboratory is supported by the Stanley Medical Research Institute and Wellcome Trust and Medical Research Council. Margaret Cousin provided secretarial assistance.

REFERENCES Akbarian, S., Bunney, W. E., Potkin, S. G. et al. (1993). Altered distribution of nicotinamide-adenine dinucleotide phosphatediaphorase cells in frontal lobe of schizophrenics implies disturbances of cortical development. Archives of General Psychiatry, 50, 169–77. Akbarian, S., Kim, J. J., Potkin, S., Hetrick, W. P., Bunney, W. E. & Jones, E. G. (1996). Maldistribution of interstitial neurons in prefrontal white matter of the brains of schizophrenic patients. Archives of General Psychiatry, 53, 425–36. Akil, M. & Lewis, D. A. (1997). Cytoarchitecture of the entorhinal cortex in schizophrenia. American Journal of Psychiatry, 154, 1010–12. Akil, M. & Weinberger, D. R. (2000). Neuropathology and the neurodevelopmental model. In: Harrison, P. J. & Roberts, G. W., eds., The Neuropathology of Schizophrenia. Progress and Interpretation, pp. 189–212. Oxford University Press, Oxford. Aleman, A., Hijman, R., De Haan, E. H. F. & Kahn, R. S. (1999). Memory impairment in schizophrenia: A meta-analysis. American Journal of Psychiatry, 156, 1358–66. Andreasen, N. C. (1999). A unitary model of schizophrenia – Bleuler’s ‘fragmented phrene’ as schizencephaly. Archives of General Psychiatry, 56, 781–7.

503

504

P. J. Harrison

Arnold, S. E. & Trojanowski, J. Q. (1996a). Cognitive impairment in elderly schizophrenia: a dementia (still) lacking distinctive histopathology. Schizophrenia Bulletin, 22, 5–9. (1996b). Recent advances in defining the neuropathology of schizophrenia. Acta Neuropathologica, 92, 217–31. Arnold, S. E., Hyman, B. T., Van Hoesen, G. W. & Damasio, A. R. (1991). Some cytoarchitectural abnormalities of the entorhinal cortex in schizophrenia. Archives of General Psychiatry, 48, 625–32. Arnold, S. E., Gur, R. E., Shapiro, R. M. et al. (1995a). Prospective clinicopathologic studies of schizophrenia: Accrual and assessment of patients. American Journal of Psychiatry, 152, 731–7. Arnold, S. E., Franz, B. R., Gur, R. C. et al. (1995b). Smaller neuron size in schizophrenia in hippocampal subfields that mediate cortical-hippocampal interactions. American Journal of Psychiatry, 152, 738–48. Arnold, S. E., Franz, B. R., Trojanowski, J. Q., Moberg, P. J. & Gur, R. E. (1996). Glial fibrillary acidic protein-immunoreactive astrocytosis in elderly patients with schizophrenia and dementia. Acta Neuropathologica, 91, 269–77. Arnold, S. E., Ruscheinsky, D. D. & Han, L. Y. (1997a). Further evidence of abnormal cytoarchitecture of the entorhinal cortex in schizophrenia using spatial point pattern analyses. Biological Psychiatry, 42, 639–47. Arnold, S. E., Joo, E., Martinoli, M. G., Roy, N. et al. (1997b). Apolipoprotein E genotype in schizophrenia: frequency, age of onset, and neuropathologic features. Neuroreport, 8, 1523–6. Arnold, S. E., Trojanowski, J. Q., Gur, R. E., Blackwell, P., Han, L. Y. & Choi, C. (1998). Absence of neurodegeneration and neural injury in the cerebral cortex in a sample of elderly patients with schizophrenia. Archives of General Psychiatry, 55, 225–32. Arnold, S. E., Trojanowski, J. Q. & Parchi, P. (1999). Protease resistant prion proteins are not present in sporadic ‘poor outcome’ schizophrenia. Journal of Neurology, Neurosurgery and Psychiatry, 66, 90–2. Baldessarini, R. J., Hegarty, J. D., Bird, E. D. & Benes, F. M. (1997). Meta-analysis of postmortem studies of Alzheimer’s diseaselike neuropathology in schizophrenia. American Journal of Psychiatry, 154, 861–3. Barch, D. M., Carter, C. S., Braver, T. S. et al. (2001). Selective deficits in prefrontal cortex function in medication-naive patients with schizophrenia. Archives of General Psychiatry, 58, 280–8. Benes, F. M. & Berretta, S. (2001). GABAergic interneurons: implications for understanding schizophrenia and bipolar disorder. Neuropsychopharmacology, 25, 1–27. Benes, F. M., Davidson, J. & Bird, E. D. (1986). Quantitative cytoarchitectural studies of the cerebral cortex of schizophrenics. Archives of General Psychiatry, 43, 31–5. Benes, F. M., Sorensen, I. & Bird, E. D. (1991a). Reduced neuronal size in posterior hippocampus of schizophrenic patients. Schizophrenia Bulletin, 17, 597–608. Benes, F. M., McSparren, J., Bird, E. D., SanGiovanni, J. P. & Vincent, S. L. (1991b). Deficits in small interneurons in prefrontal and

cingulate cortices of schizophrenic and schizoaffective patients. Archives of General Psychiatry, 48, 996–1001. Benes, F. M., Kwok, E. W., Vincent, S. L. & Todtenkopf, M. S. (1998). Reduction of nonpyramidal cells in sector CA2 of schizophrenics and manic depressives. Biological Psychiatry, 44, 88–97. Benes, F. M., Vincent, S. L. & Todtenkopf, M. (2001). The density of pyramidal and nonpyramidal neurons in anterior cingulate cortex of schizophrenic and bipolar subjects. Biological Psychiatry, 50, 395–406. Bernstein, H. -G., Krell, D., Baumann, B. et al. (1998). Morphometric studies of the entorhinal cortex in neuropsychiatric patients and controls: clusters of heterotopically displaced lamina II neurons are not indicative of schizophrenia. Schizophenia Research, 33, 125–32. Bilder, R. M., Goldman, R. S., Robinson, D. et al. (2000). Neuropsychology of first-episode schizophrenia: Initial characterization and clinical correlates. American Journal of Psychiatry, 157, 549–59. Blanchard, J. J. & Neale, J. M. (1994). The neuropsychological signature of schizophrenia: generalized or differential deficit? American Journal of Psychiatry, 151, 40–8. Bloom, F. E. (1993). Advancing a neurodevelopmental origin of schizophrenia. Archives of General Psychiatry, 50, 224–7. Bogerts, B., Häntsch, J. & Herzer, M. (1983). A morphometric study of the dopamine-containing cell groups in the mesencephalon of normals, Parkinson patients, and schizophrenics. Biological Psychiatry, 18, 951–69. Bruton, C. J., Crow, T. J., Frith, C. D., Johnstone, E. C., Owens, D. G. C. & Roberts, G. W. (1990). Schizophrenia and the brain: A prospective cliniconeuropathological study. Psychological Medicine, 20, 285–304. Buhl, L. & Bojsen-Møller, M. (1988). Frequency of Alzheimer’s disease in a postmortem study of psychiatric patients. Danish Medical Bulletin, 35, 288–90. Casanova, M. F., Stevens, J. R. & Kleinman, J. E. (1990). Astrocytosis in the molecular layer of the dentate gyrus: a study in Alzheimer’s disease and schizophrenia. Psychiatry Research, 35, 149–66. Christison, G. W., Casanova, M. F., Weinberger, D. R., Rawlings, R. & Kleinman, J. E. (1989). A quantitative investigation of hippocampal pyramidal cell size, shape, and variability of orientation in schizophrenia. Archives of General Psychiatry, 46, 1027–32. Conrad, A. J., Abebe, T., Austin, R., Forsythe, S. & Scheibel, A. B. (1991). Hippocampal pyramidal cell disarray in schizophrenia as a bilateral phenomenon. Archives of General Psychiatry, 48, 413–17. Corsellis, J. A. N. (1962). Mental Illness and the Ageing Brain., pp. 1–76. Oxford University Press, London. Cotter, D., MacKay, D., Landau, S., Kerwin, R. & Everall, I. (2001). Reduced glial cell density and neuronal size in the anterior cingulate cortex in major depressive disorder. Archives of General Psychiatry, 58, 545–53. Crow, T. J. (1990). Temporal lobe asymmetries as the key to the etiology of schizophrenia. Schizophrenia Bulletin, 16, 433–43.


Crow, T. J., Ball, J., Bloom, S. R. et al. (1989). Schizophrenia as an anomaly of development of cerebral asymmetry. Archives of General Psychiatry, 46, 1145–50. Cullen, T. J., Walker, M. A., Roberts, H., Crow, T. J., Harrison, P. & Esiri, M. (2000). The mediodorsal nucleus of the thalamus in schizophrenia: a post-mortem study [abstract]. Schizophrenia Research, 41, 5. Daniel, D. G., Goldberg, T. E., Gibbons, R. D. & Weinberger, D. R. (1991). Lack of a bimodal distribution of ventricular size in schizophrenia: a Gaussian mixture analysis of 1056 cases and controls. Biological Psychiatry, 30, 887–903. David, A. S. & Cutting, J. C. (1994). The Neuropsychology of Schizophrenia. Lawrence Erlbaum Associates, Hove, pp. 406. Davidson, M., Harvey, P. D., Powchik, P. et al. (1995). Severity of symptoms in chronically institutionalized geriatric schizophrenic patients. American Journal of Psychiatry, 152, 197–207. De Vries, P. J., Honer, W. G., Kemp, P. M. & McKenna, P. J. (2001). Dementia as a complication of schizophrenia. Journal of Neurology, Neurosurgery and Psychiatry, 70, 588–96. DeLisi, L. E. (1997). Is schizophrenia a lifetime disorder of brain plasticity, growth and aging? Schizophr. Res., 23, 119–29. Dwork, A. J., Susser, E. S., Keilp, J. et al. (1998). Senile degeneration and cognitive impairment in chronic schizophrenia. American Journal of Psychiatry, 155, 1536–43. Eastwood, S. L., Cotter, D. & Harrison, P. J. (2001). Cerebellar synaptic protein expression in schizophrenia. Neuroscience, 104, 219–29. El-Mallakh, R. S., Kirch, D. G., Shelton, R. et al. (1991). The nucleus basalis of Meynert, senile plaques and intellectual impairment in schizophrenia. Journal of Neuropsychiatric and Clinical Neuroscience, 3, 383–6. Falkai, P. & Bogerts, B. (1986). Cell loss in the hippocampus of schizophrenics. European Archives of Psychiatry and Neurological Sciences, 236, 154–61. Falkai, P., Bogerts, B. & Rozumek, M. (1988) Limbic pathology in schizophrenia: the entorhinal region – a morphometric study. Biological Psychiatry, 24, 515–21. Falkai, P., Honer, W. G., David, S., Bogerts, B., Majtenyi, C. & Bayer, T. A. (1999) No evidence for astrogliosis in brains of schizophrenic patients. A post-mortem study. Neuropathology and Applied Neurobiology, 25, 48–53. Falkai, P., Schneider-Axmann, T. & Honer, W. G. (2000). Entorhinal cortex pre-alpha cell clusters in schizophrenia: Quantitative evidence of a developmental abnormality. Biological Psychiatry, 47, 937–943. Friston, K. (1998). The disconnection hypothesis. Schizophrenia Research, 30, 115–25. Fucetola, R., Seidman, L. J., Kremen, W. S., Faraone, S. V., Goldstein, J. M. & Tsuang, M. T. (2000). Age and neuropsychologic function in schizophrenia: A decline in executive abilities beyond that observed in healthy volunteers. Biological Psychiatry, 48, 137–46. Garey, L. J., Ong, W. Y., Patel, T. S. et al. (1998). Reduced dendritic spine density on cerebral cortical pyramidal neurons in

schizophrenia. Journal of Neurology, Neurosurgery and Psychiatry, 65, 446–53. Glantz, L. A. & Lewis, D. A. (1997). Reduction of synaptophysin immunoreactivity in the prefrontal cortex of subjects with schizophrenia – regional and diagnostic specificity. Archives of General Psychiatry, 54, 943–52. (2000). Decreased dendritic spine density on prefrontal cortical pyramidal neurons in schizophrenia. Archives of General Psychiatry, 57, 65–73. Gold, S., Arndt, S., Nopoulos, P., O’Leary, D. S. & Andreasen, N. C. (1999). Longitudinal study of cognitive function in firstepisode and recent-onset schizophrenia. American Journal of Psychiatry, 156, 1342–8. Goldberg, T. E., Hyde, T. M., Kleinman, J. E. & Weinberger, D. R. (1993). Course of schizophrenia: Neuropsychological evidence for a static encephalopathy. Schizophrenia Bulletin, 19, 797– 804. Goldman-Rakic, P. S. & Selemon, L. D. (1997). Functional and anatomical aspects of prefrontal pathology in schizophrenia. Schizophrenia Bulletin, 23, 437–58. Green, M. F. (1996). What are the functional consequences of neurocognitive deficits in schizophrenia. American Journal of Psychiatry, 153, 321–30. Haroutunian, V., Davidson, M., Kanof, P. D. et al. (1994). Cortical cholinergic markers in schizophrenia. Schizophrenia Research, 12, 137–44. Harrison, P. J. (1995). On the neuropathology of schizophrenia and its dementia: Neurodevelopmental, neurodegenerative, or both. Neurodegeneration, 4, 1–12. (1997). Schizophrenia: a disorder of neurodevelopment? Commentary. Current Opinion in Neurobiology, 7, 285–9. (1999a). The neuropathology of schizophrenia – a critical review of the data and their interpretation. Brain, 122, 593–624. (1999b). The neuropathological effects of antipsychotic drugs. Schizophrenia Research, 40, 87–99. (2002). The neuropathology of primary mood disorder. Brain, 125, 1438–49. Harrison, P. J. & Eastwood, S. L. (1998). Preferential involvement of excitatory neurons in medial temporal lobe in schizophrenia. Lancet, 352, 1669–73. (2001). Neuropathological studies of synaptic connectivity in the hippocampal formation in schizophrenia. Hippocampus, 11, 508–19. Harrison, P. J. & Lewis, D. A. (2003). The neuropathology of schizophrenia. In: Hirsch, S. R., Weinberger, D. R. (Eds.) Schizophrenia. 2nd edn, pp. 306–17. Oxford: Blackwells Science. Harrison, P. J. & Roberts, G. W. (2000). The Neuropathology of Schizophrenia. Progress and Interpretation. Oxford University Press, Oxford, pp. 374. Harrison, P. J., Freeman, N. & Geddes, J. R. (2003). Meta-analysis of brain weight in schizophrenia. Schizophrenia Research 64: 25–34. Harvey, P. D. & Keefe, R. S. E. (2001). Studies of cognitive change in patients with schizophrenia following novel

505

506

P. J. Harrison

antipsychotic treatment. American Journal of Psychiatry, 158, 176–84. Harvey, P. D., Silverman, J. M., Mohs, R. C. et al. (1999). Cognitive decline in late-life schizophrenia: a longitudinal study of geriatric chronically hospitalized patients. Biological Psychiatry, 45, 32–40. Heckers, S., Heinsen, H., Geiger, B. & Beckmann, H. (1991). Hippocampal neuron number in schizophrenia: A stereological study. Archives of General Psychiatry, 48, 1002–8. Heckers, S., Rauch, S. L., Goff, D. C. et al. (1998). Impaired recruitment of the hippocampus during conscious recollection in schizophrenia. Nature Neuroscience, 1, 318–23. Hoff, A. L., Sakuma, M., Wieneke, M., Horon, R., Kushner, M. & DeLisi, L. E. (1999). Longitudinal neuropsychological followup study of patients with first-episode schizophrenia. American Journal of Psychiatry, 156, 1336–41. Holinger, D. P., Galaburda, A. M. & Harrison, P. J. (2000). Cerebral asymmetry. In: Harrison, P. J. & Roberts, G. W., eds., The Neuropathology of Schizophrenia. Progress and Interpretation, pp. 151–72. Oxford University Press, Oxford. Jakob, H. & Beckmann, H. (1986). Prenatal developmental disturbances in the limbic allocortex in schizophrenics. Journal of Neural Transmission, 65, 303–26. Jellinger, K. A. (1977). Neuropathologic findings after neuroleptic longterm therapy. In: Roizin, L., Shiraki, H. & Grçevic, M. (Eds.), Neurotoxicology, pp. 25–42. Raven Press, New York. Jellinger, K. A. & Gabriel, E. (1999). No increased incidence of Alzheimer’s disease in elderly schizophrenics. Acta Neuropathologica, 97, 165–9. Jeste, D. V. & Lohr, J. B. (1989). Hippocampal pathologic findings in schizophrenia. A morphometric study. Archives of General Psychiatry, 46, 1019–24. Johnston-Wilson, N. L., Sims, C. D., Hofmann, J.-P. et al. (2000). Disease-specific alterations in frontal cortex brain proteins in schizophrenia, bipolar disorder, and major depressive disorder. Mol Psychiatry, 5, 142–9. Johnstone, E. C., Crow, T. J., Frith, C. D., Stevens, M., Kreel, L. & Husband, J. (1978). The dementia of dementia praecox. Acta Psychiatrica Scandinavica, 57, 305–24. Jonsson, S. A. T., Luts, A., Guldberg-Kjaer, N. & Brun, A. (1997). Hippocampal pyramidal cell disarray correlates negatively to cell number: implications for the pathogenesis of schizophrenia. European Archives of Psychiatry and Neurological Sciences, 247, 120–7. Karson, C. N., Casanova, M. F., Kleinman, J. E. & Griffin, W. S. T. (1993). Choline acetyltransferase in schizophrenia. American Journal of Psychiatry, 150, 454–9. Karson, C. N., Mrak, R. E., Schluterman, K. O., Sturner, W. Q., Sheng, J. G. & Griffin, W. S. T. (1999). Alterations in synaptic proteins and their encoding mRNAs in prefrontal cortex in schizophrenia: a possible neurochemical basis for ‘hypofrontality’. Molecular Psychiatry, 4, 39–45. Konick, L. C. & Friedman, L. (2001). Meta-analysis of thalamic size in schizophrenia. Biological Psychiatry, 49, 28–38. Kovelman, J. A. & Scheibel, A. B. (1984). A neurohistological correlate of schizophrenia. Biological Psychiatry, 19, 1601–21.

Kremen, W. S., Seidman, L. J., Faraone, S. V. & Tsuang, M. T. (2001). Intelligence quotient and neuropsychological profiles in patients with schizophrenia and in normal volunteers. Biological Psychiatry, 50, 453–62. Krimer, L. S., Herman, M. M., Saunders, R. C. et al. (1997). A qualitative and quantitative analysis of the entorhinal cortex in schizophrenia. Cerebral Cortex, 7, 732–9. Kuperberg, G. & Heckers, S. (2000). Schizophrenia and cognitive function. Current Opinion in Neurobiology, 10, 205–10. Lawrie, S. M. & Abukmeil, S. S. (1998). Brain abnormality in schizophrenia – a systematic and quantitative review of volumetric magnetic resonance imaging studies. British Journal of Psychiatry, 172, 110–20. Lawrie, S. M., Whalley, H., Kestelman, J. N. et al. (1999). Magnetic resonance imaging of brain in people at high risk of developing schizophrenia. Lancet, 353, 30–3. Lewis, D. A. (2000). GABAergic local circuit neurons and prefrontal cortical dysfunction in schizophrenia. Brain Research Reviews, 31, 270–6. Lewis, D. A. & Gonzalez-Burgos, G. (2000). Intrinsic excitatory connections in the prefrontal cortex and the pathophysiology of schizophrenia. Brain Research Bulletin, 52, 309–17. Lewis, D. A., Cruz, D. A., Melchitzky, D. S. & Pierri, J. N. (2001). Lamina-specific deficits in parvalbumin-immunoreactive varicosities in the prefrontal cortex of subjects with schizophrenia: evidence for fewer projections from the thalamus. American Journal of Psychiatry, 158, 1411–22. Lieberman, J. A. (1999). Is schizophrenia a neurodegenerative disorder? A clinical and neurobiological perspective. Biological Psychiatry, 46, 729–39. McKenna, P. J. (1994). Schizophrenia and Related Syndromes. Oxford University Press, Oxford. Mohamed, S., Paulsen, J. S., O’Leary, D., Arndt, S. & Andreasen, N. (1999). Generalized cognitive deficits in schizophrenia – a study of first-episode patients. Archives of General Psychiatry, 56, 749–54. Murphy, G. M. Jr., Lim, K., Wieneke, M., Ellis, W., Forno, L., Hoff, A. L. & Nordahl, T. (1998). No neuropathologic evidence for an increased frequency of Alzheimer’s disease among elderly schizophrenics. Biological Psychiatry, 43, 205–9. Nelson, M. D., Saykin, A. J., Flashman, L. A. & Riordan, H. J. (1998). Hippocampal volume reduction in schizophrenia as assessed by magnetic resonance imaging – a meta-analytic study. Archives of General Psychiatry, 55, 433–40. Niizato, K., Arai, T., Kuroki, N., Kase, K., Iritani, S. & Ikeda, K. (1998). Autopsy study of Alzheimer’s disease brain pathology in schizophrenia. Schizophrenia Research, 31, 177–84. O’Brien, J., Thomas, A., Ballard, C. et al. (2001). Cognitive impairment in depression is not associated with neuropathologic evidence of increased vascular or Alzheimer-type pathology. Biological Psychiatry, 49, 130–6. Olney, J. W. & Farber, N. B. (1995). Glutamate receptor dysfunction and schizophrenia. Archives of General Psychiatry, 52, 998– 1007. ¨ ur, ¨ D., Drevets, W. C. & Price, J. L. (1998). Glial reducOng tion in the subgenual prefrontal cortex in mood disorders.


Proceedings of the National Academy of Sciences of the United States of America, 95, 13290–5. Owen, F., Crow, T. J., Frith, C. D. et al. (1987). Selective decreases in MAO-B activity in post-mortem brains from schizophrenic patients with type II syndrome. British Journal of Psychiatry, 151, 744–52. Pakkenberg, B. (1987). Post-mortem study of chronic schizophrenic brains. British Journal of Psychiatry, 151, 744–52. (1990). Pronounced reduction of total neuron number in mediodorsal thalamic nucleus and nucleus accumbens in schizophrenics. Archives of General Psychiatry, 47, 1023–8. (1993). Total nerve cell number in neocortex in chronic schizophrenics and controls estimated using optical disectors. Biological Psychiatry, 34, 768–72. Perrone-Bizzozero, N. I., Sower, A. C., Bird, E. D., Benowitz, L. I., Ivins, K. J. & Neve, R. L. (1996). Levels of the growth-associated protein GAP-43 are selectively increased in association cortices in schizophrenia. Proceedings of National Academy of Sciences of the United States of America, 93, 14182–7. Pierri, J. N., Volk, C. L. E., Auh, S., Sampson, A. & Lewis, D. A. (2001). Decreased somal size of deep layer 3 pyramidal neurons in the prefrontal cortex of subjects with schizophrenia. Archives of General Psychiatry, 58, 466–73. Plum, F. (1972). Prospects for research on schizophrenia. Neuropathological findings. Neuroscience Research Program Bulletin, 10, 384–8. Popken, G. J., Bunney, W. E., Jr., Potkin, S. G. & Jones, E. G. (2000). Subnucleus-specific loss of neurons in medial thalamus of schizophrenics. Proceedings of the National Academy of Sciences of the United States of America, 97, 9276–80. Powchik, P., Davidson, M., Nemeroff, C. B. et al. (1993). Alzheimer’sdisease-related protein in geriatric schizophrenic patients with cognitive impairment. American Journal of Psychiatry, 150, 1726–27. Prohovnik, I., Dwork, A. J., Kaufman, M. A. & Willson, N. (1993). Alzheimer-type neuropathology in elderly schizophrenia patients. Schizophrenia Bulletin, 19, 805–16. Purohit, D. P., Davidson, M., Perl, D. P. et al. (1993). Severe cognitive impairment in elderly schizophrenic patients: a clinicopathological study. Biological Psychiatry, 33, 255–60. Purohit, D. P., Perl, D. P., Haroutunian, V., Powchik, P., Davidson, M. & Davis, K. L. (1998). Alzheimer disease and related neurodegenerative diseases in elderly patients with schizophrenia – a postmortem neuropathologic study of 100 cases. Archives of General Psychiatry, 55, 205–11. Radewicz, K., Garey, L. J., Gentleman, S. M. & Reynolds, R. (2000). Increase in HLA-DR immunoreactive microglia in frontal and temporal cortex of chronic schizophrenics. Journal of Neuropathology and Experimental Neurology, 59, 137–50. Ragland, J. D., Gur, R. E., Klimas, B. C., McGrady, N. & Gur, R. C. (1999). Neuropsychological laterality indices of schizophrenia: Interactions with gender. Schizophrenia Bulletin, 25, 79–89. Rajkowska, G., Selemon, L. D. & Goldman-Rakic, P. S. (1998). Neuronal and glial somal size in the prefrontal cortex – a postmortem morphometric study of schizophrenia and Huntington disease. Archives of General Psychiatry, 55, 215–24.

Roberts, G. W. (1991). Schizophrenia: a neuropathological perspective. British Journal of Psychiatry, 158, 8–17. Roberts, G. W. & Harrison, P. J. (2000). Gliosis and its implications for the disease process. In: Harrison, P. J. & Roberts, G. W., eds., The Neuropathology of Schizophrenia. Progress and Interpretation, pp. 137–50. Oxford University Press, Oxford. Roberts, G. W., Colter, N., Lofthouse, R., Bogerts, B., Zech, M. & Crow, T. J. (1986). Gliosis in schizophrenia: a survey. Biological Psychiatry, 21, 1043–50. Roberts, G. W., Colter, N., Lofthoux, R., Johnstone, E. C. & Crow, T. J. (1987). Is there gliosis in schizophrenia? Investigation of the temporal lobe. Biological Psychiatry, 22, 1459–68. Saykin, A. J., Shtasel, D. L., Gur, R. E. et al. (1994). Neuropsychological deficits in neuroleptic naive patients with firstepisode schizophrenia. Archives of General Psychiatry, 51, 124– 31. Selemon, L. D. & Goldman-Rakic, P. S. (1999). The reduced neuropil hypothesis: a circuit based model of schizophrenia. Biological Psychiatry, 45, 17–25. Selemon, L. D., Rajkowska, G. & Goldman-Rakic, P. S. (1995). Abnormally high neuronal density in the schizophrenic cortex – a morphometric analysis of prefrontal area 9 and occipital area 17. Archives of General Psychiatry, 52, 805–18. Selemon, L. D., Rajkowska, G. & Goldman-Rakic, P. S. (1998). Elevated neuronal density in prefrontal area 46 in brains from schizophrenic patients: application of a three-dimensional, stereologic counting method. Journal of Comparative Neurology, 392, 402–12. Selemon, L. D., Lidow, M. S. & Goldman-Rakic, P. S. (1999). Increased volume and glial density in primate prefrontal cortex associated with chronic antipsychotic drug exposure. Biological Psychiatry, 46, 161–72. Sharma, T. & Harvey, P. (2000). Cognition in Schizophrenia. Oxford University Press, Oxford, pp. 363. Soustek, K. (1989) Ultrastructure of cortical synapses in the brain of schizophrenics. Zentralbatt fur Allgemeine Pathologie und Pathologische Anatomie, 135, 25–32. Stevens, C. D., Altshuler, L. L., Bogerts, B. & Falkai, P. (1988). Quantitative study of gliosis in schizophrenia and Huntington’s chorea. Biological Psychiatry, 24, 697–700. Stevens, J. R. (1982). Neuropathology of schizophrenia. Archives of General Psychiatry, 39, 1131–9. (1997). Anatomy of schizophrenia revisited. Schizophrenia Bulletin, 23, 373–83. Sweet, R. A., Hamilton, R. L., Lopez, O. L. et al. (2000). Psychotic symptoms in Alzheimer’s disease are not associated with more severe neuropathologic features. International Psychogeriatrics, 12, 547–58. Velakoulis, D., Pantelis, C., McGorry, P. D. et al. (1999). Hippocampal volume in first-episode psychoses and chronic schizophrenia. Archives of General Psychiatry, 56, 133–41. Weickert, T. W., Goldberg, T. E., Gold, J. M., Bigelow, L. B., Egan, M. F. & Weinberger, D. R. (2000). Cognitive impairments in patients with schizophrenia displaying preserved and compromised intellect. Archives of General Psychiatry, 57, 907– 13.

507

508

P. J. Harrison

Weinberger, D. R. (1987). Implications of normal brain development for the pathogenesis of schizophrenia. Archives of General Psychiatry, 44, 660–9. Weinberger, D. R. (1995). From neuropathology to neurodevelopment. Lancet, 346, 552–7. Wisniewski, H. M., Constantinidis, J., Wegiel, J., Bobinski, M. & Tarnawski, M. (1994). Neurofibrillary pathology in brains of elderly schizophrenics treated with neuroleptics. Alzheimer’s Disease and Associated Disorders, 8, 211–27. Woods, B. T. (1998). Is schizophrenia a progressive neurodevelopmental disorder? Toward a unitary pathogenetic mechanism. American Journal of Psychiatry, 155, 1661–70. Wragg, R. E. & Jeste, D. V. (1989). Overview of depression and psychosis in Alzheimer’s disease. American Journal of Psychiatry, 146, 577–87.

Wright, I. C., Rabe-Hesketh, S., Woodruff, P. W. R., David, A. S., Murray, R. M. & Bullmore, E. T. (2000). Meta-analysis of regional brain volumes in schizophrenia. American Journal of Psychiatry, 157, 16–25. Young, K. A., Manaye, K. F., Liang, C. L., Hicks, P. B. & German, D. C. (2000). Reduced number of mediodorsal and anterior thalamic neurons in schizophrenia. Biological Psychiatry, 47, 944–53. Zaidel, D. W., Esiri, M. M. & Harrison, P. J. (1997a). The hippocampus in schizophrenia: Lateralized increase in neuronal density and altered cytoarchitectural asymmetry. Psychological Medicine, 27, 703–13. (1997b). Size, shape, and orientation of neurons in the left and right hippocampus: Investigation of normal asymmetries and alterations in schizophrenia. American Journal of Psychiatry, 154, 812–18.

23 Other diseases that cause dementia Margaret M. Esiri Department of Neuropathology, The Radcliffe Infirmary Oxford, UK

In this chapter we consider a number of diverse pathological conditions that may occasionally give rise to a clinical picture in which dementia predominates (Table 23.1). Other neurological features are also likely to be present at some stage in the course of these diseases. In some, perhaps most, such cases these other clinical features may have led to the correct diagnosis being established before clinical dementia develops, but in a few the diagnosis may be impossible to achieve before death. Therefore, a pathologist can expect to encounter these diseases occasionally among cases of dementia presenting at autopsy. Full treatment of pathological aspects of these diseases and their illustration is beyond the scope of this chapter and for details a more general textbook of neuropathology should be consulted, e.g. Greenfield’s Neuropathology (Graham & Lantos, 2002) or Davis and Robertson’s Textbook of Neuropathology (1996).

Neurodegenerative conditions Neuronal intranuclear hyaline inclusion disease (NIHID) Neuronal intranuclear hyaline inclusion disease (NIHID) is a rare neurodegenerative disorder characterized by the presence of large eosinophilic inclusions in neuronal nuclei in a wide distribution (Sung et al., 1980; Funata et al., 1990). Whilst most cases present clinically with predominantly motor or other non-cognitive problems, some cases have had dementia (Weidenheim & Dickson, 1995). The neuronal inclusions are present in peripheral ganglionic neurons as well as at all levels of the CNS from cerebral cortex to

spinal cord. In addition to the presence of neuronal inclusions, there may be somewhat smaller inclusions in astrocytic nuclei. Neuronal inclusions consist ultrastructurally of haphazardly arranged 10 nm diameter straight filaments without any surrounding membrane. Neuronal loss with accompanying gliosis is also widespread. If anything, an inverse relation between the density of intranuclear inclusions in a given region and the extent of neuronal loss has been described (Takahashi et al., 2001). The neuronal inclusions are immunoreactive for ubiquitin and, in some cases, for an epitope specific for expanded polyglutamine tracts (Trottier et al., 1995), suggesting that polyglutaminecontaining proteins may be non-specifically recruited into neuronal nuclei in NIHID (Leiberman et al., 1999; Takahashi et al., 2001). Ataxin-3 was found in most neuronal nuclear inclusions and ataxin-2 and a TATA box binding protein in some inclusions in one study (Takahashi et al., 2001). Another study detected ataxin-1 and ataxin-3 (Leiberman et al., 1999). These findings suggest that NIHID may share pathogenetic mechanisms with diseases associated with expanded polyglutamine tracts (see Chapter 16).

Diffuse neurofibrillary tangles with calcification A neurodegenerative condition has been reported principally from Japan presenting as adult early-onset dementia usually diagnosed clinically as Alzheimer’s disease but in some cases with mild parkinsonism and, rarely, with pyramidal weakness. Autopsies have shown numerous cortical and brainstem neurofibrillary tangles, frontotemporal lobar atrophy and pronounced calcific deposits in cerebrum, basal ganglia and cerebellum (Kosaka, 1994). One case contained neurofibrillary tangles in the dentate nucleus of the cerebellum and argyrophilic inclusions in


509

510

M. M. Esiri

Table 23.1. Diseases considered in this chapter that may rarely cause dementia as a predominating clinical feature Neurodegenerative conditions Neuronal intranuclear inclusion disease (NINID) Neurofibrillary tangles with calcification Thalamic degeneration Familial encephalopathy with neuroserpin inclusion bodies (FENIB) Space-occupying lesions Subdural haematoma Intracranial neoplasms Gliomas Meningiomas Inflammatory conditions Sarcoidosis Behçet’s syndrome Paraneoplastic syndrome (limbic encephalitis) Glial nodule cerebral irradiation Other Consequences of cerebral irradiation Multiple sclerosis Epilepsy Superficial haemosiderosis of the central nervous system Whipple’s disease Metabolic and toxic diseases Other obscure conditions Coeliac disease with dementia Multifocal leukoencephalopathy with calcification Multiple cavernous angiomas

oligodendrocytes. The calcareous deposits contain traces of lead in addition to calcium, phosphorus and iron (Haraguchi et al., 2001). An absence of argyrophillic plaques as well as the calcification distinguishes the condition from Alzheimer’s disease, while the predominantly cortical localization of the tangles, calcification and a clinical picture dominated by dementia serve to distinguish it from progressive supranuclear palsy.

cases are described by Janssen et al. (2000) and GoeckeHoyer et al. (1990). A high proportion of cases are familial. The most severely affected nucleus in most cases has been the dorsomedial thalamic nucleus (Fig. 2.14). Lesser degrees of cell loss and gliosis have been found in the ventral anterior, centro median, lateral and posterior nuclei. Other subcortical sites outside the thalamus are almost invariably affected to some extent and these have included the striatum, globus pallidus, red nucleus, cerebellum and basal nucleus of Meynert. One case had pathological features of motor neuron disease with dementia as well as severe degeneration of the dorsomedial thalamic nuclei (Deymeer et al., 1989). Some cases have shown laminar spongy change, and neuron loss in the cerebral cortex and gliosis in the cerebral white matter. No inclusions of any sort are present. As with motor neuron disease and dementia, similarities are to be noted between thalamic dementia and non-specific frontal lobe dementia in which many cases do indeed have thalamic, along with other subcortical, pathology.

Familial encephalopathy with neuroserpin inclusion bodies (FENIB) This condition is a recently recognized cause of rare familial encephalopathy in which dementia and myoclonus feature prominently. Onset is usually in adult life. The distinctive pathological finding is of numerous rounded eosinophilic, PAS-positive, intracytoplasmic inclusion bodies occupying neuron cell bodies and processes. These react immunocytochemically for neuroserpin, a serine protease inhibitor of the serpin family, which is expressed mainly in neurons (Berger et al., 1998). Mutation in the gene coding for this protein have been described in affected members of families with this disease (Davis et al., 1999; Takao et al., 2000).

Space-occupying lesions

Thalamic degeneration and dementia

Subdural haematoma

There are patients described with dementia, among other neuro-psychological abnormalities, who have had pathologies of various types (vascular, neoplastic, inflammatory) largely confined to the thalamus (Martin, 1969). Vascular disease of the thalamus associated with dementia is discussed in Chapter 13 and fatal familial insomnia, in which thalamic pathology is prominent, in Chapter 17. More relevant to this chapter are rare cases of neuron loss and gliosis predominantly, though never exclusively, affecting the thalamus. Such cases are reviewed by Martin (1975) and Deymeer et al. (1989) and additional

Subdural haematomas may produce symptoms of dementia that mimic those of Alzheimer’s disease, vascular dementia or normal pressure hydrocephalus (Foster et al., 2000; Perlmutter & Gobles, 1961; Stuteville & Welch, 1958). Dementia may also develop in those operated on for subdural haematomas. Chronic subdural haematomas are particularly liable to present as dementia rather than with symptoms more suggestive of raised intracranial pressure. These may occur apparently spontaneously in the elderly, or after trivial head injury or complicate shunt insertions and other intracranial operations. However, as chronic

Other diseases that cause dementia

subdural haematomas may occur as a consequence of cerebral atrophy, which produces a less good fit of the brain inside the skull, care should be taken in ascribing dementia to a small chronic subdural haematoma. A neurodegenerative condition should first be excluded, particularly if the brain is atrophic. Large, chronic unilateral or bilateral subdural haematomas may, however, be the only significant pathological findings in an elderly demented patient with a normal sized or slightly oedematous brain. Subdural haematomas occur over the vertex and are immediately apparent when the top of the skull is removed and the dura opened. The chronic variety usually forms a part-membranous and part-fluid pad over the arachnoid from which it can be readily lifted away. Microscopically, the membranous component consists of a meshwork of thin walled sinusoidal capillaries from which blood continually seeps to maintain the chronic haematoma. Products of haemoglobin breakdown, together with macrophages and a few fibroblasts, fill the loose interstices between the vessels.

Intracranial neoplasms A number of different types of intracranial neoplasm may produce a primarily dementing clinical picture. Some 5% of cases of dementia may be associated with such lesions that also produce raised intracranial pressure (Cummings & Benson, 1992; Lishman, 1987). Tumours may cause symptoms of dementia mainly by obstructing flow of cerebrospinal fluid and producing hydrocephalus (Chapter 19). This is particularly true of posterior fossa and midline tumours such as craniopharyngiomas or extracellular pituitary adenomas, and carcinomatosis of the leptomeninges, which obstructs cerebrospinal fluid flow in the subarachnoid space. Other tumours produce dementia by interfering with widespread cerebral function, as may occur with gliomatosis cerebri, primary CNS lymphoma and multiple small cerebral metastases each of which may be surrounded by oedema. Periventricular spread of carcinoma, glioma or lymphoma deposits may also present with memory disturbance by destroying the fimbriae or fornices and invading the corpus callosum. This can occur either when the primary site of the tumour is in the fimbriae or fornices or by intra- or peri-ventricular spread. In such cases the affected parts may appear irregularly softened and necrotic but not necessarily grossly expanded. Intravascular (angiotropic) lymphoma is another rare cause of dementia, in this case producing its effect on higher brain functions by producing multiple ischaemic lesions (see Chapter 13). Intrinsic gliomas, particularly those diffusely involving a cerebral hemisphere or the frontal lobes bilaterally, with

extension through the corpus callosum, also occasionally produce dementia with few or no focal neurological signs or obvious headache. Temporal lobe gliomas may similarly present in this way. Low- or middle-grade astrocytomas are usually found to be the histological tumour cell type responsible, though there may be foci of frankly malignant astrocytoma or glioblastoma present by the time the patient dies. Gliomas arising close to the ventricular system and disseminating to other parts of the brain via cerebrospinal fluid pathways may also give rise to dementia. The presence of glioma does not necessarily provide the entire explanation for a dementing illness. We have seen two elderly demented subjects in whom Alzheimer’s disease co-existed with malignant glioma; others are reported in the literature. Extrinsic meningiomas, which predominantly compress rather than invade or infiltrate the brain, may present with dementia. Large, slowly growing, olfactory groove meningiomas in the subfrontal region are particularly notorious in this respect. Sphenoidal wing meningiomas abutting on the temporal lobe may do likewise. However, it is important to realize that careful clinicopathological correlation needs to be considered when an intracranial tumour is discovered in a demented subject. Small, benign meningiomas are unlikely to have caused dementia on their own and are more likely to be incidental findings. An elderly patient with a long history of progressive dementia, and even a large tumour at autopsy, may also have Alzheimer’s disease. We have seen a large meningioma occurring in conjunction with severe pathological changes of Alzheimer’s disease. Similarly, multiple cerebral metastases are unlikely to be the cause of long-standing dementia in an elderly patient and an additional neurodegenerative disease should be sought. On the other hand, in a younger patient with a short history of rapidly progressive dementia, multiple cerebral metastases may provide a convincing pathological explanation for the clinical picture. Likewise, a large, slowly growing frontal meningioma is an adequate explanation for dementia in a patient with clinical signs and symptoms referable predominantly to frontal lobe dysfunction. As with other forms of mixed pathology, the presence of a cerebral tumour on its own may not account for dementia, but may summate with a normally sub-clinical level of Alzheimer-type pathology to produce clinical dementia.

Inflammatory conditions Sarcoidosis This is a chronic granulomatous inflammatory condition of unknown aetiology in which the CNS may be involved on its own or together with systemic organs. Dementia may

511

512

M. M. Esiri

occasionally be a presenting symptom in this condition (Cordingley et al., 1981; Sanson et al., 1996). It can produce mass lesions within the brain, chronic meningitic foci, arteritic damage or hydrocephalus and any combination of these pathological features. These lesions should be sought in the leptomeninges and parenchyma of the cerebrum, cerebellum or brainstem. Vasculitic foci may be present in the same region. Macroscopically, there is usually some thickening and opacity of the leptomeninges and coronal slices may reveal hydrocephalus, possibly with grossly visible granulomas. The microscopic lesions resemble tuberculoid granulomata with multinucleated giant cells but caseation does not occur and no organisms can be demonstrated.

Behçet’s syndrome Clinical features of Behçet’s syndrome are recurrent ulceration of the mouth and genitalia and iridocyclitis. Some cases develop neurological disease in addition, and symptoms may include dementia. Neuropathological lesions of Behçet’s syndrome consist of multiple, focal ill-defined areas of congestion and softening in the brain, particularly the upper brain stem, deep cerebral grey matter and internal capsules. Severe cortical involvement is less common but occurs in cases with dementia (Miyakawa et al., 1976). The nature of these lesions is inflammatory and in some instances vasculitic with micro-infarcts of varying age. Affected tissues contain perivascular and parenchymal inflammatory cell infiltrates consisting predominantly of lymphocytes, and activated microglial cells, some of which form nodules. There is a more diffuse microglial and astrocyte reaction. Neuronal or axonal loss is seen within the main foci of damage. The cause of the condition is unknown.

Paraneoplastic syndrome (limbic encephalitis) Limbic encephalitis is one of a number of remote effects of carcinoma on the nervous system (Currie, 1992). It may present with dementia, personality change and prominent memory disturbance. By far the most common associated neoplasm is small cell lung carcinoma, but many other forms of carcinoma have also been associated with it: breast and gynaecological cancer in women, renal carcinoma and a variety of lymphoreticular neoplasms. Although it is common for the carcinoma to be clinically occult and small in size occasionally, despite a careful search at autopsy, no tumour is discovered. Clinically, the disease usually runs a subacute or chronic course lasting 1 to 2 years, but some cases are fatal within a few months of onset. The changes

in higher cerebral functions include prominent memory impairment, hallucinations, reduced alertness, anxiety, depression and global or patchy cognitive deficits. Commonly, there are other neurological manifestations including epilepsy, cerebellar ataxia, brainstem encephalitis, extrapyramidal movement disorder, myelopathy, radiculopathy or subacute sensory peripheral neuropathy. In rapidly progressive cases, the condition may be easily mistaken for spongiform encephalopathy clinically. Neuropathology The macroscopic appearance of the brain is usually within normal limits but there may be slight atrophy, particularly of the temporal lobes. It is necessary to take blocks for microscopic examination from a wide selection of brain regions: multiple areas of cortex, hippocampus basal ganglia, thalamus, brainstem, cerebellum, spinal cord, peripheral ganglia and nerves. Microscopically, there are perivascular mononuclear inflammatory cell infiltrates, microglial nodules, occasional neuronophagia and reactive astrocytes. These are particularly prominent in the grey and white matter of the hippocampus and medial temporal lobe, though they are not necessarily confined to these regions. The majority of the inflammatory cells are T-cells and macrophages. Deep grey matter of the basal ganglia, thalamus and brainstem are also liable to be affected. Neuron cell loss may also be prominent and out of proportion to the severity of the inflammatory cell reaction. The cerebellum may also show severe pathology either in the form of severe Purkinje cell loss or with inflammatory cell changes in dentate nuclei and white matter, with severe loss of dentate nucleus neurons and Wallerian degeneration in the superior cerebellar peduncles. The spinal cord and peripheral nerve roots, sensory ganglia and nerve trunks may show further inflammatory changes, axonal damage and neuron loss. Aetiology of paraneoplastic encephalitis The most likely explanation for this condition is that it represents an autoimmune attack directed at antigens shared between the tumour cells and the nervous system. Such shared antigens have been demonstrated in small cell lung cancer, gynaecological cancers and cerebellar Purkinje cells (Anderson et al., 1988; Graus et al., 2001; Kornguth, 1989; Sakai et al., 1993). In cases of limbic encephalitis and in cases with predominantly cerebellar ataxia associated with gynaecological cancers, antineuronal antibodies have been demonstrated in serum and cerebrospinal fluid (Antoine et al., 1999; Graus et al., 1987; Gultekin et al., 2000). The neuronal proteins involved are not as yet fully characterized. The strongest evidence available to support an


autoimmune aetiology for a paraneoplastic syndrome relates to the Lambert–Eaton myasthenic syndrome, a condition in which there is defective neuromuscular transmission at the neuromuscular junction associated with small cell lung cancer. In this condition, the clinical and neurophysiological features can be transferred to normal mice using the serum of affected patients and antibodies have been demonstrated to the voltage-gated calcium channels present both in the cell membranes of the small cell lung cancer tumour cells and in the pre-synaptic terminals of peripheral motor nerve fibres (Vincent et al., 1989). In limbic encephalitis there is as yet no clear evidence of the role of anti-neuronal antibodies in causing the syndrome (Dropcho, 1989; Hart et al., 1998).

Glial nodule encephalitis Mention should be made for the sake of completeness of glial nodule encephalitis, a condition in which multiple microglial nodules, varying in frequency from rare to numerous, may be found predominantly in grey matter of the cerebral hemispheres. The condition is largely confined to immunosuppressed patients, and it is generally considered that most cases are due to cytomegalovirus (CMV) infection. Typical intranuclear inclusion bodies, immunocytochemical staining for CMV antigens, and in situ hybridization for CMV nucleic acid in occasional cells supports this view. Most cases do not show clinical symptoms but severe cases may be expected to show some decline of cognitive function. Most cases occur now in the context of AIDS and it may be expected that dementia in such cases can usually be attributed to other lesions, particularly those associated with local presence of HIV (Chapter 21), rather than to coexisting glial nodule encephalitis.

Other inflammatory conditions associated with dementia Occasionally, inflammatory brain disease of uncertain aetiology is held responsible for isolated cases of dementia complicating systemic immune-mediated diseases. An example is a reported case of dementia in which a 15-month dementing illness in a 56-year-old woman with Sjögren’s syndrome was shown by brain biopsy to be associated with a meningoencephalitis. Treatment with corticosteroids produced resolution of the dementia (Caselli et al., 1991). Another example is provided by the case of a 47-year-old male who developed cognitive impairment and a severe memory disorder in association with cat scratch disease (Revol et al., 1992).

Consequences of cerebral irradiation A variety of different forms of brain damage following cerebral irradiation have been described (Burger et al., 1981; DeAngelis et al., 1989; Sheline et al., 1980; Zeman & Samorajski, 1971). These are uncommon and not always predictable, though they are more likely to occur with high dosages of radiation, or when radiation is combined with chemotherapy (see below). They can occur over a long time interval following the radiation, from a few months to many years later, and their pathogenesis is poorly understood. The most common of these is a multifocal leukoencephalopathy that produces insidious but progressive personality change and intellectual decline over a period of months, though not all those who show CT evidence of lesions develop well-recognized symptoms. A second pattern of pathology consists of coagulation necrosis associated with fibrinoid necrosis of blood vessels in cerebral white matter. A third pathological change is multifocal, sharply demarcated pontine lesions in which there is myelin loss, focal axonal swelling with or without deposition of calcium salts and some relatively mild macrophage infiltration (Breuer et al., 1978). These lesions resemble those described below (p. 527). Fourthly, and exceptionally, a form of focal neuronal gigantism has been described in the brain following irradiation (Caccamo et al., 1989; Lampert & Davis, 1964). In this condition, developing 6 and 13 years after brain irradiation, there is marked thickening of affected gyri which may be visible to the naked eye, disorganisation of cortical laminar architecture and the presence of large, abnormal neurons and telangiectatic blood vessels in the affected cortex.

Multiple sclerosis Multiple sclerosis is a white matter demyelinating disease of unknown aetiology. It produces generally sharply demarcated, demyelinated foci, chiefly around veins and the ventricular system of the brain and in optic nerves and spinal cord. The clinical course is one of relapses and remissions, or of more consistent progression. Most patients develop their first symptoms between the ages of 20 and 40 years, and the average duration of the disease is at least 22 years. In most cases the diagnosis is made on clinical grounds and may be backed up by neuroimaging studies, neurophysiological evidence of slowed conduction velocities in the CNS or the finding of oligoclonal bands of immunoglobulin in cerebrospinal fluid. Occasionally, in clinically atypical cases, the diagnosis can only be made at autopsy. Dementia in clinically definite cases of multiple sclerosis is common.

513

514

M. M. Esiri

Various studies have found evidence of clinical dementia in 40–65% of patients with multiple sclerosis (Brassington & Marsh, 1998; Rao et al., 1991; Surridge, 1969). Neuropsychological investigation reveals deficits in as many as 50% of patients who appear intellectually intact (Peyser et al., 1980). Rarely, the clinical presentation and course of the disease are dominated by dementia (Bergin, 1957; Fontaine et al., 1994; Mendez & Frey, 1992).

Neuropathology To external appearance the brain may seem normal or slightly atrophic. Focal plaques of demyelination may be evident, however, on the ventral surface of the pons or the optic nerves may be diffusely grey and atrophic or show focally defined plaques of demyelination. In coronal sections of the brain and transverse sections of the spinal cord, the plaques of demyelination are seen most easily in white matter where they form usually well-defined circular or more irregular patches of grey–brown discolouration measuring a few millimetres to 2 or 3 centimetres across. Periventricular white matter is particularly likely to be affected with coalescing lesions liable to be found in the walls of any parts of the lateral ventricles. Blocks for microscopic examination should be taken from macroscopically affected regions. Frozen sections stained with oil red O are useful for assessing the age of lesions, old lesions containing little stainable fat but subacute ones containing an abundance of fat in lipid phagocytes. Microscopically, recently formed lesions also have a prominent inflammatory cell infiltrate consisting chiefly of T-cells, macrophages and plasma cells while old lesions are relatively acellular and gliotic. Occasionally, small, old plaques characteristic of multiple sclerosis are found at autopsy in patients with no neurological history, or only trivial symptoms of neurological disease in the distant past (Gilbert & Sadler, 1983). If such lesions are found in a patient with a history of severe, recent dementia, another disease, most commonly Alzheimer’s disease, is likely to be also present, and appropriate microscopic examination for this should be undertaken. In such cases the mild multiple sclerosis lesions are likely to be an incidental finding. Cases of multiple sclerosis with prominent dementia are likely to show at autopsy a predominance of cerebral demyelinated plaques and considerable cerebral atrophy consequent upon extensive periventricular and diffuse or frontal lobe cerebral white matter lesions (Bergin, 1957; Filley et al., 1989; Mendez & Frey, 1992). Long-standing lesions show loss of axons, as well as complete loss of myelin in the central regions of the plaques. Demyelination of nerve fibres entering the corpus callosum and coursing

through the centrum semiovale provide a reasonable explanation for symptoms of dementia in most cases of multiple sclerosis. Subcortical U-fibres are usually spared (Prineas et al. (2002); Brownell & Hughes, 1962). However, in a personally studied case in which the diagnosis was only made at autopsy and in which a 9-year progressive history was dominated by symptoms and signs of dementia with some spastic motor weakness, the cerebral lesions chiefly affected subcortical white matter. Recent neuroimaging studies of multiple sclerosis suggest that the burden of cerebral demyelinated plaques and the severity of pathological damage, particularly to axons, in the plaques, and the extent of axon loss in normal appearing white matter all contribute to the pathogenesis of dementia in multiple sclerosis (Rovaris & Filippi, 2000). Another factor may be cortical neuronal apoptosis which was documented in chronic cortical multiple sclerosis plaques in a recent study (Peterson et al., 2001).

Differential diagnosis Other white matter lesions causing dementia need to be distinguished from those of multiple sclerosis. Usually, the clinical history makes the distinction clear, but there may be overlap clinically particularly with progressive multifocal leukoencephalopathy, diffuse ischaemic white matter damage of Binswanger type or due to vasculitis, and with post-infectious leukoencephalitis (perivenous encephalitis). Rarely, the white matter lesions of Marchiafava Bignami disease, which are virtually confined to alcoholics, may also require distinction from multiple sclerosis. Pathologically, the lesions of multiple sclerosis can be distinguished from those of progressive multifocal leukoencephalopathy by the absence of oligodendrocyte intranuclear inclusions and lack of detectable JC virus antigens or nucleic acid in multiple sclerosis and by the characteristic distribution of lesions particularly in optic nerves, periventricular regions and spinal cord in multiple sclerosis. The large, pleomorphic nuclei found in astrocytes in PML are also absent from the reactive astrocytes of MS plaques though, in both diseases, the astrocytes can be large and quite bizarre in appearance. Large lesions in PML are formed by the coalescence of multiple small lesions and this is usually evident in sections. Multiple sclerosis is most reliably distinguished from ischaemic lesions of white matter by comparing myelin and axonal stains. In multiple sclerosis lesions, some axons are preserved whereas in ischaemic lesions most or all are destroyed. It should be noted that some elements of vasculitis may be present within venular walls in multiple sclerosis, but this does not give rise to infarcts, as does the vasculitis associated with collagen and other


primarily vasculitic diseases. The clinical history in perivenous encephalitis is usually short and monophasic and presents more characteristically as an acute encephalopathy than dementia. Rare chronic cases do, however, occur. Multiple sclerosis plaques differ from the foci of perivenous encephalitis in being more irregularly distributed and larger in size. In perivenous encephalitis, there are thin sleeves of demyelination around most small veins in the cerebral white matter, but they are such narrow zones that they are barely appreciable to naked eye examination.

Aetiology The aetiology of multiple sclerosis is unknown. An immune reaction undoubtedly occurs in the acute lesions and is much less in evidence in the older ones, but the provoking antigen is unidentified. Local presence of a foreign antigen is suspected to trigger the formation of lesions. Others consider that there is no need to implicate local presence of a foreign antigen and suspect the disease to be autoimmune in origin.

Table 23.2. Conditions commonly giving rise both to epilepsy (including myoclonic epilepsy) and dementia Alzheimer’s disease (esp. late) (see Chapter 9) Paraneoplastic syndrome (see this chapter) Herpes simplex encephalitis (see Chapter 21) Superficial haemosiderosis (see this chapter) Familial encephalopathy with neuroserpin inclusion bodies (see this chapter) Kuf’s disease (see this chapter) Lafora body disease (see this chapter) Mitochondrial cytopathies (see this chapter) Polycystic lipomembraneous osteodysplasia with sclerosing encephalopathy (see this chapter) Adrenoleukodystrophy (see this chapter) Gaucher’s disease (see this chapter) Leigh’s encephalopathy (see this chapter) Methotrexate toxicity (see this chapter) Trimethyl tin toxicity (see this chapter) Pick’s disease (see Chapter 11) Some neurodegenerative conditions (Chapters 11, 12, 16) Familial oculomeningeal amyloidosis (see Chapter 14) Post-head injury (see Chapter 20) Ramsay Hunt syndrome (see this chapter) Multiple cavernous angiomas (see this chapter)

Epilepsy Readily controlled, mild epilepsy is not usually associated with dementia unless the epilepsy is symptomatic of an underlying cerebral disease that also causes dementia (Table 23.2). However, severe intractable epilepsy per se can be associated with a deteriorating intellectual function and there is evidence for a moderately increased risk of dementia in patients with epilepsy (Breteler et al., 1995). Those patients shown to have continuing epileptiform activity on the EEG despite treatment show more intellectual deterioration than otherwise matched patients who show little such EEG abnormality (Dodrill & Wilkus, 1976). In addition to this suggestive evidence that severe, repeated epileptic activity can itself cause intellectual deterioration, cognitive impairment is also a well-recognized side effect of excessive anticonvulsant medication.

Neuropathology Mild epilepsy is rarely associated with structural neuropathological changes unless there is a focal lesion provoking the seizures, such as a small cortical glioma or region of dysgenesis, arteriovenous malformation or cavernous angioma. Severe, long-standing epilepsy of the type which may be associated with intellectual impairment may show characteristic neuropathological changes which are

thought to result from the excessive electrical discharges that accompany the epilepsy. These pathological changes chiefly affect the cerebral cortex, hippocampus and cerebellum. The hippocampal lesion consists of neuron loss and gliosis largely confined to the CA1 sector of the pyramidal cell layer, the CA4 or end folium and the dentate fascia granule cell layer. Usually the hippocampus on one side only is affected and the structure may appear atrophied to naked eye examination with corresponding enlargement of the adjacent inferior horn of the ventricle. This lesion is known as Ammon’s horn sclerosis. Development of this lesion in the hippocampus may give rise to psychomotor epilepsy. There may also be secondary atrophy in the ipsilateral mamillary body with severe Ammon’s horn sclerosis. In the cortex there may be patchy loss of pyramidal neurons particularly in the temporal lobes. If the epilepsy has developed early in life, the whole temporal lobe or even the whole cerebral hemisphere on one side may be small, and the contralateral cerebellar hemisphere also small. The cerebellar lesion in epilepsy chiefly affects the cortical Purkinje cell layer where severe loss of Purkinje cells may be present together with reactive astrocytosis of the Purkinje cell or molecular layer of the cortex. Experimental evidence has shown that such a lesion can be caused

515

516

M. M. Esiri

by phenytoin intoxication, but it is not certain whether the anticonvulsant or the epilepsy itself is the principal factor in the development of epileptic cerebellar damage seen in humans (Jacobs & Le Quesne, 1992). The fully developed Ammon’s horn sclerosis and cerebral hemiatrophy with contralateral cerebellar atrophy are distinctive of severe epileptic brain damage. Milder degrees of hippocampal damage affecting CA1 and CA4 but not the dentate fascia granule cells are less specific and resemble hippocampal damage seen in severe hypoxia. The pathogenesis of hippocampal and cortical epileptic brain damage is thought to involve excessive release during the epileptic discharges of the excitotoxic neurotransmitter glutamate (Lothman, 1991; Meldrum, 1997). This excessive neurotransmitter release has the capacity to damage and ultimately destroy post synaptic neurons with glutamate receptors by causing excessive stimulation of receptors, opening of calcium-gated ion channels and the entry into the cells of excessive calcium.

Unverricht–Lundborg’s syndrome This rare condition combines myoclonic or generalized epilepsy and cerebellar ataxia with mild dementia as a progressive neurodegeneration. Myoclonus is the presenting and predominant symptom. The main pathology is usually confined to loss of cerebellar Purkinje cells possibly with some additional degeneration in thalamic nuclei (Haltia et al., 1969; Koskiniemi et al., 1974). There are mutations in the gene coding for the cysteine protease inhibitor cystatin B.

Superficial haemosiderosis of the central nervous system Superficial haemosiderosis of the central nervous system is a condition associated with repeated subarachnoid haemorrhages. It is seen in a variety of spontaneously developing or iatrogenically induced lesions including hemispherectomy, arterial aneurysms, arteriovenous malformations and vascular tumours. A dementing syndrome may eventually evolve as a consequence of the repeated bleeding episodes. A component of the pathology in such cases is often hydrocephalus (Chapter 19). In addition, the cerebellar cortex appears to be damaged directly by the toxic effects of superficial deposits of iron-containing pigments with loss of the molecular layer and of Purkinje cells from exposed surfaces (Hughes & Oppenheimer, 1969).

Whipple’s disease Whipple’s disease is a rare condition due to chronic infection with an actinomycete, Tropheryma whippelii (Relman & Falkow, 1992). It usually presents with clinical manifestations relating to small intestinal disease: malabsorption syndrome, weight loss and abdominal pain. However, neurological involvement also occurs and occasionally is seen on its own (Adams et al., 1987; Fleming et al., 1988; Swartz, 2000). The organisms can be found in macrophages in affected organs, chiefly small intestinal mucosa, lymph nodes and brain. A PCR diagnostic test based on amplification of 165 rDNA specific for the organism is now available but the DNA has also been found in some people without clinical or histological evidence of the disease so a positive result needs to be interpreted with care (Dutly & Altwegg, 2001). Progressive dementia is a prominent clinical feature of central nervous system involvement with Whipple’s disease. It may be accompanied by myoclonus, ataxia and signs and symptoms of brain stem dysfunction. Occasionally obstructive hydrocephalus develops. The pathology is usually centred on the diencephalon and upper brain stem, though lesions may also be widespread in the cerebrum. Small infarcts may be found in the cerebral cortex in association with leptomeningeal involvement. Lesions contain collections of PAS-positive macrophages together with a variable, sometimes inconspicuous, lymphocyte infiltrate. The organisms, which lie in macrophages or free in the neuropil, are 1.5–2.5 m × 0.2 m in size and are argyrophilic and gram-positive. Their bacterial form can be appreciated readily on ultrastructural examination.

Metabolic diseases Metabolic derangements of the adult brain are better known for causing acute or subacute confusion or encephalopathy than dementia. Reversible confusion in a patient with broncho-pneumonia is a commonplace occurrence in geriatric wards. However, when disturbed brain metabolism develops insidiously and persists for months or years, dementia may be the main clinical manifestation. Symptoms of impaired subcortical function are likely to be prominent: slowed responses, impaired arousal and attention, memory impairment, and mood changes. There may also be disturbance of motor function: tremor, myoclonus, chorea or alterations in limb tone, and epilepsy. The acquired disorders giving rise to those disturbances are varied and, particularly in the elderly, common (Table 23.3). In addition, there is an increasing number of


Table 23.3. Metabolic conditions associated with dementia in adults Conditions associated with hypoxia Anoxic/hypoxic anoxiaa Pulmonary insufficiency Stagnant anoxiaa Cardiac disease Hyperviscosity states Severe anaemic Post-anoxic dementiaa Chronic renal failure Uraemic encephalopathy

Gangliosidosis type III (GM1 )a Gangliosidosis type II (GM2 )a Gaucher’s diseasea Niemann–Pick diseasea Mucopolysaccharidosis type II-B (San filippo’s disease)a Mitochondrial disordersa Mitochondrial myopathy with ragged red fibres (MERRF) Mitochondrial myopathy, encephalopathy, lactic acidosis and stroke-like episodes (MELAS) Kearns-Sayre syndrome (KSS)

Dialysis dementia Hepatic diseases Porto-systemic encephalopathya Inherited hepatolenticular degeneration

Metachromatic leucodystrophya Fabry’s diseasea Krabbe’s disease (Globoid cell leukodystrophy)a Acid maltase deficiency

Pancreatic diseases Insulinoma and severe recurrent hypoclygcaemia Vitamin deficienciesa Thiamine/B1 (see Chapter 13) B12 Folate

Effects of drugs a

Nicotinic acid (see Chapter 13) Endocrine disease Thyroid disturbance Parathyroid disturbance Adrenal disturbance Pan hypopituitarism Porphyria Alexander’s diseasea Lafora diseasea Kuf ’s diseasea Cerebrotendinous xanthomatosisa Polycystic lipomembranous osteodysplasia with sclerosing leukoencephalopathya Neuronal intranuclear inclusion diseasea Adrenoleukodystrophya

anticholinergics levodopa anticonvulsants methotrexatea and other cancer chemotherapies

barbiturates neuroleptics opiatesa amphoteracin Ba cyclosporina

therapies Effects of toxins Alcohola (see Chapter 13) Aluminiuma Arsenic Bismutha Cadmium Carbon monoxidea (see Hypoxia, this chapter) Chromium Ethylene oxidea Leada Mercurya Methyl bromidea Toluenea Trimethyl tina

Note: Modified after Cumming’s and Benson (1992) and Coker (1991). a Structural neuropathology described.

inherited diseases most commonly associated with mental retardation or dementia in infancy and childhood, that are now recognized occasionally to cause dementia as a presenting feature in adult life (Table 23.3) (Coker, 1991). Some conditions – for example, thyroid, parathyroid and adrenal disturbances, hyper- and hypocalcaemia and the effects of a variety of drugs – produce no well-defined neuropathological changes. However, others do, as outlined briefly below. There are problems for the pathologist anxious not to miss cases of metabolic disease when examining brains from adult cases of dementia, some of which may not have had an adequate neurological work up. Although metabolic

diseases are common, those giving rise to dementia with structural pathology are rare, probably comprising less than 2% of adult cases of progressive dementia coming to autopsy and thus, the wide range of inherited diseases is likely to be more familiar to paediatric neuropathologists and pathologists than to those dealing with adult neuropathology. Therefore, the advice of a paediatric pathologist colleague more experienced in diagnosing these diseases can be invaluable. In making a diagnosis of many of the inherited metabolic diseases it may be often necessary to examine snap frozen cryostat sections to search for stored materials that are leached out during fixation or paraffin embedding, so that if an inherited disease is

517

518

M. M. Esiri

recognized as a possibility (e.g. there is a family history), it is important to freeze samples at autopsy. Samples taken from the fresh brain at autopsy for this purpose should include white matter as well as cerebral cortex. When taking blocks of fixed tissue for microscopy, areas of brain that appear macroscopically abnormal should be chosen and, in addition, a wide selection of grey and white matter from cerebrum, brainstem and cerebellum should be sampled. For full treatment of the neuropathology of these conditions a general textbook of neuropathology such as Greenfield’s Neuropathology (Graham & Lantos, 2002) or Davis & Robertson’s (1996) Textbook of Neuropathology should be consulted.

Conditions associated with hypoxia Usually dementia in common conditions associated with hypoxia, such as respiratory or cardiac disease, is reversible when the worst effects of the disease can be relieved. Under these conditions no obvious structural sequelae in the nervous system result. However, after profound oxygen deprivation of the brain, as may occur following cardiorespiratory arrest or carbon monoxide poisoning, there may be structural pathology and incomplete return of function. Three levels of severity of sequelae can result: brain death, in which bodily functions can only be maintained artificially, persistent vegetative state, in which autonomic and brain stem function, but no higher cortical activity, is restored, and a dementia syndrome, in which a variable degree of loss of higher mental functions is the only sequela. In the latter syndrome there is loss of neurons in vulnerable parts of the cerebral cortex and hippocampus and possibly also in basal ganglia and thalamus (Auer & Sutherland, 2002). Purkinje cells of the cerebellum are also at high risk. In the cerebral cortex the pyramidal neurons are vulnerable to damage, either in a diffuse, laminar or patchy distribution, or with accentuation at the depths of the cortical sulci and at the boundary zones of the main cortical arterial supply territories. Cortical layers 2,3 and 5 containing pyramidal neurons are those most at risk. In the hippocampus the CA1 pyramidal neurons are particularly liable to be damaged or lost. Depending on the severity and chronicity of the damage the macroscopic appearance of the brain may vary from normal to severely atrophic. Carbon monoxide poisoning tends to produce foci of necrosis in the globus pallidus which, when well established, are visible to the naked eye and later form cysts. Neuron loss at these sites is accompanied by local microglial cell activation, capillary proliferation and reactive astrocytosis. Acute neuronal damage is evidenced by eosinophilia and shrinkage of neuronal cytoplasm and basophilia of nuclei. Long after the hypoxic episode the remnants of some dead

neurons may be seen in mummified form, impregnated with iron and calcium salts and appearing deeply basophilic in haematoxylin and eosin preparations. Rarely, white matter damage may also be seen following severe anoxia, particularly after carbon monoxide poisoning (Lapresle & Fardeau, 1966). White matter damage takes one or more of three forms and may occur in the absence of grey matter lesions. These are (i) perivascular foci occurring throughout the white matter of the corpus callosum, internal and external capsules and optic tracts; (ii) diffuse damage to the centrum ovale; (iii) plaque-like foci of demyelination in posterior and deep parts of the cerebral white matter, with myelin sheaths around blood vessels being relatively spared. The third pattern tends to be seen in patients who undergo secondary deterioration with neuropsychiatric symptoms after initial recovery. The other two patterns are seen more in patients who remain comatose throughout the course of the illness.

Haematological conditions Any condition that interferes with the efficient delivery of oxygen and nutrients to the brain and impairs removal of its waste products is likely to adversely affect intellectual function. There are a number of haematological conditions which act in this way by increasing the viscosity of the blood and therefore reducing cerebral blood flow, or by reducing the oxygen-carrying capacity of the blood. Hyperviscosity syndromes which predispose to impaired cerebral blood flow include macroglobulinaemia, myeloma and polycythemia rubra vera. In cases of dementia suffering from such diseases there may be no specific neuropathological changes, or there may be evidence of hypoxia as described above, or of ischaemic damage due to thrombosis and infarction (Chapter 13). Severe, chronic anaemia may give rise to the features of hypoxic damage described above. Sickle cell anaemia is, in addition, liable to produce multiple small cerebral infarcts due to sludging of deformed erythrocytes in small blood vessels. Severe anaemia from any cause may be associated with venous thrombosis, including sagittal sinus thrombosis and consequent haemorrhagic cerebral venous infarction.

Chronic renal failure There are many possible causes of altered mental state and dementia in patients with chronic renal failure (Burn & Bates, 1998; Cummings & Benson, 1992) (Table 23.4). As indicated in Table 23.2 the neuropathology of many of these conditions can be found in other parts of this book. In uraemic encephalopathy there are no specific changes to


Table 23.4. Possible causes of dementia in patients with chronic renal failure Chronic renal failure only Uraemic encephalopathy Electrolyte imbalance Anaemia (this chapter) Hypertensive encephalopathy or cerebrovascular disease (Chapter 12) Drug accumulation CNS effects of underlying cause of renal failure, e.g. SLE Chronic renal failure with dialysis Dialysis dementia (see Aluminium – this chapter) Subdural haematoma (this chapter) Emboli from shunt (Chapter 13) Vitamin deficiency Chronic renal failure with transplantation Steroid psychosis Progressive multifocal leukoencephalopathy (Chapter 21) Lymphoma (this chapter) Chronic glial nodule meningoencephalitis (e.g. due to CMV) (Chapter 21) Source: After Cummings and Benson (1992).

be found in the brain. White matter shows a non-specific diffuse astrocytosis. Cerebrovascular disease is common in renal failure so incidental cerebral infarcts may be found or there may be vascular dementia present. Peripheral nerves may show evidence of a peripheral neuropathy with chronic axonal degeneration.

Hepatic diseases Acquired portal-systemic encephalopathy most commonly develops in those with cirrhosis of the liver. The pathogenesis of the encephalopathy is not well understood, but there are several types of compounds normally metabolised by the liver which accumulate in the blood in cirrhosis: ammonia, mercaptans, short-chain fatty acids and false neurotransmitter amines. Blood ammonia levels show the closest correlation with mental impairment, which tends to occur with blood levels above 3 g/ml. Glutamine, a metabolite of ammonia shows elevated levels in cerebrospinal fluid in encephalopathic patients. The naked eye appearance of the brain in subjects with hepatic encephalopathy is often normal. Microscopically, there are characteristic changes: an alteration in the appearance of astrocytes and microcavitation and neuronal necrosis in the cerebral cortex. The astrocyte changes are discussed in chapter 18. The changes in the cerebral cortex take the form of patchy laminar or pseudo-laminar necrosis with loss of neurons particularly in the deep layers of the cortex. The microcavitation is often well seen at

the junction between cortex and white matter and is due to focal oedema. It may also be seen in the basal ganglia and cerebellum. Wilson’s disease (inherited hepatolenticular degeneration) tends to present either in children with clinical features of liver disease, or in adults with neurological symptoms of an extrapyramidal movement disorder. However, it occasionally presents with psychosis or dementia. It is an autosomal recessive disease of copper metabolism diagnosed by finding a low blood caeruloplasmin level, high tissue and urine copper levels and a Keyser-Fleischer ring in the iris. Mutations in a gene, ATP7B, coding for a copper transporting ATPase are responsible (Brewer, 2000). The disease produces a cirrhotic liver and changes in the brain which predominantly affect the basal ganglia (Harper & Butterworth, 1997). The putamen and caudate nuclei show a brown discoloration and shrinkage and the putamen may show cystic degeneration visible in brain slices. Microscopic examination shows changes that are maximal in the putamen, consisting of foci of neuronal loss and a prominence of astrocytes which resemble the type 2 astrocytes of acquired hepatic encephalopathy described in Chapter 18. There are also occasional larger, multinucleated glial cells (Alzheimer type 1 glia). The larger necrotic foci show cystic degeneration with macrophage infiltration. The cytoplasm of the macrophages contains neutral fat, stainable in frozen sections with oil Red O or Sudan black stains, and iron. Petechial haemorrhages or their sequelae – perivascular macrophages containing iron, and perivascular fibrosis – are common, and capillary endothelial cells appear unduly prominent and swollen. Perivascular and parenchymal deposits of copper may be demonstrable with rhodamine or rubeanic acid stains. Other sites besides the putamen that are liable to be affected are the globus pallidus, thalamus and subthalamic nucleus. Opalski cells may be found occasionally, scattered in these regions. These are large cells with rounded cytoplasmic outline, granular cytoplasm, which is sometimes also vacuolated, and a small, darkly staining nucleus. The cerebellar dentate nuclei and cerebral cortex may additionally show foci of neuron loss. These changes may be considerably muted in patients treated long-term with penicillamine. Dementia has been described occasionally in subjects with a related but distinct ceruloplasmin gene mutation (Servan et al., 1998).

Pancreatic disorders Recurrent episodes of severe hypoglycaemia sufficient to cause structural damage to the brain are most often associated with excess insulin secretion by an islet cell tumour

519

520

M. M. Esiri

of the pancreas. This can cause progressive or episodic memory disturbance and personality change with aggression, apathy and emotional lability. In contrast, episodes of hypoglycaemia in diabetics only exceptionally are severe enough to produce structural damage to the brain and this is, as likely as not, after intentional overdose of insulin. Severe hypoglycaemia produces changes in the central nervous system that closely mirror those due to severe anoxia, though with some differences of emphasis (Harper & Butterworth, 1997). If damage is widespread and occurred more than a few months before death there may be severe cerebral cortical atrophy. Pyramidal cortical neurons at the depths of sulci are particularly liable to be lost, usually in a patchy distribution, with the temporal neocortex and striatum and the outer cortical laminae bearing the brunt of the damage. In contrast, Purkinje cells in the cerebellum and pyramidal neurons in the hippocampus are relatively spared. As with hypoxia, in the acute stage the affected neurons show shrinkage and eosinophilia while at a later stage loss of neurons, microglial activation, astrocytosis and vascular capillary proliferation mark the sites of damage. Diffuse demyelination has also been described.

The spinal cord lesions commence at the thoracic level and chiefly affect the posterior and lateral white matter columns. From the thoracic level they may spread upwards and downwards and into the anterior white columns. As in the cerebrum there is demyelination, with a bubbly or honeycombed appearance to those myelin sheaths that remain. Axonal degeneration may occur in severe or longstanding cases.

Vitamin B12 and folate deficiencies

Pathogenesis of vitamin B12 and folate nervous system lesions

Neurological manifestations of vitamin B12 deficiency include dementia as well as peripheral neuropathy, myelopathy and optic neuropathy (Harper & Butterworth, 2002; Pallis, 1974). In man vitamin B12 deficiency is usually due to pernicious anaemia in which there is lack of production of intrinsic factor, a glycoprotein required for vitamin B12 to bind to in order to be absorbed. The main cerebral manifestations are slowed mental reactions, confusion, memory defect and depression. These may show fluctuation in severity and can precede any haematological evidence of vitamin B12 deficiency. Vitamin B12 deficiency is present when the serum level is less than 100 pg per ml. The cerebral and, better known and more common, spinal cord (‘subacute combined degeneration’) lesions of vitamin B12 deficiency are similar and affect white matter. The cerebral lesions may not be detectable with the naked eye, but are seen in sections of the deep cerebral white matter. They have been described in only a relatively small number of cases. There is patchy but widespread loss of myelin, characteristically perivascular in distribution, with swelling between the myelin lamellae of surviving myelin sheaths and preservation of axons. Degeneration of the myelin is accompanied by focal collections of lipid-containing macrophages and a mild astrocytic reaction.

Folate deficiency Folate deficiency is present when the serum level is less than 2 g per ml. Inborn errors of folate metabolism can closely mimic the neurological manifestations of vitamin B12 deficiency. Folate and vitamin B12 both participate in the same sequence of biochemical reactions central to methyl group metabolism. Dietary folate deficiency much more frequently causes haematological than neurological disturbances, and its causal relationship to neurological disease is controversial. However, psychiatric disturbance in the elderly including Alzheimer’s disease has been associated with folate depletion (Girdwood, 1968) or with associated hyperhomocysteinaemia (Clarke et al., 1998).

A deficiency of methyl group metabolism is thought to be important in the pathogenesis of neurological disease due to vitamin B12 and folate deficiencies. Methyl groups are needed for maintenance of normal cell membrane stability and for the biosynthesis of choline, an important constituent of myelin. Vitamin B12 is an essential co-factor for two enzymes involved in methyl group metabolism: methylmalonyl CoA mutase and methionine synthetase. The latter enzyme is also folate-dependent. Hyperhomocysteinaemia which is associated with low folate and B12 levels may also exert a neurotoxic effect (Selhub et al., 2000).

Porphyria Porphyrias are genetic disorders in which there is overproduction and excretion of porphyrins and their precursors. Neurological manifestations occur in the dominantly inherited acute intermittent, variegate and coproporphyria forms. No specific pathological changes have been described in the brain, although there are abnormalities described with MRI during attacks (Aggarwal et al., 1994). There is also a ‘dying back’ peripheral neuropathy with distal axonal degeneration and loss of and chromatolysis


in anterior horn cells. Posterior column axons are lost in addition.

Alexander’s disease This rare disease usually presents in infancy or childhood. A rare case has been described in a 39-year-old with mental retardation, ataxia, dysarthria and dementia (Walls et al., 1984). The brain in Alexander’s disease may be somewhat enlarged and the white matter is soft, discoloured and jelly-like. Microscopy shows demyelination and rarefaction of white matter with little or no sparing of subcortical fibres. The distinctive feature of the condition is the presence of vast numbers of Rosenthal fibres, condensations of astrocytic glial fibres, many of them arranged around blood vessels. These stain immunocytochemically for − -crystallin. Mutations in the gene coding for glial fibrillary acid protein have recently been identified in this disease (Brenner et al., 2001).

Adult polyglucosan body disease (Lafora disease) This is a rare autosomal recessive disease with onset of symptoms usually in the second decade (Gray et al. 1988; Minassian, 2001). Presentation is most frequently with generalized seizures and myoclonus followed by dementia. Macroscopically the brain shows only mild cerebral atrophy. Microscopically there is an abundance of concentric inclusions occurring in neuronal cytoplasm, axons, dendrites and lying free in the neuropil of the brain. These bodies, which closely resemble corpora amylacea, are most numerous in the thalamus, globus pallidus, substantia nigra, superior olives, brainstem reticular formation, lateral geniculate bodies, dentate nuclei and pre- and postcentral cortical gyri. The inclusions are intensely stained by PAS and by stains for acid muco-polysaccharides. They appear basophilic with haematoxylin and eosin stain. Outside the brain they are found in skeletal muscle and liver. Biopsy of these tissues can enable a diagnosis to be made. The biochemical basis of the disease has not been elucidated. Gene mutations in the EPM2A gene coding for laforin have been identified in most cases tested.

Kuf’s disease (adult onset neuronal ceroid lipofuscinosis) Kuf’s disease is the adult form of neuronal ceroid lipofuscinosis. Most cases are recessively inherited. It presents in early adult life with dementia, behavioural change and sometimes with violent psychosis (Josephson et al., 2001). Rarely it presents in middle or old age with

dementia. Motor disturbance, myoclonus, facial dyskinesia and ataxia may also be present. Diagnosis depends on finding the characteristic curvilinear or fingerprint inclusions interspersed with eosinophilic granules in brain neurons (Suzuki and Suzuki, 2002). Unlike juvenile forms of neuronal ceroid lipofuscinosis, there is no retinal degeneration, and no storage of abnormal material outside the brain. Macroscopically the brain at autopsy appears atrophic and the leptomeninges thickened. The cerebral cortex is narrowed and the white matter abnormally firm and gliotic. Microscopic sections show some neuron loss, reactive gliosis and microglial activation. Some surviving neurons contain prominent granular material with histochemical characteristics of lipfuscin – PAS-positive and sudan black-positive with strong autofluorescence. The chemical nature of the stored material has not been fully identified.

Cerebrotendinous xanthomatosis This is an autosomal recessive condition in which there are mutations in the cytochrome P450 gene. There is decreased bile acid formation and increased levels of bile acid intermediates including the high density lipoprotein cholestanol in serum and tissues (Bjorkhem & Skrede, 1990). Xanthones form in tendons, cataracts occur and deposits of cholesterol in brain lead to dementia, ataxia, long tract signs and pseudobulbar palsy. Neurologic manifestations commence usually in the second and third decade and progress to death in the fourth or fifth decades. The diagnosis can be made by demonstrating high levels of cholestanol in serum or tissues. Similar biochemical changes have been described in a 44-year-old woman with frontal lobe dementia (Sugama et al., 2001). The cholesterol deposits in brain discolour the white matter and lead to formation of cholesterol clefts in microscopic sections of white matter. These may be accompanied by a local foreign body giant cell reaction.

Polycystic lipomembranous osteodysplasia with sclerosing leukoencephalopathy This is a rare autosomal recessive disorder presenting with bone pain due to cysts and pathological fractures in early adult life. The underlying genetic defect is mutations in the gene encoding tyrosine kinase binding protein (TYROBP) (Paloneva et al., 2000, 2001). Dementia and focal neurological symptoms develop in the fourth decade and lead to death usually in the fifth decade. The neurological deterioration is due to gliosis in subcortical white matter sometimes with calcification in vessel walls

521

522

M. M. Esiri

(Paloneva et al., 2001; Sourander, 1970). The diagnosis can be made on biopsies of bone and skin, which contain characteristic folding and wrinkling of the cell membranes of fat cells (Bird et al., 1983), or with molecular genetic studies.

Hypocalcaemia with calcification of the basal ganglia Dementia or depression may be seen in hypoparathyroidism and other conditions causing hypocalcaemia. Extrapyramidal symptoms are also sometimes present. Increased amounts of calcium are found predominantly in the basal ganglia, internal capsule, lateral thalamic nuclei and dentate nuclei of the cerebellum. The cerebral and cerebellar cortex may also be affected. The calcium deposits are detected as grittiness on slicing the brain. Microscopically they are found in the walls of capillaries as calcosphorites, and as more diffuse deposits in the walls of larger vessels. There is attenuation of adjacent neuropil but little astrocytic or microglial cell reaction.

Adrenoleukodystrophy This is a sex-linked disorder due to deficiency of a peroxisomal enzyme, lignoceroyl CoA synthetase which is needed for the beta oxidation of very long-chain fatty acids (Moser, 1997; Powers & Moser, 1998). Most cases present in childhood but a few present later with a progressive spastic paraparesis (myelopathy), peripheral neuropathy or dementia. Mild symptoms may be displayed in carrier females. Addison’s disease is clinically evident in only 20% of sufferers at the time of neurological presentation. Diagnosis can be made by estimation of very long chain fatty acids in serum and using molecular genetics. At autopsy the brain appears normal or mildly atrophic to external examination but on slicing shows marked, diffuse abnormality of the white matter which appears discoloured and gliotic. Microscopy shows diffuse demyelination of cerebral white matter with relative sparing of axons, and marked reactive astrocytosis, macrophage and microglial cell activation and, in areas of active demyelination, neutral fat in phagocytes and an intense perivascular lymphocytic infiltrate (Griffin et al., 1985; Ito et al., 2001). Areas of chronic demyelination show loss of oligodendrocytes. In the spinal cord there is loss of fibres in pyramidal tracts and, sometimes loss of fibres in posterior columns also. Ultrastructurally, macrophages in affected cerebral white matter and peripheral nerves, and cortical cells of the adrenal gland show typical trilaminar inclusions. The adrenal glands typically show atrophy of the cortex but occasionally display a

reactive hyperplasia. At the light microscope level adrenal cortical cells show a fibrillary, striated appearance of the cytoplasm which is characteristic of the disease.

GM1 and GM2 Gangliosidosis There is a rare adult form of GM1 gangliosidosis, an autosomal recessive disease which presents in adults with dementia, dysarthria, gait disturbance and limb rigidity (Nakano et al., 1985). At autopsy the brain may appear macroscopically normal but microscopy shows ballooned neurons with pale cytoplasm and eccentrically placed nuclei. Similar swollen neurons are present in myenteric and submucosal plexuses of the gut. The condition is due to deficiency of the lysosomal enzym˙e galactosidase caused by mutations in the gene coding for this enzyme (Paschke et al., 2001). Ultrastructural appearances are similar to those in GM2 gangliosidosis. An adult form of the autosomal recessive storage disorder, GM2 gangliosidosis is seen almost exclusively in Ashkenazi Jews. It is due to partial ( 20years before another successful rodent model of human neurodegenerative disorders was established. The clinical and neuropathological features of the prion diseases can be highly variable (Ironside, 1996) (Chapter 17). The most common symptoms are dementia and ataxia. Neuropathologically, most patients show spongiform degeneration and reactive astrogliosis, in many cases compact extracellular deposits of prion protein amyloid are found (for review see DeArmond & Prusiner, 1995). In most cases, cytoplasmic inclusions are rare, however, there are a few cases of prion disease where neurofibrillary tangles are a prominent pathology (Bugiani et al., 2000; Ghetti et al., 1994). The extent of cell death is highly variable and, in the different diseases (Table 24.1), may affect different regions of the brain with differing levels of severity. Although the mechanisms of cell death are not entirely clear (for review see Chiesa & Harris, 2001), there is evidence for activation of apoptotic death pathways (for review see Giese & Kretzschmar, 2001). Because of the early availability of rodent models for prion disorders, much of the pioneering work in human neurodegenerative disease has involved the study of prions. In the early 1980s Prusiner and colleagues became the first to purify and characterize the protein component of neuropathological lesions in diseased brain. Although these studies were directed towards purifying the ‘infectious scrapie agent’ (McKinley et al., 1983; Prusiner et al., 1983), it quickly became apparent that the prion protein was also a component of the amyloid plaques, also called ‘Kuru’ plaques, which may occur in the brains of mammals affected by prion disease (Bendheim et al., 1984, 1985). Purification of this protein permitted the generation

of antibodies and the molecular cloning of the prion protein (PrP) gene, establishing the relationships between a normal cellular protein (PrPC ) and a disease-specific conformer (PrPSc ) (Oesch et al., 1985). Approaches in cell biology, biochemistry, and structural biology have combined to demonstrate that when an apparently normally folded PrPC comes in contact with PrPSc (either as monomer, small oligomer, or larger aggregate), a transition in conformation results, generating a new PrPSc molecule (for review see Prusiner et al., 1998). Similar examples of protein conformation transitions in proteins of normal sequence include the assembly of tau and -synuclein into cytoplasmic inclusions and the assembly of -amyloid peptide into extracellular deposits. The identification of the prion protein gene provided the key to the rare forms of prion disease that occurred in families as dominantly inherited disease; familial Creutzfeldt– Jakob, Gerstmann–Sträussler–Scheinker syndrome, and fatal familial insomnia (Table 24.1) (for reviews see Collinge, 2001; Aguzzi et al., 2000). Molecular characterization of the prion protein gene in affected members of these families demonstrated that mis-sense mutations tracked with disease; to date, more than 24 disease causing mutations have been described (www.mad-cow.org). Similarly, mutations in the constituents of protein aggregate lesions have been found in multiple neurodegenerative settings, including Alzheimer’s disease (amyloid precursor protein), Frontotemporal Dementia with Parkinson’s (tau), Parkinson’s disease (-synuclein), and familial amyotrophic lateral sclerosis (superoxide dismutase 1). Hence, there is a growing body of evidence to suggest the mis-folding of certain proteins can initiate processes that result in the death and dysfunction of specific populations of neurons. The first successful uses of transgenic and knockout mice in the study of human neurodegenerative diseases involved studies of prions (Scott et al., 1989, 1993; Prusiner et al., 1990). Drs Scott, Prusiner, and colleagues used transgenic mice expressing prion proteins derived from heterologous species to study the molecular basis for a phenomenon termed the ‘species barrier’, which is based on the observation that prions from one species do not readily infect animals from another species (very long incubation period and low penetrance). To examine this phenomenon, transgenic mice were created that express the prion protein gene of hamsters in addition to their endogenous prion protein. When inoculated with hamster scrapie, these mice could support the replication of the hamster prion efficiently. Moreover, the newly formed pathogen (containing hamster PrPSc ) retained the specificity for hamsters. In nontransgenic mice, inoculation with hamster scrapie was associated with a very long incubation period and mice that

Transgenic mouse models

did develop disease produced prions that were much more transmissible to other mice, not hamsters. These studies provided some of the first definitive evidence that fundamental properties of prion infectivity were dictated by the prion protein itself, and that this protein must be intimately associated with the infectious agent. In addition, when inoculated with hamster scrapie, the neuropathology of these transgenic mice revealed the intimate association between the prion protein and disease lesions; mice expressing hamster prion proteins developed amyloid plaques composed of hamster prion protein. One of the clinching experiments involved the production of mice with targeted deletions of the prp gene. These mice, which developed normally and otherwise appeared unaltered, were highly ¨ resistant to ‘infection’ with mouse prions (Bueler et al., 1993). Collectively, these studies demonstrated the pivotal role the prion proteins play in the pathogenesis of prion disorders. More recently, transgenic approaches have been used to define the minimal elements within the prion protein that are required to support prion disease [for review see (Scott et al., 2000)]. Perhaps one of the most innovative uses of transgenic and knockout mice has been the work of Dr. Adriano Aguzzi and colleagues, in which PrP knockout mice have been the recipients of transplanted tissues from normal mice (for review see Aguzzi, 2001). These landmark studies have probed the basis for prion neurotoxicity and the mechanisms of prion ‘infection’. For example, neural grafts of normal brain have been transplanted into the brains of PrP knockout mice. Upon infection by mouse prions, the grafts composed of normal tissues degenerate with classic spongiform degeneration. In areas around the graft, including host brain tissues, extracellular deposits of PrP amyloid are evident. Despite this robust pathology, host, PrP null, neurons remained resistant to degeneration; data suggesting that prion toxicity is completely cell autonomous. Only neurons that produce PrPC , which then acquires the toxic conformation, are affected. The first transgenic mouse models of familial forms of human neurodegenerative disease were created by expressing mutant human prion proteins in transgenic mice (Hsiao et al., 1990, 1994). As mentioned above, mutations in the prion protein gene were identified in affected members of families exhibiting inherited forms of prion disease (Table 24.1). The first mutation identified was in families afflicted with Gerstmann–Sträussler–Scheinker Syndrome (Hsiao et al., 1991), and hence prion proteins encoding this mutation were the first to be expressed in transgenic mice. Taking guidance from previous studies on the nature of prion species barriers, Hsaio, Prusiner, and colleagues chose to introduce the disease-associated

mutation (P102L) into the mouse prion protein gene, and to express high levels of this mutant mouse gene (Hsiao et al., 1990). Mice expressing this variant developed a neurological disease characterized by abundant spongiform encephalopathy and a strong astrocytic response. Moreover, it was possible to transmit the disease to other mice (expressing mouse PrP-P102L at levels too low to initiate spontaneous disease) (Hsiao et al., 1994). Despite this early success, it has proven to be difficult to model the familial human prion diseases in mice. To date, only one other mutant PrP has induced disease in transgenic mice (Chiesa et al., 2000, 2001). In the transgenic models of prion disease, neuropathological features of the mice have not been a major focus. Initial work focused on the factors that govern prion infection, such as host susceptibility and prion structure. The most important elements in these studies were evidence of neurologic disease, fatality, and transmission of disease. However, neuropathological evaluations were pivotal in confirming the presence of spongiform degeneration, reactive gliosis, and PrP amyloid plaques, the hallmark features of prion disorders. In some of the other mouse models, there has been a much greater focus on the pathological features. Notably, many of the technological advances first applied to the study of the prion disorders have been modified and adapted for use in other disease settings. One of the most successful examples of this process is the modification of prion protein gene elements to create some of the most widely used transgene vectors. These vectors derive from the hamster (Scott et al., 1992) or mouse (Borchelt et al., 1996a)† prion protein genes and direct expression of foreign genes in both neurons and glia throughout the CNS. Multiple laboratories have found these vectors to be highly useful in a variety of neurodegenerative disease settings.

CAG repeat disorders The expansion of polymorphic tracts of CAG repeats (coding for consecutive glutamine residues) within a diverse set of genes is the cause of a number of neurodegenerative disorders, including Huntington’s Disease, several types of spinocerebellar ataxia (SCA), spinobulbar muscular atrophy (SBMA), Machado–Joseph Disease (MJD), and dentato-rubral pallido-luysian atrophy (DRPLA) (La Spada et al., 1991; Huntington’s Disease Collaborative Research Group, 1993; Orr et al., 1993; Kawaguchi et al., 1994; Koide et al., 1994; Nagafuchi et al., 1994). Although these disorders †

The MoPrP.Xho vector is now available from the ATCC at www.ATCC. org, ATCC # JHU-2.

535

536

D. R. Borchelt, J. Jankowsky, A. Savonenko, G. Schilling, J. Wang and G. Xu

are caused by a similar type of genetic mutation, the populations of affected neurons, the clinical syndromes, and the neuropathological lesions are quite distinct for each disorder (for reviews see Ross et al., 1997; Sieradzan & Mann, 2001; Orr, 2001; Fischbeck, 2001). Unlike the prion diseases, where different clinical/pathological syndromes are associated with different mutations in a single gene (PrP), each of the CAG repeat disorders tracts with a different gene product; huntingtin for HD, ataxins for SCA-1 to 13 and MJD (ataxin-3), androgen receptor for SBMA, and atrophin-1 for DRPLA. In most cases, the normal function of the diseaseassociated gene is either unknown or incompletely characterized; hence many of the gene products have been given a disease-specific name. The length of the glutamine repeat correlates inversely with disease onset; the longer the repeat the earlier the age of onset (for reviews see Ross, 1995; Ross et al., 1997) (see also Chapter 16). The identification of the disease-associated genes in these diverse disorders has facilitated the generation of transgenic mice that recapitulate many of the behavioural and neuropathological features of the disease. Indeed, the analysis of transgenic mice designed to model HD provided new insights into the pathology of the polyglutamine disorders. The first description of structures termed neuronal intranuclear inclusions (NII) came in the analysis of transgenic mice expressing the first exon of huntingtin, encoding an exceptionally long glutamine stretch. These inclusions, detected by antiserum to N-terminal residues of huntingtin or ubiquitin, were visible in most neurons of the brain (Davies et al., 1997; Mangiarini et al., 1996). The initial description of these structures in the mice prompted the re-examination of human tissues, and it was quickly recognized that intranuclear aggregation of mutant huntingtin was a pathognomonic feature of HD (Fig. 24.1) (DiFiglia et al., 1997; Becher et al., 1998). Moreover, intranuclear neuronal inclusions have been described in most of the polyglutamine disorders. HD is one of the rare exceptions where aggregates of mutant huntingtin are also found in cytoplasmic compartments (Gutekunst et al., 1999). Studies with a variety of antibodies have suggested that the huntingtin immunoreactive aggregates are composed of N-terminal fragments of the protein (Fig. 24.2); the glutamine repeat domain starts at residue 18. In many cases, the nuclear aggregates are also immunoreactive to antibodies against ubiquitin. The polyglutamine repeat disorders have been one of the easiest to model in genetically modified mice. At the time of writing, no less than 18 distinct reports of transgenic or knock-in mouse models of polyglutamine disorders could be found in the literature; 8 of these reports described variations of mice that model HD. Mice have been

Fig. 24.1. Nuclear and cytoplasmic inclusion in HD-N171-82Q mice. Brain tissues from HD-N171–82Q (line 81) mice were harvested and stained with antibodies to N-terminal domains of huntingtin as previously described (Schilling et al., 2001). Inset – cytoplasmic aggregate found in the subthalamic nucleus. Reprinted from Neurobiology of Disease Vol 8, 2001 (Schilling et al.), with permission from Elsevier.

created that express full-length huntingtin, a large protein of more than 3000 amino acids, via cDNA-based expression plasmids (Reddy et al., 1998), Yeast Artificial Chromosomes (containing the entire human gene) (Hodgson et al., 1999), and by knock-in of repeat expansions in the mouse huntingtin gene (Lin et al., 2001; Levine et al., 1999; Wheeler et al., 2000). Three different groups have expressed a fragment of mutant huntingtin, ranging from only the first exon (Mangiarini et al., 1996; Yamamoto et al., 2000), the first and second exons (Schilling et al., 1999a), to nearly one-half of the protein (Laforet et al., 2001). Comparing and contrasting the neuropathology of these different models has been very informative. Mice expressing short N-terminal fragments of mutant huntingtin have the most severe behavioural disturbances and exhibit abbreviated life spans (Mangiarini et al., 1996; Schilling et al., 1999a); a finding that seems to confirm the notion that truncated fragments of mutant huntingtin possess greater toxicity (for review see Ross, 1995). Mice expressing truncated fragments of huntingtin show severe behavioural abnormalities, significant weight-loss, and premature death. For the most part, mice expressing full-length versions of mutant huntingtin have less severe phenotypes; behavioural disturbances can be elicited with


mHD549 [549-679] AP81[650-663] IC2 [TBP-38Q]

P2 [1187-1207]

Non-reactive Anti-hunt 2 [2110-2121] HF1 [1981-2580]

HP1 [80-113] CAG53b [1-118 (53Q)] HD1 [1-173 (23Q)]

Reactive to Nuclear/Cytoplasmic Aggregates

EM48 [1-212 (dQ)] Fig. 24.2. Diagrammatic summary of epitopes represented in full-length huntingtin (above) and huntingtin fragments found in nuclear and cytoplasmic compartments (below). The monoclonal antibody 1C2 recognizes expanded polyglutamine domains, but does not readily recognize aggregated forms of polyglutamine.

extremely long glutamine repeats (150 consecutive glutamine repeats) (Lin et al., 2001). There are two examples where full-length, mutant, huntingtin has been expressed in the brains of FVB/N mice. In both cases, these animals show a tendency to circle (Reddy et al., 1998; Hodgson et al., 1999), however, in both cases, circling was observed in only a subset of animals. Neither of these models appears to show alterations in coordination. In another study targeted recombination was used to expand the polyglutamine of mouse huntingtin, C57BL/6J × 129/J mice. These animals show no overt behavioural abnormalities (Wheeler et al., 2000). Surprisingly, this higher degree of toxicity in mice expressing truncated forms of mutant huntingtin does not equate to more extensive cell death. Mice expressing the first exon of mutant huntingtin with > 100 glutamine repeats show evidence of neuronal atrophy, but lack the severe loss of neurons in the caudate and striatum that occurs in the human disease. However, as occurs in human disease, intranuclear inclusions containing mutant huntingtin are found in nearly all populations of neurons that express the mutant huntingtin (Schilling et al., 1999a). In mice expressing full-length protein, either via a cDNA vector, a YAC transgenic, or knock-in, the neurons most affected in HD are the first to accumulate, aggregated, Nterminal fragments of huntingtin in nuclei (Hodgson et al., 1999; Lin et al., 2001; Levine et al., 1999; Wheeler et al., 2000). Moreover, all of the full-length mutant huntingtin

models appear to show a more selective pathology to the striatum in the form of frequent neuronal intranuclear inclusions. However, there is ample evidence that, in HD patient brains, multiple populations of neurons show abnormalities (Becher et al., 1998; Gutekunst et al., 1999) (Chapter 16). One synthesis of the data obtained in the study of these different strains of transgenic mice would hold that in the striatum, full-length mutant huntingtin is more readily fragmented, and that once fragmented, the polyglutamine containing fragment is prone to aggregate/accumulate in the nucleus and cytoplasm. The data from mice expressing truncated versions of mutant huntingtin suggest that, once generated, the fragments are highly toxic. Although it remains controversial as to whether the aggregated form of mutant huntingtin fragments, mis-folded monomers, or some intermediate oligomer is the toxic species, there is no question that fragmentation facilitates aggregation, and the propensity to aggregate equates with toxicity. Some of the best evidence to suggest that fragmentation/truncation of polyglutamine proteins can play a significant role in the generation of toxic species comes from the study of transgenic mouse models for DRPLA, which is caused by a polyglutamine expansion in atrophin1. Mice expressing mutant full-length atrophin-1 accumulate C-terminally truncated products, containing the polyglutamine domain, in nuclear compartments of affected cells (Schilling et al., 1999b). Sequence analysis of

537

538


PolyQ Atrophin-1 N

C NES

NLS PolyQ

Cleavage in nucleus C NES

N NLS

N

Accumulation and aggregation in the nucleus

N NLS Fig. 24.3. Model for processing of atrophin-1.

atrophin-1 identified a consensus nuclear localization signal (NLS) near the N-terminus of atrophin-1 and a nuclear export signal (NES) near the C-terminus (Fig. 24.3). The pathogenic truncation event removes the nuclear export signal, leaving a fragment that is targeted to the nucleus. The first polyglutamine disorder to be modeled in transgenic mice was SCA-1. Drs Orr and Zoghbi produced a mouse model of SCA-1 by expressing selectively a mutant version of ataxin-1 in Purkinje cells of the cerebellum. This model system was then subsequently used to probe the mechanisms of disease by expressing experimentally modified forms of mutant ataxin-1. For example, normally ataxin-1 is localized to the nucleus, and targeting the mutant form of ataxin-1 to the cytoplasm diminished the toxicity of the mutant protein (Klement et al., 1998). This model system was also used to examine the role of aggregation and intranuclear inclusions in the pathogenesis of SCA-1 by generating transgenic mice expressing an altered form of mutant ataxin-1 (deletion of a self association signal (125 aa)) with 77 glutamines. These animals were found to develop a disease phenotype that is similar to transgenic mice expressing ataxin-1 with 82 glutamines, but the affected cells in these animals lacked nuclear aggregates (Klement et al., 1998); data that were interpreted to suggest that aggregates, per se, are not required for toxicity. The controversy over the role of aggregates in the pathogenesis of polyglutamine disease is further fuelled by the observation that, in mice expressing fragments of mutant huntingtin, many neurons with large visible nuclear and/or cytoplasmic aggregates do not degenerate. However, there are several reports indicating at least some harmful role for aggregation and inclusions in the pathogenesis of the polyglutamine disorders. Intranuclear

inclusions can trap components of the proteasome pathway (Cummings et al., 1998; Chai et al., 1999b), molecular chaperones (Cummings et al., 1998; Chai et al., 1999a; Stenoien et al., 1999), or other proteins with short stretches of polyglutamine such as the TATA-binding protein (Perez et al., 1998; Suhr et al., 2001). It has been suggested that the sequestration of these factors could be a mode of injury in these diseases. One example of this mode of injury has been work demonstrating that a transcription regulatory factor, termed CREB-binding-protein (CBP), is recruited to nuclear intraneuronal inclusions and that the transcriptional activating activity of CBP is diminished in the presence of polyglutamine-containing proteins (Steffan et al., 2000, 2001; McCampbell et al., 2000; McCampbell & Fischbeck, 2001; Nucifora, Jr. et al., 2001). However, it is possible that the recruitment of CBP to inclusions is not required for toxicity, but rather its appearance in inclusions is just one of the downstream sequela of an abnormal interaction between mutant huntingtin and CBP; an interaction that alters the normal function of CBP. Recent studies in other diseases have also provided evidence to suggest that alterations in transcription may play a role in the pathogenesis of polyglutamine diseases. Ataxin-1 has been reported to interact, via the polyglutamine domain, with a protein termed PQBP-1, which also functions in regulating transcription (Okazawa et al., 2002). Ataxin-7 interacts, via its polyglutamine domain, with cone-rod homeobox protein (CRX), which is a transcription factor expressed in retinal photoreceptor cells (La Spada et al., 1991). Atrophin-1 has been shown to interact with transcriptional repressor proteins (Wood et al., 1998, 2000), and recent work in Drosophila has identified an interaction between atrophin-1 and a transcriptional repressor protein that functions in early embryonic patterning (Zhang et al., 2002). Thus, there is mounting evidence to suggest that transcriptional dysregulation could play a role in these disorders. To study gene expression in pathological settings, investigators have turned to some of the newest tools in pathology, the gene arrays. These new tools facilitate molecular pathology, allowing investigators to study changes in cells at the molecular, and systems, level. In a study of transgenic mouse models of HD, Luthi-Carter and colleagues demonstrated reductions in the levels of expression of < 1% of the genes expressed in striatum (Luthi-Carter et al., 2000), indicating that disease is not due to wholesale dysregulation of transcription in HD. As this technology improves, and as the database of transcriptional profiling expands, these powerful techniques may prove invaluable in deciphering the molecular events in disease pathogenesis.


No. of Mutations

6

www.alsod.org

5 4 3 2

* **d * *

1

I Ex 1

II

23

III Ex 2

IV

56

Ex 3

Cu ligand Zn ligand stop d in-frame deletion Glu 133 t

*

V

80

VI

Ex 4

119

VII Ex 5

153 aa

Loop domains (white) Beta strand structures (black)

Fig. 24.4. Map of mutations associated with familial ALS in SOD1. A comprehensive list of mutations in SOD1 that have been associated with familial ALS is available at www.alsod.org. The locations of beta structure, loop domains, Cu2+ binding, and Zn binding sites are based on the published descriptions of SOD1 structure (Parge et al., 1992).

Amyotrophic lateral sclerosis Amyotrophic lateral sclerosis (ALS) presents as both sporadic and familial (FALS) illness. In a subset of FALS cases, mutations in Cu/Zn superoxide dismutase 1 (SOD1) have been identified as the cause (Rosen et al., 1993). SOD1 is the principal superoxide scavenging activity of the cytoplasm (McCord & Fridovich, 1969; Reaume et al., 1996). To date, over 70 different point mutations at more than 50 residues in Cu/Zn SOD1 have been linked to FALS (Fig. 24.4) (www.alsod.org). Five additional mutations that cause C-terminal truncations of the protein have also been described (www.alsod.org). Early studies of FALS-SOD1, a very abundant enzyme, demonstrated that only a minority of mutations render the enzyme either totally inactive or severely unstable (Borchelt et al., 1994, 1995; Gurney et al., 1994; Ratovitski et al., 1999). Moreover, mutant proteins do not exhibit evidence of dominant negative action towards wild-type enzyme; the hyper-expression of mutant protein did not appear to significantly alter the stability or activity of co-expressed wild-type protein (Gurney et al., 1994; Borchelt et al., 1995; Wong et al., 1995b).

Among the various attempts to model human neurodegenerative disease in genetically modified mice, the SOD1ALS mice have been one of the most successful in recapitulating the clinical/pathological features of the disease. The faithfulness of these mice to the human disease may be due, in part, to the use of genomic fragments of the human SOD1 gene to produce the transgenic animals. Three different mutations (G37R, G85R, and G93A) have been expressed in transgenic mice, yielding animals that reproduce some of the classic symptoms of ALS (lower limb weakness progressing to paralysis and upper limb weakness) and much of the classic pathology (selective loss of large motor neurons in the ventral horn of the spinal column) (Gurney et al., 1994; Ripps et al., 1995; Wong et al., 1995b; Bruijn et al., 1997). By and large, the affected motor neurons in these mice lack the accumulations of neurofilament proteins that have been widely described in sporadic ALS patients (for review see Hirano et al., 1969). The transgenics do, however, show accumulations of ubiquitin immunoreactivity, which is common in affected tissues from ALS patients (Watanabe et al., 2001). The SOD1 transgenic mice have been used extensively to ask basic questions regarding the mechanisms of

539

540


disease, utilizing crosses with other genetically modified mice. One example of the more extensively studied questions, using this approach, has been to examine the role of neurofilaments in the pathogenesis of SOD1-linked ALS. Mice expressing mutant SOD1 have been crossed with mice lacking the light-chain neurofilament protein, mice that over-express neurofilament H (heavy chain), and mice that express a neurofilament-H-fusion with beta-galactosidase. The outcomes of these various crosses have suggested that neurofilament proteins are only minimally involved in the pathogenesis of SOD1-linked ALS (for review see Rosenthal et al., 1999). Another important example of the use of genetically modified mice has been in investigations designed to delineate the molecular mechanisms by which mutant SOD1 kills motor neurons. At present there are three prevailing hypotheses to explain the toxicity of FALS-SOD1 variants; gain of peroxidase-like activity (Wiedau-Pazos et al., 1996), enhanced reaction with peroxynitrite and tyrosine nitration (Beckman et al., 1993; Estévez et al., 1999), and the formation of aggregates (Bruijn et al., 1998; Johnston et al., 2000). The first two mechanisms may be viewed as related because the chemistry of the toxic reactions involves the copper cofactor of the enzyme. There is much evidence in favour of the notion that copper-mediated chemistry could be the basis for the toxicity of FALS mutant SOD1. It has been known for many years that wild-type SOD1 can react with H2 O2 (Yim et al., 1990, 1993). In vitro, reactivity between SOD1-wt and H2 O2 can result in damage to SOD1, reducing its activity and stability (Hodgson & Fridovich, 1975; Salo et al., 1990). Investigators quickly moved to test whether FALS-linked mutations in SOD1 may augment this natural activity. Three independent studies examined a small number of mutants, reporting that FALS mutations increase the reactivity of SOD1 with H2 O2 , resulting in the production of radical species (Wiedau-Pazos et al., 1996; Sankarapandi & Zweier, 1999a,b; Yim et al., 1996; Liochev et al., 1997). It is hypothesized that these radicals cause oxidative stress in spinal motor neurons and this stress would play a role in the pathogenesis of disease. In addition to peroxidase activity, SOD1 has previously been demonstrated to react with peroxynitrite (−ON OO) to generate nitronium ions that are capable of covalently modifying tyrosine residues in protein substrates (Ischiropoulos et al., 1992; Crow et al., 1997a,b). More recently, studies by Estévez and colleagues suggested that FALS mutant SOD1 are prone to lose the zinc cofactor, and that enzymes lacking zinc (but retaining copper) were much more prone to catalyze the nitration of protein substrates (Estévez et al., 1999). Importantly, Estévez and colleagues demonstrated that wild-type enzyme (in a zinc deficient state) also showed enhanced

nitration activity, providing a potential link between the familial and sporadic forms of ALS (Estévez et al., 1999). To examine the importance of these potential disease pathways, several laboratories have turned to genetically modified mice as tools to ask basic questions. One group, headed by Dr Philip Wong (JHU School of Medicine), has spearheaded the effort to reduce the loading of copper to mutant SOD1 by eliminating the protein that appears to be responsible for delivering copper from plasma membrane transporters to SOD1; a protein termed the copper chaperone for SOD1 (CCS) (Culotta et al., 1997; Wong et al., 2000). It has become clear from Dr Wong’s work that eliminating CCS leads to a dramatic reduction in the levels of active SOD1 in nervous tissues (Wong et al., 2000). Using radiolabelled 64 Cu, Dr Wong and colleagues have demonstrated that mutant SOD1 acquires copper slowly (or not at all) in the absence of CCS (Subramaniam et al., 2002). Mice that express the G37R, G85R, and G93A variants of FALS-SOD1, in the absence of CCS, develop motor neuron disease with no apparent delay in the age of appearance or severity of disease (Subramaniam et al., 2002). On the surface, these data would appear to have resolved the issue of whether copper-mediated chemistry plays a role in the pathogenesis of SOD1-linked FALS. However, proponents of the ‘copper hypothesis’ have suggested that it may be possible that motor neurons possess alternative means to deliver copper to SOD1. It has also been suggested that the toxic species of mutant SOD1 are not molecules that acquired copper by the normal route (involving CCS) but instead acquire it via an alternative route, and that only enzymes that acquire the metal by this alternative route are toxic. Hence eliminating CCS would have little impact on the toxicity of the enzyme. In favour of these arguments are data in yeast demonstrating that human SOD1 (when expressed at high levels) can acquire copper in the absence of CCS (Corson et al., 1998). Moreover, it has recently been demonstrated that a minor fraction of SOD1 is translocated into the mitochondrial intermembrane space (Sturtz et al., 2001), and it is possible that once in the mitochondria, mutant SOD1 may acquire copper by non-CCS mediated pathways. Thus, although there are compelling data to suggest that copper-mediated chemistries may not underlie the toxicity of FALS mutant SOD1, the complete story is not yet written. Another approach to examining the role of copper chemistry in the pathogenesis of disease is to study enzymes that harbour disease-causing mutations at histidine residues crucial for the coordinated binding of copper. Previous studies have demonstrated that the H46R and H48Q mutations each dramatically reduce the ability of mutant enzyme to scavenge superoxide as assayed by the ability of these mutants to rescue the growth defects of SOD1 null


A

B

H&E

C

Thio-S

Ubq H46R/H48Q mice

D G85R mice

H&E

Ubq

Fig. 24.5. Ubiquitin immunoreactive, fibrillar, inclusions in mice expressing FALS mutant SOD1. (a)–(c) The spinal cords from paralyzed mice expressing H46R/H48Q human SOD1 were stained with hematoxylin/eosin (a), Thioflavine-S (b), and ubiquitin antibodies (c) as previously described (Wang et al., 2002b). (d ) The images shown in panel D are from mice expressing the G85R variant of human SOD1, and were originally published by Bruijn et al. (1997). Reprinted from Neuron Vol. 18 (1997) with permission from Elsevier.

yeast (Ratovitski et al., 1999). To reduce the ability of SOD1 further to bind copper, Drs Wang, Borchelt and colleagues engineered the human SOD1 cDNA to encode a double mutation of histidines 46 (H46R) and 48 (H48Q); this double mutant has no detectable superoxide scavenging activity (Wang et al., 2002b). Mice expressing the H46R/H48Q double mutant develop, in a transgene product dosedependent manner, motor neuron disease that is identical to that previously described in mice expressing the G37R, G85R, and G93A variants of human SOD1 (Gurney et al., 1994; Bruijn et al., 1997; Wong et al., 1995b). The spinal cords of paralyzed mice contained numerous structures that resemble hyaline inclusions (Fig. 24.5(a)) and are fibrillar in nature as revealed by staining with Thioflavine-S (Fig. 24.5(b)). Similar structures were visible in paralyzed mice expressing the G37R, G85R, and G93A variants of mutant SOD1 (Wang et al., 2002b). In common with these other mouse models of ALS (Watanabe et al., 2001), the inclusions found in the spinal cords of paralyzed H46R/H48Q mice stain with antibodies to ubiquitin (Fig. 24.5(c) and (d )). Collectively, these data provide additional evidence to support the notion that the toxicity of mutant SOD1 may not involve copper-mediated chemical reactions. As is the case in the polyglutamine disorders, the role of mutant SOD1 aggregation in the pathogenesis of familial

ALS has been a controversial issue. In mice expressing the G37R and G93A variants of mutant SOD1, immunoreactive SOD1 inclusions have not been described as a prominent feature (Dal Canto & Gurney, 1995; Gurney et al., 1994; Wong et al., 1995b). We have recently described an alternative method for detecting aggregates of mutant SOD1 in tissues that involves the filtration of tissue homogenates through cellulose acetate membranes (Wang et al., 2002a); a technique first developed to study aggregates of mutant huntingtin (Scherzinger et al., 1997, 1999). Using this approach, we have become convinced that aggregation of mutant SOD1 is a common feature of each of the models. Moreover, we have demonstrated that the levels of filterable SOD1 aggregates correlate well with disease severity in the mice (Wang et al., 2002a). Thus, there is growing evidence that the aggregation of mutant SOD1 may play a role in disease pathogenesis. It is noteworthy that several other groups have produced animal models of motor neuron degeneration by expressing other abundant cytosolic proteins; in some cases the outcome was unexpected. Two groups generated transgenic mice that develop some of the features of motor neuron disease by over-expressing neurofilament proteins (NF-H, NF-M, or NF-L). Mice over-expressing mouse NF-L develop hindlimb weakness and muscle atrophy, characteristic of ALS (Xu et al., 1993). Similarly, over-expressing

541

542


human NF-H causes disturbances in the axonal transport of neurofilament, tubulin, and actin, and motorneuronopathy (Collard et al., 1995; Côté et al., 1993). Interestingly, mice expressing high levels of NF-M develop neurofilamentous inclusions in motor neurons that resemble those found in human ALS without obvious consequence on motor neuron function (Wong et al., 1995a). Notably, however, significant motor neuron death does not occur in any of the models generated by the over-expression of wild-type NF proteins. By contrast, the expression of an experimental mutant NF results in a classic ALS-like phenotype, characterized by abundant neurofilamentous inclusions in spinal cord motor neurons (Lee et al., 1994). Motor neuron disease has also been induced in transgenic mice by overexpressing an abundant intermediate filament protein, termed peripherin (Beaulieu et al., 1999). Similarly, death of motor neurons and prominent inclusion pathology has been described in transgenic mice that over-express wildtype tau and tau encoding mutations linked to fronto temporal dementia with Parkinson’s (FTDP) (Götz et al., 2001a,b; Ishihara et al., 1999; Probst et al., 2000; Spittaels et al., 1999; Lewis et al., 2000). Most recently, inclusion pathology and motor neuron degeneration has been observed in trans genic mice expressing high levels of synuclein with a mutation linked to familial Parkinson’s disease (Giasson et al., 2002; Lee et al., 2002). Importantly, the transgene expression constructs utilized in both the tau and -synuclein mice used a pan-neuronal promoter (MoPrP.Xho), and yet pathology was most prominent in motor neurons of the brainstem and spinal cord. Moreover, a common theme in the SOD1, NF, tau, and -synuclein mice is that inclusions often appear to occur in the axonal projections of motor neurons. One potential explanation for these outcomes is that these inclusions cause disturbances in the trafficking of membrane and cytosolic proteins in the axon, and that motor neurons are particularly vulnerable to these kinds of insults. In favor of this view is the recent demonstration of motor neuron disease in mice over-expressing dynamitin (p50), a subunit of the dynein/dynactin molecular motor, which functions in retrograde axonal transport (LaMonte et al., 2002). Dynamitin over-expression has been shown to disrupt the dynein/dynactin interaction and disable the molecular motor. Given the frequency with which motor neuron disease can be induced in transgenic mice by over-expressing abundant proteins that can form inclusion bodies (e.g. tau and -synuclein), and the recent demonstration of abundant SOD1 aggregates in the SOD1ALS mouse models (Wang et al., 2002b), it appears that the propensity of mutant SOD1 to aggregate plays a significant role in the toxicity of the protein. Coupling these observations with those of others involving mice that express a

variety of proteins, it can be argued strongly that the aggregation of proteins within axons of motor neurons is likely to play a very important role in the pathogenesis of motor neuron disease.

Alzheimer’s disease AD is the most common type of progressive dementia in the elderly. Neuropathologically, AD is characterized by neurofibrillary tangles, reactive astrocytosis, activation of microglial cells, and parenchymal deposits of amyloid (for review see Terry & Katzman, 1983a,b) (Chapter 9). Among these lesions, the extracellular deposition of amyloid, composed of a 4 kDa peptide termed A, and the intracellular accumulation of neurofibrillary tangles, composed of hyperphosphorylated tau, are the two diagnostic pathologies of AD. Most cases of AD are classified as sporadic disease; no definable aetiologic factor. However, in rare cases, the disease is inherited in an autosomal dominant fashion, and traditional molecular genetics have been used to identify the gene mutations responsible for disease; amyloid precursor protein (APP), presenilin 1 (PS1), and presenilin 2 (PS2) (Chapter 6). Model systems, including transgenic and knock-out mice, have been extraordinarily valuable tools in deciphering how mutations in these genes may cause disease. Over 15 years ago, two groups of investigators determined the sequence of a 40 residue peptide, termed amyloid or A, which had been purified from amyloid plaques (Glenner & Wong, 1984; Masters et al., 1985). Molecular cloning demonstrated that A is derived from a larger precursor protein designated the amyloid precursor protein (APP) (Masters et al., 1985; Weidemann et al., 1989). Subsequent study of the biology of APP demonstrated that normal cleavage of APP produces multiple species of A peptides (for reviews see Haass & Selkoe, 1993; Younkin, 1995), which differ at their N- and C-termini. The N- and C-termini of A are generated by two endoproteases; the N-termini are generated by BACE-1 (Cai et al., 2001) whereas the C-termini are generated by an entity termed -secretase. The principal peptide species, ranked in order of relative abundance, carry the designations of A11–40, 1–40, 11–42, and 1–42 (Fig. 24.6). In healthy individuals, A1–42 is a relatively rare peptide, however in patients with AD, the levels of A1–42 are increased greatly as is the level of A-1–40 (Saido et al., 1996; Iwatsubo et al., 1994, 1996). Shortly after the discovery of APP, mutations in the app gene were found in familial AD (FAD) kindreds. For the most part, investigators used cultured cell models to determine that one of the consequences of FAD-linked


presenilinase Dibasic endoproteases

ADAM-metalloproteinases

α

β−Secretase (BACE)

Presenilins

γ−Secretase x11/FE65


Lipoprotein-receptor related protein

Aβ

** *

* Aβ11-40

Aβ1-40 Aβ17-40

Aβ11-42

Aβ17-42

Aβ1-42

Fig. 24.6. Processing of amyloid precursor protein. APP is subject to multiple cleavage events by four different endoproteases, which are themselves subject to proteolytic processing. Mutations in APP, linked to early onset FAD (*), occur near the N- and C-termini of the A domain. The N-terminus of A1–40 and 1–42 is generated by BACE-1, which is itself subject to cleavage by dibasic endoproteases to produce mature, active, enzyme. The majority of BACE-1 cleavages occur at residue 11 within A (Cai et al., 2001). Within the A domain, APP may be cleaved by BACE-2 and by members of the ADAM-metalloproteinase family (-secretase cleavage). The C-terminus of A is produced by -secretase, which contains presenilin, cleavage within the transmembrane domain of APP. Presenilin is subject to endoproteolytic cleavage by a putative factor termed presenilinase (identity unknown, possibly self-cleaving). Interactions between APP and lipoprotein receptor related protein (LRP) and FE65 may influence the metabolism and processing of APP.

mutations in APP was to alter its cleavage by either BACE1 or -secretase, with one of the net effects being an increase in the frequency of cleavages that generate A1– 42. Moreover, mutations in the endoproteases themselves have also been linked to FAD. Although presenilin (PS) 1 and 2 were first identified through linkage analysis in kindreds of FAD, it appears that PS1 (and/or 2) may be integral components of the -secretase (Esler et al., 2000; Li et al., 2000; Seiffert et al., 2000; Xia et al., 2000). To date, more than 70 mutations in PS1 (or 2) have been associated with FAD (http://www.alzforum.org/members/resources/ pres mutations/ps1/pres1table.html), and extensive analyses of more than 20 disease-associated variants has shown that -secretases, in the presence of these mutant PS, are more prone to produce peptides with C-termini at residue 42 (Murayama et al., 1999). A number of laboratories have used transgenic mouse approaches to determine the in vivo consequences of expressing mutant APP. Although there are numerous reports on transgenic mice in the literature, for the purposes of this chapter we will focus on those mice that produce pathological features resembling most closely the human disease. The first transgenic mouse model that developed fulminant amyloid pathology was described by Games et al.

in 1995. These mice expressed very high levels of human APP, encoding the V717F mutation linked to familial AD. Shortly thereafter, Hsiao and colleagues described amyloid pathology in mice that express human APP with the so-called Swedish mutation (Hsiao et al., 1996), followed by two other reports on mice expressing APPswe by Sturchler-Pierrat et al., (1997), and by Borchelt et al. (1997). In each of these models, investigators noted abundant amyloid pathology including: argentophilic neuritic deposits of A that are immunoreactive for A1–40, A1– 42, and Apo E; ubiquitin, APP, and PS1-immunoreactive neurites; displacement of dendrites (MAP-2); and activation of astrocytes (Fig. 24.7). Generalized tau pathology as seen in AD in the form of neurofibrillary tangles is not a feature, although neurites adjacent to neuritic deposits often contain accumulations of tau immunoreactivity (Xu et al., 2002a). In addition, the levels of neuronal loss, even in mice with high amyloid burdens, was found to be modest or absent (Calhoun et al., 1998; Irizarry et al., 1997). Mice expressing mutant APP have been extremely valuable animals to probe the pathogenesis of amyloid deposition. One of the first important uses of these animals was in studies to define the mechanisms by which mutations in presenilin 1 and 2 may cause early-onset AD.

543

544


A

B

Aβ1-40

C

D

APP

E

hu PS1

F

GFAP

G

Ubq

H

MAP-2

Apo E

Fig. 24.7. Architecture of amyloid deposits in mice co-expressing APPswe and mutant PS1. The images shown depict neuritic amyloid deposits that occur in the cortex and hippocampus of mice that co-express Mo/HuAPPswe and mutant PS1 (A246E or dE9 variants). (a) High power view of deposits revealed by silver stain. (b)–(h) High power views of deposits revealed by various immunostains (antibodies noted in Figure). MAP-2 is a protein associated with microtubules that is exclusively localized to dendritic compartments of the neuron. Note the exclusion of dendrites from neuritic deposits of A.

Presenilin-1, discovered by classic genetic linkage to earlyonset AD, was a gene of unknown function bearing some homology to a protein in C.elegans that had been implicated in the Notch-1 differentiation pathway (Sherrington et al., 1995, 1996). Presenilins also bore homology to a protein implicated as a potential regulator of apoptosis (Vito et al., 1996). Several laboratories produced strains of transgenic mice that expressed different disease-linked variants of PS1, noting that expression of these mutant proteins was not, per se, particularly deleterious to the murine nervous system (Borchelt et al., 1996b; Thinakaran et al., 1996; Duff et al., 1996; Citron et al., 1997; Dewachter et al., 2000; Janus et al., 2000; Lamb et al., 1999). However, several of these groups demonstrated that co-expressing mutant PS1 with APP led to a dramatic acceleration in the rate of amyloid deposition, which correlated with a selective increase in the levels of A1–42. Couple these observations with data demonstrating that A1–42 is a primary component of AD plaques (Roher et al., 1993; Iwatsubo et al., 1994; Gravina et al., 1995; Lemere et al., 1996), and with in vitro studies demonstrating that A1–42 more rapidly aggregates into amyloid fibrils (Jarrett et al., 1993; Burdick et al., 1992), and it becomes increasingly clear that one of the pathways to AD involves the production and deposition of A1–42. Transgenic mice expressing mutant APP have also been extremely useful in studies to examine the mechanisms behind the prevalence of the ε4 allele of Apolipoprotein E (ApoE) in patients with AD (for review see Roses, 1996). Apolipoprotein E is a polymorphic gene with three common allelic variants, differing at 2 amino acid residues. Although the ε3 allele is by far the most common allele in the human population, the ε4 allele is more commonly found in patients developing AD. Multiple laboratories have reported that individuals with either AD or Down’s syndrome, who inherit an Apo ε4 allele, show a more rapid onset of dementia and an increased amyloid burden (Schmechel et al., 1993; Nagy et al., 1995; Ohm et al., 1995; Gearing et al., 1996; Hyman et al., 1996). Most of the increase in amyloid burden appears to be caused by elevations in levels of A1– 40 peptides (Gearing et al., 1996; Ishii et al., 1997; Mann et al., 1997). In contrast to Apo ε4, the presence of an Apo ε2 allele appears to provide some protection from amyloid deposition; amyloid burden is diminished in individuals with AD who inherit Apo ε2 alleles (Lippa et al., 1997; Nagy et al., 1995). To examine how Apo E may influence amyloid deposition, investigators have used genetically modified mice extensively. Drs Bales, Holtzman, and coworkers have created mice transgenic for APP (V717F) and co-expressing either no (Apo E -/-), endogenous mouse (wild-type), or


human Apo E (driven by the GFAP promoter on an APOE -/-background) to examine the effect of manipulating Apo E expression on A deposition in a physiological setting (Bales et al., 1997, 1999; Holtzman et al., 1999, 2000). Ablation of APOE from the V717F APP transgenic mice reduced A deposition, prevented the appearance of neuritic, thioflavine-S positive, plaques, and shifted the localization of diffuse plaques within the hippocampus when compared to transgenic (V717F) mice expressing wild-type Apo E. Replacing endogenous mouse Apo E with a human transgene encoding either APOE ε3 or ε4 restored the appearance of neuritic plaques, and returned the distribution of diffuse plaques within the hippocampus to the pattern seen in the presence of mouse Apo E (Holtzman et al., 2000). Thus, there is ample data to support the idea that some aspect of Apo E function plays a critical role in modulating the character and distribution of A deposits in transgenic mouse models. Although the data supporting the notion that Apo E, in some manner, modulates A deposition are compelling, the studies in transgenic mice have not delineated clearly whether the different allelic variants of Apo E differ qualitatively in their ability to modulate A deposition. Holtzman and colleagues noted a 10-fold greater density of A deposits in mice expressing human Apo E4 than in mice transgenic for human Apo E3 (Holtzman et al., 2000). Moreover, when examined at 15 months of age, the Apo E4-expressing mice displayed marked neocortical A deposition that was not observed in the ε3-transgenic mice. However, the studies of Holtzman and colleagues focused exclusively on the effects of human Apo E proteins in mice with targeted deletions of the endogenous Apo E genes. To ask whether Apo E, in a dose-dependent fashion, influences amyloid deposition, we produced transgenic mice that express human Apo E4 at high levels (> 5–10-fold over endogenous levels) in the brains of transgenic mice. In these animals, human Apo E4 is expressed in both neurons and astrocytes, but studies in primary cell culture demonstrated that the protein is efficiently secreted only from astrocytes. The level of human Apo E4 secreted from astrocytes was estimated to be at least five times the normal level and could be much higher. These Apo E4 transgenic mice were mated to mice that express mutant APP (APPswe) and to mice that co-express APPswe with mutant PS1 (PS1dE9) to examine whether human Apo E4 can in a dominant and dose-dependent fashion alter the rate, character, or distribution of A deposition. Surprisingly, we found that the presence of Apo E4 had no effect on any parameter of A deposition (overall amyloid burden, numbers of neuritic deposits, and distribution of deposited amyloid) (Lesuisse et al., 2001). One interpretation of these data would hold

that the mechanism by which Apo E4 modulates A deposition is not augmented by hyper-expression. An alternative interpretation, and a hypothesis that is testable, is that Apo E4 is not a promoter of A deposition, but rather Apo E2 and E3 are inhibitors. Transgenic and knock-out mice have also been used to test the hypothesis that Apo E may interact with tau in a manner that promotes the formation of neurofibrillary tangles; a hypothesis that was put forth shortly after APOE ε4 alleles were associated with increased risk of AD (Strittmatter et al., 1994a,b). One recent study demonstrated that in the FVB/N strain of mice, the expression of human Apo E4, via a neuron specific enolase promoter, leads to an increase in the levels of phosphorylated tau in mice aged > 18 months (Tesseur et al., 2000). To study whether the presence of human Apo E4, in the context of relatively high amyloid burden, leads to changes in tau metabolism, Drs Lesuisse, Xu and colleagues examined mice that co-express APPswe, PS1dE9, and Apo E4. Mice expressing all three proteins showed no greater level of tau phosphorylation than controls expressing other combinations of transgenes or APPswe and PS1dE9 alone. To analyse further whether the expression of Apo E4 leads to an increase in tau aggregation, Dr Xu utilized the cellulose filtration assay described above. In the brains of AD patients, aggregates of A, tau, synuclein, and other proteins can be ‘trapped’ by cellulose acetate filtration (Fig. 24.8(a)) (Xu et al., 2002b). Similarly, aggregates of A in the brains of APPswe/PS1dE9/Apo E4 transgenic mice are readily retained on cellulose acetate filters, whereas there was no evidence of aggregated forms of tau in these mice (Fig. 24.8(b)) (Lesuisse et al., 2001). Thus, unlike the studies of presenilin and APP in transgenic mice where multiple groups reported consistent findings of increased amyloid deposition in the presence of mutant presenilin, the outcomes of studies on Apo E in mice have been less consistent and thus whether, and how, Apo E modulates amyloid and neurofibrillary pathology in AD remains unclear. Similar to the work on Apo E, knockout mice have been extremely useful in studies to determine the role of a protein termed BACE-1 in the processing of APP. BACE-1 (amyloid converting enzyme-1) was first identified in an expression screen as an activity that could cleave APP to generate the N-terminus of the A peptide (Vassar et al., 1999). To determine whether this activity was the principal player in the processing of APP, two groups of investigators produced mice with targeted deletions of BACE-1 (Cai et al., 2001; Luo et al., 2001). Mice lacking BACE-1 develop normally and show no overt phenotypes. Analyses of primary neuronal cultures from these mice demonstrated dramatic reductions in the levels of A 1–40 and 1–42, data

545

the roles of specific gene products in the pathogenesis of AD.

NF-H (SMI 31)

α-synuclein

PrP(3F4)

Tau (PHF-1)

AA

APP (22C11)


Aβ (6E10)

546

Synuclein- and tau-opathies

AD with 1645 ith NFTs 1647 ith with AD few NFTs1407 1653 ith with AD 1578 Lewyies bodies 1281 1613 Controls 1577

B +- +- +- +- ++ +- ++ ++ - - + - + ++ 1 2 3 4 5 6 7 8

APPswe PS1dE9 Apo E4 (11-12 mo of age) Aβ mAb 6E10 tau mAb AD C PHF-1

Fig. 24.8. Filter-trap of protein aggregates in the brains of AD patients and transgenic mouse models. (a) Homogenates of brain tissue from AD patients with differing levels of tau pathology, or with the Lewy body variant of AD, dispersed in buffer with detergent and filtered through cellulose acetate membranes (0.22 m) as previously described (Wang et al., 2002a). Aggregated forms of protein are retained on the filters and then revealed by immunonstaining. The antibody used detect the protein aggregated is noted on the figure. Reprinted from Alzheimer’s Disease and Associated Disorders, Vol. 16 (2002) (Xu et al., 2002a) with permission from Lippincott, Williams and Wilkins. (b) Brain tissues from transengic mice co-expressing APPswe, PS1dE9, and Apo E4 were homogenized in detergents and filtered as previously described (Lesuisse et al., 2001). Mice co-expressing APPswe and PS1dE9 contain aggregated forms of A that can be retained on these filters. Aggregated forms of tau are not present, even in mice that co-express human Apo E4. Reprinted from Human Molecular Genetics, Vol. 10 (2001) (Lesuisse et al., 2001).

suggesting that BACE-1 is an extremely good target for therapeutic intervention. As additional genes are identified, either by linkage analysis to familial cases of AD, by genomic approaches (such as gene profiling), or by proteomic approaches (such as protein-interaction assays), genetically modified mice will continue to play important roles in clarifying

The tau- and -synuclein-opathies are a diverse set of degenerative disorders, which are characterized by inclusion body pathology in multiple populations of neurons (Chapters 11, 12 and 15). Cytoplasmic aggregates of tau, in both neurons and glia, are found in multiple disorders, including AD, fronto-temporal dementia with Parkinson’s (FTDP), progressive supranuclear palsy, and corticobasal degeneration. With -synuclein, inclusions (in the form of Lewy bodies and Lewy neurites) are a prevalent pathology in Parkinson’s disease, Lewy-body Dementia, multisystem atrophy, and a variant of AD. In a subset of these illnesses, mutations in the culprit proteins have been associated with disease. FTDP is largely a genetic, autosomal dominant, disorder that is caused by mutations in tau (Hong et al., 1998; Hutton et al., 1998; Poorkaj et al., 1998; Spillantini et al., 1998a). Parkinson’s disease, which occurs most frequently as a sporadic disease, also can show autosomal dominant patterns of inheritance and some of these have been linked to mutations in -synuclein (Polymeropoulos et al., 1997). Interestingly, a fragment of -synuclein had been identified previously as a non-amyloid component of senile plaques in Alzheimer’s patients (Ueda et al., 1993; Chen et al., 1995). With the identification of mutations in synuclein as a cause of Parkinson’s disease, there came increased interest in this protein and the rapid realization that one of the pathognomonic lesions of Parkinson’s disease (the Lewy body) is composed of aggregated forms of -synuclein (Spillantini et al., 1997, 1998b,c, Baba et al., 1998; Trojanowski & Lee, 1998; Mezey et al., 1998). To model the tau- and synuclein-opathies, several laboratories have generated transgenic mice that express wild-type or mutant versions of these proteins. The outcomes of these studies have revealed new linkages between these proteins and motor neuron disease. Mice expressing high levels of wild-type or mutant tau develop neurofibrillary pathology in neurons of the brainstem and spinal cord, and develop an ALS-like phenotype that includes muscle atrophy and paralysis (Spittaels et al., 1999; Ishihara et al., 1999, 2001; Probst et al., 2000; Lewis et al., 2000). Although affected neurons contain structures resembling neurofibrillary tangles, the distribution of these lesions within the CNS is not what it is to be expected for FTDP. For the most part, the pathology is restricted to the ventral midbrain, the brainstem, and the spinal cord despite the use of panneuronal promoter elements to drive the expression of the transgene. Thus, it would appear that motor neurons are


Fig. 24.9. Pathological and biochemical characteristics of transgenic mice expressing wild-type human tau. (a) Spinal cord of a 6-month-old tau transgenic mouse shows axonal inclusions that are immunolabelled with monoclonal anti-human tau antibody T14. (b) High-power view of the tau immunoreactive spheroid indicated by the arrow in (a). (c) Immunostaining of spheroidal inclusions (arrow) with polyclonal antibodies against alpha-tubulin. (d )–( f ) Triple immunofluorescence staining demonstrates that the T14-positive inclusion (d ) is also stained with polyclonal anti-NFL antibodies (e) and monoclonal anti-NFH antibody DP1 ( f ). (g) Transmission EM image shows that the intra-axonal inclusion (arrow) consists of randomly oriented 10–20 nm straight filaments. Arrowhead indicates a degenerating organelle. The section was generated using spinal cord of a 6-month-old tau transgenic mouse. (h) High-power view of the filamentous inclusion indicated by the arrow in (g). Mitochondria are contained at the centre. (I) The inclusions are shown to contain tau filaments by post-embedding immuno-EM with the anti-tau antibody T14. Images taken from Trojanowski et al., 2002. Reprinted from Experimental Neurology Vol. 176 (2002), with permission from Elsevier.

particularly prone to mis-handle tau in a manner that induces the formation of cytoplasmic aggregates. Moreover, motor neurons appear to be particularly sensitive to the toxicity that results from mis-folding/aggregation of tau. Recent work by Ishihara and colleagues (Ishihara et al. 2002) has demonstrated the similarity between the pathology in mice expressing tau and forms of ALS with Parkinson’s that occur in Guam (Fig. 24.9) (for review see Trojanowski et al., 2002). However, in the more common forms of ALS, tau

pathology has not been recognized as a major component of the disease. To create models of Parkinson’s disease, several groups have created mice that express wild-type and mutant -synuclein. Early studies focused on mice that express high levels of wild-type human -synuclein. Immunocytochemical and histological studies of these mice indicated injury to dopaminergic systems and the formation of inclusions within the nuclei of cortical neurons (Masliah

547

548


Fig. 24.10. Synuclein pathology in human PD and transgenic mice expressing the A53T variant of human -synuclein. (a) and (b) -synuclein immunoreactive neurites and inclusions in the entorhinal cortex of a patient harboring the A53T mutation. (c)–(m) Transgenic mice expressing human A53T -synuclein (line M83 homozygous) (Giasson et al., 2002). (c)–(e) Inclusions in the spinal cords of mice expressing human A53T -synuclein immunostained with antibodies to neurofilament (c), ubiquitin (d ) and (e). ( f ) GFAP immunostains of spinal cord in affected mice. (g) and (h) Inclusions in the pons stained with antibodies to nitrated -synuclein (nSyn 823 antibody). (i)–(k) Silver stained sections from spinal cords of affected mice. (l)–(m) Thioflavine-S stained spinal cords.

et al., 2000). However, in mice where -synuclein was expressed via promoter elements derived from tyrosine hydroxylase, a system that is designed to target expression to dopaminergic neurons, there were no changes in dopaminergic markers (Matsuoka et al., 2001). Recently, two groups have reported severe pathology and motor neuron disease in mice expressing the A53T variant of -synuclein (linked to Parkinson’s disease) (Giasson et al., 2002; Lee et al., 2002). In affected spinal cords of these mice immunostains with antibodies to -synuclein recognized cytoplasmic inclusion bodies that were strikingly similar to structures found in patients with the same mutation (Fig. 24.10) (Giasson et al., 2002). It is important to note that the motor neuron phenotypes occurred in lines of mice expressing the transgene via a pan-neuronal promoter. Hence, as was observed

in mice expressing tau, it is clear that the motor neurons of mice are particularly sensitive to mis-folded, and/or aggregated, forms of tau and -synuclein. As mentioned above, the pathology of AD includes both extracellular aggregates of A and intracellular aggregates of tau. In a subset of Alzheimer’s cases, the deposition of A is accompanied by the appearance of Lewy bodies in addition to, or instead of, neurofibrillary tangles. To model this aspect of AD in mice, investigators have crossed mice that express mutant APP (and which develop amyloid deposits) to mice expressing mutant tau. In mice developing cortical and hippocampal amyloid pathology, structures resembling neurofibrillary tangles were also found (Lewis et al., 2001), indicating that the metabolism of tau is altered in cortical and


hippocampal neurons exposed to extracellular deposits of amyloid. Similar conclusions were drawn when it was found that injecting aggregated forms of A into the brains of mice expressing mutant tau also induced cortical/hippocampal neurofibrillary pathology (Götz et al., 2001a,b). Synergy between extracellular amyloid pathology and intracellular protein aggregates has also been observed in crosses of mice that express mutant APP and wild-type -synuclein. In this latter case, the accumulation of aggregated forms of -synuclein in cortical neurons was augmented and altered to produce a greater degree of fibrillar structures, still within nuclei. Based on studies with transgenic and knockout mice, a putative roadmap for AD pathogenesis can be drawn (Fig. 24.11). Genes with critical roles in the processing of APP are depicted in red and are targets of mutations that cause early onset FAD. In yellow are genes whose products are found within pathological structures in the brains of AD patients. Mice developing amyloid deposits, alone, show only modest levels of neurodegeneration (Calhoun et al., 1998; Irizarry et al., 1997). By contrast, mice expressing mutant or wild-type tau and -synuclein show striking levels of neuronal degeneration, albeit in populations of neurons not expected for the human disease (see below). Nevertheless, these data demonstrate a critical role for cytoplasmic pathology in neuronal death, and based on studies where amyloid pathology is combined with the expression of mutant tau, one can hypothesize that amyloid is a critical initiator of disease while events within the cytoplasm, impacting of the metabolism of tau and/or -synuclein, are critical in neuronal dysfunction and death.

Ubiquitin in neurodegenerative disease In AD, amyotrophic lateral sclerosis (ALS), and other neurodegenerative disorders, an accumulation of ubiquitin immunoreactive material (usually in the form of cytoplasmic inclusions) is a common pathologic feature (for reviews see Leroy et al., 1998; Lowe et al., 1993; Manetto et al., 1988; Pines, 1994). The nuclear inclusions, evident in affected neurons of HD and DRPLA cases, are often immunoreactive to ubiquitin antibodies (Becher et al., 1998). In AD, the accumulation of ubiquitin immunoreactivity has been viewed as a hallmark lesion. Ubiquitin is a pivotal regulatory moiety in protein metabolism (for reviews see Hershko, 1988; Ciechanover, 1994). Through a series of enzymatic reactions, ubiquitin is covalently linked to lysine residues within the target protein. Once the initial ubiquitin is bound, other ligases catalyse the covalent linkage

549

Presenilins β−Secretase (BACE)

γ−Secretase


Apo E

Aβ Aβ 1-42

Amyloid Deposits

α− synuclein

Lewy Bodies

Tau Neurofibrillary Tangles

Neuronal Dysfunction/Death Fig. 24.11. Alzheimer’s disease: a genetic roadmap.

of additional ubiquitin chains (to the initial ubiquitin), leading to a polyubiquitin chain attached to the protein substrate. This polyubiquitin chain is then recognized by the proteasome, which is responsible for the hydrolysis of the targeted protein. This process of ubiquitination and chain elongation is in equilibrium with de-ubiquitination by ubiquitin isopeptidases. The ubiquitin/proteasome system is well conserved from yeast to mammals and is highly evolved as there are multiple isoforms of ubiquitin ligase and hydrolase. In addition to ubiquitin, a host of ubiquitin homologues have been described (for reviews see Weissman, 2001; Muller et al., 2001; Ohsumi, 2001). One of these homologues is called SUMO, which also exists in multiple isoforms (for review see Muller et al., 2001). At least one form of SUMO appears to act as a dominant inhibitor of polyubiquitination and may serve to stabilize proteins (for review see Muller et al., 2001). Importantly for human diseases, is the observation that in normal cells and tissues polyubiquitinated proteins do not normally accumulate as these molecules appear to be rapidly degraded by the proteasome with the polyubiquitin chains rapidly recycled into monomeric ubiquitin by isopeptidases. Hence, the accumulation of ubiquitin immunoreactivity in specific populations of neurons in neurodegenerative disease can be construed as evidence for a disturbance in the ubiquitin/proteasome system in these cells. The ubiquitin pathology that is prevalent in human disease is recapitulated in transgenic mouse models of AD, ALS, FTDP, and -synculeinopathy. For example, in

550


transgenic mice that co-express APPswe and PS1dE9, accumulations of ubiquitin immunoreactivity occur adjacent to neuritic plaques (see Fig. 24.7). Similarly, in the mouse models of FALS, ubiquitin immunoreactive inclusions are a prominent pathologic feature (Bruijn et al., 1997; Wang et al., 2002a). Ubiquitin immunoreactive spheroids have been described in mice expressing the A53T variant of synuclein (Giasson et al., 2002). The NFT-like structures seen in mice expressing mutant tau are often immunoreactive with ubiquitin antibodies (Lewis et al., 2000). Collectively, these data support the idea that a dysfunction of the ubiquitin/proteasome pathway could play a role in the pathogenesis of neurodegenerative disease. Perhaps the strongest evidence that disturbances in the ubiquitin/proteasome system could play a pivotal role in neurodegeneration comes from the analysis of a rare, juvenile, recessive form of Parkinson’s disease. Disabling mutations in a gene termed parkin have been associated with this rare form of Parkinson’s disease and studies of the product of the parkin gene have revealed its function as a ubiquitin ligase; a crucial enzyme in targeting the degradation of proteins (Zhang et al., 2000). Among the substrates for parkin are a novel form of -synuclein (Shimura et al., 2001) and a protein that interacts with -synuclein called synphilin (Chung et al., 2001), which has been found in Lewy body lesions (Engelender et al., 1999; Wakabayashi et al., 2000). Although not a ‘classic’ neurodegenerative disease, a disorder call Angelman’s syndrome is another example of a neurologic illness that involves the loss of a ubiquitin ligase through mutation (Kishino et al., 1997; Malzac et al., 1998). In addition to these neurological diseases, there are multiple examples of systemic disease that arise from the mutation or loss of specific components of the ubiquitin/proteasome system (for reviews see Schwartz & Ciechanover, 1999; Ciechanover & Schwartz, 2002; Glickman & Ciechanover, 2002). Based on these important examples, it is anticipated that the ubiquitin/proteasome system will continue to be a target of intense study in neurodegenerative settings.

Concluding remarks For the most part, it can be said that modelling human neurodegenerative diseases in genetically modified mice has provided a wealth of new information about the pathogenesis of these largely incurable human illnesses. In many cases, these animals have been useful in studies to test new therapeutic strategies, in addition to their use in studies to examine pathological processes. It is noteworthy that, in those cases where the culprit proteins have been

implicated in diverse neurological conditions (e.g. tau and -synuclein), the transgenic mouse models have produced unexpected distributions of pathology. By contrast, in cases where mutations in a single protein have been implicated in a specific disease (e.g. superoxide dismutase 1, amyloid precursor protein, presenilins, and polyglutamine proteins) the models have been much more faithful to the human pathology. It is interesting, however, to note that the extent of neuronal cell death (a key feature of the human diseases) is highly variable in these models. Mice expressing mutant superoxide dismutase 1 show patterns of neuronal death similar to what is found in familial ALS. Mice expressing mutant APP and mice co-expressing presenilin with APP develop amyloid pathology in cortical and hippocampal brain regions as expected, but do not progress to develop significant losses of neurons in the disease-relevant populations. Similarly, the extent of neuronal loss in the striatum of mice expressing mutant huntingtin is far less than what is observed in patients, even in mice with severe behavioral disturbances (Mangiarini et al., 1996; Schilling et al., 1999a). Thus, in some cases it is possible to study all components of the pathologic process, up to the point of cell death, in mouse models, whereas in other cases the study may be limited to the study of specific lesions such as amyloid deposition. Hence, it is important to recognize the limitations of some of these models in relating outcomes to processes that may occur in the human illness.

REFERENCES Aguzzi, A. (2001). Peripheral prion pursuit. J Clin Invest, 108: 661–2. Aguzzi, A., Klein, M. A., Montrasio, F. et al. (2000). Prions: pathogenesis and reverse genetics. Ann N Y Acad Sci, 920: 140–57. Alpers, M., Gajdusek, D. C. (1965). Changing patterns of kuru: epidemiological changes in the period of increasing contact of the Fore people with western civilization. Am J Trop Med Hyg, 14: 852–79. Baba, M., Nakajo, S. & Tu, P. H. (1998). Aggregation of -synuclein in lewy bodies of sporadic Parkinson’s disease and dementia with lewy bodies. Am J Pathol, 152: 879–84. Bales, K. R., Verina, T., Dodel, R.C. et al. (1997). Lack of apolipoprotein E dramatically reduces amyloid-peptide deposition. Nature Genet, 17: 263–4. Bales, K. R., Verina, T., Cummins, D. J. et al. (1999). Apoplipoprotein E is essential for amyloid deposition in the APPV717F transgenic mouse model of Alzheimer’s disease. Proc Natl Acad Sci USA, 96: 15233–8. Beaulieu, J. M., Nguyen, M. D. & Julien, J. P. (1999). Late onset death of motor neurons in mice overexpressing wild-type peripherin. J Cell Biol, 147: 531–44.


Becher, M. W., Kotzuk, J. A., Sharp, A. H. et al. (1998). Intranuclear neuronal inclusions in Huntington’s disease and dentatorubral and pallidoluysian atrophy: correlation between the density of inclusions and IT15 CAG triplet repeat length. Neurobiol Dis, 4: 387–97. Beckman, J. S., Carson, M., Smith, C. D. & Koppenol, W. H. (1993). ALS, SOD and peroxynitrite. Nature, 364: 584. Bendheim, P. E., Barry, R. A., DeArmond, S. J., Stites, D. P. & Prusiner, S. B. (1984). Antibodies to a scrapie prion protein. Nature, 310: 418–21. Bendheim, P. E., Bockman, J. M., McKinley, M. P., Kingsbury, D. T. & Prusiner, S. B. (1985). Scrapie and Creutzfeldt–Jakob disease prion proteins share physical properties and antigenic determinants. Proc Natl Acad Sci USA, 82: 997–1001. Borchelt, D. R., Lee, M. K. & Slunt, H. H. et al. (1994). Superoxide dismutase 1 with mutations linked to familial amyotrophic lateral sclerosis possesses significant activity. Proc Natl Acad Sci USA, 91: 8292–6. Borchelt, D. R., Guarnieri, M., Wong, P. C. et al. (1995). Superoxide dismutase 1 subunits with mutations linked to familial amyotrophic lateral sclerosis do not affect wild-type subunit function. J Biol Chem, 270: 3234–8. Borchelt, D. R., Davis, J., Fischer, M. et al. (1996a). A vector for expressing foreign genes in the brains and hearts of transgenic mice. Genet Anal (Biomed Eng), 13: 159–63. Borchelt, D. R., Thinakaran, G., Eckman, C. B. et al. (1996b). Familial Alzheimer’s disease-linked presenilin 1 variants elevate A1– 42/1–40 ratio in vitro and in vivo. Neuron, 17: 1005–13. Borchelt, D. R., Ratovitski, T., Van Lare, J. et al. (1997). Accelerated amyloid deposition in the brains of transgenic mice coexpressing mutant presenilin 1 and amyloid precursor proteins. Neuron, 19: 939–45. Bruijn, L. I., Becher, M. W., Lee, M. K. et al. (1997). ALS-linked SOD1 mutant G85R mediates damage to astrocytes and promotes rapidly progressive disease with SOD1-containing inclusions. Neuron, 18: 327–38. Bruijn, L. I., Houseweart, M. K., Kato, S. et al. (1998). Aggregation and motor neuron toxicity of an ALS-linked SOD1 mutant independent from wild-type SOD1. Science, 281: 1851–1854. ¨ Bueler, H., Aguzzi, A., Sailer, A. et al. (1993). Mice devoid of PrP are resistant to scrapie. Cell, 73: 1339–47. Bugiani, O., Giaccone, G., Piccardo, P., Morbin, M., Tagliavini, F. & Ghetti, B. (2000). Neuropathology of Gerstmann–Straussler– Scheinker disease. Microsc Res Tech, 50: 10–15. Burdick, D., Soreghan, B., Kwon, M. et al. (1992). Assembly and aggregation properties of synthetic Alzheimer’s A4/ amyloid peptide analogs. J Biol Chem, 267: 546–54. Cai, H., Wang, Y., McCarthy, D. et al. (2001). BACE1 is the major secretase for generation of A peptides by neurons. Nat Neurosci, 4: 233–4. Calhoun, M. E., Wiederhold, K. H., Abramowski, D. et al. (1998). Neuron loss in APP transgenic mice. Nature, 395: 755–6. Chai, Y., Koppenhafer, S. L., Bonini, N. M. & Paulson, H. L. (1999a). Analysis of the role of heat shock protein (Hsp) molecular chaperones in polyglutamine disease. J Neurosci, 19: 10338–47.

Chai, Y., Koppenhafer, S. L., Shoesmith, S. J., Perez, M. K., Paulson, H. L. (1999b). Evidence for proteasome involvement in polyglutamine disease: localization to nuclear inclusions in SCA3/MJD and suppression of polyglutamine aggregation in vitro. Hum Mol Genet, 8: 673–82. Chandler, R. L. (1967). Cytopathology of scrapie in the rat: an electron microscopie study of thalamic and hippocampal areas. Res Vet Sci, 8: 98–102. Chen, X., de Silva, H. A. R., Pettenati, M. J. et al. (1995). The human NACP/-synuclein gene: chromosome assignment to 4q21.3q22 and TaqI RFLP analysis. Genomics, 26: 425–7. Chiesa, R. & Harris, D. A. (2001). Prion diseases: what is the neurotoxic molecule? Neurobiol Dis, 8: 743–63. Chiesa, R., Drisaldi, B., Quaglio, E. et al. (2000). Accumulation of protease-resistant prion protein (PrP) and apoptosis of cerebellar granule cells in transgenic mice expressing a PrP insertional mutation. Proc Natl Acad Sci USA, 97: 5574–9. Chiesa, R., Pestronk, A., Schmidt, R. E. et al. (2001). Primary myopathy and accumulation of PrPSc-like molecules in peripheral tissues of transgenic mice expressing a prion protein insertional mutation. Neurobiol Dis, 8: 279–88. Chung, K. K. K., Zhang, Y., Lim, K. L. et al. (2001). Parkin ubiquitinates the -synuclein-interacting protein, synphilin-1: implications for lewy body formation in Parkinson’s disease. Nature Med, 7: 1140–50. Ciechanover, A. (1994). The ubiquitin-proteasome proteolytic pathway. Cell, 79: 13–21. Ciechanover, A. & Schwartz, A. L. (2002). Ubiquitin-mediated degradation of cellular proteins in health and disease. Hepatology, 35: 3–6. Citron, M., Westaway, D., Xia, W. et al. (1997). Mutant presenilins of Alzheimer’s disease increase production of 42-residue amyloid -protein in both transfected cells and transgenic mice. Nature Med, 3: 67–72. Collard, J.-F., Côté, F. & Julien, J.-P. (1995). Defective axonal transport in a transgenic mouse model of amyotrophic lateral sclerosis. Nature, 375: 61–4. Collinge, J. (2001). Prion diseases of humans and animals: their causes and molecular basis. Annu Rev Neurosci, 24: 519–50. Corson, L. B., Culotta, V. C. & Cleveland, D. W. (1998). Chaperonefacilitated copper binding is a property common to several classes of familial amyotrophic lateral sclerosis-linked superoxide dismutase mutants. Proc Natl Acad Sci USA, 95: 6361–6. Côté, F., Collard, J.-F. & Julien, J.-P. (1993). Progressive neuronopathy in transgenic mice expressing the human neurofilament heavy gene: a mouse model of amyotrophic lateral sclerosis. Cell, 73: 35–46. Crow, J. P., Sampson, J. B., Zhuang, Y., Thompson, J. A. & Beckman, J. S. (1997a). Decreased zinc affinity of amyotrophic lateral sclerosis-associated superoxide dismutase mutants leads to enhanced catalysis of tyrosine nitration by peroxynitrite. J Neurochem, 69: 1936–44. Crow, J. P., Ye, Y. Z., Strong, M., Kirk, M., Barnes, S. & Beckman, J. S. (1997b). Superoxide dismutase catalyzes nitration of tyrosines

551

552


by peroxynitrite in the rod and head domains of neurofilamentL. J Neurochem, 69: 1945–53. Culotta, V. C., Klomp, L. W. J., Strain, J., Casareno, R. L. B, Krems, B. & Gitlin, G. D. (1997). The copper chaperone for superoxide dismutase. J Biol Chem, 272: 23469–72. Cummings, C. J., Mancini, M. A., Antalffy, B., DeFranco, D. B., Orr, H. T. & Zoghbi, H. Y. (1998). Chaperone suppression of aggregation and altered subcellular proteasome localization imply protein misfolding in SCA1. Nat Genet, 19: 148–54. Dal Canto, M. C. & Gurney, M. E. (1995). Neuropathological changes in two lines of mice carrying a transgene for mutant human Cu, ZN SOD, and in mice overexpressing wild type human SOD: a model of familial amyotrophic lateral sclerosis (FALS). Brain Res, 676: 25–40. Davies, S. W., Turmaine, M., Cozens, B. A. et al. (1997). Formation of neuronal intranuclear inclusions underlies the neurological dysfunction in mice transgenic for the HD mutation. Cell, 90: 537–48. DeArmond, S. J. & Prusiner, S. B. (1995). Etiology and pathogenesis of prion diseases. Am J Pathol, 146: 785–811. Dewachter, I., van Dorple, J., Spittaels, K. et al. (2000). Modeling Alzheimer’s disease in transgenic mice: effect of age and presenilin 1 on amyloid biochemistry and pathology in APP/London mice. Exp Gerontol, 35: 831–41. DiFiglia, M., Sapp, E., Chase, K. O. et al. (1997). Aggregation of huntingtin in neuronal intranuclear inclusions and dystrophic neurites in brain. Science, 277: 1990–3. Duff, K., Eckman, C., Zehr, C. et al. (1996). Increased amyloid42(43) in brains of mice expressing mutant presenilin 1. Nature, 383: 710–13. Engelender, S., Kaminsky, Z., Guo, X. et al. (1999). Synphilin-1 associates with alpha-synuclein and promotes the formation of cytosolic inclusions. Nat Genet, 22: 110–14. Esler, W. P., Taylor Kimberly, W., Ostaszewski, B. L. et al. (2000). Transition-state analogue inhibitors of -secretase bind directly to presenilin-1. Nature Cell Biol., 2: 1–7. Estévez, A. G., Crow, J. P., Sampson, J. B. et al. (1999). Induction of nitric oxide-dependent apoptosis in motor neurons by zincdeficient superoxide dismutase. Science, 286: 2498–500. Field, E. J. (1967). Invasion of the mouse nervous system by scrapie agent. Br J Exp Pathol, 48: 662–4. Field, E. J., Raine, C. S. & Joyce, G. (1967). Scrapie in the rat: an electron-microscope study. I. Amyloid bodies and deposits. Acta Neuropathol (Berl), 8: 47–56. Fischbeck, K. H. (2001). Polyglutamine expansion neurodegenerative disease. Brain Res Bull, 56: 161–3. Gajdusek, D. C., Gibbs, C. J., Jr. & Alpers, M. (1966). Experimental transmission of a kuru-like syndrome to chimpanzees. Nature, 209: 794–7. Games, D., Adams, D., Alessandrini, R. et al. (1995). Alzheimer-type neuropathology in transgenic mice overexpressing V717F amyloid precursor protein. Nature, 373: 523–7. Gearing, M., Mori, H. & Mirra, S. S. (1996). A-peptide length and apolipoprotein E genotype in Alzheimer’s disease. Ann Neurol, 39: 395–9.

Ghetti, B., Tagliavini, F., Giaccone, G. et al. (1994). Familial Gerstmann–Sträussler–Scheinker disease with neurofibrillary tangles. Mol Neurobiol, 8: 41–8. Giasson, B. I., Duda, J. E., Quinn, S. M., Zhang, B., Trojanowski, J. Q. & Lee, V. M. Y. (2002). Neuronal -synucleinopathy with severe movement disorder in mice expressing A53T human -synuclein. Neuron, 34: 521–33. Gibbs, C. J., Jr., Gajdusek, D. C., Asher, D. M. et al. (1968). Creutzfeldt–Jakob disease (spongiform encephalopathy): transmission to the chimpanzee. Science, 161: 388–9. Giese, A. & Kretzschmar, H. A. (2001). Prion-induced neuronal damage – the mechanisms of neuronal destruction in the subacute spongiform encephalopathies. Curr Top Microbiol Immunol, 253: 203–17. Glenner, G. G. & Wong, C. W. (1984). Alzheimer’s disease: initial report of the purification and characterization of a novel cerebrovascular amyloid protein. Biochem Biophys Res Commun, 120: 885–90. Glickman, M. H. & Ciechanover, A. (2002). The ubiquitinproteasome proteolytic pathway: destruction for the sake of construction. Physiol Rev, 82: 373–428. Götz, J., Chen, F., Van Dorpe, J. & Nitsch, R. M. (2001a). Formation of neurofibrillary tangles in P3011 tau transgenic mice induced by Abeta fibrils. Science, 293: 1491–5. Götz, J., Chen, F., Barmettler, R. & Nitsch, R. M. (2001b). Tau filament formation in transgenic mice expressing P301L tau. J Biol Chem, 276: 529–34. Gravina, S. A., Ho, L., Eckman, C. B. et al. (1995). Amyloid protein (A) in Alzheimer’s disease brain. J Biol Chem, 270: 7013–16. Gurney, M. E., Pu, H., Chiu, A. Y. et al. (1994). Motor neuron degeneration in mice that express a human Cu, Zn superoxide dismutase mutation. Science, 264: 1772–5. Gutekunst, C. A., Li, S. H., Yi, H. et al. (1999). Nuclear and neuropil aggregates in Huntington’s disease: relationship to neuropathology. J Neurosci, 19: 2522–34. Haass, C. & Selkoe, D. J. (1993). Cellular processing of -amyloid precursor and the genesis of amyloid -peptide. Cell, 75: 1039–42. Hershko, A. (1988). Ubiquitin-mediated protein degradation. J Biol Chem, 263: 15237–40. Hirano, A., Malamud, N., Kurland, L. T. & Zimmerman, H. M. (1969). A review of the pathologic findings in amyotrophic lateral sclerosis. In Motor Neuron Diseases. Research on Amyotrophic Lateral Sclerosis and Other Disorders (F. H. Norris, & L. T. Kurland, eds.), pp. 51–60. New York: Grune & Stratton. Hodgson, E. K. & Fridovich, I. (1975). The interaction of bovine erythrocyte superoxide dismutase with hydrogen peroxide: inactivation of the enzyme. Biochemistry, 14: 5294–303. Hodgson, J. G., Agopyan, N., Gutekunst, C. A. et al. (1999). A YAC mouse model for Huntington’s disease with full-length mutant huntingtin, cytoplasmic toxicity, and selective striatal neurodegeneration. Neuron, 23: 181–92. Holtzman, D. M., Bales, K. R., Wu, S. et al. (1999). Expression of human apolipoprotein E reduces amyloid- deposition in a mouse model of Alzheimer’s disease. J Clin Invest, 103: R15–R21.


Holtzman, D. M., Bales, K. R., Tenkova, T. et al. (2000). Apolipoprotein E isoform-dependent amyloid deposition and neuritic degeneration in a mouse model of Alzheimer’s disease. Proc Natl Acad Sci USA, 97: 2892–7. Hong, M., Zhukareva, V., Vogelsberg-Ragaglia, V. et al. (1998). Mutation-specific functional impairments in distinct tau isoforms of hereditary FTDP-17. Science, 282: 1914–17. Hsiao, K., Chapman, P., Nilsen, S. et al. (1996). Correlative memory deficts, A elevation and amyloid plaques in transgenic mice. Science, 274: 99–102. Hsiao, K. K., Scott, M., Foster, D., Groth, D. F., DeArmond, S. J. & Prusiner, S. B. (1990). Spontaneous neurodegeneration in transgenic mice with mutant prion protein. Science, 250: 1587–90. Hsiao, K. K., Cass, C., Schellenberg, G. D. et al. (1991). A prion protein variant in a family with the telencephalic form of Gerstmann– Sträussler–Scheinker syndrome. Neurology, 41: 681–4. Hsiao, K. K., Groth, D., Scott, M. et al. (1994). Serial transmission in rodents in neurodegeneration from transgenic mice expressing mutant prion proteins. Proc Natl Acad Sci USA, 91: 9126–30. Huntington’s Disease Collaborative Research Group (1993). A novel gene containing a trinucleotide repeat that is expanded and unstable on Huntington’s disease chromosomes. Cell, 72: 971–83. Hutton, M., Lendon, C. L., Rizzu, P. et al. (1998). Association of missense and 5 -splice-site mutations in tau with the inherited dementia FTDP-17. Nature, 393: 702–5. Hyman, B. T., Gomez-Isla, T., West, H. et al. (1996). Clinical and neuropathological correlates of apolipoprotein E genotype in Alzheimer’s disease – window on molecular epidemiology. Ann NY Acad Sci, 777: 158–65. Irizarry, M. C., McNamara, M., Fedorchak, K., Hsiao, K. & Hyman, B. T. (1997). AppSw transgenic mice develop age-related A deposits and neuropil abnormalities, but no neuronal loss in CA1. J Neuropathol Exp Neurol, 56: 965–73. Ironside, J. W. (1996). Human prion diseases. J Neural Transm, 103 Suppl. 47: 231–46. Ischiropoulos, H., Zhu, L., Chen, J. et al. (1992). Peroxynitritemediated tyrosine nitration catalyzed by superoxide dismutase. Arch Biochem Biophys, 298: 431–7. Ishihara, T., Hong, M., Zhang, B. et al. (1999). Age-dependent emergence and progression of a tauopathy in transgenic mice overexpressing the shortest human tau isoform. Neuron, 24: 751–62. Ishihara, T., Zhang, B., Higuchi, M., Yoshiyama, Y., Trojanowski, J. Q. & Lee, V. M. (2001). Age-dependent induction of congophilic neurofibrillary tau inclusions in tau transgenic mice. Am J Pathol, 158: 555–62. Ishii, K., Tamaoka, A., Mizusawa, H. et al. (1997). A-1–40 but not A-1–42 levels in cortex correlate with apolipoprotein E epsilon-4 allele dosage in sporadic Alzheimer’s disease. Brain Res, 748: 250–252. Iwatsubo, T., Odaka, A., Suzuki, N., Mizusawa, H., Nukina, N. & Ihara, Y. (1994). Visualization of A42(43)-positive

and A40-positive senile plaques with end-specific Amonoclonal antibodies: evidence that an initially deposited A species is A1–42(43). Neuron, 13: 45–53. Iwatsubo, T., Saido, T. C., Mann, D. M. A., Lee, V. M. Y. & Trojanowski, J. Q. (1996). Full-length amyloid-(1–42(43)) and amino-terminally modified and truncated amyloid-42(43) deposit in diffuse plaques. Am J Pathol, 149: 1823–30. Janus, C., D’Amelio, S., Amitay, O. et al. (2000). Spatial learning in transgenic mice expressing human presenilin 1 (PS1) transgenes. Neurobiol Aging, 21: 541–9. Jarrett, J. T., Berger, E. P. & Lansbury P. T., Jr. (1993). The carboxy terminus of the amyloid protein is critical for the seeding of amyloid formation: implications for the pathogenesis of Alzheimer’s disease. Biochemistry, 32: 4693–7. Johnston, J. A., Dalton, M. J., Gurney, M. E. & Kopito, R. R. (2000). Formation of high molecular weight complexes of mutant Cu, Zn-superoxide dismutase in a mouse model for familial amyotrophic lateral sclerosis. Proc Natl Acad Sci USA, 97: 12571–6. Kawaguchi, Y., Okamoto, T., Taniwaki, M. et al. (1994). CAG expansions in a novel gene for Machado-Joseph disease at chromosome 14q32.1. Nat Genet, 8: 221–7. Kimberlin, R. H. & Millson, G. C. (1967). Some biochemical aspects of mouse scrapie. J Comp Pathol, 77: 359–66. Kishino, T., Lalande, M. & Wagstaff, J. (1997). UBE3A/E6-AP mutations cause Angelman syndrome. Nat Genet, 15: 70–3. Klement, I. A., Skinner, P. J., Kaytor, M. D. et al. (1998). Ataxin-1 nuclear localization and aggregation: role in polyglutamineinduced disease in SCA1 transgenic mice. Cell, 95: 41–53. Koide, R., Ikeuchi, T., Onodera, O. et al. (1994). Unstable expansion of CAG repeat in hereditary dentatorubral-pallidoluysian atrophy (DRPLA). Nat Gene, 6: 9–13. La Spada, A. R., Wilson, E. M., Lubahn, D. B., Harding, A. E. & Fischbeck, K. H. (1991). Androgen receptor gene mutations in Xlinked spinal and bulbar muscular atrophy. Nature, 352: 77–9. Laforet, G. A., Sapp, E. & Chase, K. et al. (2001). Changes in cortical and striatal neurons predict behavioral and electrophysiological abnormalities in a transgenic murine model of Huntington’s disease. J Neurosci, 21: 9112–23. Lamb, B. T., Bardel, K. A., Kulnane, L. S. et al. (1999). Amyloid production and deposition in mutant amyloid precursor protein and presenilin-1 yeast artificial chromosome transgenic mice. Nature Neurosci, 2: 695–697. LaMonte, B. H., Wallace, K. E., Holloway, B. A. et al. (2002). Disruption of dynein/dynactin inhibits axonal transport in motor neurons causing late-onset progressive degeneration. Neuron, 34: 715–27. Lee, M. K., Marszalek, J. R. & Cleveland, D. W. (1994). A mutant neurofilament subunit causes massive, selective motor neuron death: implications for the pathogenesis of human motor neuron disease. Neuron, 13: 975–88. Lee, M. K., Stirling, W., Xu, Y. et al. (2002). Human -synucleinharboring familial Parkinson’s disease-linked Ala-53-Thr mutation causes neurodegenerative disease with -synuclein aggregation in transgenic mice. Proc Natl Acad Sci USA, 99: 8968–73.

553

554


Lemere, C. A., Blusztajn, J. K., Yamaguchi, H., Wisniewski, T., Saido, T. C. & Selkoe, D. J. (1996). Sequence of deposition of heterogeneous amyloid -peptides and APO E in Down syndrome: implications for initial events in amyloid plaque formation. Neurobiol Dis, 3: 16–32. Leroy, E., Boyer, R., Auburger, G. et al. (1998). The ubiquitin pathway in Parkinson’s disease. Nature, 395: 451–2. Lesuisse, C., Xu, G., Anderson, J. et al. (2001). Hyper-expression of human apolipoprotein E4 in astroglia and neurons does not enhance amyloid deposition in transgenic mice. Hum Mol Gen, 10: 2525–37. Levine, M. S., Klapstein, G. J., Koppel, A. et al. (1999). Enhanced sensitivity to N-methyl-d-aspartate receptor activation in transgenic and knockin mouse models of Huntington’s disease. J Neurosci Res, 58: 515–32. Lewis, J., McGowan, E., Rockwood, J. et al. (2000). Neurofibrillary tangles, amyotrophy and progressive disturbance in mice expressing mutant (P301L) tau protein. Nat Genet, 25: 402–5. Lewis, J., Dickson, D. W., Lin, W.-L. et al. (2001). Enhanced neurofibrillary degeneration in transgenic mice expressing mutant tau and APP. Science, 293: 1487–91. Li, Y. M., Xu, M., Lai, M. T. et al. (2000). Photoactivated secretase inhibitors directed to the active site convalently label presenilin 1. Nature, 405: 689–93. Lin, C. H., Tallaksen-Greene, S., Chien, W. M. et al. (2001). Neurological abnormalities in a knock-in mouse model of Huntington’s disease. Hum Mol Genet, 10: 137–44. Liochev, S. I., Chen, L. L., Hallewell, R. A. & Fridovich, I. (1997). Superoxide-dependent peroxidase activity of H48Q: a superoxide dismutase variant associated with familial amyotrophic lateral sclerosis. Arch Biochem Biophys, 346: 263–8. Lippa, C. F., Smith, T. W., Saunders, A. M., Hulette, C., Pulaski-Salo, D. & Roses, A. D. (1997). Apolipoprotein E-ε2 and Alzheimer’s disease: genotype influences pathologic phenotype. Neurology, 48: 515–19. Lowe, J., Mayer, R. J. & Landon, M. (1993). Ubiquitin in neurodegenerative diseases. Brain Pathol, 3: 55–65. Luo, Y., Bolon, B., Kahn, S. et al. (2001). Mice deficient in BACE1, the Alzheimer’s -secretase, have normal phenotype and abolished -amyloid generation. Nature, 4: 231–2. Luthi-Carter, R., Strand, A., Peters, N. L. et al. (2000). Decreased expression of striatal signaling genes in a mouse model of Huntington’s disease. Hum Mol Genet, 9: 1259–71. McCampbell, A. & Fischbeck, K. H. (2001). Polyglutamine and CBP: fatal attraction? Nat Med, 7: 528–30. McCampbell, A., Taylor, J. P., Taye, A. A. et al. (2000). CREB-binding protein sequestration by expanded polyglutamine. Hum Mol Genet, 9: 2197–202. McCord, J. M., Fridovich, I. (1969). Superoxide dismutase. An enzymic function for erythrocuprein (hemocuprein). J Biol Chem, 244: 6049–55. McKinley, M. P., Bolton, D. C. & Prusiner, S. B. (1983). A proteaseresistant protein is a structural component of the scrapie prion. Cell, 35: 57–62.

Malzac, P., Webber, H., Moncla, A. et al. (1998). Mutation analysis of UBE3A in Angelman syndrome patients. Am J Hum Genet, 62: 1353–60. Manetto, V., Perry, G., Tabaton, M. et al. (1988). Ubiquitin is associated with abnormal cytoplasmic filaments characteristic of neurodegenerative diseases. Proc Natl Acad Sci USA, 85: 4501–5. Mangiarini, L., Sathasivam, K. & Seller, M. et al. (1996). Exon 1 of the HD gene with an expanded CAG repeat is sufficient to cause a progressive neurological phenotype in transgenic mice. Cell, 87: 493–506. Mann, D. M. A., Iwatsubo, T., Pickering-Brown, S. M., Owen, F., Saido, T. C. & Perry, R. H. (1997). Preferential deposition of amyloid protein (A) in the form A40 in Alzheimer’s disease is associated with a gene dosage effect of the apolipoprotein E E4 allele. Neurosci Lett, 221: 81–4. Masliah, E., Rockenstein, E., Veinbergs, I. et al. (2000). Dopaminergic loss and inclusion body formation in alpha-synuclein mice: implications for neurodegenerative disorders. Science, 287: 1265–9. Masters, C. L., Simms, G., Weinman, N. A., Multhaup, G., McDonald, B. L. & Beyreuther, K. (1985). Amyloid plaque core protein in Alzheimer disease and Down syndrome. Proc Natl Acad Sci USA, 82: 4245–9. Matsuoka, Y., Vila, M., Lincoln, S. et al. (2001). Lack of nigral pathology in transgenic mice expressing human alpha-synuclein driven by the tyrosine hydroxylase promoter. Neurobiol Dis, 8: 535–9. Mezey, E., Dehejia, A. M., Harta, G. et al. (1998). Alpha synuclein is present in Lewy bodies in sporadic Parkinson’s disease. Mol Psychiatry, 3: 493–9. Muller, S., Hoege, C., Pyrowolakis, G. & Jentsch, S. (2001). SUMO, ubiquitin’s mysterious cousin. Nat Rev Mol Cell Biol, 2: 202–10. Murayama, O., Tomita, T. & Nihonmatsu, N. et al. (1999). Enhancement of amyloid beta 42 secretion by 28 different presenilin 1 mutations of familial alzheimer’s disease. Neurosci Lett 265: 61–3. Nagafuchi, S., Yanagisawa, H., Sato, K. et al. (1994). Dentatorubral and pallidoluysian atrophy expansion of an unstable CAG trinucleotide on chromosome 12p. Nat Genet, 6: 14–18. Nagy, Z., Esiri, M. M., Jobst, K. A. et al. (1995). Influence of the apolipoprotein E genotype on amyloid deposition and neurofibrillary tangle formation in Alzheimer’s disease. Neuroscience, 69: 757–61. Nucifora, F. C., Jr., Sasaki, M. Peters, M. F. et al. (2001). Interference by huntingtin and atrophin-1 with CBP-mediated transcription leading to cellular toxicity. Science, 291: 2423– 8. Oesch, B., Westaway, D., Wälchli, M. et al. (1985). A cellular gene encodes scrapie PrP 27–30 protein. Cell, 40: 735–46. Ohm, T. G., Kirca, M., Bohl, J., Scharnagl, H., Gross, W. & Marz, W. (1995). Apolipoprotein E polymorphism influences not only cerebral senile plaque load but also Alzheimertype neurofibrillary tangle formation. Neuroscience, 66: 583–7.


Ohsumi, Y. (2001). Molecular dissection of autophagy: two ubiquitin-like systems. Nat Rev Mol Cell Biol, 2: 211–16. Okazawa, H., Rich, T., Chang, A. et al. (2002). Interaction between mutant ataxin-1 and PQBP-1 affects transcription and cell death. Neuron, 34: 701–13. Orr, H. T. (2001). Beyond the Qs in the polyglutamine diseases. Genes Dev, 15: 925–32. Orr, H. T., Chung, M.-Y., Banfi, S. et al. (1993). Expansion of an unstable trinucleotide CAG repeat in spinocerebellar ataxia type 1. Nat Genet, 4: 221–6. Parge. H. E., Hallewell, R. A. & Tainer, J. A. (1992). Atomic structures of wild-type and thermostable mutant recombinant human Cu, Zn superoxide dismutase. Proc Natl Acad Sci USA, 89: 6109–13. Paulson, H. L. (1999). Protein fate in neurodegenerative proteinopathies: polyglutamine diseases join the (mis)fold. Am J Hum Genet, 64: 339–45. Perez, M. K., Paulson, H. L., Pendse, S. J., Saionz, S. J., Bonini, N. M. & Pittman, R. N. (1998). Recruitment and the role of nuclear localization in polyglutamine-mediated aggregation. J Cell Biol, 143: 1457–70. Pines, J. (1994). Ubiquitin with everything. Nature, 371: 742–3. Polymeropoulos, M. H., Lavedan, C., Leroy, E. et al. (1997). Mutation in the -synuclein gene identified in families with Parkinson’s disease. Science, 276: 2045–7. Poorkaj, P., Bird, T. D., Wijsman, E. et al. (1998). Tau is a candidate gene for chromosome 17 frontotemporal dementia. Ann Neurol, 43: 815–25. Price, D. L., Sisodia, S. S. & Borchelt, D. R. (1998). Genetic neurodegenerative diseases: the human illness and transgenic models. Science, 282: 1079–83. Probst, A., Götz, J., Wiederhold, K. H. et al. (2000). Axonopathy and amyotrophy in mice transgenic for human four-repeat tau protein. Acta Neuropathol, 99: 469–81. Prusiner, S. B., McKinley, M. P., Bowman, K. A. et al. (1983). Scrapie prions aggregate to form amyloid-like birefringent rods. Cell, 35: 349–58. Prusiner, S. B., Scott, M., Foster, D. et al. (1990). Transgenic studies implicate interactions between homologous PrP isoforms in scrape prion replication. Cell, 63: 673–86. Prusiner, S. B., Scott, M. R., DeArmond, S. J. & Cohen, F. E. (1998). Prion protein biology. Cell, 93: 337–48. Ratovitski, T., Corson, L. B., Strain, J. et al. (1999). Variation in the biochemical/biophysical properties of mutant superoxide dismutase 1 enzymes and the rate of disease progression in familial amyotrophic lateral sclerosis kindreds. Hum Mol Genet, 8: 1451–60. Reaume, A. G., Elliott, J. L., Hoffman, E. K. et al. (1996). Motor neurons in Cu/Zn superoxide dismutase-deficient mice develop normally but exhibit enhanced cell death after axonal injury. Nat Genet, 13: 43–7. Reddy, P. H., Williams, M., Charles, V. et al. (1998). Behavioural abnormalities and selective neuronal loss in HD transgenic mice expressing mutated full-length HD cDNA. Nat Genet, 20: 198–202.

Reiss, C., Lesnik, T., Parvez, H., Parvez, S. & Ehrlich, R. (2000). Conformational toxicity and sporadic conformational diseases. Toxicology, 153: 115–21. Ripps, M. E., Huntley, G. W., Hof, P. R., Morrison, J. H. & Gordon, J. W. (1995). Transgenic mice expressing an altered murine superoxide dismutase gene provide an animal model of amyotrophic lateral sclerosis. Proc Natl Acad Sci USA, 92: 689–93. Roher, A. E., Lowenson, J. D., Clarke, S. et al. (1993). Structural alterations in the peptide backbone of -amyloid core protein may account for its deposition and stability in Alzheimer’s disease. J Biol Chem, 268: 3072–83. Rosen, D. R., Siddique, T., Patterson, D. et al. (1993). Mutations in Cu/Zn superoxide dismutase gene are associated with familial amyotrophic lateral sclerosis. Nature, 362: 59–62. Rosenthal, G., Walter, P., Bartz-Schmidt, K. U. & Heimann, K. (1999). Primary non familial amyloidosis of the vitreous and subretinal amyloid bodies [In Process Citation]. Int Ophthalmol, 23: 1–2. Roses, A. D. (1996). Apolipoprotein E alleles as risk factors in Alzheimer’s disease. Annu Rev Med, 47: 387–400. Ross, C. A. (1995). When more is less: pathogenesis of glutamine repeat neurodegenerative diseases. Neuron, 15: 493–6. Ross, C. A., Margolis R. L., Rosenblatt, A., Ranen, N. G., Becher, M. W. & Aylward, E. (1997). Huntington’s disease and the related disorder, dentatorubral-pallidoluysian atrophy (DRPLA). Medicine, 76: 305–38. Saido, T. C., Yamao-Harigaya, W., Iwatsubo, T. & Kawashima, S. (1996). Amino- and carboxyl-terminal heterogeneity of amyloid peptides deposited in human brain. Neurosci Lett, 215: 173–6. Salo, D. C., Pacifici, R. E., Lin, S. W., Giulivi, C. & Davies, K. J. A. (1990). Superoxide dismutase undergoes proteolysis and fragmentation following oxidative modification and inactivation. J Biol Chem, 265: 11919–27. Sankarapandi, S. & Zweier, J. L. (1999a). Bicarbonate is required for the peroxidase function of Cu, zn-superoxide dismutase at physiological pH. J Biol Chem, 274: 1226–32. (1999b). Evidence against the generation of free hydroxyl radicals from the interaction of copper, zinc-superoxide dismutase and hydrogen peroxide. J Biol Chem, 274: 34576–83. Scherzinger, E., Lurz, R., Turmaine, M. et al. (1997). Huntingtinencoded polyglutamine expansions form amyloid-like protein aggregates in vitro and in vivo. Cell, 90: 549–58. Scherzinger, E., Sittler, A., Schweiger, K. et al. (1999). Self-assembly of polyglutamine-containing huntingtin fragments into amyloid-like fibrils: implications for Huntington’s disease pathology. Proc Natl Acad Sci USA, 96: 4604–9. Schilling, G., Becher, M. W., Sharp, A. H. et al. (1999a). Intranuclear inclusions and neuritic pathology in transgenic mice expressing a mutant N-terminal fragment of huntingtin. Hum Mol Genet, 8: 397–407. Schilling, G., Wood, J. D., Duan, K. et al. (1999b). Nuclear accumulation of truncated atrophin-1 fragments in a transgenic mouse model of DRPLA. Neuron, 24: 275–86. Schilling, G., Jinnah, H. A., Gonzales, V. et al. (2001). Distinct behavioral and neuropathological abnormalities in transgenic

555

556


mouse models of HD and DRPLA. Neurobiol Dis, 8: 405– 18. Schmechel, D. E., Saunders, A. M., Strittmatter, W. J. et al. (1993). Increased amyloid -peptide deposition in cerebral cortex as a consequence of apolipoprotein E genotype in late-onset Alzheimer’s disease. Proc Natl Acad Sci USA, 90: 9649–53. Schwartz, A. L. & Ciechanover, A. (1999). The ubiquitin-proteasome pathway and pathogenesis of human diseases. Annu Rev Med, 50: 57–74. Scott, M., Foster, D., Mirenda, C. et al. (1989). Transgenic mice expressing hamster prion protein produce speciesspecific scrapie infectivity and amyloid plaques. Cell, 59: 847–57. Scott, M., Groth, D., Foster, D. et al. (1993). Propagation of prions with artificial properties in transgenic mice expressing chimeric PrP genes. Cell, 73: 979–88. Scott, M. R., Köhler, R., Foster, D. & Prusiner, S. B. (1992). Chimeric prion protein expression in cultured cells and transgenic mice. Protein Sci, 1: 986–97. Scott, M. R., Supattapone, S., Nguyen, H. O., DeArmond, S. J. & Prusiner, S. B. (2000). Transgenic models of prion disease. Arch Virol Suppl, 113–24. Seiffert, D., Bradley, J. D., Rominger, C. M. et al. (2000). Presenilin-1 and -2 are molecular targets for gamma-secretase inhibitors. J Biol Chem, 275: 34086–91. Sherrington, R., Rogaev, E. I. & Liang, Y. et al. (1995). Cloning of a gene bearing missense mutations in early-onset familial Alzheimer’s disease. Nature, 375: 754–60. Sherrington, R., Froelich, S., Sorbi, S. et al. (1996). Alzheimer’s disease associated with mutations in presenilin 2 is rare and variably penetrant. Hum Mol Genet, 5: 985–8. Shimura, H., Schlossmacher, M. G., Hattori, N. et al. (2001). Ubiquitination of a new form of -synuclein by parkin from human brain: implications for Parkinson’s disease. Science, 293: 263–9. Sieradzan, K. A. & Mann, D. M. (2001). The selective vulnerability of nerve cells in Huntington’s disease. Neuropathol Appl Neurobiol, 27: 1–21. Spillantini, M. G., Schmidt, M. L., Lee, V. M. Y., Trojanowski, J. Q., Jakes, R. & Goedert, M. (1997a). -synuclein in lewy bodies. Nature, 388: 839–40. Spillantini, M. G., Bird, T. D. & Ghetti, B. (1998a). Frontotemporal dementia and parkinsonism linked to chromosome 17: a new group of tauopathies. Brain Pathol, 8: 387–402. Spillantini, M. G., Crowther, R. A., Jakes, R., Cairns, N. J., Lantos, P. L. & Goedert, M. (1998b). Filamentous alpha-synuclein inclusions link multiple system atrophy with Parkinson’s disease and dementia with Lewy bodies. Neurosci Lett, 251: 205–8. Spillantini, M. G., Crowther, R. A., Jakes, R., Hasegawa, M. & Goedert, M. (1998c). alpha-Synuclein in filamentous inclusions of Lewy bodies from Parkinson’s disease and dementia with lewy bodies. Proc Natl Acad Sci USA, 95: 6469–73. Spittaels, K., Van den Haute, C., Van Dorpe, J. et al. (1999). Prominent axonopathy in the brain and spinal cord of transgenic

mice overexpressing four-repeat human tau protein. Am J Pathol, 155: 2153–65. Steffan, J. S., Kazantsev, A., Spasic-Boskovic, O. et al. (2000). The Huntington’s disease protein interacts with p53 and CREB-binding protein and represses transcription. Proc Natl Acad Sci USA, 97: 6763–8. Steffan, J. S., Bodai, L., Pallos, J. et al. (2001). Histone deacetylase inhibitors arrest polyglutamine-dependent neurodegeneration in Drosophila. Nature, 413: 739–43. Stenoien, D. L., Cummings, C. J., Adams, H. P. et al. (1999). Polyglutamine-expanded androgen receptors form aggregates that sequester heat shock proteins, proteasome components and SRC-1, and are suppressed by the HDJ-2 chaperone. Hum Mol Genet, 8: 731–41. Strittmatter, W. J., Saunders, A. M., Goedert, M. et al. (1994a). Isoform-specific interactions of apolipoprotein E with microtubule-associated protein tau: implications for Alzheimer disease. Proc Natl Acad Sci USA, 91: 11183–6. Strittmatter, W. J., Weisgraber, K. H., Goedert, M. et al. (1994b). Hypothesis: microtubule instability and paired helical filament formation in the Alzheimer disease brain are related to apolipoprotein E genotype. Exp Neurol, 125: 163–71. Sturchler-Pierrat, C., Abramowski, D., Duke, M. et al. (1997). Two amyloid precursor protein transgenic mouse models with Alzheimer disease-like pathology. Proc Natl Acad Sci USA, 94: 13287–92. Sturtz, L. A., Diekert, K., Jensen, L. T., Lill, R. & Culotta, V. C. (2001). A fraction of yeast Cu, Zn-superoxide dismutase and its metallochaperone, CCS, localize to the intermembrane space of mitochondria. A physiological role for SOD1 in guarding against mitochondrial oxidative damage. J Biol Chem, 276: 38084– 9. Subramaniam, J. R., Lyons, W. E., Liu, J. et al. (2002). Mutant SOD1 causes motor neuron disease independent of copper chaperone-mediated copper loading. Nat Neurosci, 5: 301– 7. Suhr, S. T., Senut, M. C., Whitelegge, J. P., Faull, K. F., Cuizon, D. B. & Gage, F. H. (2001). Identities of sequestered proteins in aggregates from cells with induced polyglutamine expression. J Cell Biol, 153: 283–94. Terry, R. D. & Katzman, R. (1983a). Senile dementia of the Alzheimer type: defining a disease. In: The Neurology of Aging (R., Katzman, R., Terry eds), pp. 51–84. Philadelphia: F. A. Davis. (1983b). Senile dementia of the Alzheimer type. Ann Neurol, 14: 497–506. Tesseur, I., Dorpe, J. V., Spittaels, K., Van den Haute, C., Moechars, D. & Van Leuven, F. (2000). Expression of human apolipoprotein E4 in neurons causes hyperphosphorylation of protein tau in the brains of transgenic mice. Am J Pathol, 156: 951–64. Thinakaran, G., Borchelt, D. R., Lee, M. K. et al. (1996). Endoproteolysis of presenilin 1 and accumulation of processed derivatives in vivo. Neuron, 17: 181–90. Trojanowski, J. Q. & Lee, V. M. Y. (1998). Aggregation of neurofilament and -synuclein proteins in Lewy bodies. Arch Neurol, 55: 151–2.


Trojanowski, J. Q., Ishihara, T., Higuchi, M. et al. (2002). Amyotrophic lateral sclerosis/parkinsonism dementia complex: transgenic mice provide insights into mechanisms underlying a common tauopathy in an ethnic minority on Guam. Exp Neurol, 176: 1–11. Ueda, K., Fukushima, H., Masliah, E. et al. (1993). Molecular cloning of cDNA encoding an unrecognized component of amyloid in Alzheimer disease. Proc Natl Acad Sci USA, 90: 11282–6. Vassar, R., Bennett, B. D., Babu-Khan, S. et al. (1999). -secretase cleavage of alzheimer’s amyloid precusor protein by the transmembrane aspartic protease BACE. Science, 286: 735–41. Vito, P., Wolozin, B., Ganjei, J. K., Iwasaki, K., Lacaná, E. & D’Adamio, L. (1996). Requirement of the familial Alzheimer’s disease gene PS2 for apoptosis – opposing effect of ALG-3. J Biol Chem, 271: 31025–28. Wakabayashi, K., Engelender, S., Yoshimoto, M., Tsuji, S., Ross, C. A. & Takahashi, H. (2000). Synphilin-1 is present in Lewy bodies in Parkinson’s disease. Ann Neurol, 47: 521–3. Walker, L. C. & LeVine, H., III (2000). The cerebral proteopathies. Neurobiol Aging, 21: 559–61. Wang, J., Xu, G. & Borchelt, D. R. (2002a). High molecular weight complexes of mutant superoxide dismutase 1: age-dependent and tissue-specific accumulation. Neurobiol Dis, 9: 139–48. Wang, J., Xu, G., Gonzales, V. et al. (2002b). Fibrillar inclusions and motor neuron degeneration in transgenic mice expressing superoxide dismutase 1 with a disrupted copper-binding site. Neurobiol Dis, 10: 128–38. Watanabe, M., Dykes-Hoberg, M., Cizewski, C. V., Price, D. L., Wong, P. C. & Rothstein, J. D. (2001). Histological evidence of protein aggregation in mutant SOD1 transgenic mice and in amyotrophic lateral sclerosis neural tissues. Neurobiol Dis, 8: 933–41. Weidemann, A., König, G., Bunke, D. et al. (1989). Identification, biogenesis, and localization of precursors of Alzheimer’s disease A4 amyloid protein. Cell, 57: 115–26. Weissman, A. M. (2001). Themes and variations on ubiquitylation. Nat Rev Mol Cell Biol, 2: 169–78. Wheeler, V. C., White, J. K., Gutekunst, C. A. et al. (2000). Long glutamine tracts cause nuclear localization of a novel form of huntingtin in medium spiny striatal neurons in HdhQ92 and HdhQ111 knock-in mice. Hum Mol Genet, 9: 503–13. Wiedau-Pazos, M., Goto, J. J., Rabizadeh, S. et al. (1996). Altered reactivity of superoxide dismutase in familial amyotrophic lateral sclerosis. Science, 271: 515–8. Wong, P. C., Marszalek, J., Crawford, T. O. et al. (1995a). Increasing neurofilament subunit NF-M expression reduces axonal NF-H, inhibits radial growth, and results in neurofilamentous accumulation in motor neurons. J Cell Biol, 130: 1413–22. Wong, P. C., Pardo, C. A., Borchelt, D. R. et al. (1995b). An adverse property of a familial ALS-linked SOD1 mutation causes motor neuron disease characterized by vacuolar degeneration of mitochondria. Neuron, 14: 1105–16.

Wong, P. C., Waggoner, D., Subramaniam, J. R. et al. (2000). Copper chaperone for superoxide dismutase is essential to activate mammalian Cu/Zn superoxide dismutase. Proc Natl Acad Sci USA, 97: 2886–91. Wood, J. D., Yuan, J., Margolis, R. L. et al. (1998). Atrophin-1, the DRPLA gene product, interacts with two families of WW domain-containing proteins. Mol Cell Neurosci, 11: 149–60. Wood, J. D., Nucifora, F. C., Jr., Duan, K. et al. (2000). Atrophin-1, the dentato-rubral and pallido-luysian atrophy gene product, interacts with ETO/MTG8 in the nuclear matrix and represses transcription. J Cell Biol, 150: 939–48. Xia, W., Ostaszewski, B. L., Kimberly, W. T. et al. (2000). FAD mutations in presenilin-1 or amyloid precursor protein decrease the efficacy of a gamma-secretase inhibitor: evidence for direct involvement of PS1 in the gamma-secretase cleavage complex. Neurobiol Dis, 7: 673–81. Xu, G., Gonzales, V. & Borchelt, D. R. (2002a). Rapid detection of protein aggregates in the brains of Alzheimer patients and transgenic mouse models of amyloidosis. Alzheimer Dis Assoc Disord, 16: 191–5. (2002b). A deposition does not cause the aggregation of endogenous tau in transgenic mice. Alzheimer Dis Assoc Disord, 16: 196–201. Xu, Z., Cork, L. C., Griffin, J. W. & Cleveland, D. W. (1993). Increased expression of neurofilament subunit NF-L produces morphological alterations that resemble the pathology of human motor neuron disease. Cell, 73: 23–33. Yamamoto, A., Lucas, J. J. & Hen, R. (2000). Reversal of neuropathology and motor dysfunction in a conditional model of huntington’s disease. Cell, 101: 57–66. Yim, M. B., Chock, P. B. & Stadtman, E. R. (1990). Copper, zinc superoxide dismutase catalyzes hydroxyl radical production from hydrogen peroxide. Proc Natl Acad Sci USA, 87: 5006–10. (1993). Enzyme function of copper, zinc superoxide dismutase as a free radical generator. J Biol Chem, 268: 4099–105. Yim, M. B., Kang, J. H., Yim, H. S., Kwak, H. S., Chock, P. B. & Stadtman, E. R. (1996). A gain-of-function of an amyotrophic lateral sclerosis-associated Cu, Zn-superoxide dismutase mutant: an enhancement of free radical formation due to a decrease in Km for hydrogen peroxide. Proc Natl Acad Sci USA, 93: 5709–14. Younkin, S. G. (1995). Evidence that A42 is the real culprit in Alzheimer’s disease. Ann Neurol, 37: 287. Zhang, S., Xu, L., Lee, J. & Xu, T. (2002). Drosophila atrophin homolog functions as a transcriptional corepressor in multiple developmental processes. Cell, 108: 45–56. Zhang, Y., Gao, J., Chung, K. K., Huang, H., Dawson, V. L. & Dawson, T. M. (2000). Parkin functions as an E2-dependent ubiquitinprotein ligase and promotes the degradation of the synaptic vesicle-associated protein, CDCrel-1. Proc Natl Acad Sci USA, 97: 13354–9.

557

Appendix: Dementia brain banks

Dementia brain banks in Australia Department of Neuropathology Institute of Medical & Veterinary Science Frome Road Adelaide SA 5000 Australia Dr P Blumbergs [email protected] Department of Biochemistry The University of Queensland St Lucia, Queensland 4072 Australia Dr P Dodd [email protected] Neuropathology Unit Department of Pathology University of Sydney Sydney, NSW 200 6 Australia Professor C Harper [email protected] Department of Pathology University of Melbourne Parkville Victoria 3052 Australia Professor C Masters [email protected]

Dementia brain banks in Canada Canadian Brain Tissue Bank University of Toronto

558

University Health Network – Western Division 399 Bathurst Street Fell Wing 5–222 Toronto, ON M5T 2S8 Canada J R Wherrett, MD FRCP(C) PhD [email protected] www.utoronto.ca/neuropathology/cbb.html Kinsmen Laboratory of Neurological Research The University of British Columbia 2255 Wesbrook Mall Vancouver British Columbia V6T 1Z3 Canada Professor E McGeer [email protected]

Dementia brain banks in Europe Kuopio University Department of Neuroscience and Neurology Section of Neuropathology PO Box 1627 Fin 70211 Kuopio Finland Dr I Alafuzoff [email protected] Laboratoire de Neuropathologie R Escourolle Inserm U360 Hôpital de La Salpêtrière 47 Blvd de l’Hôpital F-75651 Paris Cedex 13 France

Dementia brain banks in Europe

Professor J J Hauw and Dr V Sazdovitch Réseau Français de Cérébrothèques Maladie de Parkinson Inserm U289, Hôpital de La Salpêtrière 47 Blvd de l’Hôpital F-75651 Paris Cedex 13 France Professor C Duyckaerts and Dr E Hirsch [email protected] Austro German Brain Bank Clinic and Polyclinic of Psychiatry and Psychotherapy Clinical Neurochemistry University of Wuerzburg Germany Dr Peter Riederer [email protected]

Sweden Dr Nenad Bogdanovic [email protected] MRC Cognitive Function and Ageing Study (CFAS) Brain Bank Biostatistics Unit Institute of Public Health Cambridge CB2 2SR UK Mrs F Matthews [email protected] www.mrc-bsu.cam.ac.uk/cfas/ The Cambridge Brain Bank Laboratory University of Cambridge Department of Pathology Box 231, Addenbrooke’s Hospital Cambridge CB2 2QQ UK Dr John Xuereb [email protected]

Department of Neurology Abt. fur Neurologie, Universitatsklinikum Rudolf Virchow Augustenburger Platz 1 D-W-1000 Berlin 65 Germany Dr K Jendroska

UK Multiple Sclerosis Brain Bank Neuropathology Department Charing Cross Hospital Fulham Palace Road London W6 8RF UK

The Netherlands Brain Bank Meiberdreof 33 1105 AZ Amsterdam The Netherlands Dr R Ravid [email protected] www.brainbank.nl

Parkinson’s Disease Society Brain Bank Neuropathology Department Charing Cross Hospital Fulham Palace Road London W6 8RF UK Dr S Daniel

Neurological Tissue Bank Servico de Neurologia Hospital Clinico y Provincial Villarroel 170 Barcelona 08036 Spain Dr F F Cruz-Sanchez

MRC London Neurodegenerative Diseases Brain Bank Institute of Psychiatry Denmark Hill London SE5 8AF UK Dr Nadeem Khan [email protected]

Huddinge Brain Bank Department of Geriatric Medicine NEUROTEC, Karolinska Institutet NOVUM, KFC, plan 4 S-141 86 Huddinge

Dementia Brain Bank University of Manchester Laboratory Medicine Academic Group Department of Medicine

559

560

Dementia Brain Banks

Manchester M13 9PT UK Professor DMA Mann [email protected] MRC Brain Bank Neurochemical Pathology Unit Newcastle General Hospital Newcastle upon Tyne NE4 6BE UK Professor J Edwardson Neuropathology Department Radcliffe Infirmary Oxford OX2 6HE UK Professor M M Esiri and Dr J H Morris [email protected]

Dementia brain banks in North America

Massachusetts General Hospital Department of Neuropathology Warren 321, Fruit Street Boston, MA 02114 USA E Tessa Hedley-Whyte MD Department of Pathology and Neuroscience Albert Einstein College of Medicine 1300 Morris Park Avenue Bronx, NY 10461 USA Peter Davies PhD Rush Alzheimer’s Disease Center 710 S. Paulina 8 North JRB Chicago, IL 60612 USA David A Bennett MD [email protected] www.rush.edu/patients/radc/index.html

Department of Neurology Michigan Alzheimer’s Disease Research Center University of Michigan 1500 E Medical Center Drive Ann Arbor, MI 48109 USA Roger L Albin MD

Department of Pathology University of Cincinnati College of Medicine 231 Bethesda Avenue Cincinnati, OH 45267-0524 USA Frank P Zemlan PhD

Sun Health Research Institute 10515 W Santa Fe Drive Sun City Arizona 85351 USA Thomas Beach MD PhD [email protected] www.shri.org

Alzheimer’s Disease Research Center and National Prion Disease Pathology Surveillance Center Institute of Pathology Case Western Reserve University 2085 Adelbert Road Cleveland, OH 44106 USA Pierluigi Gambetti MD

Brain Tissue Resource Center McLean Hospital 115 Mill Street Belmont, MA 021178 USA Edward D Bird MD Center for Neurologic Diseases Brigham and Women’s Hospital 75 Francis Street Boston, MA 02115 USA Dennis J Selkoe MD

Division of Neuropathology Ohio State University College of Medicine N-112B Upham Hall 473 W 12th Avenue Columbus, OH 43210 USA Leopold Liss MD Department of Pathology Neuropathology Laboratory University of Texas Southwestern Medical Center


5323 Harry Hines Boulevard Dallas, TX 75235–9072 USA Charles L White III MD Geriatric Neurobehavior & Alzheimer’s Center Rancho Los Amigos National Rehabilitation Center University of Southern California 7601 E Imperial Highway Medical Science Building, Room 26 Downey, CA 90242 USA Chris Zarow PhD Helena Chui MD [email protected] Kathleen Price Bryan Brain Bank Department of Pathology Box 3712 Duke University Medical Center Durham, NC 27710 USA Barbara J Crain MD PhD 206 Psychology Building University of Texas at El Paso El Paso TX 79968 USA Donald E Moss PhD Department of Pathology Baylor College of Medicine Mail Station 205, Methodist Hospital 6565 Fannon Houston, TX 77030 USA Joel B Kirkpatrick MD Indiana University Medical Center Division of Neuropathology 635 Barnhill Drive MS A142 Indianapolis, IN 46223 USA Bernardino Ghetti MD The University of Iowa College of Medicine Department of Anatomy1 Department of Neurology2

200 Hawkins Drive Iowa City, IA 52242 USA Gary Van Hoesen PhD1 [email protected] Antonio R Damasio MD PhD2 Mayo Clinic Jacksonville 4500 San Pablo Road Jacksonville, FL 32224 USA Dr Dennis Dickson Department of Pathology UMKC Neurodegeneration Autopsy Program Truman Medical Center 2301 Holmes Street Kansas City, MO 64108 USA Joseph L Parker MD Sanders-Brown Center on Aging University of Kentucky 101 Sanders-Brown Building Lexington, KY 40536–0230 USA William R Markesbery MD [email protected] www.mc.uky.edu/coa Neuropathology Autopsy Core University of Southern California Medical Center 2011 Zonal Avenue Los Angeles, CA 90033 USA Carol A Miller MD Human Brain and Spinal Fluid Resource Center VA Greater Los Angeles Healthcare System West Los Angeles Healthcare Center Neurology Service (127A) 11301 Wilshire Blvd Los Angeles, CA 90073 USA Wallace W Tourtellotte MD PhD [email protected] [email protected] www.loni.ecla.edu/∼nnrsb/NNRSB/

561

562

Dementia Brain Banks

NPF/UM Brain Endowment Bank University of Miami (D4-5) 1501 NW 9th Avenue Miami, FL 33136 USA Deborah C Mash PhD [email protected] www.parkinson.org/brainbank.htm Froedtert Memorial Lutheran Hospital Neurology Department 9200 W Wisconsin Avenue Milwaukee, WI 53226 USA Piero Antuono MD [email protected] Department of Psychiatry Mount Sinai School of Medicine One Gustave L Levy Place New York, NY 10029–6571 USA Kenneth L Davis MD [email protected] Alzheimer’s Disease Research Center Columbia University 630 West 168th Street New York, NY 10032 USA James M Powers MD Department of Pathology New York University Medical Center 550 First Avenue New York, NY 10016 USA Douglas C Miller MD PhD Western Psychiatric Institute and Clinic 3811 O’Hara Street, Room E-1230 Pittsburg, PA 15213 USA George S Zubenko, MD PhD Division of Neuropathology Oregon Health Sciences University 3181 SW Sam Jackson Park Road Portland L113, OR 97201 USA Melvyn J Ball MD

Department of Pathology Mayo Clinic 200 First Street, SW Rochester MN 55905 USA Joseph E Parisi MD University of Rochester School of Medicine Box 673 601 Elmwood Avenue Rochester, NY 14642 USA UCSD National Alzheimer’s Disease Brain Bank Department of Pathology M-012, BSB Rm 1004 University of California San Diego, La Jolla CA 92093 USA George G Glenner MD Department of Neurosciences and Psychiatry University of California San Diego, La Jolla, CA 92093 USA Robert D Terry MD and Lawrence Hansen MD Center for Alzheimer’s Disease and Related Disorders PO Box 19643 Southern Illinois School of Medicine Springfield, IL 62794–9643 USA Robert G Struble PhD St Louis University/Alzheimer’s Association Brain Bank Department of Psychiatry St Louis University School of Medicine 1221 S Grand Blvd St Louis, MO 63104 USA George T Grossberg MD [email protected] Department of Pathology Box 8118 Washington University School of Medicine


660 South Euclid Avenue St Louis, MO 63110 USA Daniel W McKeel Jr MD Alzheimer’s Research Center Regions Hospital

640 Jackson Street St Paul, MN 55101 USA William H Frey II PhD [email protected] www.alzheimersinfo.org

563

Index

Pages with tables are shown in italics, those with figures are in bold. (Figures are indexed when not covered by the text reference) acetylcholine basal nucleus (of Meynert) 40–1 neostriatum 378 ACT gene 215 adrenoleukodystrophy 521 adult onset neuronal ceroid lipofuscinosis 521 adult polyglucosan body disease (Lafora disease) 521 ageing brain 113–27 centenarians, pathology 117–20 cortical disconnection evidence 121 non-human primates as models 121–3 normal ageing 113–17, 120–1 vs AD 68 cerebrovascular changes 120–1 distribution of NFT and SP 113–15 major studies 114 neuronal loss 115–17 AIDS-related dementia see HIV infection and AIDS alcoholism 427–41 hepatic encephalopathy 434–6 pellagra 436–7 primary alcoholic dementia 428–9 Wernicke–Korsakoff syndrome 429–33 Alexander’s disease 521 alien limb 236 -galactosidase deficiency, Fabry’s disease 524 -syn A53T mutation 365, 367 model of inclusion formation (fibrillogenesis) 366 non-amyloid component of SP precursor protein (NACP) 353, 365–8 transgenic models 548–51 -synuclein inclusion diseases 353–68 see also Parkinson’s disease; synucleinopathies aluminium toxicity 525–6 Alzheimer II astrocytes 434–6

564

Index

Alzheimer’s disease (AD) 161–206 AD 1-4 categories 100 amyloid in AD 188–95 anatomic basis 10–16 brainstem involvement 19, 20 cholinergic hypothesis 10 mesial temporal NFTs 16 pathological hallmarks 10–13 temporal distribution of pathology 13–15 clinical diagnosis 183 accuracy 49 correlation of clinical and pathological stages 14–15 criteria 4–5 differential diagnosis 7–8 motor and sensory impairments 18 neuroimaging 166–8 presentation 10 variant presentations of cortical involvement 19, 21 see also CERAD development 179–80 Down’s syndrome 207–26 see also Down’s syndrome and Alzheimer’s disease epidemiology 162–6 clinical studies 162–3 clinico-pathologic studies 163–4 coexistence with Huntington disease 392 risk factors 164–6 genetics 164–5, 185–8 APOE polymorphisms 467 early onset familial AD 185–7 genetic testing 103–4 late onset AD 187–8 summary 100 gross changes 168–70 head injury 165, 459–61 histopathological criteria 180–4 Braak staging 13–15, 181–2 CERAD 181–3 choice 182–3 differential diagnosis 183–4 Khachaturian criteria 180–1, 499 microscopy 170–9 minimalist approach to pathological diagnosis 184 mixed pathology, CVD/Parkinsonism 67 staging severity of pathology 181–2 see also CERAD histopathology 170–9 congophilic angiopathy 175–8 neocortical involvement 19–20 neurofibrillary changes 172–4 senile plaques 170–2 spatial pattern of atrophy 14 subcortical pathology 175

white matter changes 178–9 historical aspects 161–2 imaging see neuroimaging late-onset AD, ApoE (chr-19) 103, 187 Lewy body variant see diffuse Lewy body disease (DLB) pathogenesis 184–95 amyloid in AD 188–95 cell division cycle 194–5 molecular genetics 185–8 pathology see Alzheimer’s disease (AD), histopathology prevalence in schizophrenia 496, 499 risk factors 164–6 transgenic mouse models 542–6 see also brain atlases Ammon’s horn see cornu ammonis (CA) amnestic syndrome, paradigm for 16 amphotericin B, leukoencephalopathy 525 amygdala AD changes 17 anatomy 16, 40–1, 42 atrophy 58 basalateral nuclei 40 corticomedial nuclei 40 diffuse plaques vs neuritic plaques 171–2 hippocampal area 35 amygdaloid nucleus 378 amyloid peptide (A) 188–95, 542–6 allelic forms of ApoE 215–16 amyloid cascade hypothesis 190 cell cycle hypothesis 194–5 chemistry and genetics familial British dementia 339–42 familial Danish dementia 344–5, 347 HCHWA-D 337 HCHWA-I 334–5 chromosome-21, APP gene overexpression 214 composition in AD and DS 191, 209 deposition, long-term post head injury 464–5 immunohistochemistry 209 metabolism of APP and production of A 188–90 neurofibrillary pathology 191 neurotoxicity 190 nidus theory 165 phosphorylation of tau and formation of PHF 191–3 regulation of tau and its association with amyloid 193–4 staining and ultrastructure 170–1 see also congophilic amyloid angiopathy (CAA); senile plaques amyloid precursor protein (APP) 103, 542–6 APP gene chr-21 185–6 and Notch3 307 processing 189, 543 production of A 188–90, 209, 543 amyloid Bri precursor protein (ABriPP) 331, 341 amyloid Dan precursor protein (ADanPP) 331, 341

565

566

Index

amyotrophic lateral sclerosis transgenic mouse models 539–42 ubiquitin in 549–50 amyotrophic lateral sclerosis/Parkinsonism–dementia complex 274–9 aetiology/pathogenesis 278–9 clinical features 274–5 epidemiology 278 Guam diseases 274–7 neuropathology 275–6 anatomy 10, 34–47 amygdala 16, 40–1 basal ganglia 41–3 brainstem 44–5 cerebellum 45–6 cerebral cortex 34–9 cerebral white matter 39–40 normal functional neuroanatomy 8–9 thalamus and hypothalamus 43–4 see also neuroanatomy angiitis, primary, of CNS (PACNS) 315 angiomas, multiple cavernous 315, 316, 525 angiotropic (intravascular) large cell lymphoma 315 animal models ageing in non-human primates 121–3 tau FTDP-17 273 see also transgenic mouse models of neurodegenerative disease aphasia, progressive non-fluent 24 apolipoproteins, ApoE polymorphism 215–16, 544–5 chromosome-19, late-onset AD 103, 187 and head injury 467 APoE gene 215, 187 RFLP analysis 97, 103 apoptosis vs necrosis 465–6, 485 see also neuronal loss argyrophilic grain disease (AGD) 245–50 clinical features 245–6 differential diagnosis 248 neuronal and tau pathology spectrum 249 genetics 246 neuroimaging 246 pathological features 247–50 biochemical findings 250 gross findings 247 microscopic findings 247–9 ultrastructural findings 249 arylsulfatase A deficiency, metachromatic leukodystrophy 523 association cortex in moderately advanced Alzheimer’s disease 17–18 in primary dementia 17 astrocytic lesions CBD 237, 238 hepatic encephalopathy 434–6 PiD 229

PML 478–9 PSP 244 schizophrenia 498–50, 501 see also gliosis ataxins 536, 538 atherosclerosis 312 atrophin-1 536 processing model 538 autonomic function, cortical and subcortical regions 18–19 autopsy see post-mortem examination axonal injury, concussion syndromes 458 Balint’s syndrome 19, 20 ‘ballooned’ neurons in Pick’s disease 229, 331 in corticobasal degeneration 235 in argyrophobic gram disease 247 basal ganglia anatomy 41–3, 378 calcification, with hypocalcaemia 521 in Huntington disease 378 nomenclature 378 post-mortem examination 59 see also caudate; globus pallidus; putamen; subthalamic nucleus basal nucleus (of Meynert) 40–1, 42 CBD 238 cholinergic brain pathways, head injury 466 multipolar neurons 43 Pick’s disease 231–2, 233–4 post-mortem examination 59, 60 neurofibrillary tangless 174 Behcet’s syndrome 511 -amyloid see amyloid Betz cells 37 Binswanger’s disease 299, 300, 318 see also CADASIL; leukoencephalopathy biopsy 69–70 laboratory safety 86–8 prion diseases 407 usefulness 70 blood–brain interfaces, hydrocephalus 443–5 Braak and Braak criteria, pathological staging of AD 13–15, 181–2 brain atlases 128–60 brain averaging 145–51 coordinate systems 131 cortical averaging 145 cortical modelling 143–5 deformable, computing differences 133 dynamic (4D) brain atlases 121, 151–4 4D coordinate systems 151–2 mapping brain development and degeneration 152 tissue loss in dementia 152–4 early anatomic templates 131 elastic registration 132

Index

individualized brain atlases 133–5 analysing brain data with an atlas 134 anatomic variations 133–4 continuum–mechanical devices 134 continuum–mechanical warping 135 statistical templates 134 model-driven deformable atlases 135–6 anatomical models 136 surface parameterization 136 modelling anatomy 137 multi-modality atlases 131–2 pathology detection 140–3 deficit maps 140–1 deformable probabilistic atlases 140 hemispheric differences 141–3 mapping asymmetrics 143 mapping grey matter loss 140, 142 statistical map of grey matter loss 141 population-based brain mapping 128–31 3D structural variation and asymmetry 138 averaged ventricular anatomy 138 cortical patterns 130 demographic factors 130 disease-specific atlases 128–30 dynamic brain atlases 131 pathology detection 130 statistical brain templates 130 post-mortem data fusion 132 probabilistic atlases 136–40 brain asymmetry 139 corpus callosum differences 139–40 deformable 140 encoding anatomic variability 136 parametric mesh modelling 136–9 brain atrophy 54–8, 229 neuroimaging 51 sulcal widening 53 brain averaging 145–51 average image templates 145–50 disease progression 150 image distortion and registration accuracy 150 other average templates 150–1 brain banks 53–4, 558–63 Australia 558 Europe 558–9 Finland 558 France 558–9 Germany 559 Netherlands 559 Spain 559 Sweden 559 UK 559–60 North America 560–3 Canada 558 USA 560–3

online resources 558–63 safety issues 87–8 brain blocks choice of stains 63–5 selection of tissue blocks 61–9 see also post-mortem examination brain mapping see brain atlases brain weight and volume 54–5, 59 Cavalieri principle 76 determination 76–8, 428 see also cerebral atrophy brainstem anatomy 44–5 involvement in Alzheimer’s disease 19, 20 post-mortem examination 60 substantia nigra 44 BRI dementias familial British dementia 338–43 amyloid chemistry and genetics 339–42 clinical features 338–9 neuroimaging 339 neuropathology 342–3 familial Danish dementia 343–6, 347 amyloid chemistry and genetics 344–5 clinical features 343–5 neuroimaging 344 neuropathology 345–6 British dementia familial 338–43 see also BRI dementias Brodmann’s maps 39 area-2 see entorhinal cortex area-35 see transentorhinal cortex and NFTs 39 Pick’s disease (PiD) 37 Buerger’s disease 313 CADASIL 106–7, 305–12 biopsy findings 310 clinical picture 306 cognitive decline and dementia 306 diagnosis and differential diagnosis 311–12 epidemiology 306 genetics 100 and Notch signalling 106–7, 307–12 historical aspects 305 neuroimaging 306–7 onset and duration 306 pathogenesis 311 pathology 310–11 post-mortem brain findings 310–11 small vessel diseases 305–12 white matter lesions (WML) 302 CAG repeat disorders 100–1 transgenic mouse models 535–8

567

568

Index

cancer chemotherapy 525 caudate nucleus 41–3, 379 diffuse plaques vs neuritic plaques 171–2 Huntington disease 381–3 post-mortem examination 59 schematic representation 388 causes of dementia 4–7 Cavalieri principle, brain volume 76 cavernous angiomas 317, 318, 527 celiac disease 526 cell cycle hypothesis, pathogenesis of AD 194–5 centenarians, normal ageing 117–20 CERAD (Consortium to Establish a Registry for Alzheimer’s Disease) protocol 53, 164, 181–3 microscopic examination 65–6 selection of tissue blocks 61–9 cerebellar ataxia 361 cerebellum anatomy 45–6 atrophy, in HD 388 dentate nucleus 45 diffuse plaques vs neuritic plaques 171–2 disease associations 45 Huntington disease 387–8 post-mortem examination 61 cerebral amyloid angiopathies (CAAs) 330–8 clinical features, genes and amyloid proteins 331 differential diagnosis and neuropathological recommendations 348–9 familial and sporadic forms, clinical features, genes and amyloid proteins 331 gelsolin 331, 348 hereditary cerebral haemorrhage with amyloid Dutch type (HCHWA-D) 335–8 Icelandic type (HCHWA-I) 333–5 historical aspects 332 morphological aspects 332–3 other examples of familial A CAA 337–8 prion protein (PRP) 331, 348 sporadic forms 331, 333–5 transthyretin 346–8 see also familial British dementias cerebral atrophy 56, 168 spatial pattern of atrophy 14 cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy see CADASIL cerebral cortex anatomy 34–9 vascular supply 39 Huntington disease 385–6 see also cortical cerebral irradiation 513 cerebral ischaemia 298–9, 406 global (GCI) 301 cerebral lobar atrophy, patterns 18–19, 21

cerebral vasculature changes in normal ageing 120–1 organization in CVD 296–7, 298 cerebral white matter anatomy 39–40 appearance 59 vacuolation, conditions resembling spongiform change 406, 407 volume and texture 59, 428 Wallerian degeneration in dementias 40 see also neuronal loss; white matter lesions (WML) cerebrospinal fluid flow pathways 444 physiology in hydrocephalus 442–5 cerebrotendinous xanthomatosis 521 cerebrovascular disease and vascular dementias 289–327 assessment scheme (proposed) 319–20 associated diseases 67, 312–15 clinical criteria for VaD 6, 290–1 DSM-III-R and ICD-10 292–3 NINDS-AIREN criteria 291 subcortical vascular dementia 291–2 vascular cognitive impairment (VCI) 292 dementia following stroke 293–4 differential diagnosis 7 epidemiological studies 294–6 Asian countries 296 Western countries 294–6 familial VaDs 290 haemorrhagic dementia 301–2 hereditary vascular dementias 305–12 liability to intracerebral haemorrhages 305 liability to ischaemic stroke 305 see also CADASIL historical aspects 289–90 hypoperfusion (global cerebral ischaemia) 301 hypoxia–ischaemia and infarction 298–9 lesions imaging and autopsy studies 302–3 significance of site, infarctions 303–4 significance of volume 302–3 mechanisms producing dementia 296–305 cerebral vasculature organization 296–7 NINDS-AIREN 297–9 neuropathological criteria and assessment of brains 318–20 pathology 315–18 diffuse small vessel lesions 317–18 lacunar infarcts 316–17 lesions causing VCI or VaD 297–303 multi-infarct and single strategic infarct dementia 299, 315–16 small vessel disease 316 post-mortem examination, extracranial organs 52 progressive dementing syndrome 293–4 risk factors 296 small vessel disease 299–301, 304–5, 316

Index

subcortical vascular dementia 291–2 Chamorros, Guam diseases 274–7 cholinergic brain pathways, pathological mechanisms in head injury 466 cholinergic hypothesis anatomic basis, AD 10 dementia with Lewy bodies (DLB) 25–6 chorea–acanthocytosis, differential diagnoses, Huntington disease 393–4 choroid plexus, cerebrospinal fluid formation 442 chromosomes chr-1, early onset familial AD 186–7 chr-4, Huntington disease 376 chr-14, early onset familial AD 186 chr-19, late onset AD 103, 187 chr-21 APP gene overexpression 214 early onset familial AD 185–6 mosaicism 214–16 triplication in Down’s syndrome 185, 214–16 chronic renal failure 518 chronic subdural haematomas 301–2 cisternography, SPECT, shunt results in NPH 449 classification of dementia 3–4 clinical diagnosis, accuracy, AD 49 clinical evaluation 3–4 coeliac disease 526 cognitive function, effects of head injury 458–9 cognitive impairment anatomic basis in CJD and FTLD 23–4 domains 4 collagen vascular disease 315–16 concussion syndromes, axonal injury 458 confusional state, defined 2 congophilic amyloid angiopathy (CAA) 11 histopathology of AD 175–8 Consortium to Establish a Registry for Alzheimer’s Disease see CERAD cornu ammonis (CA) 34–6 alveus 36 CA1/subiculum, occurrence of NFTs 13 fimbria 36 fornix 36 neuronal disarray 502 subdivisions 34–5 see also hippocampus corpus callosum, probabilistic atlases in AD 139 corpus striatum 378 cortical atrophy 56–7, 229 posterior, patterns 18–19, 21 cortical averaging 145 cortical variability 145 tensor maps of directional variation 145 cortical disconnection theory 121 cortical modelling 143–5

averaging images or geometry 144 cortical matching 144–5 cortical neurons connectivity 38 pyramidal/non-pyramidal cells 37 cortical pattern anomalies, neuroimaging 130 cortical and subcortical regions, autonomic function 18–19 corticobasal degeneration (CBD) 234–40 anatomic basis 24–5 clinical criteria 5 clinical features 234 differential diagnosis 248 Huntington disease 393 neuronal and tau pathology spectrum 249 genetics 234–5 neuroimaging 234 pathological features 235–40 biochemical findings 239–40 gross findings 235 microscopic findings 235–9 ultrastructural findings 239 corticodentatonigral degeneration, with neuronal achromasia 393 COSHH Regulations, safety 82, 88 Cowdry inclusions 486 Creutzfeldt–Jakob disease (CJD) acquired (iatrogenic and variant) 405, 417–21, 422 anatomic basis of cognitive impairment 23–4 clinical diagnostic criteria 7, 412, 419 familial CJD 409, 410, 414–16 MM, MV, VV polymorphism 413–14 genetics 100 genotypes and phenotypes, summary 420 Heidenhain variant 7 iatrogenic CJD, clinical features and neuropathology 405, 417–18, 422 pathological diagnostic criteria 50, 412, 419 post-mortem examination 48–51 sporadic (sCJD) 407–14 clinical features 407–13 investigation 412 MM, MV, VV polymorphism 413–14 neuropathology 413–14 variant (vCJD) 418–21 clinical features 418 diagnostic criteria 419 neuropathology 411, 418–21 relationship with BSE 421 cryptococcal meningitis 477–9 Cu/Zn SOD-1 gene 215, 216, 539–42 cyanocobalamin (vitamin B[U]12[u]) deficiency 519–20 cycads 278 cyclosporin 525 encephalopathy 525 cystatin C 331, 334–5

569

570

Index

cytomegalovirus (CMV) infection 475, 476 cytoskeletal neurodegenerative pathology, head injury 463–4 Danish dementia familial 343–6, 347 see also BRI dementias definitions and clarifications 2–3 vascular dementias 290–1 dementia causes 4–7 differential diagnosis 7 etiology 4–7 dementia with Lewy bodies see diffuse Lewy body (DLB) disease dementia pugilistica 461–62 neurofibrillary tangles (NFT) 464 dentate fascia 34, 35, 36 granule cells and mossy fibres 37 and Pick body formation 37 dentate nucleus, cerebellum 45 dentato–rubro–pallido–luysian atrophy (DRPLA) 394–5, 535–6 differential diagnosis 7 diffuse Lewy body (DLB) disease 358–60 anatomic basis 25–6 Brodmann’s area-24 39 Brodmann’s maps 39 cholinergic hypothesis 25–6 clinical criteria 5 hallucinations 25 in Parkinsonism 360 disease-specific brain atlases, population-based mapping 128–30 dopamine (DA) neurotransmission, striatum 379–80 Down’s syndrome and Alzheimer’s disease (AD) 207–26 causative factors 216–17 chromosome-21 185 dementia in 217 genetic factors 214–16 gross changes in brain in Down’s syndrome 212–13 neuronal loss and neurochemical changes in elderly 212 NFT appearance 211 pathological similarities 207–8, 211–12 senile plaque appearance 208–11 temporal progression 213–14 drug-related conditions, illicit drugs 315, 474, 525 drugs and toxins, effects 525 Dutch type (HCHWA-D) hereditary cerebral haemorrhage with amyloid 335–8 East Boston study (AD) 163 education, and AD 165 embolic disease 314–15 encephalitis von Economo encephalitis 363 glial nodule encephalitis 512 herpes simplex, anatomy 18 limbic encephalitis 511–12

paraneoplastic syndrome (limbic encephalitis) 511–12 subacute sclerosing panencephalitis (SSPE) 486–9 see also herpes simplex virus encephalitis (HSVE); HIV-associated dementia entorhinal cortex anatomy 16, 34–7 convergence zone 9–10 NFTs 13, 115–20 criterion for dementia 15, 116 schizophrenia 502 stellate cells 38 epilepsy 514–15 list 515 neuropathology 514–15 Unverricht–Lundborg’s syndrome 515 ethical and social issues, genetic testing 94 Fabry’s disease 523 familial CAAs, British and Danish types see BRI dementias familial fatal insomnia 100 genotypes and phenotypes, summary 420 pathological diagnostic criteria 412 familial frontotemporal dementia 261–2 folate deficiency 519–20 fourth ventricle, progressive supranuclear palsy (PSP) 56 fractionator, stereology-based morphometric methods 79–80 frontotemporal dementias (FTD) 257–88 see also frontotemporal lobar degeneration (FTLD) amyotrophic lateral sclerosis/Parkinsonism–dementia complex 274–9 clinical features 5, 258 familial 261–2 FTD with Parkinsonism linked to chr-17 (FTDP-17) 262–73 animal models 273 biology of tau protein 263–4 clinical features 264–5 effects of mutations 268–73 genetic testing 104–6 neuropathology 265–8 tau mutations identified 262 tau pathology 266 transgenic models 546–9 genetic testing 104–6 frontotemporal lobar degeneration (FTLD) 260–1 anatomic basis of cognitive impairment 23–4 clinical criteria 5 differential diagnosis 7 frontal variant 24 neuropathology 260–1 presentation 10 progressive subcortical gliosis 261 frontotemporal lobar degeneration with motor neuron disease-type inclusions (FTLD-MND) 258–60 neuropathology 258–60 relationship between FTLD-MND and classical MND 260

Index

FTDP-17 see frontotemporal dementias (FTD) with Parkinsonism linked to chr-17 functional neuroimaging, specific disorders 22–3 fungal meningitis 477–9 galactocerebroside galactosidase deficiency, Krabbe’s disease 523 -galactosidase deficiency, Fabry’s disease 523 gangliosidoses, Gm1 and GM2 522 Gaucher’s disease, type 1 522 gelsolin 331, 348 gender and AD 165 vascular dementias and AD, Western countries 295 general paresis of insane (GPI) 490 genetic testing 94–112 carrier testing 95 categories 95–6 clinical tests 95 development 99–100 diagnostic testing 94 ethical and social issues 94 indications 94–5 laboratory issues 96–100 allele-specific mutation detection 96–8 insertions, expansions and deletions 98 mutation detection 96–9 RFLP analysis 96 screening for heterogeneous mutations 98–9 preimplantation diagnosis 95 prenatal testing 95 presymptomatic testing 94–5 prion diseases 107–8 research tests 95–6 single nucleotide polymorphisms (SNPs) 93 specific disorders 100–8 AD 103–4 CADASIL 106–7 frontotemporal dementia (FTD) 104–6 FTDP-17 105–6 Huntington disease (HD) 101–3 trinucleotide repeat diseases 98, 100–4 genetics early onset familial AD chr-1 186–7 chr-14 186 chr-21 185–6 late onset AD apolipoproteins 187–8 chr-19 103, 187 mutations allele-specific 96–8 laboratory issues in genetic testing 96–9 transgenic mouse models 533–57 variation and disease 92–4 Gerstmann–Straussler–Scheinker (GSS) syndrome 405

differential diagnosis 183 genetics 100, 107–8 genotypes and phenotypes, summary 420 GSS A117V 409 GSS D202N 409 GSS P102L 409 pathological diagnostic criteria 412 giant cell (temporal) arteritis 313 Glasgow Outcome Scale (GOS) 457 glial cells interleukins, neuroinflammation 466–7 and vascular supply of cortex 39 glial cytoplasmic inclusions (GCIs) 360–62 glial fibrillary acidic protein 434, 435 glial nodule encephalitis 512 gliosis defined 498–500 progressive subcortical, in FTLD 261 schizophrenia 498–500, 501 global cerebral ischaemia (GCI) 301 globoid cell leukodystrophy 523 globus pallidus 43 Huntington disease 385 post-mortem examination 59 glutamate neurotransmission, striatum 379–80 granular osmophilic material (GOM) 310–12 granule cells, and mossy fibres, dentate fascia 37 granulomatous angiitis (PACNS) 313 granulovacuolar degeneration 10 Guam diseases 274–7 haematological conditions 518 haematomas, subdural 301–2, 510 haemorrhagic dementia 301–2 haemosiderosis of CNS 515 Hallovorden–Spatz disease 362 hallucinations, diffuse Lewy body (DLB) disease 25 HD-IT15 CAG repeats 376 head injury 457–71 and AD 165, 459–61 animal models 463 boxers and dementia pugilistica 461–62 clinical studies 459–67 effects on cognitive function 458–9 genetic influences on outcome 467–8 long-term outcome and chronic neurodegeneration 459–66 pathological mechanisms amyloid deposition 464–5 cholinergic brain pathways 466 cytoskeletal neurodegenerative pathology 463–4 neuroinflammation 466–7 neuronal loss 465–6 hemispheric dominance 8 heparan sulphate proteoglycan (HSPG), and A 209 hepatic encephalopathy 434–6, 518–19

571

572

Index

herbicides 363 hereditary cerebral haemorrhage with amyloid Danish type see BRI dementias Dutch type (HCHWA-D) 335–8 amyloid genetics 337 neuropathology 336–7 Icelandic type (HCHWA-I) 333–5 amyloid genetics 334–5 clinical features 333–5 neuropathology 334 hereditary vascular dementias 305–12 see also CADASIL heredopathia ophthalmo-otoencephalica (familial Bri dementia Danish type) see BRI dementias herpes simplex virus encephalitis (HSVE) 17–18, 37, 476, 486–7, 488 anatomy 18 destruction of hippocampus 37 herpesviruses, HIV-1 infection and AIDS 476 heteromodal (higher order) association areas 9 hippocampus anatomy 16 atrophy 58 connections with limbic system 36, 37 diffuse plaques vs neuritic plaques 172 head 41 isolation in early AD 17 large pyramidal cells 37 stratum pyrimidale 34, 35 see also cornu ammonis (CA); dentate fascia Hirano bodies, Pick’s disease 231 histology 61–9 see also brain blocks HIV-associated dementia (HAD) 473–86 cellular/molecular pathogenesis 485–6 clinical features and neuroimaging 481–83 clinicopathologic features 483–5 HIV encephalitis (HIVE) 483–5 neuropathological manifestations 474 secondary to opportunistic infections, lymphoma and vascular disease 473–81 cytomegalovirus (CMV) infection 475 fungal meningitis 477–9 herpesviruses 476 mass CNS lesions 475 primary CNS lymphoma (PCNSL) 479–81 progressive multifocal leukoencephalopathy (PML) 476–7 human genome 92–4 huntingtin 389–41, 536–7 Huntington disease 376–401 basal ganglia system 378–80 glutamate and dopamine (DA) neurotransmission in striatum 379–80 neostriatal neurons, classification 379 pathways 378 striosome-matrix compartments 378

clinical and genetic features 376–8 clinicopathological discrepancies and differential diagnoses 392–5 corticodentatonigral degeneration with neuronal achromasia 393 dentato–rubro–pallido–luysian atrophy 394–5 multiple system atrophy and corticobasal degeneration 393 neuroacanthocytosis (chorea–acanthocytosis, McLeod syndrome) 393–4 Pick disease 393 progressive supranuclear palsy 393 coexistence with AD 392 concomitant neuropathological findings 391–92 genetic testing 101–3 genetics 100 grading 385, 388 mutant huntingtin putative pathogenic mechanisms 390–92 wild type characteristics 389–90 neuropathology 380–89 cerebellum 387–8 cerebral cortex 385–6 general features 380–81 globus pallidus 385 grading system 389 historical aspects 380 neostriatal degeneration, relationship to other brain changes 383–5 neostriatal neurons, relative vulnerability 388–9 striatal neuropathy grading 381–83 thalamus, substantia nigra and subthalamic nucleus 387 nonagenarians 392 hydrocephalus 442–56 blood–brain interfaces 443–5 cerebrospinal fluid physiology 442–3 non-communicating/communicating 445 causes 446 normal pressure 6–7, 445–9 causes 446 hypertension 312 hypertrophic (craniocervical) pachymeningitis 491 hypocalcaemia with calcification of basal ganglia 521 hypoglycaemia 519 hypoperfusion (global cerebral ischaemia) 301 hypothalamus see thalamus and hypothalamus hypoxia 517–18 hypoxia–ischaemia and infarction 298–9 Icelandic type (HCHWA-I) hereditary cerebral haemorrhage with amyloid 333–5 idiopathic hypertrophic (craniocervical) pachymeningitis 491 illicit drugs, drug-related conditions 315, 474, 525 immunocytochemistry, viral/parasitic organisms 65 infarction angular gyrus infarcts 303

Index

hypoxia–ischaemia 298–9 lacunar infarcts 300, 304–5 multi-infarct and single strategic infarct dementia 299, 315–16 thalamus 304 infectious (and inflammatory) conditions 472–96, 511–12 Behcet’s syndrome 511 glial nodule encephalitis 512 HIV-associated dementia 473–86 idiopathic hypertrophic (craniocervical) pachymeningitis 491 list 473 neurosyphilis and Lyme disease 489–40 other bacterial infections and meningitides 491 other viral infections 486–9 paraneoplastic syndrome (limbic encephalitis) 511–12 safety precautions 82–90 sarcoidosis 511 inflammatory conditions 511–12 insertional mutations, inherited prion diseases 416 interleukins IL-1 polymorphisms 467 neuroinflammation 466–7 internet see online resources intracerebral haemorrhage (ICH) 301–2 intracranial neoplasms 510–11 iron, neurodegeneration with brain iron accumulation type-1 (NBAI-1) 362–3 Khachaturian criteria, Alzheimer’s disease (AD) 180–1, 499 Korsakoff’s psychosis 429–33 Krabbe’s disease (globoid cell leukodystrophy) 523 Kuf’s disease (adult onset neuronal ceroid lipofuscinosis) 520–21 Kuru, clinical features and neuropathology 416–17 lacunar state 447–8 lacunes, lacunar infarcts 300, 304–5, 317 Lafora (adult polyglucosan body) disease 520 Leigh’s encephalopathy 524–5 ´ L’Etat Criblé 317 leukoaraiosis 300, 302 leukodystrophy, globoid cell 523 leukoencephalopathy amphotericin B 525 multifocal, with calcification 526–7 progressive multifocal (PML) 476–7 sclerosing, polycystic lipomembranous osteodysplasia 521 subcortical 299–300, 318, 319 see also CADASIL; white matter lesions (WML) Lewy bodies (LBs) classical 357–60, 359 in DS and AD 211 models 548–9 see also diffuse Lewy body (DLB) disease lignoceroyl CoA synthetase deficiency (adrenoleukodystrophy) 521 limbic encephalitis 511–12

limbic regions 9 medial temporal 16–17 neuritic plaques in early AD 15 limbic system, connections with hippocampus 36 locus ceruleus 60–1 lupus, systemic lupus erythematosus (SLE) 313–14 Lyme disease 490–91 lymphoma angiotropic large cell lymphoma 315 primary CNS lymphoma (PCNSL), HIV-1 infection 479–81 lymphomatoid granulomatosis 315 McLeod syndrome (neuroacanthocytosis, chorea–acanthocytosis), differential diagnoses, Huntington disease 393–4 mamillary bodies 40, 44 mamillo-thalamic tract 36 Marchiafava–Bignami disease 436 Marianas dementia 274, 278 measles, subacute sclerosing panencephalitis (SSPE) 486–9 medulla, nuclei 45 meningioma-like mass lesion 491 meningitides 491 fungal 477–9 idiopathic hypertrophic (craniocervical) pachymeningitis 491 meningovascular syphilis 490 mesial temporal limbic region in AD 16–17 anatomy 16–17 MCI 15 metabolic diseases 516–26 metachromatic leukodystrophy 523 methionine, homozygosity, and CJD 7 methotrexate and other cancer therapy 525 microglia, interleukins, neuroinflammation 466–7 microscopy, CERAD protocol 65–6 mild cognitive impairment (MCI) amnesic, pathology spread to entorhinal cortex 17 defined 2 pathology limited to mesial temporal region 15, 16 mitochondrial disorders 524 mitochondrial cytopathies (encephalomyopathies) 524 models see animal models; transgenic mouse models of neurodegenerative disease molecular diagnosis 91–112 future technologies 108 genetic testing 94–108 online resources 108–9 see also genetic testing morphometric methods 75–81 profile-counting 75–6 stereology-based 76–80 mossy fibres, and granule cells, dentate fascia 37 motor neuron disease (MND)-type inclusions 258–60 MPTP 363 mucopolysaccharidosis type III-B (San-Filippo’s disease) 522–3

573

574

Index

multi-infarct dementia (VaD) 290, 315–16 and single strategic infarct dementia 299, 315–16 multifocal leukoencephalopathy with calcification 526–7 progressive 526–7 multiple cavernous angiomas 315, 316, 527 multiple sclerosis 513–14 aetiology 514 differential diagnosis 514 neuropathology 513–14 multiple system atrophy 360–62 and corticobasal degeneration, vs Huntington disease 393 neocortex neuroanatomy 37–9 vascular supply 39 neostriatal neurons classification 379 dark neurons (NDN) 382 glutamate and dopamine (DA) neurotransmission 379–81 Huntington disease 388–9 neuritic plaques see senile plaques neuroacanthocytosis (chorea–acanthocytosis, differential diagnosis, Huntington disease 393–4 neuroanatomy of dementia 8–26 convergence zones 9 cortices, connectivity 9–10 normal functional neuroanatomy 8–9 see also anatomy neurodegenerative conditions with brain iron accumulation type-1 (NBAI-1) 362–3 thalamic degeneration 509–10 transgenic mouse models 533–57 see also specific conditions neurofibrillary tangles (NFT) 113–20, 172–4 anatomic distribution 12 characteristics 113–15 composition 191 conformation 173 correlation with AD 116–20, 172–4 vs plaques 22 dementia pugilistica 464 hierarchy of region involvement 13 histology 11–13 limited to mesial temporal region in MCI 15 and neuropil threads 11 normal ageing 13 centenarians 117–20 distribution 113–15 number, and patient age 13 stellate cells 38 neuroimaging 51, 128–60 analysing brain data 132–3 atlas-based pathology detection 140–3

brain atrophy 51 brain averaging 145–51 clinical diagnosis 166–8 cortical averaging 145 cortical modelling 143–5 cortical pattern anomalies 130 corticobasal degeneration (CBD) 234 dynamic (4D) brain atlases 151–4 evaluation 4 functional 22–3 individualized brain atlases 133–5 model-driven deformable atlases 135–6 MRI 166–7, 306–7 population-based brain mapping 128–31 probabilistic atlases 136–40 specific disorders AD 22–3 argyrophilic grain disease (AGD) 246 BRI dementias 339, 344 CADASIL 306–7 HIV-associated dementia (HAD) 481–83 Pick’s disease (PiD) 229 progressive supranuclear palsy (PSP) 240 SPECT 51, 166–7 prediction of shunt results 449 types of brain atlases 131–2 see also brain atlases neuroinflammation head injury 466–7 see also infectious (and inflammatory) conditions neuronal achromasia, with corticodentatonigral degeneration 393 neuronal ceroid lipofuscinosis, adult onset 520–21 neuronal loss in AD 11, 80, 116–20 mild vs severe 116 alcohol-related 428–9, 433 apoptosis 465–6, 485 CBD 235 in elderly with DS and AD 212 head injury 465–6 neurodegeneration in head injury 465–6 neuroinflammation 467 neuron counting, stereology-based optical disector 77, 78–80 NMDA-mediated neuronal death 430 normal ageing 115–17 Pick’s disease (PiD) 231 programmed cell death vs necrosis 465–6 neuropil threads 11 neurosyphilis 489–90 general paresis of insane (GPI) 490 meningovascular syphilis 490 neurotrophic factor S-100 215 neurotropism 485 niacin deficiency (pellagra) 436–7

Index

Niemann–Pick disease 522 NMDA receptors 379–80 NMDA-mediated neuronal death 430 non-amyloid component of SP precursor protein (NACP) 353 normal pressure hydrocephalus (NPH) 445–55 clinical criteria 6–7 diagnosis 447–9, 450 neuropathological findings 449–53 post-mortem examination 451–52 procedure for diagnosis 453–5 shunting 448–9 Notch signalling, Notch3 mutations in CADASIL 106, 107, 307–12, 309 nucleus basalis (of Meynert) see basal nucleus (of Meynert) olivopontocerebellar atrophy (OPCA) 360 online resources brain banks 558–63 genetic counselling 109 genetic testing resources 109 laboratory resources 109 mutation databases 108–9 scientific and clinical societies 109 optical disector counting frame 77, 78–80 coefficient of error 80 osteodysplasia, polycystic lipomembranous, with sclerosing leukoencephalopathy 521 oxidative injury, Parkinson’s disease 363 paired helical filaments (PHF), formation, phosphorylation of tau 191–3 pancreatic disorders 519 pantothenate kinase-2 (PANK2) 362 paraneoplastic disease vs Creutzfeldt–Jakob disease 50 limbic encephalitis 511–12 post-mortem examination 52, 53 parkin gene 364–5 Parkinsonism–dementia complex FTD-linked see frontotemporal dementia with Parkinsonism linked to chr-17 (FTDP-17) mixed pathology with AD 67–8, 274–9 see also amyotrophic lateral sclerosis/Parkinsonism–dementia complex Parkinson’s disease 355–8, 363–5 differential diagnosis 5–6 etiology 363–5 environmental factors 363 genetics 363–5, 364 and multiple system atrophy 360–62 oxidative injury 363 substantia nigra in 62, 355 see also diffuse Lewy body (DLB) disease pathological diagnosis 48–74

mixed pathology 66–7 see also biopsy; post-mortem examination PCR-RFLP, genetic testing 96, 97 pellagra 436–7 pericerebral space, and brain volume 428 perirhinal cortex, occurrence of NFTs 13 pesticides 363 Pick bodies 229–33 Pick’s disease (PiD) 37, 227–34 Brodmann’s maps 37 clinical features 228–34 cortical atrophy 56–7 differential diagnosis 248 Huntington disease 393 neuronal and tau pathology spectrum 249 genetics 229 neuroimaging 229 neuronal loss 231 pathological features 229–34 biochemical findings 233–4 gross findings 229 microscopic findings 229–32 ultrastructural findings 232–3 subtypes 228 polycystic lipomembranous osteodysplasia with sclerosing leukoencephalopathy 521 polyglutaminopathies 376, 535–8 pons, anatomy 45 porphyria 520 portal–systemic encephalopathy 434–6, 518–19 post-mortem examination 48–69, 168–70 accidents protocol 88 assessment scheme (proposed) for CVD and VaD 321–22 brain blocks, selection 61–9 brain slices 168–70 brain weight 168 cerebral atrophy 14, 56, 168 consent issues 51–2 cortical atrophy 56–8 external appearance 168 extracranial organs 52 fixed brain 54–6 fresh brain 52–4 other changes 58–61 paraneoplastic disease 52, 53 prion diseases 48–51 procedure for diagnosis (NPH) 453–5 safety precautions 82–90 preimplantation genetic diagnosis 95 prenatal genetic testing 95 presenilins 542 PS-2 103, 186 presymptomatic genetic testing 94–5 primary dementia 3

575

576

Index

prion diseases 402–26, 534 acquired 416–21, 422 cases classified at autopsy 84 classification 404, 405 decontamination procedures 87 diagnostic criteria 50–1 genetic basis 107–8, 421–22 genotypes and phenotypes, summary 420 iatrogenic CJD 405, 417–18, 422 idiopathic 421 inherited 421–22 clinical features and neuropathology 414–16 insertional mutations 416 nosocomial transmission 50 investigation and neuropathology 405–7 Kuru, clinical features and neuropathology 416–17 post-mortem examination 48–51 transgenic mouse models 533–5 Western blot analysis of CJD brain 423 see also Creutzfeldt–Jakob disease; familial fatal insomnia; Gerstmann–Straussler–Scheinker (GSS) syndrome; prion protein prion protein cerebral amyloid angiopathies (CAAs) 331, 348 PrPc 403 conversion to PrPSc 403, 405 PrPRes isotypes 404, 421–23 acquired forms 422 familial forms 421–22 idiopathic forms 421 immunocytochemistry on paraffin-embedded sections 421 Western blot analysis 404, 423 PrP Sc 403–4 structural variation 404 prion protein gene PRNP 402–3, 420, 421–22, 534–5 profile-counting morphometric methods 75–6 programmed cell death, vs necrosis, neuronal loss 465–6, 485 progressive dementing syndrome CVD-related 293–4 and head injury 165 progressive multifocal leukoencephalopathy (PML) 476–7 progressive non-fluent aphasia 24 progressive supranuclear palsy (PSP) 240–5 clinical criteria 5–6 clinical features 240 differential diagnosis 248 Huntington disease 393 neuronal and tau pathology spectrum 249 genetics 240–1 neuroimaging 240 pathological findings 241–5 fourth ventricle 56 gross findings 241 microscopic findings 241–4

thalamus 61 ultrastructural findings 244–5 PS-1 gene 215 putamen 41–3 diffuse plaques vs neuritic plaques 171–2 post-mortem examination 59 pyramidal/non-pyramidal cells cortical neurons 37 neurotransmitters 37–8 radiotherapy, cerebral irradiation damage 511 recreational drugs, drug-related conditions 315, 474, 525 renal failure 518 restriction fragment length polymorphism (RFLP) analysis, laboratory issues in genetic testing 96, 97 rotenone 363 safety 82–90 brain banks 87–8 COSHH Regulations 82, 88 hazards and risks 82–6 laboratories, categories 86–8 Western blot analysis, prion diseases 404, 423 see also prion diseases San-Filippo’s disease 522–3 sarcoidosis 511 Schaffer collaterals 37 schizophrenia 497–508 antipsychotic drugs and neurofibrillary pathology 500 cognitive impairment 497 cytoarchitectural neuropathology 500, 503 frontal cortex 500 mediodorsal thalamus 500 gliosis 498–500 prevalence of Alzheimer’s disease 498 sclerosing leukoencephalopathy, with polycystic lipomembranous osteodysplasia 521 secondary dementia 3–4 semantic dementia 24 senile plaques (SP) 11, 170–2 anatomic distribution 12 characteristics 115 correlation with AD, vs neurofibrillary tangles 22 diffuse vs neuritic plaques 170–1, 190–1 formation 190 historical aspects 161–2 non-amyloid component of SP precursor protein (NACP) 353, 353, 365–8 normal ageing, distribution 113–15 short tandem repeat (STR) 100–1 Shy–Drager syndrome 360–61 single nucleotide polymorphisms (SNPs) 93 small vessel disease 299–301, 304–5, 316 diffuse lesions 317–18 hereditary 305–12

Index

lacunes 300, 304–5 white matter lesions 300–1, 305 Sneddon’s syndrome 314 SOD-1 gene 215, 216, 539–42 space-occupying lesions 510–1 intracranial neoplasms 510–11 subdural haematomas 301–2, 510 sphingomyelin, Niemann–Pick disease 522 spinal cord, post-mortem examination 52 spongiform change, conditions resembling 407 spongiform encephalopathies see prion diseases sport injury, dementia pugilistica 461–62 stains in histology 61–9 stellate cells, entorhinal cortex, NFT formation 38 stereology-based morphometric methods 76–80 brain volume determination 76–8 fractionator 79–80 optical disector 77, 78–80 systematic random sampling 76 strategic single infarct dementia 299, 315–16 striatum see neostriatum striosome-matrix compartments 378 stroke dementia following stroke 293–4 clinically certified stroke 293 hereditary vascular dementias 305–12 liability to ischaemic stroke 305 see also CADASIL subacute sclerosing panencephalitis (SSPE) 486–9 subarachnoid haemorrhage (SAH) 301–2 subcortical leukoencephalopathy (Binswanger’s disease) 299, 300, 318, 448 subcortical vascular dementia 291–2 subdural haematomas 301–2, 510 subiculum anatomy 34–7 connections 37 substantia nigra (SN) 44, 378 CBD 237 head injury 463 Huntington disease 387 Parkinson’s disease 62, 355, 356 Pick’s disease 232 progressive supranuclear palsy 241 subthalamic nucleus 41–3, 378 Huntington disease 387 progressive supranuclear palsy 241 superficial haemosiderosis of central nervous system 515 superoxide dismutase, Cu/Zn SOD-1 gene 215, 216, 539–42 -syn 353, 365–8 synuclein protein family 353–5 amino acid sequences 354 synucleinopathies (-synuclein inclusion diseases) 355–67 mechanisms of polymerization and inclusion formation 365–8

future direction and therapeutic interventions 368 in vitro modelling of fibrillogenesis 365–7 transgenic animal models 367–8 neurodegeneration with brain iron accumulation type-1 (NBAI-1) 362–3 other synucleinopathies 362–3 transgenic mouse models 546–9 see also diffuse Lewy body disease; multiple system atrophy; Parkinson’s disease syphilis see neurosyphilis systemic lupus erythematosus (SLE) 313–14 tau biology of tau protein 263–4 characteristics 11, 263–4 fibrillar 227 FTDP-17 105–6 animal models 273 mutations 268–73 mutations identified 262 pathology 266 function as MT binding protein 264 gene, representation 264 isoforms 227 mRNA 104–5 mutations 105 pathology ALS/PDC 277 FTDP-17 266, 269 phosphorylation, and formation of PHF 191–3, 264 Pick’s disease vs AD 233 regulation, association with amyloid 193–4 Western blot analysis, banding patterns 270, 271 see also neurofibrillary tangles; neuropil threads tauopathies 227–56, 257–88 hereditary 257–88 see also frontotemporal dementias sporadic argyrophilic grain disease 245–50 compared 228 corticobasal degeneration 234–40 differential diagnosis 248 Pick’s disease 227–34 progressive supranuclear palsy 240–5 transgenic mouse models 546–9 ‘thalamic dementias’ 290 thalamus and hypothalamus anatomy 43–4 Huntington disease 387 infarction 304 nuclei 44 post-mortem examination 60 thalamic degeneration 509–10 thiamin deficiency 429–30 toxoplasmosis, HIV-associated, mass CNS lesions 475

577

578

Index

transentorhinal cortex 16 transgenic mouse models of neurodegenerative disease 533–57 AD 542–6 amyotrophic lateral sclerosis 539–42 CAG repeat disorders 535–8 prion disorders 533–5 synucleinopathies 367–8, 546–9 tauopathies 546–9 ubiquitin 549–50 transmissible spongiform encephalopathies see prion diseases transthyretin 331, 346–8 trauma see head injury trinucleotide repeat diseases 535–8 categories 101 genetic basis 100–1 transgenic mouse models of CAG repeat disorders 535–8 see also Huntington disease trinucleotide repeats 98 tumour-related vascular disease 315 tyrosine kinase binding protein 521 ubiquitin immunoreactivity 358, 541 nuclear inclusions 384 transgenic mouse models of neurodegenerative disease 549–50 UBOs 51 Unverricht–Lundborg’s syndrome 515 valine, homozygosity, and CJD 7 varicella zoster virus (VZV) 476, 487 vascular cognitive impairment (VCI) 292 vascular dementias see cerebral amyloid angiopathies (CAAs); cerebrovascular disease and vascular dementias vascular smooth muscle cells (VSMC), GOM 310–12

vasculitides 301, 313 vegetative state 458–9 ventricular size, variations 57, 58 viral infections 486–9 herpes simplex virus encephalitis (HSVE) 17–18, 37, 476, 486–7, 488 subacute sclerosing panencephalitis (SSPE) 486–9 varicella zoster virus (VZV) 487 Virchow–Robin spaces, dilatation 60, 177 vitamin B1 , deficiency 429–30 vitamin B12 , and folate deficiencies 519–20 von Economo’s encephalitis 363 Wallerian degeneration, white matter lesions (WML) 40, 458, 460, 461 web see online resources Wernicke–Korsakoff syndrome alcoholism 429–33, 436 neuropathology 431–33 Western blot analysis prion diseases 404, 423 tau 270, 271 Whipple’s disease 516 white matter lesions (WML) small vessel disease 300–1, 305 diffuse/rarefaction 317–18 vacuolation, conditions resembling spongiform change 406, 407 Wallerian degeneration in dementias 40, 458, 460, 461 see also CADASIL; cerebral white matter; leukoWilson’s disease 519 xanthomatosis 521 zoonoses 418–21 see also Creutzfeldt–Jakob disease, variant CJD

The Neuropathology of Dementia, Second edition

Forensic Neuropathology, Second Edition

Dementia, 3rd edition

Neuropathology Review

Dementia

Dementia

Dementia

Dementia

Dementia

Dementia

Neuropathology Review

Neuropathology Review

Forensic Neuropathology

Forensic Neuropathology,

Principles and Practice of Neuropathology

Forensic Neuropathology and Neurology

Forensic Neuropathology and Neurology

The Dementia Care Workbook

Management of Dementia

Management of Dementia

Neurobiology of Dementia

The Prince (Second Edition)

The Ovary, Second Edition

The Lymphomas, Second Edition

The Fungi, Second Edition

The Stoics, Second Edition

The Myocardium, Second Edition

The Mechanics of the Circulation, Second Edition

Atlas of Post-Mortem Techniques in Neuropathology

Severe Dementia

The Neuropathology of Dementia, Second edition

Forensic Neuropathology, Second Edition

Dementia, 3rd edition

Neuropathology Review

Dementia

Dementia

Dementia

Dementia

Dementia

Dementia

Neuropathology Review

Neuropathology Review

Forensic Neuropathology

Forensic Neuropathology,

Principles and Practice of Neuropathology

Forensic Neuropathology and Neurology

Forensic Neuropathology and Neurology

The Dementia Care Workbook

Management of Dementia

Management of Dementia

Neurobiology of Dementia

The Prince (Second Edition)

The Ovary, Second Edition

The Lymphomas, Second Edition

The Fungi, Second Edition

The Stoics, Second Edition

The Myocardium, Second Edition

The Mechanics of the Circulation, Second Edition

Atlas of Post-Mortem Techniques in Neuropathology

Severe Dementia

Recommend Documents