Infection and Autoimmunity

List of Contributors Mahmoud Abu-Shakra Autoimmune Rheumatic Diseases Unit Soroka Medical Center Ben-Gurion University...

Author: Y. Shoenfeld (Editor) | Noel Richard Rose (Editor)

64 downloads 1271 Views 78MB 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

List of Contributors

Mahmoud Abu-Shakra Autoimmune Rheumatic Diseases Unit Soroka Medical Center Ben-Gurion University Beer-Sheva Israel

Anat Achiron Multiple Sclerosis Center Sheba Medical Center Tel-Hashomer Israel Miranda K. Adelman Arizona Arthritis Center The University of Arizona Tucson, AZ USA Marina Afanasyeva Department of Physiology and Biophysics Faculty of Medicine University of Calgary Calgary, Alberta Canada Jun Akaogi Division of Rheumatology & Clinical Immunology Center for Systemic Autoimmune Diseases University of Florida Gainesville, FL USA Jorge Alcocer-Varela Department of Immunology and Rheumatology Instituto Nacional de Ciencias M6dicas y Nutrici6n Salvador Zubir~in Mexico City Mexico

Amedeo Amedei Department of Internal Medicine University of Florence Florence Italy

Ben J. Appelmelk Department of Medical Microbiology Vrije Universiteit, Medical School Amsterdam The Netherlands

Anabel Aron-Maor Center for Autoimmune Diseases Department of Medicine 'B' Sheba Medical Center Tel Hashomer Israel

Ronald A. Asherson Rheumatic Diseases Unit University of Cape Town School of Medicine Cape Town South Africa

Leonard H. van den Berg Department of Neurology University Medical Center Utrecht Utrecht The Netherlands Mathijs P. Bergman Department of Medical Microbiology Vrije Universiteit, Medical School Amsterdam The Netherlands

~

Vll

M. Blank Internal Medicine 'B' The Center of Autoimmune Diseases Sheba Medical Center Tel Hashomer Israel

Stefano Bombardieri Rheumatology Unit Department of Intemal Medicine University of Pisa, Medical School Pisa Italy

Milan Buc Department of Immunology School of Medicine Bratislava Slovak Republic

James Bussei Pediatric Hematology/Oncology Weill Medical College of Cornell University New York, NY USA

Irun R. Cohen Department of Immunology The Weizmann Institute of Science Rehovot Israel

Pascal Cohen Department of Internal Medicine H6pital Cochin University of Paris Paris France

J.C. Crispin Acufia Department of Immunology and Rheumatology Instituto Nacional de Ciencias M6dicas y Nutrici6n Salvador Zubirdn Mexico City Mexico

Edecio Cunha-Neto Laboratory of Immunology Heart Institute (Incor) University of Sao Paulo, School of Medicine S~o Paulo Brazil

Franceseo Cainelli Section of Infectious Diseases Department of Pathology University of Verona Verona Italy

Maurizio Cutolo Department of Internal Medicine University of Genova Genova Italy

Ricard Cervera Department of Autoimmune Diseases Institut Clinic d'Infeccions i Immunologia Hospital Clinic Barcelona, Catalonia Spain

Michael David Rabin Medical Center Department of Dermatology Petah-Tiqva Israel

Alexander J. Chou Pediatric Hematology/Oncology Weill Medical College of Comell University New York, NY USA

Mario M. D'Elios Department of Internal Medicine University of Florence Florence Italy

viii

Gianfranco Del Prete Department of Internal Medicine University of Florence Florence Italy

A.M. Denman Northwick Park Hospital Harrow England Barbara Detrick School of Medicine The Johns Hopkins University Baltimore, MD USA Andrea Doria Division of Rheumatology Department of Medical and Surgical Science University of Padova Padova Italy C. Dugu6 Laboratory of Immunology Brest University Medical School Brest France

Clodoveo Ferri Rheumatology Unit Department of Internal Medicine University of Modena, Medical School Modena Italy

Johan FrostegArd Department of Medicine Unit of Rheumatology Karolinska Hospital Stockholm Sweden

Pier Franca Gambari Division of Pdaeumatology Department of Medical and Surgical Science University of Padova Padova Italy

Marjorie A. Garvey National Institute of Mental Health Department of Health and Human Services Bethesda, MD USA

Malarvizhi Durai Department of Microbiology and Immunology University of Maryland, School of Medicine Baltimore, MD USA

M. Eric Gershwin Division of Rheumatology, Allergy and Clinical Immunology University of California at Davis Davis, CA USA

Alan Ebringer Division of Life Sciences King's College University of London London UK

Anna Ghirardello Division of Rheumatology Department of Medical and Surgical Science University of Padova Padova Italy

Miriam Eisenstein Department of Chemical Services The Weizmann Institute of Science Rehovot Israel

Roberto Giacomelli Department of Internal Medicine University of L'Aquila L'Aquila Italy

ix

Serena Guiducci Department of Rheumatology University of Florence Florence Italy

Simone Gonqaives Fonseca Laboratory of Immunology Heart Institute (Incor) University of S~o Paulo, School of Medicine Sao Paulo Brazil Luiza Guilherme Heart Institute (InCor) University of S~o Paulo, School of Medicine Sao Paulo Brazil Lo'ic Guillevin Department of Internal Medicine H6pital Cochin University of Paris Paris France David A. Hailer Laboratory of Molecular Immunology Center for Neurologic Diseases Brigham and Women's Hospital and Harvard Medical School Boston, MA USA Giilen Hatemi Department of Internal Medicine Division of Rheumatology Cerrahpasa Medical School Istanbul University Istanbul Turkey Xiao-Song He Division of Rheumatology, Allergy and Clinical Immunology University of California at Davis Davis, CA USA

John J. Hooks Laboratory of Immunology National Eye Institute National Institutes of Health Bethesda MD USA

Luca Iaccarino Division of Rheumatology Department of Medical and Surgical Science University of Padova Padova Italy

Steven Jacobson Viral Immunology Section NIH/NINDS Bethesda, MD USA

M.D. Jansen Department of Neurology University Medical Center Utrecht Utrecht The Netherlands

Luis J. Jara Clinical Research Unit Hospital de Especialidades Centro M6dico La Raza, IMSS Mexico City Mexico Hee-Sook Jun Center for Immunologic Research and Department of Microbiology and Immunology The Chicago Medical School North Chicago, IL USA Jorge Kalil Heart Institute- InCor University of S~o Paulo, School of Medicine S~o Paulo Brazil

Efstathia K. Kapsogeorgou Department of Pathophysiology Medical School National University of Athens Athens Greece

Dimitrios A. Liakos Department of Pathophysiology Medical School National University of Athens Athens Greece

Leo Kei Iwai Laboratory of Immunology Heart Institute (Incor) University of Sao Paulo, School of Medicine S~o Paulo Brazil

Dong-Gyun Lim Laboratory of Molecular Immunology Center for Neurologic Diseases Brigham and Women's Hospital and Harvard Medical School Boston, MA USA

Kindra M. Kelly Division of Rheumatology & Clinical Immunology Center for Systemic Autoimmune Diseases University of Florida Gainesville, FL USA

Burkhard Ludewig Kantonal Hospital St. Gallen Research Department St. Gallen Switzerland

Ilan Krause Department of Medicine E Rabin Medical Center Petah-Tikva Israel

John J. Marchalonis Microbiology and Immunology The University of Arizona Tucson, AZ USA

Philippe Krebs Kantonal Hospital St. Gallen Research Department St. Gallen Switzerland Yoshiki Kuroda Division of Rheumatology & Clinical Immunology Center for Systemic Autoimmune Diseases University of Florida Gainesville, FL USA Aaron Lerner Department of Pediatrics Carmel Medical Center Haifa Israel

Marco Matucci Cerinic Department of Rheumatology University of Florence Florence Italy Gabriela Medina Clinical Research Unit Hospital de Especialidades Centro Mrdico La Raza, IMSS Mexico City Mexico Stephen D. Miller Department of Microbiology-Immunology Feinberg School of Medicine Northwestern University Chicago, IL USA

xi

Gui Milo Department of Medicine E Rabin Medical Center Petah-Tikva Israel

Carmen Navarro Molecular Biology Department Instituto Nacional de Enfermedades Respiratorias Mexico City Mexico

Daniel Mimouni Rabin Medical Center Department of Dermatology Petah-Tiqva Israel

Robert Nussenblatt Laboratory of Immunology National Eye Institute National Institutes of Health Bethesda MD USA

Juan M. Miranda Rheumatology Department Hospital de Especialidades Centro M6dico La Raza, IMSS Mexico City Mexico

Angelina Morand B. Bilate Laboratory of Immunology Heart Institute (Incor) University of S~o Paulo, School of Medicine S~o Paulo Brazil

Kamal D. Moudgil Department of Microbiology and Immunology University of Maryland, School of Medicine Baltimore, MD USA Haralampos M. Moutsopoulos Department of Pathophysiology Medical School National University of Athens Athens Greece Dina C. Naeionales Division of Rheumatology & Clinical Immunology Center for Systemic Autoimmune Diseases University of Florida Gainesville, FL USA

xii

Christian Pagnoux Department of Internal Medicine HOpital Cochin University of Paris Paris France

Michael P. Pender Department of Medicine Clincial Sciences Building Royal Brisbane and Women's Hospital Herston Queensland Australia

Jaques-Olivier Pers Laboratory of Immunology Brest University Medical School Brest France W-Ludo van der Pol Department of Neurology University Medical Center Utrecht Utrecht The Netherlands Mark E Prummel Department of Endocrinology & Metabolism Academic Medical Center University of Amsterdam The Netherlands

Francisco J. Quintana Department of Immunology The Weizmann Institute of Science Rehovot Israel

B. Rager-Zisman Department of Microbiology and Immunology The University Center for Cancer Research Ben Gurion University of the Negev Beer Sheva Israel

Taha Rashid Division of Life Sciences King's College University of London London UK

Westley H. Reeves Division of Rheumatology & Clinical Immunology Center for Systemic Autoimmune Diseases University of Florida Gainesville, FL USA

Shimon Reif Division of Pediatric Gastroenterology Dana Children's Hospital Tel Aviv Israel

Jozef Rovensk~ National Institute for Rheumatic Diseases Pie~t' any Slovakia

Minoru Satoh Division of Rheumatology & Clinical Immunology Center for Systemic Autoimmune Diseases University of Florida Gainesville, FL USA

Reinhold E. Schmidt Department of Clinical Immunology Medical School Hannover Hannover Germany

Bruno Seriolo Department of Internal Medicine University of Genova Genova Italy

Marianne C. Severin Center for Autoimmune Diseases Department of Medicine 'B' Sheba Medical Center Tel Hashomer Israel

Yves Renaudineau Laboratory of Immunology Brest University Medical School Brest France

Yehuda Shoenfeld Center for Autoimmune Diseases Department of Medicine 'B' Sheba Medical Center Tel Hashomer Israel

Noel R. Rose Center for Autoimmune Disease Research Department of Pathology The John Hopkins Medical Institutions Baltimore, MD USA

Jean Sibilia Rheumatology department H6pitaux Universitaire de Strasbourg Universit6 Louis Pasteur Strasbourg France

xiii

Lisa A. Snider National Institute of Mental Health Department of Health and Human Services Bethesda, MD USA

Moshe Tishler Department of Medicine 'B' Asaf Harofe Medical Center Zerifin Israel

Samantha S. Soldan Viral Immunology Section NIH/NINDS Bethesda, MD USA

Yaron Tomer Division of Endocrinology Department of Medicine Mount Sinai School of Medicine New York, NY USA

N.M. van Sorge Departments of Neurology and Immunology University Medical Center Utrecht Utrecht The Netherlands

Allen C. Steere Center for Immunology and Inflammatory Diseases Division of Rheumatology, Allergy and Immunology Massachusetts General Hospital Harvard Medical School Boston, MA USA

Matthias Stoll Department of Clinical Immunology Medical School Hannover Hannover Germany

Alberto Sulli Department of Internal Medicine University of Genova Genova Italy Susan E. Swedo National Institute of Mental Health Department of Health and Human Services Bethesda, MD USA

xiv

Alan Tyndall Department of Rheumatology University of Basle Basle Switzerland

Christina M. Vandenbroucke-Grauls Department of Medical Microbiology Vrije Universiteit, Medical School Amsterdam The Netherlands

Carol L. VanderLugt-Castaneda Department of Biology Indiana University Northwest Gray, IN USA

Dimitrios Vassilopoulos Department of Medicine Hippokration General Hospital University of Athens School of Medicine Athens Greece Sandro Vento Section of Infectious Diseases Department of Pathology University of Verona Verona Italy

Olga Vera-Lastra Internal Medicine Department Hospital de Especialidades Centro M6dico La Raza, IMSS Mexico City Mexico

Ronald Villanueva Division of Endocrinology Department of Medicine Mount Sinai School of Medicine New York, NY USA

Dominique Wachsmann Unit6 INSERM U392 "Immunit6-Infection" Universit6 Louis Pasteur Illkirch-Graffenstaden France

Joel V. Weinstock University of Iowa Hospitals and Clinics Division of Gastroenterology Iowa City, IA USA

Georg Wick Institute for Pathophysiology University of Innsbruck Medical School Innsbruck Austria Wilmar M. Wiersinga Department of Endocrinology & Metabolism Academic Medical Center University of Amsterdam The Netherlands Clyde Wilson Division of Life Sciences King's College University of London London UK

Jan G.J. van de Winkel Department of Immunology University Medical Center Utrecht Utrecht The Netherlands

Kai W. Wucherpfennig Department of Cancer Immunology & Aids Dana Farber Cancer Institute Boston, MA USA

Qingbo Xu Department of Cardiological Sciences St. George's Hospital Medical School London UK

Hasan Yazici Department of Internal Medicine Division of Rheumatology Cerrahpasa Medical School Istanbul University Istanbul Turkey

David E. Yocum Arizona Arthritis Center University of Arizona Tucson, AZ USA

Ji-Won Yoon Center for Immunologic Research and Department of Microbiology and Immunology The Chicago Medical School North Chicago, IL USA Pierre Youinou Laboratory of Immunology Brest University Medical School Brest France

XV

Sandra Zampieri Division of Rheumatology Department of Medical and Surgical Science University of Padova Padova Italy

xvi

Gisele Zandman-Goddard Center for Autoimmune Diseases Department of Medicine 'B' Sheba Medical Center Tel Hashomer Israel

9 2004 Elsevier B. V. All rights reserved. Infection and Autoimmunity Y. Shoenfeld and N.R. Rose, editors

Introduction: Infection and Autoimmunity Yehuda Shoenfeld ~and Noel R. Rose 2

1Centerfor Autoimmune Diseases, Department of Medicine 'B', Sheba Medical Center, Tel-Hashomer, Sackler Faculty of Medicine, Incumbent of the Laura Schwarz-Kipp Chair for Research of Autoimmune Diseases, Tel-Aviv University, Israel; 2Centerfor Autoimmune Disease Research, Department of Pathology, and Department of Molecular Microbiology & Immunology, The John Hopkins Medical Institutions, Baltimore, MD, USA

About 80 recognized autoimmune diseases fulfill the Rose-Bona criteria (Table 1). Yet many other conditions are claimed to be of autoimmune origin. While some would say that "everything is autoimmune until proven otherwise", reading the chapters in this book written by world leaders in autoimmunity brings one to the conclusion that everything after all is infectious until proven otherwise (including autoimmune diseases). The story started many years ago with infectious etiology of rheumatic fever (RF) (See Guilherme and Kalil). RF was long known to be induced by infection with beta-hemolytic streptococci. During the last decades the molecular mimicry between the M-protein of the streptococci and its peptides was delineated as a classical mechanism for autoimmunity. The structural similarities between the M-protein on the streptococcal membrane and on the cells of the heart, brain and joint synovium could explain the typical clinical manifestations in patients encountering RF (i.e. migrating arthritis, valve involvement and chorea). Furthermore, a genetic predisposition to RF is implied by the fact that only few infected individuals will be affected. The RF paradigm epitomizes the environmental agent-genetic background interplay in induction of autoimmunity as shown in Fig. 1.

Environmental factor

Autoimmune

Genetic

predisposition

,

I~ disease

Figure 1. The RF paradigm epitomizes the environmental agent-genetic background interplay in induction of autoimmunity.

1. THE OLD AND THE NEW It has taught us many valuable lessons that may be applicable to other autoimmune diseases. Nowadays RF is not only rarely seen in the industrialized countries of Europe and North America, a dramatic change usually attributed to the prompt use of antimicrobials in treating streptococcal infections. This experience suggests that if we can identify microbial agents that trigger an autoimmune disease, it may be possible to prevent the disease, even in genetically susceptible individuals. The classical autoimmune conditions, have been joined by newly described diseases having possible infectious etiology such as the anti-phospholipid syndrome (APS). Characterized originally (1982) by the triad of manifestations i.e. repeated thromboemboric phenomenon, recurrent pregnancy loss and thrombocytopenia [1], this disease became even-

Table 1. The Rose-Bona criteria for recognizing autoimmune diseases (Immunology Today 14: 426-430, 1993) DIRECT EVIDENCE A. Autoantibody-mediated 1. Circulating autoantibodies affecting function a. Destruction or sequestration of a target cell b. Interaction with receptor 1) Stimulated function 2) Impaired function c. Interaction with hormones or enzymes 2. Localized autoantibodies a. Demonstration of immunoglobulin and/or complement components at site of lesion b. Ability to elute antibodies from lesions c. Reproduction of lesions by immunoglobulin eluates 3. Localized immune complexes at site of lesion a. Elution of antibody-antigen complex b. Transfer to experimental animals 4. Reproduction of disease by passive transfer a. Maternal-fetal transfer b. Transfer to experimental animals c. Demonstration of in vitro injury to target cell B. Cell-mediated 1. Transfer of T cells to immunodeficient mice implanted with target organ 2. In vitro cytotoxicity of T cells with cells of target organ II. INDIRECT EVIDENCE A. Reproduction of Autoimmune Disease by Experimental Immunization 1. Identification of initiating antigen 2. Immunization of susceptible experimental animal with analogous antigen 3. Production of characteristic lesions 4. Reaction of antibody or T cells with an analogous antigen or epitope B. Reproduction of Autoimmune Disease through Idiotype Network 1. Identification of disease-associated idiotype 2. Immunization of susceptible host with the idiotype 3. Production of characteristic lesions C. Spontaneous Models in Experimental Animals 1. Identification of disease in an experimental animal 2. Breeding and selection to increase disease frequency 3. Demonstration of self-reactive antibodies/T cells 4. Passive transfer/adoptive transfer of disease to syngeneic recipients 5. Passive transfer/adoptive transfer of disease to syngeneic recipients D. Animal Models Produced by Dysregulation of the Immune System 1. Neonatal thymectomy 2. Irradiation with thymectomy 3. Cytokine-deficient homologous inbred animals 4. Transgenic animals with altered: a. cytokine production b. antigen expression c. co-stimulatory factor expression 11I. CIRCUMSTANTIAL EVIDENCE A. Presence of Autoantibodies B. Association with Other Autoimmune Diseases C. Association with MHC Haplotype D. Lymphocyte Infiltration of Target Organ 1. Presence of germinal centers in lesions 2. Restricted V-gene usage of infiltrating T cells E. Favorable Response to Immunosuppression 1. Nonspecific 2. Specific (including oral tolerance)

tually one of the most common systemic diseases [2] entailing valve involvement (Liebman-Sacks endocarditis) as well as CNS affliction (chorea and cognitive impairment). The APS recently was found to be induced by diverse infectious agents [3]. The syndrome for years was thought to be clinicially "associated" with infections. Yet, only employment of modem techniques such as phage library display, monoclonal autoantibodies and proteinomics led to the conclusive results that APS can be induced by peptides residing on infecting agents' membranes and on the autoantigen ~2-glycoprotein-I (~2GPI) [4]. The genetic predilection for the disease was reported by several groups [5]. Interestingly enough, some kind of cross-reactivity exists between the autoantigens and streptococci of both RF and APS which affects the heart and the brain. This is an example of how the old diseases like RF may relate to the new and modem aspects of others. Despite clinical and epidemiologic evidence that many, if not most, autoimmune diseases are triggered by infection, there are very few instances where the underlying mechanisms have been explored. Autoimmune myocarditis in humans often follows infection by enteroviruses. The disease can be reproduced in genetically susceptible strains of mice by infection with a cardiotropic strain of Coxsackievirus B3 [6]. The antigen in this disease has been identified as cardiac myosin, and the autoimmune disease reproduced without virus in susceptible mice by immunization with purified cardiac myosin heavy chain or its myocarditogenic peptide [7]. The antigen arises from the host itself, and the virus serves both to deliver endogenous intracellular myosin and to provide a favorable inflammatory microenvironment [8]. The book intends to encompass the different mechanisms involved in the infection-autoimmunity association/induction. In addition to molecular mimicry one will find mechanisms such as polyclonal activation, tissue damage and the adjuvant effect of the inflammatory process itself. Kai W. Wucherpfennig discusses the ability of T cells to recognize a number of peptides from different viral and bacterial antigens that are remarkably distinct in their primary sequence, thus indicating the degeneracy of TCR recognition. Kamal D. Moudgil and Malaruithi Du Rai discuss the diversification of

the immune system by an antigen to new T cell and/or antibody specificities during the course of an autoimmune disease known as "epitope spreading". Special attention will be given to HSPs and to trangenic mouse models to better understand infection-induced autoimmunity. Due to the fact that vaccine may contain infectious antigens, the controversial issue of vaccineautoimmunity relationship is critically reviewed.

2. INFECTING AGENTS INDUCING DIVERSE AUTOIMMUNE DISEASES There are notorious infecting agents i.e. EBV, hepatitis-C, parwovirus-19 and others which were reported to be linked to many autoimmune diseases. These are detailed in Section 2. But other infecting agents such as HIV, Theiler's murine encephalomyelitis virus (causing demyelinating disease), endogenous retroviruses (leading to SLE), parasites (i.e. trypanosoma c r u z i - causing Chagas' disease; Yersinia enterocolitica leading to autoimmune thyroid disease and the newly appearing players the Saccharomyces cervisiae), and endogenous retroviruses, will be detailed.

3. AUTOIMMUNE DISEASES INDUCED BY A VARIETY OF INFECTING AGENTS

Another viewpoint of the infection autoimmunity inter-relationship entails the various diseases associated with a variety of infections. Those are summarized in Section 3. Among the diseases having enough data to support an infectious origin, we include rheumatic fever, SLE, APS, Sjrgren's syndrome, polymyositis, IBD, reactive arthritis, autoimmune thyroid diseases, type I diabetes mellitus, autoimmune heart diseases, autoimmune liver diseases, vasculitides and multiple sclerosis. We hope that the readers of this book will gain a deeper insight into these increasingly important etiological aspects of autoimmunity.

REFERENCES 5. 1.

2. 3.

4.

Asherson RA, Cervera R, Piette JC, Shoenfeld Y, eds. The Antiphospholipid Syndrome II: Autoimmune Thrombosis. Amsterdam: Elsevier, 2002; 1457. Shoenfeld Y. Systemic antiphospholipid syndrome. Lupus 2003;12:497--498. Blank M, Krause I, Fridkin M, Keller N, Kopolovic J, Goldberg I, Tobar A, Shoenfeld Y. Bacterial induction of autoantibodies to 132-glycoprotein-I accounts for the infectious etiology of antiphospholipid syndrome. J Clin Invest 2002; 109:797-804. Blank M, Shoenfeld Y, Cabilly S, Heldman Y, Fridkin M, Katchalski-Katzir E. Prevention of experimental antiphospholipid syndrome and endothelial cell activation by synthetic peptides. Proc Natl Acad Sci (USA)

6.

7.

8.

1999;96:5164-5168. Minisola G, Galeazzi M, Sebastiani DS. HLA class 1I alleles and genetic predisposition to the antiphospholipid syndrome. Autoimmun Rev 2003;2:387-394. Rose NR, Wolfgram LJ, Herskowitz A, Beisel KW. Postinfectious autoimmunity: two distinct phases of Coxsackievirus B3-induced myocarditis. Ann NY Acad Sci 1986;475:146-156. Neu N, Rose NR, Beisel KW, Herskowitz A, GurriGlass G, Craig SW. Cardiac myosin induces myocarditis in genetically predisposed mice. J Immunol 1987; 139:3630-363. Rose NR, Afanasyeva M. The inflammatory process in experimental myocarditis. In: Feuerstein GZ, Libby P, Mann DL, eds. Inflammation and Heart Disease. Basel: Birkhiiuser, 2003;325-333.

9 2004 Elsevier B. E All rights reserved. Infection and Autoimmunity Y. Shoenfeld and N.R. Rose, editors

Implications of T Cell Receptor Crossreactivity for the Pathogenesis of Autoimmune Diseases Kai W. Wucherpfennig

Department of Cancer Immunology & AIDS, Dana-Farber Cancer Institute & Department of Neurology, Harvard Medical School, Boston, MA, USA

1. HOW SPECIFIC IS TCR RECOGNITION OF MHC/PEPTIDE COMPLEXES? The experimental practice of isolating "antigenspecific" T cells by in vivo or in vitro selection with a particular antigen led to the widely held notion that TCR recognition is highly specific since T cell clones selected in such a fashion are typically not activated by control antigens. Within this conceptual framework it was thought that rare microbial antigens in which the peptide sequences were closely related to a self-peptide could represent mimics and be responsible for the induction of autoimmune diseases. Since it was assumed that such peptides would have to possess substantial sequence identity with the self-antigen, sequence alignments with self-peptides were used to identify candidate sequences. This approach was used to identify a peptide from the Hepatitis B virus DNA polymerase that induced histological signs of experimental autoimmune encephalomyelitis (EAE) following immunization of rabbits [1]. However, a large number of groups used this method to identify potential mimicry peptides with largely disappointing results since the vast majority of peptides had no biological activity. It thus appeared that TCR crossreactivity might be a very rare event. Given the paucity of definitive experimental data, the biological relevance of TCR crossreactivity was questioned by many basic scientists. However, structural studies on the interaction of MHC molecules with peptides and on TCR recognition of MHC/peptide complexes suggested that this view of TCR recognition might have to be

reconsidered. Peptide elution studies demonstrated that a single MHC molecule could bind a very large and diverse set of peptides since several hundred distinct peptide masses could be defined with mass spectrometry techniques [2, 3]. Investigation of the structural requirements for peptide binding by MHC class II molecules demonstrated that five peptide side chains contributed to binding [4]. However, each of these 'anchor residues' could typically be substituted by a number of other amino acids so that the resulting peptide binding motifs were highly degenerate [5]. For the multiple sclerosis (MS) associated HLA-DR2 molecule (DP~a., DRB 1" 1501) that has been the focus of our studies, three of these five peptide positions (P6, P7 and P9 pockets) could be substituted by many different amino acids. Even though a higher degree of specificity was observed for the P1 and P4 anchor residues, substitutions by structurally related amino acids were permitted [6]. The crystal structure of HLA-DR1 with a bound peptide from influenza hemagglutinin (HA, 306318) elucidated how peptides are bound with high affinity, despite such degenerate sequence motifs: a significant fraction of the binding energy is derived from interactions between the backbone of the peptide and conserved residues of the MHC class II binding cleft. This structure also demonstrated that the peptide is buried in the binding site such that substitutions of peptide side chains located in deep pockets might not interfere with TCR recognition of the MHC/peptide surface [4]. Analysis of TCR recognition of MHC-bound peptides in light of this structural information demonstrated that specificity was typically confined to a

small number of peptide side chains. For the myelin basic protein (MBP) specific T cell clones that we have studied, three peptide side chains in the core of the MBP peptide (P2 His, P3 Phe and P5 Lys) were particularly relevant for TCR recognition [6, 7]. This observation appeared to be general since similar findings were made for murine TCR reactive with microbial or self-antigens. Global amino acid replacements in the moth cytochrome C (93-103) peptide recognized by murine I-E k restricted T cells demonstrated a strong preference for a particular amino acid at three peptide positions [8]. For the Ac(1-11) peptide of MBP that is encephalitogenic in PL/J mice, only four native MBP residues were required for activation of MBP-specific T cells [9]. Based on these considerations, we developed the hypothesis that TCR recognition is characterized by a considerable degree of crossreactivity and that a TCR could recognize a number of different peptides that may be rather distinct in their sequence. This hypothesis was supported by a reported case of crossreactivity where no obvious sequence similarity was present between the two peptides, as well as the observation that many T cell clones recognized alloreactive MHC/peptide complexes [ 10, 11 ].

2. D E V E L O P M E N T OF A STRATEGY TO SYSTEMATICALLY EXAMINE T C R CROSSREACTIVITY The challenge therefore was to develop a systematic approach that would allow us to identify such peptides even though their structural similarity might not be evident by conventional sequence alignments. Formation of the trimolecular complex of MHC, peptide and TCR is based on two independent binding events: high affinity binding of peptide to an MHC molecule and the more shortlived association of TCR with this MHC/peptide surface. We decided to base our strategy on the minimal structural requirements for each of these two binding events. This approach took advantage of the fact that T cell epitopes can be mapped to short peptide segments and that the contribution of individual peptide side chains to MHC binding and TCR recognition can be evaluated with a series of peptide analogs. Experimentally, this work focused on T cell recognition of a peptide from human MBP

(residues 85-99) that is bound with high affinity by the MS-associated HLA-DR2 molecule (DRA, DRB 1"1501) [6]. Analysis of TCR crossreactivity for T cell clones activated by this HLA-DR2/MBP peptide complex could thus provide insights into disease mechanisms in MS. We defined the minimal structural requirements for binding of the MBP peptide to HLA-DR2 and for TCR recognition of this MHC/peptide complex and searched available sequence databases with this motif information. This approach permitted the identification of many examples of TCR crossreactivity, not only for the human MBP specific T cell clones that we have studied but also for CD4 and CD8 T cells of both human and murine origin, as described in some of the other chapters of this book. A more recent variant of this approach has been to determine the search motif with peptide analogs in which all neighboring positions carry mixtures of amino acids (so called combinatorial peptide libraries/positional scanning libraries) [ 12]. These libraries provide less detailed motif information, but can be used on any MHC class II restricted T cell clone.

3. T CELL CLONES CAN R E C O G N I Z E MULTIPLE PEPTIDES W I T H L I M I T E D SEQUENCE SIMILARITY The search criteria focused on the two major HLADR2 anchor residues of the peptide (P1 and P4) and four putative TCR contact residues (at P-1, P2, P3 and P5), all of which were located in a six amino acid core segment of the peptide. In the initial study, we synthesized a panel of 129 microbial peptides that matched these criteria and tested them for their ability to activate human MBP specific T cell clones that had been isolated from the peripheral blood of two patients with MS. Even though we only analyzed T cell clones that recognized a single selfpeptide, we could identify eight microbial peptides that activated MBP reactive T cell clones [7]. These peptides were remarkably distinct in their sequence from each other and the MBP peptide and only one of the eight peptides had obvious sequence similarity with the MBP peptide in the core segment. The motif-based strategy was thus essential for the identification of these peptides. In the initial study, such peptides were identified for three of

the seven T cell clones that we studied. In order to determine whether microbial peptides could activate the majority of these clones, we examined the recognition motif for two of the four remaining clones in detail and synthesized a panel of peptides that included sequences from recently characterized microbial genomes. A total of five bacterial peptides were identified in this set of experiments [13]. Thus, we have identified a total of thirteen microbial peptides that can activate human T cell clones specific for a single myelin peptide. Since we could only synthesize a subset of peptides identified in each search and a number of microbial genomes have not yet been sequenced, it is evident that a substantial number of microbial peptides can activate T cell clones that recognize this MBP peptide. These experiments thus demonstrated that T cell clones could recognize a variety of different peptides with limited sequence similarity. A number of different terms have been used to describe this property of TCR recognition. TCR crossreactivity describes the basic biological observation, while plasticity and degeneracy suggest structural m e c h a n i s m s - mobility of TCR CDR loops (plasticity) or a relatively poor fit of the TCR on the MHC/peptide surface (degeneracy). Molecular mimicry has been widely utilized to describe the specialized case where TCR crossreactivity involves peptides from an infectious agent and a self-antigen.

4. ACTIVATION OF MBP SPECIFIC T CELLS BY NATURALLY PROCESSED VIRAL ANTIGEN One of the viral peptides was derived from the EBV DNA polymerase and we examined whether the viral peptide is presented by infected antigen presenting cells to MBP specific T cells. The EBV DNA polymerase gene is part of the lyric cycle and is therefore not transcribed in EBV transformed B cells. However. the lytic cycle and expression of the EBV DNA polymerase gene can be induced by treatment of EBV transformed B cells with phorbol esters [14]. A HLA-DR2 § B cell line that had been treated with a phorbol ester activated a MBP specific T cell clone that recognized both MBP and EBV DNA polymerase peptides. T cell activa-

tion was blocked by a mAb specific for HLA-DR but not by a control mAb against HLA-DQ, and was not observed when MHC-mismatched EBV transformed B cells were used as antigen presenting cells. These results demonstrated that the MBP specific T cell clone recognized not only the EBV peptide but also antigen presenting cells in which the viral gene was transcribed [7].

5. CRYSTAL STRUCTURE OF THE HLADR2/MBP PEPTIDE COMPLEX Since the peptides that can be recognized by the same TCR are quite distinct in their primary sequence, an important question relates to the structural basis of TCR crossreactivity. Structural information is also required to determine why only a subset of peptides that match the MHC binding/ TCR recognition motif activate the appropriate T cell clones. We determined the crystal structure of HLA-DR2 (DRA, DRBI*I501) with the bound MBP peptide as a step towards defining molecular mimicry at a structural level [ 15]. Fig. I gives an overview of the structure and illustrates features of HLA-DR2 that are important for MBP peptide binding as well as TCR recognition of the HLA-DR2/ MBP peptide complex. The MBP peptide is bound in an extended conformation as a type II polyproline helix and MBP peptide side chains occupy the P I, P4, P6 and Ix) pockets of the binding groove (Fig. l a). The two major anchor residues of the MBP peptide (Val and Phe) occupy the P I and P4 pockets of HLA-DR2. The P4 pocket of HLA-DR2 is distinct from DR molecules associated with other autoimmune diseases. In HLA-DR2, this pocket is large and predominantly hydrophobic due to the presence of a small residue (Aia) at a key polymorphic position (DRI3 71), which permits an aromatic side chain of the MBP peptide to be accommodated (Phe) (Fig. Ic). The peptide residues shown to be important for TCR recognition of the MBP peptide (P2 His, P3 Phe and P5 Lys) are solvent exposed in the structure of the HLA-DR2/MBP peptide complex (Fig. l b, l d: Fig. 2). Comparison with the published high resolution crystal structures of human MHC class I/peptide/TCR complexes [ 16-18] suggests that P5 Lys binds in a TCR pocket formed by the CDR3 loops of the r and [3 chains.

Figure 1. Crystal structure of the complex of HLA-DR2 (DRA, DRB I * 1501 ) and the MBP (85-99) peptide. (A) Overview of the structure. MBP peptide side chains that are located in four pockets of the binding site are indicated, PI Val, P4 Phe, P6 Asn and P9 Thr. (B) Solvent exposed residues that are important for TCR recognition of the MBP (185-99) peptide, IY2 His, P3 Phe and P5 Lys. (C) P4 pocket of the HLA-DR2 binding site. This large and hydrophobic pocket is occupied by P4 Phe of the MBP peptide. The necessary room for this aromatic side chain is created by the presence of a small residue (Ala) at the polymorphic DRI$ 71 position. (D) Close-up view of MBP peptide residues recognized by the TCR of MBP reactive T cell clones, P-1 Val, P2 His. P3 Phe and P5 Lys. Preferences at these positions were considered in the search criteria for the identification of microbial peptides that activate MBP reactive T cell clones. Reprinted from The Journal of Experimental Medicine [151 with permission of the publisher.

6. C O M M O N F E A T U R E S O F M I C R O B I A L P E P T I D E S T H A T A C T I V A T E MBP SPECIFIC T CELLS The structural information on the H L A - D R 2 / M B P peptide complex was used to dissect the crossreactive peptides identified for the MBP specific T cell clone ( O b . l A I 2 ) for which the largest number of stimulatory microbial peptides were identified. The crossreactive peptides were aligned with the MBP

10

peptide in order to determine residues that may be located in pockets of the HLA-DR2 binding site (Table 1). This analysis shows that the HLA-DR2 contact surface of this set of peptides is highly diverse, No sequence identity with the MBP peptide is required on the HLA-DR2 binding surface, as illustrated by comparison of the MBP and Bacillus subtilis peptides. C o m m o n features of putative HLA-DR2 contact residues are the presence of an aliphatic residue in the P I pocket and a large hydro-

Figure 2. Electron density and model of the MBP peptide in the binding site of HLA-DR2. (A) Experimental electron density of the MBP peptide bound to HLA-DR2. (B) Model of the MBP peptide based on the elecLron density. The DR2/MBP peptide complex crystallized as a dimer of dimers, as other HLA-DR molecules [45, 41. and both peptide copies are shown (blue and yellow). The peptide backbones superimpose in the P-I to P4 segment and are more divergent in the C-terminal segment due to different crystal contacts. Reprinted from The Journal of Experimental Medicine [ 15] with permission of the publisher.

phobic residue in the P4 pocket. At P6, a preference for asparagine is observed, while the residues that occupy the P7 and P9 positions are diverse. These data are in agreement with HLA-DR2 binding studies, which indicated that only two positions of the peptide (Pl and P4) were critical for binding and that they could be substituted with other aliphatic residues or phenylalanine (Pl pocket) or other hydrophobic residues (P4 pocket). Analysis of the residues that may be solvent exposed and that could interact with the TCR shows a higher degree of sequence similarity/identity, in particular in the center of the epitope (Table 2). All peptides that activate this T cell clone carry the two primary TCR contact residues of the MBP peptide (His and Phe at P2 and P3). Also, a preference for a positively charged residue (Lys and Arg) is observed at PS. A high degree of sequence diversity

is observed in the N- and C-terminal flanking segments, even though they are required for efficient T cell stimulation. The data also indicate that combinatorial effects shape the peptide surface that can be recognized by a TCR. Analysis of double-amino acid substitutions of the MBP peptide demonstrated that certain combinations of amino acids at TCR contact residues were stimulatory, even though the individual analogs had no activity [19]. This notion is supported by the observation that the majority of microbial peptides that match the MHC binding/TCR recognition motif did not stimulate the MBP specific T cell clones. Identification of a complete set of peptide sequences that represent agonists for a TCR will therefore require analysis of complex peptide libraries. At present, such analyses represent a technical challenge since a large number of peptides may

II

Table 1. Sequences of microbial peptides that activate human MBP specific T cell clones Organism

Source protein

Sequence

Peptides that activate T cell clone Ob.lAl2 (HLA-DR2 restricted)

Homo sapiens Staphylococcus aureus Mycobacterium avium Mycobacterium tuberculosis Bacillus subtilis Haemophilus influenzae/E, coli

Myelin basic protein VgaB Transposase Transposase YqeE HI0136/ORF

ENPVVHFFKNIVTPR VLARLHFYRNDVHItE QRCRVHFLRNVLAQV QRCRVHFMRNLYTAV ALAVLHFYPDKGAKN DFARVHFISALHGSG

Peptides that activate T cell clone Hy. 1B 11 ( HLA-DQ i restricted)

Homo sapiens Herpes simplex virus type 1 Adenovirus type 12 Human papillomavirus type 7 Pseudomonas aeruginosa

Myelin basic protein ULI5 protein ORF L2 protein Phosphomannomutase

ENPVVHFFKNIVTPR FRQLVHFVRDFAQLL DFEVVTFLKDVLPEF IGGRVHFFKDIS P IA DRLLMLFAKDWSRN

Microbial peptides that activate two human T cell clones reactive with the MBP (85-99) peptide are shown. Clone Ob.lAl2 is restricted by HLA-DR2 (DRA, DRB 1" 1501) and five microbial peptides have been identified that activate this T cell clone. Clone Hy.IB 11 is HLA-DQI restricted and activated by four microbial peptides that are quite distinct in their primary sequence. The peptide from human papillomavirus type 7 is the only peptide with obvious sequence similarity to the MBP peptide within the entire set of identified microbial peptides. Residues that are identical between the MBP peptide and microbial peptides are highlighted. The HLA-DQI restricted T cell clone recognizes the same core segment of the MBP peptide as HLA-DR2 restricted clones [7, 6, 13]. ORF, open reading frame.

need to be sequenced from phage display libraries or peptide libraries on beads. However, the complexity of the peptide repertoire that is recognized by an individual TCR may be underestimated unless the combinatorial nature of peptide recognition by the TCR is taken into consideration.

7. STRUCTURAL FEATURES OF AUTOREACTIVE TCR THAT C O N T R I B U T E TO THE DEGREE OF CROSSREACTIVITY Comparison of the MBP (85-99) reactive T cell clones demonstrated obvious differences in the degree of specificity/crossreactivity. For some of the T cell clones, such as Ob.Al2, amino acid identity was required at two TCR contact residues of the MBP peptide while every TCR contact residue of the MBP peptide could be substituted by at least one structurally related amino acid for other T cell clones.

12

Two of the T cell clones (Ob.lAl2 and Ob.2F3) differed only in the CDR3 loops of TCR ct and 13 since they had identical Vo~-Jt~ and V~-J~ rearrangements [20]. Nevertheless, the clones differed in the level of crossreactivity since only three of the five microbial peptides identified for clone Ob. 1A 12 activated the other clone. We examined the basis for the different level of crossreactivity and found that the clones had a very similar fine specificity for the MBP peptide, except for the P5 position of the peptide (P5 Lys). The microbial peptides that activated clone Ob. I A I 2 were characterized by conservative or non-conservative changes at P5 (Lys to Arg, Ser or Pro). In contrast, clone Ob.2F3 was only stimulated by the peptides that had a conservative lysine to arginine substitution. The degree of specificity in recognition of the P5 side chain was the key difference between these TCR since the Haemophilus influenzae/E, coli peptide stimulated both clones when the P5 position was substituted from serine to arginine [ 13]. In the crystal structure of the HLA-DR2/MBP peptide complex, P5 Lys was a

Table 2. Alignment of microbial peptides based on the crystal structure of the HLA-DR2/MBP peptide complex Organism

Residues located in HLA-DR2 binding pockets Homo sapiens Staphylococcus aureus Mycobacterium avium Mycobacterium tuberculosis Bacillus subtilis Haemophilus influenzae/E, coli

Source protein

Sequence

Myelin basic protein VgaB

....

V-

- F-NI-T-

-

....

L- - Y-ND-H-

-

Transposase

....

V-

-L-NV-A--

Transposase

....

V-

-M-NL-

YqeE

....

L- - Y- DK-A-

-

HI0136/ORF

....

V-

-

1

Solvent exposed residues

4 6 7 9

T- -

- I -AL-G-

23

5

8

HF-

K-

- V-

Homo sapiens Staphylococcus aureus

Myelin basic protein

ENPV-

VgaB

VLAR-HF-

Mycobacterium avium Mycobacterium tuberculosis

Transposase Transposase

QRCR-HF-R-

-L-QV

QRCR-HF-R-

-Y-AV

Bacillus subtilis

YqeE

ALAV-

Haemophilus influenzae/E, coli

HI0136/ORF

DFAR-HF-S-

HF

PR

R- -V-KE

- P - - G - KN -H-SG

Peptide sequences were dissected in terms of putative HLA-DR2 anchor residues and solvent accessible residues that are available for interaction with the TCR. The HLA-DR2 contact surface of these peptides is highly diverse, and no sequence identity is observed at these five positions between the MBP and Bacillus subtilis peptides. A higher degree of sequence similarity/identity is observed for peptide residues that are likely to interact with the TCR (P2, P3 and P5). Numbers above the sequences represent positions in a nine-amino acid peptide core, starting with the anchor residue for the P1 pocket of the binding site.

prominent, solvent exposed residue in the center of the DR2/MBP peptide surface (Fig. 1). Structures of MHC class I/peptide complexes have shown that this peptide side chain occupies a central TCR pocket created by the CDR3 loops of TCRt~ and [16, 18]. This suggests that the two MBP specific TCR differ in the size and shape of the TCR pocket that accomodates the central lysine of the MBP peptide. Similar findings were made for MHC class I restricted T cell clones that recognize the HTLV-1 Tax (11-19) peptide bound to HLA-A2. The crystal structure of the MHC/peptide/TCR complex has been determined for both of these TCRs (A6 and B7), which demonstrated major differences in the shape and charge of the TCR pocket for the P5 peptide residue [16, 18]. The B7 TCR was exquisitely specific for P5 tyrosine of Tax (11-19) since only aromatic substitutions were tolerated. In contrast, the A6 TCR was much more degenerate at the P5 position since 10 of 17 analog peptides induced lysis of target cells at low peptide concentrations

[21]. The absolute requirement for an aromatic side chain by the B7 T cell clone could be explained based on the structure of the P5 pocket: the P5 Tyr represented a tight fit for this TCR pocket and a favorable stacking interaction between the aromatic ring of P5 Tyr and an aromatic residue of the CDR3 loop of the TCR 13chain (Y 104 13) was observed. In contrast, the A6 TCR had a larger P5 pocket and the P5 Tyr did not interact with an aromatic TCR residue. These data demonstrate that the TCR CDR3 loops can determine the degree of specificity and degeneracy of the central TCR pocket. Further characterization of the peptide from the EBV DNA polymerase gene demonstrated a second structural mechanism for TCR crossreactivity. The HLA-DR2 haplotype that confers susceptibility to MS encodes two DR~ chain genes (DRB 1"1501 and DRB5*0101), both of which can pair with the non-polymorphic DRot chain to form functional DR heterodimers [22]. Both DR molecules are expressed by antigen presenting cells in subjects with the HLA-DR2 haplotype. The Hy.2E11 clone recog-

13

nized the MBP peptide bound to DRA, DRB 1"1501 molecules, but surprisingly the EBV peptide bound to DRA, DRB5*0101 molecules. Comparison of the two crystal structures demonstrated a striking degree of similarity, in particular for the peptide positions previously shown to be required for TCR recognition [7, 23]. TCR crossreactivity can therefore involve the recognition of different peptides bound to the same MHC molecule, or recognition of different peptides on other self-MHC molecules.

8. TCR CROSSREACTIVITY AND IMMUNOPATHOLOGY Several different experimental autoimmune diseases have been induced by immunization with microbial peptides, indicating that crossreactive T cell populations can be pathogenic. EAE has been induced in different strains of mice and in Lewis rats with mimicry peptides of MBP and myelin oligodendrocyte glycoprotein [24-28]. However, the physiologically more relevant question is whether autoimmune disease can also result from infection with pathogens that carry such T cell epitopes. To address this question, Olson et al [29] generated recombinant Theiler's viruses in which the candidate sequences were placed into the leader segment of the virus. The first recombinant virus carded the sequence of the PLP (139-151) peptide that is immunodominant in SJL mice and infection with this virus resulted in the rapid development of central nervous system (CNS) inflammation and vigorous CD4 T cell responses to the PLP peptide. It is important to note that the wildtype Theiler's virus also induced CNS pathology, but disease onset was significantly later (day 30, rather than day 10) permitting the two disease states to be distinguished. Also, tolerance induction with the PLP (139-151) peptide prevented induction of the early disease process with the PLP (139-151) expressing virus, but not the late disease caused by wild-type Theiler's virus [30]. Importantly, CNS autoimmunity could not only be induced by a virus that carried the self peptide, but also by a recombinant virus that expressed a peptide from Hemophilus influenzae shown to stimulate PLP (139-151) specific T cells [29]. The relationship between autoimmune disease and viral infection has also been examined with

14

natural pathogens, in particular using the Herpes simplex keratitis (HSK) model [31]. Infection of the eye with Herpes simplex virus (HSV-1, KOS strain) triggers a T cell-mediated autoimmune process that persists after the virus has been cleared. In order to rigorously test the role of molecular mimicry in this disease process, a single amino acid substitution was made in the crossreactive T cell epitope of the viral UL-6 protein. The mimicry T cell epitope was found to be important for disease induction in C.AL-20 mice since 1000-fold larger quantifies of the mutant virus were required than of wild-type HSV-1, even though both viruses replicated at the same rate. However, in mice that expressed the C1-6 TCR and therefore harbored large numbers of autoreactive T cells, infection with HSV-1 was not required since scratching of the cornea or local application of LPS were sufficient for the induction of disease. The crossreactive T cell epitope was thus required for the expansion of autoreactive T cells, but activation of the innate immune system was sufficient when large numbers of such T cells were already present. This model therefore clarified the potential contributions of TCR crossreactivity and 'bystander' activation in the induction of autoimmune diseases. The diverse nature of the viral/bacterial peptides that stimulate autoreactive T cell clones suggests that different infectious agents could initiate autoimmunity by molecular mimicry. However, it is important to keep in mind that a number of other mechanisms could also result in the activation of autoreactive T cells. The diverse nature of the mimicry peptides and the ubiquitous presence of some of these pathogens may make it difficult to establish a direct epidemiological link between infectious agents and the occurrence of certain autoimmune diseases. In particular, the temporal relationship between an infection and development of an autoimmune process may in many cases not be clear because of the time that frequently elapses until clinical symptoms become obvious and a diagnosis is made. Such epidemiological relationships may be more readily established for autoimmune disorders with a rapid disease onset since early diagnosis can greatly increase the likelihood of establishing a link with a preceding infection. Recent data have demonstrated a relationship between an inflammatory, demyelinating disease of the peripheral nervous

system (Guillain-Barr6 syndrome) and preceding infections [32]. Patients with this disease acutely develop severe symptoms and rapid diagnosis permitted isolation of Campylobacter jejuni from approximately a third of new cases, compared to 2% of household controls. A better understanding of the epidemiology of infectious agents and autoimmunity could thus help to advance our understanding of the molecular mechanisms that trigger human autoimmune diseases.

9. TCR CROSSREACTIVITY AS A GENERAL PROPERTY OF T CELL RECOGNITION A large number of studies have now demonstrated TCR crossreactivity for a variety of human and murine T cells [10, 12, 25, 33-37]. An interesting example is the melanoma/melanocyte-derived peptide MART- 1 (res. 27-35) since it demonstrates how crossreactivity can shape the T cell repertoire. In normal human donors with the HLA-A2 haplotype, T cells specific for this melanocyte peptide were detected at a surprisingly high frequency, and such T cells could be visualized directly ex vivo with HLA-A2/MART- 1 tetramers (frequency of -0.1%). This suggested that these T cells crossreacted with microbial peptides and a motif search similar to the one that we had performed for human MBP specific T cell clones yielded twelve peptides that were able to sensitize target cells for lysis. One of these peptides was derived from the glycoprotein C of Herpes simplex virus (HSV) and anti-MART effectors lysed cells infected with a recombinant vaccinia virus encoding HSV-1 glycoprotein C [35, 38]. TCR crossreactivity can also have a profound effect on protective immune responses to viral pathogens in vivo, as shown in a murine model where CD8 T cells crossreacted with peptides from two different v i r u s e s - lymphocytic choriomeningitis virus (LCMV) and Pichinde virus (PV) [37]. LCMV and PV are members of the Arenaviridae family, but the two viruses are only distantly related as shown by sequence comparison. Prior infection with either LCMV or PV provided partial protection against the heterologous virus and LCMV-immune mice showed a 97% reduction in viral titer compared to naive mice when challenged with PV. CD8 T cells

from mice infected with either virus crossreacted with a nucleoprotein-derived peptide (NP 205-212) from the other virus; these two nucleoprotein peptides shared six of eight residues and differed at positions 5 (Tyr versus Phe) and 8 (Leu versus Met). In LCMV infected mice the NP 205-212 epitope was subdominant and 3.6% of CD8 T cells responded to this epitope on day 8 following infection. However, CD8 T cells specific for this NP 205-212 peptide became the predominant CD8 T cell population (30% of all CD8 T cells) when mice that had previously encountered PV were infected with LCMV. These experiments demonstrated that TCR crossreactivity influences the hierarchy of CD8 T cell responses and shapes the pool of memory T cells. TCR crossreactivity is also a critical aspect of T cell development in the thymus and weaker TCR signals are required for positive selection in the thymus compared to activation of mature T cells. Positive selection in the thymus is peptidedependent and is affected both by the density of a particular MHC/peptide complex and the affinity of the TCR for this complex. In thymic organ cultures, positive selection was observed with peptides that represented weak agonists or antagonists for the corresponding mature T cells, or with low densities of the agonist peptide [39-41]. The creation of transgenic mice that expressed a single MHC/ peptide ligand in the thymus provided a striking demonstration of crossreactive TCR recognition in thymic development [42]. In this experiment, a peptide was covalently linked to the N-terminus of the MHC class II ~ chain so that all MHC class II molecules were occupied with this peptide. The total numbers of CD4 T cells in these mice were -20% compared to wild-type mice and these CD4 T cells expressed a wide variety of different VI3 segments, indicating that a relatively diverse T cell repertoire could develop in the presence of a single MHC class II/peptide ligand. T cell hybridomas isolated from these mice reacted with peptides that had no primary sequence identity with the selecting peptide [43]. These experiments demonstrated that T cells could be activated by peptides that were unrelated in sequence to their selecting peptide. The examples described above indicate that TCR crossreactivity is common and represents an

15

important aspect of TCR recognition. The balance between specificity and crossreactivity is likely to represent a compromise that permits a sufficient number of T cells to recognize a pathogen novel to the individual's immune system. The potentially negative impact of TCR crossreactivity may be in part balanced by in vivo selection of T cells with high avidity TCR for the relevant MHC/peptide complexes. It has been postulated that a single TCR can recognize 106 different peptide ligands, and this estimate is based on the observation that a subset of T cell clones can be activated by complex peptide mixtures in which only one peptide position is specified [44]. While the number of peptide variants that can be recognized may be very large, the number of natural ligands from microbial and self-antigens is likely to be considerably smaller. Nevertheless, a number of different peptides can act as agonists for a given T cell and a considerably larger number of peptide ligands may induce weak signals, such as those that promote positive selection in the thymus and survival of naive T cells in the periphery. TCR specificity and crossreactivity thus represent important aspects of T cell biology.

3.

4.

5.

6.

7.

8.

.

ACKNOWLEDGEMENTS I would like to acknowledge the important contributions that my colleagues and collaborators have made towards this research. In particular, I would like to acknowledge the contributions of Stefan Hausmann, Katherine Smith, Laurent Gauthier, Heiner Appel, Jason Pyrdol, David A. Hailer, Don C. Wiley and Jack L. Strominger. This work was supported by grants from the National Multiple Sclerosis Society and the NIH (RO1 NS39096).

10.

11.

12.

REFERENCES 1.

2.

16

Fujinami RS, Oldstone MB. Amino acid homology between the encephalitogenic site of myelin basic protein and virus: mechanism for autoimmunity. Science 1985;230:1043-5. Hunt DF, Michel H, Dickinson TA et al. Peptides presented to the immune system by the murine class II major histocompatibility complex molecule I-Ad. Science 1992;256:1817-20.

13.

14.

15.

Chicz RM, Urban RG, Gorga JC, Vignali DA, Lane WS, Strominger JL. Specificity and promiscuity among naturally processed peptides bound to HLA-DR alleles. J Exp Med 1993;178:27-47. Stern LJ, Brown JH, Jardetzky TS et al. Crystal structure of the human class II MHC protein HLA-DR1 complexed with an influenza virus peptide. Nature 1994;368:215-21. Hammer J, Valsasnini P, Tolba K et al. Promiscuous and allele-specific anchors in I-Jd_~A-DR-binding peptides. Cell 1993;74:197-203. WucherpfennigKW, Sette A, Southwood Set al. Structural requirements for binding of an immunodominant myelin basic protein peptide to DR2 isotypes and for its recognition by human T cell clones. J Exp Med 1994; 179:279-90. Wucherpfennig KW, Strominger JL. Molecular mimicry in T cell-mediated autoimmunity: viral peptides activate human T cell clones specific for myelin basic protein. Cell 1995;80:695-705. Reay PA, Kantor RM, Davis MM. Use of global amino acid replacements to define the requirements for MHC binding and T cell recognition of moth cytochrome c (93-103). J Immunol 1994;152:3946-57. Gautam AM, Lock CB, Smilek DE, Pearson CI, Steinman L, McDevitt HO. Minimum structural requirements for peptide presentation by major histocompatibility complex class II molecules: implications in induction of autoin-trnunity. Proc Natl Acad Sci USA 1994;91:767-71. Bhardwaj V, Kumar V, Geysen HM, Sercarz EE. Degenerate recognition of a dissimilar antigenic peptide by myelin basic protein-reactive T cells. Implications for thymic education and autoimmunity. J Immunol 1993;151:5000-10. Burrows SR, Khanna R, Burrows JM, Moss DJ. An alloresponse in humans is dominated by cytotoxic T lymphocytes (CTL) cross-reactive with a single Epstein-Barr virus CTL epitope: implications for graftversus-host disease. J Exp Med 1994;179:1155-61. Hemmer B, Fleckenstein BT, Vergelli Met al. Identification of high potency microbial and self ligands for a human autoreactive class H-restricted T cell clone. J Exp Med 1997;185:1651-9. Hausmann S, Martin M, Gauthier L, Wucherpfennig KW. Structural features of autoreactive TCR that determine the degree of degeneracy in peptide recognition. J Immunol 1999;162:338-44. Datta AK, Feighny RJ, Pagano JS. Induction of Epstein-Barr virus-associated DNA polymerase by 12O-tetradecanoylphorbol-13-acetate. Purification and characterization. J Biol Chem 1980;255:5120-5. Smith led, Pyrdol J, Gauthier L, Wiley DC, Wucherpfen-

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

nig KW. Crystal structure of HLA-DR2 (DRA*0101, DRBI*1501) complexed with a peptide from human myelin basic protein. J Exp Med 1998;188:1511-20. Garboczi DN, Ghosh P, Utz U, Fan QR, Biddison WE, Wiley DC. Structure of the complex between human T-cell receptor, viral peptide and HLA-A2 [comment]. Nature 1996;384:134-41. Garcia KC, Degano M, Pease LR et al. Structural basis of plasticity in T cell receptor recognition of a self peptide-MHC antigen. Science 1998;279:1166-72. Ding YH, Smith KJ, Garboczi DN, Utz U, Biddison WE, Wiley DC. Two human T cell receptors bind in a similar diagonal mode to the HLA-A2/Tax peptide complex using different TCR amino acids. Immunity 1998;8:403-11. Ausubel LJ, Kwan CK, Sette A, Kuchroo V, Hailer DA. Complementary mutations in an antigenic peptide allow for crossreactivity of autoreactive T-cell clones. Proc Natl Acad Sci USA 1996;93:15317-22. Wucherpfennig KW, Zhang J, Witek C et al. Clonal expansion and persistence of human T cells specific for an immunodominant myelin basic protein peptide. J Immunol 1994;152:5581-92. Hausmann S, Biddison WE, Smith KJ et al. Peptide recognition by two HLA-A2/Taxl 1-19-specific T cell clones in relationship to their MHC/peptide/TCR crystal structures. J Immunol 1999; 162:5389-97. Sone T, Tsukamoto K, Hirayama K et al. Two distinct class II molecules encoded by the genes within HLADR subregion of HLA-Dw2 and Dw 12 can act as stimulating and restriction molecules. J Immunol 1985; 135: 1288-98. Lang HL, Jacobsen H, Ikemizu S et al. A functional and structural basis for TCR cross-reactivity in multiple sclerosis. Nat Immuno12002;3:940-3. Ufret-Vincenty RL, Quigley L, Tresser N e t al. In vivo survival of viral antigen-specific T cells that induce experimental autoimmune encephalomyelitis. J Exp Med 1998;188:1725-38. Grogan JL, Kramer A, Nogai A et al. Cross-reactivity of myelin basic protein-specific T cells with multiple microbial peptides: experimental autoimmune encephalomyelitis induction in TCR transgenic mice. J Immunol 1999;163:3764-70. Gautam AM, Liblau R, Chelvanayagam G, Steinman L, Boston T. A viral peptide with limited homology to a self peptide can induce clinical signs of experimental autoimmune encephalomyelitis. J Immunol 1998;161: 60-4. Mokhtarian E Zhang Z, Shi Y, Gonzales E, Sobel RA. Molecular mimicry between a viral peptide and a myelin oligodendrocyte glycoprotein peptide induces autoimmune demyelinating disease in mice. J Neuroim-

munol 1999;95:43-54. 28. Lenz DC, Lu L, Conant SB et al. A Chlamydia pneumoniae-specific peptide induces experimental autoimmune encephalomyelitis in rats. J Immunol 2001; 167: 1803-8.

29. Olson JK, Croxford JL, Calenoff MA, Dal Canto MC, Miller SD. A virus-induced molecular mimicry model of multiple sclerosis. J Clin Invest 2001 ;108:311-8. 30. Olson JK, Eagar TN, Miller SD. Functional activation of myelin-specific T cells by virus-induced molecular mimicry. J Immunol 2002; 169:2719-26. 31. Panoutsakopoulou V, Sanchirico ME, Huster KM et al. Analysis of the relationship between viral infection and autoimmune disease. Immunity 2001; 15:137-47. 32. Rees JH, Soudain SE, Gregson NA, Hughes RA. Campylobacter jejuni infection and Guillain-Barre syndrome [see comments]. N Engl J Med 1995;333: 1374-9. 33. Evavold BD, Sloan-Lancaster J, Wilson KJ, Rothbard JB, Allen PM. Specific T cell recognition of minimally homologous peptides: evidence for multiple endogenous ligands. Immunity 1995;2:655-63. 34. Hagerty DT, Allen PM. Intramolecular mimicry. Identification and analysis of two cross-reactive T cell epitopes within a single protein. J Immunol 1995;155: 2993-3001. 35. Loftus DJ, Castelli C, Clay TM et al. Identification of epitope mimics recognized by CTL reactive to the melanoma/melanocyte-derived peptide MART-l(2735). J Exp Med 1996;184:647-57. 36. Misko IS, Cross SM, Khanna R et al. Crossreactive recognition of viral, self, and bacterial peptide ligands by human class I-restricted cytotoxic T lymphocyte clonotypes: implications for molecular mimicry in autoimmune disease. Proc Natl Acad Sci USA 1999;96: 2279-84. 37. Brehm MA, Pinto AK, Daniels KA, Schneck JP, Welsh RM, Selin LK. T cell immunodominance and maintenance of memory regulated by unexpectedly crossreactive pathogens. Nat Immunol 2002;3:627-34. 38. Dutoit V, Rubio-Godoy V, Pittet MJ et al. Degeneracy of antigen recognition as the molecular basis for the high frequency of naive A2/Melan-a peptide multimer(+) CD8(+) T cells in humans. J Exp Med 2002;196:207-16. 39. Jameson SC, Hogquist KA, Bevan MJ. Specificity and flexibility in thymic selection. Nature 1994;369:750-2. 40. Sebzda E, Wallace VA, Mayer J, Yeung RS, Mak TW, Ohashi PS. Positive and negative thymocyte selection induced by different concentrations of a single peptide. Science 1994;263:1615-8. 41. Alam SM, Travers PJ, Wung JL et al. T-cell-receptor affinity and thymocyte positive selection. Nature

17

1996;381:616-20. 42. Ignatowicz L, Kappler J, Marrack P. The repertoire of T cells shaped by a single MHC/peptide ligand. Cell 1996;84:521-9. 43. Ignatowicz L, Rees W, Pacholczyk R et al. T cells can be activated by peptides that are unrelated in sequence to their selecting peptide. Immunity 1997;7:179-86.

18

44. Mason D. A very high level of crossreactivity is an essential feature of the T-cell receptor. Immunol Today 1998; 19:395-404. 45. Brown JH, Jardetzky TS, Gorga JC et al. Three-dimensional structure of the human class II histocompatibility antigen HLA-DR1. Nature 1993;364:33-9.


Epitope Spreading Kamal D. Moudgil and Malarvizhi Durai

Department of Microbiology and Immunology, University of Maryland School of Medicine, Baltimore, MD, USA

1. INTRODUCTION The phenomenon of "epitope spreading" (or "determinant spreading") is characterized by broadening or diversification of the initial immune response induced by immunization with a single peptide antigen or a multi-determinant antigen [1-3]. The new T cell and/or antibody responses are directed to different epitopes either within the same antigen as that used for immunization (intra-molecular spreading) or another antigen (inter-molecular spreading). The spreading of initial immune reactivity has been shown to occur during the course of a variety of experimentally-induced and spontaneously arising autoimmune diseases in animal models (Tables 1 and 2) [2, 3]. Studies in patients with certain autoimmune diseases, in recipients of organ transplants, and in cancer patients given peptide vaccination (described below) (Table 1) have further validated the significance of epitope spreading in disease pathogenesis. Depending on the disease process, epitope spreading can contribute either to the progression or to the control of an autoimmune disease (Fig. 1) [1, 3, 4]. The timing of epitope spreading during the course of disease and its functional attributes are of significance in designing appropriate immunotherapeutic approaches.

2. EPITOPE SPREADING IN AUTOIMMUNE DISEASES 2.1. Multiple Sclerosis and Experimental Autoimmune Encephalomyelitis Multiple sclerosis (MS) is a human autoimmune disease characterized by mononuclear cell infiltration and discrete areas of demyelination (plaques) within the central nervous system (CNS), and neurological dysfunction. Experimental autoimmune encephalomyelitis (EAE) is an experimental model for MS, and it can be induced in different mouse/rat strains by immunization (in adjuvant) with myelin antigens such as myelin basic protein (MBP), proteolipid protein (PLP), or myelin oligodendrocyte protein (MOG) [3, 5, 6] (Table 2). Epitope spreading was first demonstrated by Lehmann et al in the EAE model using (SJL X B 10.PL) F1 mice [1]. It was shown that the initial T cell response of mice with acute EAE was directed to MBP A c l - l l , but spreading of the T cell response to new determinants of MBP, namely 35-47, 81-100, and 121-140, occurred during the chronic stage of EAE [1, 7]. This broadening of the T cell response was attributed to priming of new T cells by determinants within endogenous MBP following initial CNS damage. Miller's group has established the role of epitope spreading in the pathogenesis of relapsing-EAE (R-EAE). R-EAE can be induced in SJL mice by immunization with PLP 139-151 [8, 9]. Using this model, it was observed that the T cell response to the disease-initiating epitope, PLP 139-151, was maintained in SJL mice throughout the course of

19

Table 1. Examples of epitope spreading in animal models and human diseases References

Diseases A. Autoimmune diseases a) Animal models of autoimmune diseases Experimental autoimmune encephalomyelitis (EAE) Diabetes in the non-obese diabetic (NOD) mouse Adjuvant-induced arthritis (AA) Lupus or Systemic lupus erythematosus (SLE) Experimental autoimmune myasthenia gravis (EAMG) Experimental autoimmune neuritis (EAN) Equine recurrent uveitis (ERU) Experimental autoimmune gastritis (EAG) Autoimmune oophoritis

[1, 8, 13, 17, 19, 20, 22, 24, 27]

[35,36] [4, 52, 63-65] [71, 72, 78, 80] [89,91,92] [96] [97] [98] [99, 100]

b) Human autoimmune diseases Multiple sclerosis (MS) Insulin-dependent diabetes mellitus (IDDM) or Type I diabetes Rheumatoid arthritis (RA) Systemic lupus erythematosus (SLE) or Lupus Myasthenia gravis (MG) Pemphigus

[31-33] [39-42] [671 [83-87] [93, 94] [95]

B. Other diseases

a) Organ transplantation / graft rejection

[102-106]

b) Tumors

[107-110]

c) Infection/Vaccination

[111, 112, 115, 116]

EAE. However, spreading of the T cell response to non-crossreactive PLP 178-191 and MBP 84-104 epitopes occurred after the first and second relapses, respectively. The sequential appearance of the T cell response to the 3 epitopes of PLP/MBP tested was hierarchical (PLP 139-151, PLP 178-191, and MBP 84-104, in order of decreasing level of immunodominance). Furthermore, the T cells against the spreading epitope (PLP 178-191) could transfer disease to naive syngeneic recipients [8]. Interest-

ingly, induction of tolerance in SJL mice against relapse-associated epitopes after the acute episode blocked disease progression and decreased the frequency of subsequent relapses [8]. In addition, short-term blockade of either CD28-CD80 (B7.1) costimulation by anti-CDS0 F(ab) fragment [10] or CD40-CD154 (CD40L) interaction using monoclonal anti-CD154 antibody [1 l] during remission from acute disease significantly reduced both the incidence of disease relapse and the T cell response

aUnless specified, all epitopes mentioned in Table 2 refer to T cell responses. The abbreviations are given in the order of their appearance in Table 2. M B P - Myelin basic protein; P L P - Proteolipid protein; T M E V - Theiler's murine encephalomyelitis virus; S E A Staphylococcal enterotoxin A; MOG - Myelin oligodendrocyte protein; GAD65/67 - Glutamic acid decarboxylase 65/67; HSP65 - Heat shock protein 65; Mtb - heat-killed Mycobacterium tuberculosis H37Ra; Bhsp65 - mycobacterial hsp65; A A Adjuvant-induced arthritis; Sm B/B'- Smith Ag B/B'; La & R o - Antigens within the ribonucleoprotein complex; AChRacetylcholine receptor; CTLA-4 - Cytotoxic T-lymphocyte antigen-4; P0 - Peripheral nervous system myelin P0 glycoprotein; IRBP- Interphotoreceptor retinoid binding protein; S-Ag - retinal S-antigen.

20

Table 2. The antigen specificity of epitope spreading in experimental models of autoimmune diseases Disease model

Animals tested

Disease-inducing antigen/agent

Antigen/epitopes targeted during epitope spreading a

Ref.

Experimental autoimmune encephalomyelitis (EAE)

(SJLxB10.PL)F1 mice SJL/J mice (SWR x SJL)F1 mice PL/J B 10.RIII mice Lewis rats SJL/J mice

MBP A c l - l l

[ 1]

Callithrix jacchus (the common marmoset)

MP4 fusion protein (PLP - MBP)

MBP 35-47, MBP 81-100, and MBP 121-140 PLP 178-191 and MBP 84-104 PLP 249-273, MBP 87-99, and PLP 137-198 MBP 81-100 and MBP 120-140 Multiple T cell epitopes within MBP Multiple T cell epitopes within MBP PLP 139-151, PLP 178-191, PLP 5670, and MOG 92-106 Anti-MOG antibodies

NOD mice

Spontaneous

[35]

NOD mice

Spontaneous

T cell response to GAD65, Carboxypeptidase H, Insulin, and HSP65 T cell and antibody response to GAD65/67, Peripherin, Carboxypeptidase H, and HSP60

Lewis rats

Mtb (H37Ra)

Lewis rats

Mtb (H37Ra)

NZW rabbits

Sm B/B' peptide

Mice

La (or Ro) antigen

(SWR x NZB)F1 Mice (NZB/NZW)F1 mice

Spontaneous

Diabetes in the non-obese diabetic (NOD) mouse

Adjuvant-induced arthritis

PLP 139-151 PLP 139-151 MBP 100-120 + SEA MBP 89-101 MBP TMEV

(AA)

[8] [ 13] [17] [ 19] [20] [24] [27]

[36]

417--431,441-455, 465-479, 513-527, [4] and 521-535 of Bhsp65 Multiple B cell epitopes within Bhsp65 [63] after recovery of AA Antibody response to other epitopes of Sm B/B' antigen and other spliceosomal proteins Antibody response to both La/SS-B and Ro/SS-A proteins Antibody response to nucleosomal components Response to epitopes in V. region of anti-DNA antibody

[71]

Human AChR czsubunit peptides

Antibodies to rabbit AChR

[89]

C57BL/6 mice

tz (146-162) of AChR + anti-CTLA-4 antibody

[91 ]

Rats

AChR {x-subunit

Antibody and T cell response to other subdominant epitopes of AChR txsubunit Antibody response to cytoplasmic region of AChR

Experimental autoimmune neuritis (EAN)

Lewis rats

Peptides 56-71 and 180-199 of P0 protein

T cell response to other epitopes of P0 protein

[96]

Equine recurrent uveitis (ERU)

Horse

IRBP

Multiple T cell epitopes within IRBP and S-Ag

[97]

Experimental autoimmune gastritis (EAG)

Mice

H/K ATPase ~l-subunit T cell and antibody response to {xsubunit of H/K ATPase

Systemic lupus erythematosus (SLE)

Experimental autoimmune NZW rabbits myasthenia gravis (EAMG)

Spontaneous

[72] [77a,b] [78]

[92]

[981

21

to relapse-associated epitopes. Unexpectedly, mice treated with intact anti-B7.1 antibody [10] or antiCTLA-4 antibody [12] during ongoing disease suffered from exacerbated relapsing disease and revealed enhanced epitope spreading. However, the above studies have revealed a window of therapeutic intervention in the face of epitope spreading. Tuohy and colleagues have studied the determinant specificities and functional significance of epitope spreading in R-EAE inducible in (SWR X SJL) F1 mice by injection with PLP 139-151 [13]. It was observed that the T cell response to the disease-inducing peptide, PLP 139-151, gradually diminished, whereas spreading of responses to MBP and new epitopes of PLP occurred with progression of EAE in a sequential pattern (PLP 249-273, MBP 87-99, and PLP 137-198, in that order). Interestingly, the T cells specific for the spreading determinant could passively transfer EAE to naive syngeneic recipients, whereas induction of peptide-specific tolerance to spreading epitopes after onset of EAE could prevent progression of EAE [13]. Furthermore, interferon-I] treatment of mice not only reduced the frequency/severity of disease relapses but also suppressed epitope spreading [14]. The results of another study by the same investigators revealed that the T cell response to the spreading-associated determinant (MBP 87-99) in (SWR X SJL) F1 mice had a proinfiammatory Thl cytokine profile, and that splenocytes activated with MBP 87-99 could adoptively transfer acute EAE to na'fve recipients [15]. Using a novel therapeutic approach based on immune deviation, T cells specific for MBP 87-99 were genetically modified to secrete high level of immunoregulatory cytokine, IL-10 and then transferred into (SWR X SJL) F1 mice after onset of EAE [16]. These recipient mice showed a marked inhibition of both disease progression and demyelination. Thus, induction of a disease-regulating T cell response targeted to relapse-associated epitope provided therapeutic benefit. Taken together, these results strongly support the idea that determinant spreading is involved in the progression of EAE. Using another model of EAE, it was shown that staphylococcal enterotoxin superantigen A (SEA) can reactivate EAE in PL/J mice recovered from MBP-induced EAE, and this re-activation of the disease was associated with intramolecular

22

spreading to epitope 100-120 of MBP [17]. No inter-molecular spreading was observed during the course of reactivated disease. In another set of experiments by the same investigators, it was observed that MBP peptide 100-120 per se failed to induce EAE in PL/J mice [ 17]. However, an additional challenge of these mice with SEA rendered MBP 100-120 encephalitogenic, and the diseased mice developed T cell responses to new epitopes namely, 81-100 and 120-140. These results provide insight into reactivation of autoimmunity following exposure to microbial agents containing the relevant superantigen. Another aspect of epitope spreading was highlighted in the SJL model of MBP-induced EAE: epitope spreading was found to occur in active but not passive EAE [ 18]. These results point to the fine differences in the pathogenesis of active vs. passive disease resulting from activation of T cells against the same autoantigen. Diversification of response to MBP epitopes has also been observed during the course of EAE in B10.RIII mice [19] and the Lewis rat [20]. The initial T cell response of B 10.RIII mice with relapsing EAE was focused predominantly on the immunogen (MBP 89-101), but later spreading of the response occurred to other epitopes within MBP [19]. In the case of Lewis rats with EAE, the dominant encephalitogenic T cells in the induction phase of the disease were directed to epitope 71-90, whereas T cell responses to new epitopes within MBP appeared during the recovery phase of the disease [20]. Another study revealed that Wistar Kyoto (WKY) rats having the same MHC haplotype as the Lewis rat were resistant to EAE despite raising potent T cell responses to the dominant encephalitogenic T cell epitope within MBP [21]. However, the antigen-specific T cell response in WKY rats was skewed towards a predominantly Th2 type compared to predominantly Thl type in Lewis rats. These results along with those of another study in AA using the same rat strain [4] demonstrate that a positive T cell response to the pathogenic epitope may not correlate with the presence or absence of clinical signs of the disease. Furthermore, these results point towards the significance of analysis of the cytokine responses of epitope-specific T cells in addition to measuring the proliferative T cell response in evaluation of epitope spreading and related aspects of the disease process.

In a recent study, autoantibody responses of mice with EAE were measured using a large set of autoantigens in a protein microarray [22]. Different patterns of autoantibody reactivity were observed in acute vs. chronic phase of EAE. Chronic EAE was characterized by both intra- and inter-molecular epitope spreading. Furthermore, attenuation of established EAE by a tolerizing DNA vaccination based on defined autoantigens was associated with reduced epitope spreading of autoantibody responses. Epitope spreading has also been implicated in the induction of autoimmunity in Theiler's murine encephalomyelitis virus (TMEV)-induced demyelinating disease in SJL/J mice [23-26]. TMEV causes chronic infection of the CNS. The host immune response (predominantly CD4+ T cells) to the virus is responsible for the initial myelin damage [23]. Interestingly, however, the T cell response specific for self myelin antigens appears during the progression of the disease. After the disease onset, response to the dominant epitope of PLP, PLP139-151, is observed followed by that to relatively less dominant epitopes ofPLP (PLP 178-191 and PLP 56-70)and MOG 92-106 [24, 25]. These results demonstrate that epitope spreading targeting myelin antigens contributes to TMEV-induced autoimmunity. This conclusion is further supported by the findings that tolerance induction against defined T cell determinants within myelin antigens in SJL mice with ongoing TMEV-induced disease results in marked reduction in demyelination [26]. However, unlike in PLP 139-151-induced EAE, costimulation blockade in TMEV-infected mice was found to enhance the disease severity instead of reducing it. Like in rodent EAE, epitope spreading has also been observed during the course of EAE in the common marmoset, Callithrix jacchus, and in patients with MS. The Callithrix jacchus marmoset develops chronic relapsing-remitting form of EAE following challenge with myelin antigens, and both T cells and antibodies serve as immune effector mechanisms in the disease process [27, 28]. Interestingly, treatment of these non-human primates with anti-CD40 antibody prevented intramolecular spreading and afforded protection against EAE [29, 30]. In a study based on patients with isolated monosymptomatic demyelinating syndrome (IMDS), the T cell reactivity to PLP epitopes was found to

decrease over time. However, spreading of T cell responses to other PLP epitopes was observed in those IMDS patients who progressed to clinically definite MS (CDMS) [31-33].

2.2. Insulin-Dependent Diabetes Mellitus (IDDM) or Type I Diabetes IDDM is an autoimmune disease involving mononuclear cell infiltration of the pancreatic islets (insulitis), destruction of the 13-islet cells, and insulin deficiency. Spontaneously-developing diabetes in the non-obese diabetic (NOD) mouse serves as a model for human IDDM [34]. Glutamic acid decarboxylase (GAD65) has been invoked as one of the early target antigens in the pathogenesis of autoimmune diabetes in the NOD mouse [35, 36]. With the progression of disease in the NOD mouse, the T cell response spreads to additional epitopes within GAD65 and to other ~-cell antigens (e.g., to carboxypeptidase-H, insulin, and heat shock protein 65 (Hsp65) in one study [35], and to GAD 67, carboxypeptidase-H, peripherin, and Hsp60 in another study [36]). Furthermore, tolerization of GAD65reactive T cells suppressed the development of insulitis, disease progression, and the spreading of T cell responses [35, 36]. In the NOD mouse, Thl spreading leads to disease progression, whereas Th2 spreading is associated with protection from disease [37]. Induction of a Th2 response to a single I]-cell antigen (GAD/Hsp peptide 277/Insulin B chain) or a peptide of GAD (e.g., p35/p6) led to spreading of the Th2 cellular and antibody response to other I]-cell antigens/epitopes along with protection from diabetes [37]. In another study on emergence of T cell responses to GAD during the course of spontaneously developing disease in the NOD mouse, new potential target epitopes within GAD65 and GAD67 were described [38]. The patterns of T cell responses to epitopes within GAD65/GAD67 were found to vary with the age and disease-status of mice. In addition, NOD mice given a disease-protective regimen (e.g., adjuvant challenge) revealed a different pattern of response to GAD65/67 compared to unmanipulated control NOD mice. These results validate the role of T cell responses to GAD in the pathogenesis of autoimmune diabetes. Studies on epitope spreading in human IDDM have revealed the pattern of autoantibody responses

23

to [3-cell antigens in children of diabetic patients [39]. The anti-islet autoantibody response in these subjects was characterized by appearance of an early IgG1 response to one or more islet antigens, particularly insulin. Thereafter, coupled with a decline in the titer of these antibodies, there was a sequential appearance of antibodies against other I]-cell antigens over a period of several years [39]. In other studies in preclinical childhood type I diabetes, it was observed that the initial antibody response to GAD of offspring of diabetic patients was directed primarily to epitopes within the middle portion of GAD65, but later it spread to epitopes in other regions of GAD65 and GAD67 [40, 41]. Similarly, intermolecular spreading of the T cell reactivity and antibody responses to islet antigens was found to have occurred during the pre-clinical phase of type I diabetes in subjects at risk (as defined by positivity for autoantibodies to l-islet antigens) for developing clinical diabetes [42]. Taken together, the above studies lend support to the role of epitope spreading in the pathogenesis of autoimmune diabetes in humans, and provide impetus for development of new therapies for individuals at risk for type I diabetes. The observations that patients with IDDM have activated T cells against GAD65 [43], and that diabetic siblings possess relatively lower frequency of Valpha24JalphaQ+ T cells compared to non-diabetic twins [44], have provided new insights into the autoreactive T cell repertoire and its regulation in this disease. The precise relationship of the activity of these T cell subsets to the levels and epitope specificity of anti-GAD autoantibodies in diabetes remains to be determined. 2.3. Arthritis

Rheumatoid arthritis (RA) is a human autoimmune disease that primarily affects the joints. It is characterized by persistent inflammatory synovitis, generally involving peripheral joints in a symmetrical fashion. The etiology of RA is not known. Adjuvant-induced arthritis (AA) is an experimental model of human RA, and it can be induced in Lewis rats by immunization with heat-killed Mycobacterium tuberculosis H37Ra in mineral oil [45, 46]. The T cell response to the 65-kD mycobacterial heat shock protein (Bhsp65) has been implicated in the pathogenesis of AA as well as RA [47-50]. We have

24

previously shown that there is a shift in the epitope specificity of the T cell response to Bhsp65 during the course of AA in the Lewis rat [4]. In the acute phase of AA, the T cell response of arthritic Lewis rats was focused on peptide 177-191 (which contains the arthritogenic determinant 180-188) and other epitopes in the middle and N-terminal regions of Bhsp65. However, during the recovery phase of AA, appearance of new T cell responses directed to the 5 C-terminal epitopes of Bhsp65 (namely, 417431, 441455, 465-479, 513-527, and 521-535) was observed. Interestingly, pretreatment of na'fve Lewis rats with the synthetic peptides representing these 5 Bhsp65 C-terminal determinants (BCTD) significantly reduced the severity of subsequent AA [4, 51 ]. Furthermore, T cell responses to BCTD were observed early following M. tuberculosis (H37Ra, heat-killed) challenge in the Wistar Kyoto (WKY) rats that possess the same MHC haplotype as the AA-susceptible Lewis rat but are resistant to induction of AA [4]. The simultaneous emergence of T cell responses to the pathogenic (180-188/ 177-191 determinant) and regulatory (BCTD) epitopes could explain, in part, the AA-resistance of WKY rats. The results of our recent study demonstrate that the C-terminal epitopes of self hsp65 are also disease-regulating in nature (Durai, Gupta and Moudgil, manuscript in press [51 a]). The above results suggest that spreading of the T cell responses to BCTD during the course of AA might be involved in natural recovery from acute AA in the Lewis rat. Furthermore, these findings demonstrate that epitope spreading in the course of an autoimmune disease is not always pathogenic; instead, in another situation, it can also be disease-regulating in nature. This is the first study [4] reporting the disease-regulating aspect of epitope spreading in the course of an autoimmune disease. Another aspect of emergence and spreading of T cell responses was revealed in a study on the immunological basis of environmental modulation of AA in the Fischer F344 (F344) rat [52]. We observed that F344 rats kept in a barrier facility (BF-F344) were susceptible to AA, whereas those maintained in a conventional facility (CV-F344) spontaneously acquired protection (or resistance) against induction of AA. Testing of the T cell responses to peptides of Bhsp65 of naive F344 rats showed that CV-F344 but not BF-F344 rats raised T cell response to mul-

tiple epitopes of Bhsp65, including BCTD. In addition, the level of these spontaneously-arising T cell responses gradually increased with the duration and extent of exposure of F344 rats to the conventional environment. The functional significance of these BCTD-directed T cell responses was evident from results of adoptive transfer experiments: BCTD-restimulated (in vitro) splenic cells of naive CV-F344 rats could adoptively transfer protection against AA to naive BF-F344 recipients [52]. The above results present one of the mechanisms underlying the influence of housing environment on protection against an autoimmune disease. The role of conventional environment in facilitating the induction of an autoimmune disease has been observed in various models of autoimmunity (e.g., EAE, thyroiditis, hemolytic anemia, Pristane-induced arthritis (PIA) etc.) [53-58]. In contrast to this, our study described above [52] along with others in animal models of arthritis and diabetes [59-62a] reflect upon the protective effect of environment on autoimmunity. AA is believed to be a T cell-mediated disease. There is meager information about the role of antibodies to Bhsp65 in the pathogenesis of this disease. However, recent studies have shown that in addition to possessing disease-regulating T cell epitopes, Bhsp65 also harbors protective B cell epitopes. During the course of AA, Lewis rats develop antibodies against Bhsp65, and the number of epitopes within Bhsp65 recognized by these antibodies gradually increases during the recovery phase of the disease [63] (Kim and Moudgil, manuscript in preparation). Thus, like spreading of the T cell response to BCTD in AA described above, arthritic Lewis rats also show spreading of anti-Bhsp65 antibody response. Interestingly, challenge of naive Lewis rats with peptides comprising the B cell epitopes (e.g., peptides 31-46 and 37-52) represented in the diversified response afforded protection against subsequent disease [63]. Similarly, passive immunization with the antibodies directed against one of these epitopes (peptide 31-46) also suppressed subsequent AA [63]. Furthermore, the resistance to AA of BN rats correlates with natural antibody response to the same B cell epitopes as those involved in epitope spreading in the susceptible Lewis rats. These results demonstrate the role of spreading of antibody response to Bhsp65 in regulation of AA. Spreading of the tolerogenic effect of disease-

related epitope of Bhsp65, p176-190, and that of the suppressive effect of antigen-specific anergic T cells have been invoked in downmodulation of the course of avridine-induced arthritis and/or AA [64, 65]. It was observed that induction of nasal tolerance against p176-190 provided protection against subsequent AA as well as avridine-induced arthritis [64]. It was proposed that tolerance of T cells recognizing p176-190 or its mimic spread to T cells of other specificities that are involved in induction of arthritis. Similarly, it was suggested that a subset of anergic T cells, in the presence of the specific antigen recognized by these cells, exerted spreading suppressive activity on T cells of other antigen specificities [65]. Furthermore, this amplification of suppressive effect was attributed to modulation of the activity of the antigen presenting cell (APC) by anergic T cells. The influence of a disease-regulating antigenic challenge on diversification of response has been described in the Pristane-induced arthritis (PIA) model [66]. A comparison of the T cell response of arthritic mice compared to mice protected against disease by pre-treatment with Bhsp65 revealed that there was 'repertoire limitation' (the opposite of diversification of response) in Bhsp65 pre-treated mice [66]. The protective effect of Bhsp65 was attributed in part to prevention of diversification of response via induction of a Th2 response. A study of the T cell repertoire in RA patients showed that several dominant T cell clones were found in the synovial membrane but not in the peripheral blood [67]. Analysis of the complementarity-determining region 3 (CDR3) region following sequencing of the T cell receptor (TCR) V~ V-D-J junctional regions showed evidence for antigen-driven selection of the TCR. This TCR selection was attributed to determinant spreading during the course of RA. More similar studies in RA would help define the fine characteristics of the pathogenic T cell repertoire in this disease.

2.4. Systemic Lupus Erythematosus Systemic lupus erythematosus (SLE) is a multisystem human autoimmune disease characterized by development of autoantibodies against a variety of autoantigens: dsDNA, Ro/La ribonucleoprotein complex, histones and other nucleosomal compo-

25

nents, spliceosomal proteins, ribosomal proteins, etc [68-70b]. Antibodies against these autoantigens have also been found to develop during the course of disease in various experimental models of lupus (Table 2). Studies focused on autoantibodies in lupus patients and animal models of lupus have revealed both inter-molecular and intra-molecular epitope spreading involving one or more of the above-mentioned autoantigens. For example, James et al [71] demonstrated that NZW rabbits immunized with an Sm B/B' peptide (representing a C-terminal epitope) developed antibodies directed against the immunogen and other epitopes within the middle and amino-terminal regions of the Sm B/B' antigen. In addition, these animals also raised antibody response to other spliceosomal proteins (e.g., D, 70K, A, and C). This new experimental model of lupus would help define the role of autoantibodies to Sm B/B' in the pathogenesis of the disease in lupus patients [71]. In another study, mice immunized with La protein developed autoantibodies not only to the immunogen but also to 60-kD Ro, whereas mice immunized with 60-kD Ro produced anti-Ro antibodies as well as anti-La antibodies [72]. These results and those of other studies [73-77] demonstrate that development of antibodies to multiple components of the L a ~ o ribonucleoprotein complex occurs after challenge with a single component of the antigenic complex. In addition to experimentally-induced epitope spreading described above, spreading of the T helper and antibody response to components of the nucleosome in (SWR x NZB)F1 mice [77a, 77b], and intramolecular spreading of the T cell response to T helper (Th) epitopes within the V Hregion of a pathogenic anti-DNA antibody in NZB/NZW F1 (BWF1) mice [78] has been observed during spontaneously-arising disease. The patterns of autoantibody responses observed in animal models of lupus have been shown to be influenced by multiple factors including, genetic makeup (strain differences), the MHC class II haplotype of the host, age of the animal, the level of self tolerance to a particular autoantigen, and nature of the immunogen (homology between the corresponding self and foreign antigenic determinants, immunogenicity of endogenous antigens, the level of crossreactivity between different antigens, sharing of crossreactive epitopes between unrelated and physically-separated antigens, molecular mimicry,

26

etc.) [70a, 70b, 72-77]. Interestingly, tolerization of lupus-prone mice against either autoantibodyderived peptides or the protein/peptides (e.g., nucleosomal peptides) representing antigenic determinants involved in epitope spreading can successfully halt the progression of epitope spreading as well as the disease process [77c, 79, 80]. Alternatively, blockade of costimulation by administration of antiB7.1 and anti-B7.2 antibodies into BW F1 mice has also been shown to suppress the development of SLE [81]. Similarly, induction in lupus-susceptible mice of inhibitory CD8+ T cells that are capable of controlling the activity of autoreactive B cells can downmodulate the course of disease and prolong survival of the vaccinated mice [82]. SLE patients develop autoantibodies to a variety of autoantigens described above. Studies on antigen reactivity of sera of lupus patients have demonstrated temporal shifts either in recognition of another antigen (e.g., inter-molecular spreading from Sm antigen to RNP reactivity) or in reactivity to different epitopes within the same antigen (e.g., intra-molecular spreading within a given antigen depending on the model system: Sm B/B', Sm D, ribosomal protein L7, caspase-8, etc.) [83-87].

2.5. Myasthenia Gravis Myasthenia gravis (MG) is an autoimmune disease characterized by weakness and fatigability of skeletal muscles owing to impaired neuromuscular transmission resulting from antibody-mediated autoimmune attack against the nicotinic acetylcholine receptor (AChR). Experimental autoimmune myasthenia gravis (EAMG) is inducible in animals following immunization with either AChR or its peptide along with an immunomodulator (e.g., antiCTLA-4 antibody that can enhance T cell activation by interfering with inhibitory signal via CTLA-4 molecule), and it serves as a model for human MG [88]. Epitope spreading has been reported in different models of EAMG. NZW rabbits challenged with heterologous (human) AChR t~-subunit peptides developed clinical MG and antibody response to self (rabbit) AChR [88, 89]. These antibodies reacted strongly to rabbit AChR but only weakly to human AChR. These results show that the observed spreading of antibody response to self AChR in rabbits challenged with human AChR is attributable to

processing and presentation of endogenous AChR. EAMG can also be induced in mice. C57BL/6 (B6) mice immunized with AChR and boosted with ~146-162 peptide of AChR develop EAMG [90]. In another study, it was observed that B6 mice challenged with the immunodominant peptide o~146-162 of AChR and given anti-CTLA-4 antibody developed clinical signs of MG [91]. These mice developed both antibody and T cell response against the peptide immunogen (~ 146-162) and other epitopes of AChR cx-subunit [91]. In another recent study, it has been shown that rats immunized with extracellular region of self (rat) AChR ~-subunit initially developed antibodies to the immunogen followed by antibody response to the intracellular cytoplasmic part of AChR [92]. There is some suggestive evidence for epitope spreading in patients with MG. The T cell clones generated from patients with MG were specific for peptide epsilon 201-219 and epsilon subunit of adult AChR, but did not show any reactivity to fetal AChR [93]. However, the serum antibodies of these patients showed higher reactivity with fetal AChR compared to adult AChR. These results suggest that anti-AChR antibodies in MG patients might have been generated following epitope spreading triggered by T cells specific for adult AChR. In this context, it has recently been described that primary human myoblasts treated with IFN-~, for induction of MHC class II expression can present endogenous AChR epitope to an AChR-specific CD4+ T cell clone [94]. These results suggest that during the course of clinical MG, myoblasts can serve as APC and amplify the ongoing T cell response. Furthermore, these myoblasts may also become targets of cytotoxic immune attack, and the resulting release of self antigens can in turn contribute to epitope spreading.

2.6. Other Autoimmune Diseases Pemphigus is an autoimmune disease affecting the skin. Epitope spreading has been observed in pemphigus vulgaris, pemphigus foliaceus, and other cutaneous autoimmune disorders (reviewed in Ref. [95]). Similarly, diversification of the autoimmune response has also been reported in animal models of neuritis [96], uveitis [97], gastritis [98], and oophoritis [99, 100] (Tables 1 and 2).

3. EPITOPE SPREADING IN OTHER IMMUNE-MEDIATED DISEASES 3.1. Organ Transplant/Graft Rejection Alloreactive T cells recognize foreign (allogeneic) MHC molecules either as intact antigens expressed by the donor APC (direct allorecognition) or as epitopes derived from donor MHC but presented by the recipient APC (indirect allorecognition) [ 101 ]. In a study on heart allograft recipients [ 102], the epitope specificity of self-restricted alloreactive T cells (indirect allorecognition) was examined. Sequential blood samples collected from these patients over a 3 year period were tested for allopeptide reactivity using a panel of peptides representing sequences derived from 32 HLA-DR alleles. The higher incidence of complications (in this case, coronary artery vasculopathy; CAV) in these patients directly correlated with persistent alloreactivity as well as epitope spreading (both intra- and inter-molecular). These results demonstrate the role of epitope spreading in chronic graft rejection. In other studies on heart as well as liver transplant recipients [ 103, 104], a differential T cell response to the alloantigens within the graft was observed in recipients undergoing primary acute rejection compared to those having either recurring episodes of rejection or a chronic rejection. The T cell responses in the former were directed to a single dominant epitope in one of the two mismatched HLA-DR antigens, whereas those in the latter were found to spread to both HLA-DR antigens as well as to other alloantigens of the transplanted tissue. Thus, spreading of the alloresponses apparently was involved in the pathogenesis of graft rejection. We have described above that both intra-molecular and inter-molecular epitopes spreading occur during the course of spontaneously developing autoimmune diabetes in the NOD mouse [35, 36]. Interestingly, deliberate deviation (from Thl to Th2 type) of the GAD-specific T cell response in young NOD mice was successful in inhibiting the progression of the disease process [37]. Furthermore, a similar experimental manipulation also prolonged survival of syngeneic islet grafts in NOD mice with diabetes [105]. These results show that tolerance induction to the key autoantigen could also benefit the outcome of organ graft. Similarly, in another

27

study on skin graft tolerance, it was shown that tolerance induction (by donor-specific transfusion; DST) against one MHC molecule (La)-mismatch spread to skin grafts with more than one MHC-mismatch [106]. Thus, tolerance to allogeneic MHC showed spreading that was beneficial for graft survival. 3.2. Tumors

The study of anti-tumor immunity in experimental models as well as patients vaccinated with defined peptide antigens has revealed the occurrence of epitope spreading involving CD8+ cytolytic T cell (CTL) and/or CD4+ T cell responses [107-110]. In a study using ovalbumin (OVA) as a model tumor antigen and EL4 thymoma cells with or without OVA expression (the former is named EG.7OVA) as the tumor-inducing cell line, it was observed that B6 mice immunized with OVA or EG.7OVA raised CTL response directed to the immunodominant epitope 257-264 of OVA [ 107]. Furthermore, CTL response induced by OVA afforded protection against tumor induction by challenge with EG.7OVA cell line. Interestingly, mice that survived EG.7OVA challenge raised additional CTL responses to two other epitopes within OVA, namely 55-62 and 176-183, and to other endogenous antigens within EL4 cells. The emergence of new CTL responses was attributed to release of tumor antigens following tumor rejection and subsequent cross-presentation by the APC of cryptic epitopes within these tumor antigens. Similarly, epitope spreading involving CTL response to tumor antigens was observed in another study based on P815 tumor model [108]. Immunization of DBA/2J mice with a single P815-derived tumor peptide, P1A, induced potent CTL response that could successfully induce rejection of P1A § P511 tumor. Interestingly, mice undergoing tumor rejection developed new CTL responses directed to another P815-derived peptide antigen, P1E, and also could reject P1A- P1.204 tumor. The induction of new CTL response was attributed to processing and presentation of released tumor antigen by the host APC. These results demonstrate that diversification of the CTL response could be beneficial in control or elimination of antigen-loss tumor cell variants that arise during the course of tumor progression and tumor therapy. Diversification of anti-tumor CTL/CD4+

28

immune response has also been observed in patients undergoing peptide-based immunotherapy [109, 110]. In one study, induction of immunity against peptides derived from HER-2/neu (a self tumor antigen that is quantitatively over-expressed in certain adenocarcinomas) was tested in patients with breast and ovarian cancer [ 109]. Immunization of patients with HER-2/neu peptides was performed i.d. using GM-CSF as the adjuvant. Peptide-specific CD4+ T cell responses were observed in all patients, and in most of them, the peptide-primed T cells were crossreactive with the native HER-2/neu protein. The T cell response against the HER-2/neu protein included reactivity against new epitopes within the protein (epitope spreading). Induction of these new T cell specificities was attributed to processing and presentation of endogenous HER-2/neu protein. In another study, induction of CTL response against HER-2/neu and MUCl-derived peptides was tested in patients with advanced breast and ovarian cancer [ 110]. Vaccination of patients was performed using autologous dendritic cells pulsed with the appropriate peptides. One of the patients immunized with HER-2/neu peptides raised T cell response directed against MUC-1 protein, whereas another patient vaccinated with MUC-1 peptides raised T cell response against CEA and MAGE-3 peptides. These studies demonstrate that epitope spreading invariably accompanies anti-tumor vaccination in both tumor models and cancer patients, and therefore, it is of significance in determining the outcome of the vaccination regimen. 3.3. Infection/Vaccination

(i) Lymphocytic choriomeningitis virus (LCMV) infectio.n. It has been observed that the cytolytic T cell (CTL) response in mice infected with LCMV is focused on the immunodominant epitope of the viral nucleoprotein during the acute phase of the infection, but it diversifies to sub-dominant epitopes of the viral glycoprotein in the chronic phase of the disease [111]. Furthermore, CTL response to the immunodominant epitopes apparently plays a critical role in induction of acute disease, whereas responses to sub-dominant epitopes are involved in inducing protective immunity and clearing of the infection [ 111, 112]. The density of MHC-peptide complexes on the APC surface and the composi-

tion of the T cell repertoire are among the major factors determining the hierarchy of CTL response [112, 113]; (ii) Lyme disease is caused by infection by Borrelia burgdorferi (Bb), and some of the clinical features of the disease are believed to be of autoimmune origin [114]. For example, molecular mimicry between epitopes within the outer surface protein A (Osp A) of Borrelia burgdorferi and leukocyte function-associated antigen- 1 (LFA-1) has been invoked in the pathogenesis of Lyme arthritis [114]. In one study, diversification of the antibody response to different antigens of the pathogen was observed with progression of Lyme disease [115]. However, the precise significance of this antibody spreading in pathogenesis of the disease has not yet been fully defined; and (iii) DNA vaccination

using human immunodeficiency virus-1 (HIV-1) r~ulatory genes: Both protein and DNA encoding the regulatory genes nef, rev, and tat of HIV-1 were found to be immunogenic in mice, and immunization with DNA plasmids induced higher magnitude of antibody response (compared to protein antigens) along with epitope spreading [116]. However, the protective efficacy against HIV- 1 of these antibodies remains to be determined.

4. M E C H A N I S M S UNDERLYING E P I T O P E SPREADING DURING T H E COURSE OF AN A U T O I M M U N E DISEASE Considering diverse experimental models of autoimmune diseases involving different target organs and predominantly either T cell (CD4+/CD8+)- or antibody-response to one or more disease related antigens as the pathogenic effector mediators (Table 2), various mechanisms have been proposed to explain the phenomenon of epitope spreading observed during the course of different autoimmune diseases (Table 3). These include inter-related factors or conditions that operate in concert in different combinations, depending on the disease process, to induce epitope spreading (Fig. 1, Table 3). These are described below.

1. Upregulation of the display of cryptic~subdominant epitopes within a self antigen under inflammatory conditions. Native self and foreign antigens possess potential T cell epitopes that are

Table 3. Proposed mechanisms underlying the phenomenon of epitope spreading 1. Upregulationof the display of cryptic/sub-dominant epitopes within a self antigen under inflammatory conditions 2. Releaseof self antigens and their processing and presentation following tissue damage in the course of an autoimmune disease or a chronic microbial infection 3. The frequency and avidity of epitope-specific precursor T cells within the mature T cell repertoire favoring responsiveness to certain antigenic determinants over others 4. Presentationof neo-epitopes within a particular self antigen by the B cells specific for that antigen 5. The influence of antigen-bound antibodies on processing and presentation of T cell epitopes within that antigen

processed and presented either efficiently (dominant determinants) or poorly (cryptic/sub-dominant determinants) by the antigen presenting cells (APC) [117]. However, both sets of determinants are immunogenic in the peptide form. In the case of a self antigen, tolerance is readily induced to its dominant but not cryptic/subdominant epitopes [118-121]. For this reason, unlike a foreign dominant epitope that is generally immunogenic, a self dominant determinant generally fails to induce a response owing to self tolerance. However, the T cells against cryptic/sub-dominant epitopes escape tolerance induction in the thymus and therefore, are available in the mature T cell repertoire. These T cells can be activated provided the otherwise poorly processed cryptic/sub-dominant epitopes within the native self antigen are efficiently presented to the T cells by professional APC. This could happen under conditions of upregulated antigen processing and presentation events as in the case of inflammation and/or infection [3, 122-125]. The T ceils specific for cryptic/sub-dominant epitopes of an endogenous self antigen thus activated (constituting epitope spreading) can participate in further propagation of the ongoing disease process (Fig. 1). In this manner, the inflammatory and cytokine milieu created during the initial phase of the disease induced by T cells specific for one self antigen or one of its epitopes can facilitate induction of new T cell responses

29

Experimentally-induced autoimmune disease

Spontaneously-developing autoimmune disease

Initiation of the autoimmune process

Tissue damage ~

Inflammation

Upregulation of antigen ~ processing and presentation

Microbial infection

Release of self antigens

1

Display of previously cryptic self epitopes

Priming of new subsets of self-reactive T cells

Propagation of autoimmunity

1 Worsened and chronic disease (Pathogenic epitope spreading)

Control of activity of pathogenic T cells

l

Remission from acute phase of the disease (Regulatory epitope spreading)

Figure 1. Epitope spreading: the underlying mechanisms and role in the disease process. Initiation of an autoimmune disease, either spontaneously or following an antigenic challenge, creates a local inflammatory milieu that is conducive to upregulation of antigen processing and presentation. In addition, the inflammatory events, either of autoimmune origin or induced by a microbial infection, cause tissue damage leading to the release of self antigens. Under these circumstances, self antigens are processed efficiently by the antigen presenting cells (APC) revealing previously cryptic epitopes to potentially self-reactive T cells available in the mature T cell repertoire. The activation of new subsets of autoreactive T cells enhances the ongoing inflammatory events and further amplifies the processes described above. The outcome of priming of the self-reactive T cells depends on multiple factors: the antigen/epitopes targeted during epitope spreading, the nature of the T cell response (e.g., Thl/Th2), the genetic make up of the individual, etc. In this context, epitope spreading could either further perpetuate (pathogenic epitope spreading) or attenuate (protective epitope spreading) the ongoing disease process. The above scheme depicts only induction of the T cell response. However, the activated T cells may provide help to multiple autoreactive B cells displaying one of the epitopes of the antigen recognized by T cells and thereby, lead to spreading of antibody responses. Similarly, activated B cells displaying epitopes of an autoantibody can prime the appropriate T cells, which upon activation help other B cells to effect spreading. In addition, the processing and presentation of certain T cell epitopes within the native self antigen can be modulated by antigen-bound autoantibodies. These processes are described in detail under Section 4.

30

directed to other epitopes within the same antigen (intra-molecular spread) and/or another endogenous self antigen (inter-molecular spread). This also is one of the mechanisms by which self antigens released following tissue damage (described below) could contribute to epitope spreading. 2. Release of self antigens and their processing and presentation following tissue damage in the course of a microbial infection or an autoimmune disease. The etiology of most of the human autoimmune diseases is not known. However, one important factor that is believed to constitute a trigger or precipitating factor for induction of autoimmunity is microbial infection. However, the precise role of microbial agents in induction or perpetuation of autoimmunity is not yet fully defined. Some of the mechanisms proposed to explain this association include - a) molecular mimicry: a microbial antigen/epitope structurally mimicking a self antigen/epitope such that the T cells primed following microbial infection can be re-stimulated by the endogenous self antigen, and thereby, can target cells/tissue expressing that self antigen leading to tissue damage [126-129]; b) bystander activation: stimulation of potentially self-reactive T cells under the immune environment where priming of microbial antigen-specific T cells is taking place; the autoreactive T cells can then cause tissue damage [3]; and c) tissue injury. leading to induction of a_utoimmune response: the tissue damage caused by microbial infection results in the release of endogenous self antigens that can then be processed and presented by local as well as peripheral APC leading to priming of self-reactive T cells [3, 130, 131] (Fig. 1). The T cells activated locally can then induce autoimmune damage within the target organ [130, 131], whereas the T cells activated in the periphery can re-enter the target tissue and damage the cells expressing a particular self antigen [3]. The role of virus-mediated tissue damage in induction of autoimmunity has been best demonstrated by Rose and colleagues in the model of autoimmune myocarditis [132, 133], by Sarvetnick and colleagues in the NOD model of type I diabetes [134], and by Miller and colleagues in the TMEV-induced EAE model [23-26]. Autoimmune myocarditis induced by coxsackievirus B3 has been shown to be a biphasic disease: an early 'infection' phase and a subsequent 'autoimmune'

phase characterized by T cell and antibody response to cardiac myosin [133]. In TMEV-induced EAE, the induction of autoimmunity has in addition been linked to epitope spreading resulting from release of self antigens [23-26]. (In a viewpoint different from release of antigens within the target organ, it has been suggested that epitope spreading in EAE might be initiated in the peripheral lymphoid tissues instead of the CNS; this proposition is based on the observation that myelin antigens are expressed in the lymph node, spleen and thymus of SJL mice [135].) An additional aspect relevant to release of tissue antigens is the relative susceptibility/resistance to acute infection of the target organ [136]; for example, the cytokines IFN-~ and IFN-7 produced within the pancreatic islets can significantly influence the susceptibility to coxsackievirus B4 infection of the cells. Another mechanism for the release of intracellular self antigens has been proposed by Rosen and colleagues [137, 138]; in this process, apoptotic cells serve as an important source of self antigens, and the novel antigenic fragments produced in apoptotic surface blebs are implicated in reversal of self tolerance leading to induction of autoimmunity. Taken together, irrespective of the mechanism causing tissue damage (whether the initial phase of an autoimmune damage or a microbial infection), the resulting release of self antigens and their processing and presentation leads to activation of new subsets of self-reactive T cells (Fig. 1) constituting epitope spreading. 3. The frequency and avidity of epitope-specific precursor T cells within the mature T cell repertoire favoring responsiveness to certain antigenic determinants over others. The T cell responses to various epitopes within a native antigen, or to individual antigens within a mixture of antigens, are hierarchical, and accordingly, th.e corresponding antigen/ epitope can be categorized as immunodominant, sub-dominant, or cryptic [ 117]. A similar hierarchy is also observed during the induction of epitopespecific T cell tolerance [139, 140]. The observed hierarchy of T cell response to the immunogen/ tolerogen is influenced by multiple factors operating at the level of the APC (described above) as well as those relating to the composition of the mature T cell repertoire [6, 111-113, 140]. Both the size (frequency) and composition (e.g., the relative levels of

31

high avidity vs. low avidity T cells) of the T cell repertoire can significantly influence the magnitude as well as the timing of appearance of response to an antigenic determinant following challenge with the native antigen. In the context of epitope spreading, the above-mentioned characteristics of the T cell repertoire have been invoked, in part, in explaining the hierarchy as well as ordered sequential appearance of response to different antigens/epitopes involved in inter- or intra-molecular epitope spreading [3, 8, 9, 13, 141].

4. Presentation of neo-epitopes within a particular self antigen by the B cells specific for that antigen. Most of the above discussion on priming of the T cells is based on APC implying primarily dendritic cells and macrophages. However, activated B cells serving as potent APC also can participate in induction and propagation of epitope spreading. Mamula and Janeway proposed an interesting model based on the role of B cells as APC in the diversification of the T cell and antibody response [ 142]. According to this model, the initial T cell priming to self epitopes is done by APC like dendritic cells, and these activated T cells then provide help to the appropriate B cells. The activated B cells in turn can take up antigen by specific interaction between the antigenic determinants and the B cell receptors, and then process and present that antigen to the T cells. The newly displayed antigenic determinants by B cells (as APC) can now activate new subsets of T cells, which then can provide help to new population of B cells. These T-B interactions thus lead to diversification of both T cell and antibody responses. In a parallel situation, B cells could present epitopes of an autoantibody to the T cells, and these T cells could then render help to multiple clones of B cells, each displaying an epitope crossreactive with the original determinant (reciprocal T-B determinant spreading) [78]. The role of B cells as APC in induction of selfdirected antibody response in the setting of epitope spreading has been demonstrated in experimental models of lupus [143, 144].

5. The influence of antigen-bound antibodies on processing and presentation of T cell epitopes within that antigen. We have described above the role of B cells in diversification of the T cell/antibody response. Another mechanism by which

32

components of humoral immunity can influence the induction of epitope-specific T cell response is through antigen-specific antibodies. Through an interesting series of experiments, it has been demonstrated that antigen-bound antibodies can significantly influence the processing and presentation of T cell epitopes within that antigen [122, 145]. Depending on the nature and site (in context of the antigenic structure) of interaction between the antigenic determinant and the antibody, antibodies bound to specific epitopes of an antigen can either enhance or suppress the T cell response to the epitope involved. These results provide one of the mechanisms explaining a shift in the epitope specificity of T cell response (as observed in epitope spreading) during the course of an immune response directed towards a microbial or self antigen.

5. THE E F F E C T O R MECHANISMS INVOLVED IN MEDIATING THE FUNCTIONAL O U T C O M E OF EPITOPE SPREADING: THE COMPLEX INTERPLAY BETWEEN TH1-TH2 CYTOKINES We have described above that epitope spreading can be disease-propagating in some diseases (e.g., EAE and diabetes in the NOD mouse) but disease-regulating in others (e.g., AA). In a simplistic model, diseases that are believed to be primarily Th 1-mediated can be perpetuated by new Thl responses arising during the course of disease, whereas those dependent on a Th2 response for disease initiation can be further propagated by new Th2 responses. Similarly, considering both the regulation of an ongoing immune response and the protective aspect of epitope spreading, Thl-mediated diseases can be regulated by a Th2 response, whereas diseases characterized by a predominantly Th2 response can be controlled by a Thl response. However, in recent years, it has increasingly been realized that comprehension of the pathogenesis of autoimmune diseases on the above-mentioned paradigm of Thl-Th2 cross-regulation might be an over-simplification [146]. For example, certain cytokines that are secreted by Thl (e.g., IFN-T and TNF-ct) or Th2 (e.g., IL-4 and IL-10) cells may have dual and opposing function. Taking the example of IFN- 7 in

more detail, in a particular animal model of autoimmunity, IFN-y may be disease-inducing in one set of conditions but disease-regulating in others. It is known that IFN-y plays a critical role in induction of Thl-mediated diseases. However, the results of studies in various experimental models of autoimmunity (e.g., AA [147], EAE [148-152], collageninduced arthritis (CIA) [153, 154], diabetes in the NOD mouse [155, 156], autoimmune myocarditis [157, 158], autoimmune uveitis [159, 160], and glomerulonephritis [ 161 ]) using diverse experimental approaches (e.g., use of IFN- 7 -/- mice, IFN-TR -/- mice, and challenge of mice/rats either with IFN- 7 or with neutralizing anti-IFN-y antibodies) have revealed that IFN-y plays a regulatory role in Thl/IL-12-mediated diseases. It has been suggested that control of the activity of pathogenic T cells by IFN-y might involve one or more of the following mechanisms [156, 160, 162, 163]: suppression of proliferation of target pathogenic T cells, induction of apoptosis in pathogenic T cells, a change in the activity of APC, influence on migration of T cells into the target organ, etc. In addition to the abovementioned information from animal models, limited clinical trials conducted in patients with RA several years ago also pointed towards a beneficial effect of IFN- 7 [ 164-166]. In a contrary situation, it has been suggested that IFN-yplays an important role in mediating the tissue damage in murine SLE [ 167]; lupus is characterized by development of autoantibodies and a predominance of Th2 cytokine response. In view of the above information, both induction and regulation of autoimmune disorders would need a fresh look beyond the simple cross-regulatory function of Thl and Th2 cytokines [ 146].

6. P H Y S I O L O G I C A L SIGNIFICANCE OF EPITOPE SPREADING: I N V O L V E M E N T OF EPITOPE S P R E A D I N G IN THE PATHOGENESIS OF AN A U T O I M M U N E DISEASE.

Experimental evidence from studies in different models of autoimmune diseases supports the role of epitope spreading in the pathogenesis of the disease process. The results of these studies can be categorized into 3 functional outcomes- (i) Pathogenic epitope spreading" most of the published studies

summarized above (Table 2) describe that the new T cell responses emerging via epitope spreading are involved in progression of the initial autoimmune process, and thereby, in perpetuation and chronicity of the disease process. Some of the supportive evidence establishing the pathogenic role of epitope spreading consists of the following observations - a) the intensity and extent of epitope spreading correlates with the severity and/or duration of the disease process, and that the diseased vs. non-diseased animals differ in the pattern of responses to epitopes involved in diversification of response [3, 8, 13, 35, 36]; b) sirrfilarly, the induction of epitope spreading is associated with the frequency and/or chronicity of graft rejection [103, 104], and with complications of organ transplantation [102]; c) tolerization of the T cells specific for the antigenic determinants involved in epitope spreading and relapse of disease, and cytokine modulation can limit the progression of the disease and prevent clinical relapses [3, 8, 13, 16, 35, 36]; and d) blockade of co-stimulation at the appropriate time before or after disease induction prevents disease progression [3, 10, 11]; (ii) Protective epitope spreading: in contrast to the above, other studies provide evidence favoring a regulatory or protective role for T cell/ antibody responses comprising diversification of the initial immune response [4, 63]. The protective role of epitope spreading is evident from the findings showing that- a) pre-treatment of animals with the antigenic determinants involved in epitope spreading using an immunogenic regimen (leading to priming and expansion of antigen-specific T cells) instead of a tolerogenic regimen affords protection from disease [4]; b) passive transfer of antibodies from an animal in recovery phase of the disease into a naive syngeneic recipient leads to suppression of subsequently induced disease in the recipient [63]; and c) epitope spreading occurring during the course of tumor rejection affords protection against subsequent challenge with the same tumorigenic cell line or its antigen negative variant [107, 108]; and (iii) Epitope spreading unrelated to the dis.ease process or no spreading at all: in a couple of studies in EAE, epitope spreading either was evident but without any functional relationship with the disease process [168] or did not occur at all [169]. In another study, a clinically relapsing disease was observed in a single-TCR-transgenic mice that lack

33

all T cell specificities except the one required for initiation of the disease process [ 170]. Taken together, the above-mentioned results suggest that there is no single functional outcome that can a priori be assigned to epitope spreading, and therefore, each disease and the antigenic response associated with it needs to be examined objectively and without any preformed notion or bias. In this regard, any prediction regarding the contribution of epitope spreading to the disease process in vastly heterogeneous human population poses an important challenge for both clinical prognosis and custom designing of the therapeutic regimens.

7. IMPLICATIONS OF EPITOPE SPREADING IN IMMUNOTHERAPY OF AUTOIMMUNE DISEASES: HINDRANCE VS. FACILITATION OF THE CONTROL OF THE AUTOIMMUNE PROCESS. A great deal of effort that has been invested in developing immunotherapeutic approaches for autoimmune diseases has centered on inactivation of the potentially pathogenic T cells. It is evident from extensive studies in animal models that it is relatively easier to modulate the antigen-specific immune response for preventing the development of autoimmunity than for controlling the ongoing disease process. However, from the viewpoint of treatment of patients with autoimmune diseases, the development of effective 'therapeutic' approaches has a much higher priority over devising 'preventive' approaches. In the long run, of course, preventive approaches targeted to a potentially susceptible human population would also be very rewarding. In regard to treatment of ongoing disease, epitope spreading that is disease-propagating in nature poses a major hurdle [ 1, 3, 14]. For a successful treatment, the patient would have to be treated very early in the course of disease prior to occurrence of epitope spreading. However, it might be a rather daunting task to make a prediction about the timing of epitope spreading during the natural course of disease in individual members within a patient population. Many more clinical studies are needed to determine the timing of onset of epitope spreading during the course of MS or IDDM, for example. On the other hand, epitope spreading that is disease-regulating in

34

nature can readily be exploited for therapeutic purposes [4, 51 ]. In this situation, therapeutic regimens aimed at priming and expanding the T cells specific for the new epitopes arising during the course of disease could help expedite recovery from acute phase of the disease. In this situation, the precise timing of onset of epitope spreading might not be that much of a concern. However, in either situation, carefully planned clinical trials would be warranted to insure that the programmed modulation of the immune response delivers the expected outcome. Otherwise, strategies aimed at suppression of the disease might rather exacerbate the disease. This has been exemplified in animal models. For example, the use of two versions of the same immunomodulator (in this case, anti-B7.1 antibody) [10] or administration of the same antibody (e.g., anti-CTLA-4 antibody) at different time during the disease course [ 12] had an opposite outcome in mice with EAE.

8. CONCLUDING REMARKS Epitope spreading represents a dynamic quantitative/qualitative change in the T cell and/or antibody specificities evolving during the course of an immune response initiated by a dominant antigen/epitope associated with a pathological condition. The primary event may either be triggered experimentally or arise spontaneously. The subsequently developing new T cell and/or antibody responses then participate in perpetuation of the initial pathological damage leading to chronicity of the disease. However, depending on the disease process, spreading of response to potentially disease-regulating antigens/epitopes can be protective in nature, and therefore, epitope spreading also represents a mechanism by which initial pathological immune responses are limited to effect natural recovery from acute phase of the disease. We suggest that like many other physiological processes in the body, epitope spreading represents a snapshot of dynamic events attempting to strike a balance between the pathogenic and regulatory components of the antigen-specific T cell responses, and that the picture obtained would vary depending on the time that the responses are sampled and tested during the disease process. Study of the temporal pattern of appearance of pathogenic vs. regulatory T cell and/

or antibody responses in relation to epitope spreading would significantly advance our understanding of the pathogenesis of a variety of immune disorders, particularly autoimmune diseases. Such studies would be facilitated by application of new tools like MHC-peptide tetramers [171, 172], MHC-Ig dimers [173] and autoantibody profiling using protein microarrays [22, 174], etc. As evident from results of the studies discussed above, the implications of the phenomenon of epitope spreading cannot be generalized; instead, these need to be evaluated individually in the context of a particular disease, genetic make up of the individual, and the antigen involved. Awareness of these aspects is critical for developing appropriate immunotherapeutic approaches for immune-mediated disorders. At first, the idea of treatment of an ongoing disease in the face of epitope spreading might seem discouraging. This issue is further compounded by the observations that modulation of immune responses associated with epitope spreading could, under some circumstances, exacerbate the disease process instead of suppressing it. However, identification of the 'window' of therapeutic opportunity in terms of selecting the fight target antigen and the timing of intervention in animal models has offered hope for developing better approaches for treatment of human diseases [3, 175]. In addition, the regulatory aspects of epitope spreading could be reinforced for therapeutic advantage. We have described above examples of epitope spreading in a wide spectrum of diseases ranging from highly prevalent to relatively less prevalent diseases. In the near future, more information regarding the nature and function of epitope spreading is expected to be obtained from several infectious diseases and from diseases that are believed to possess an autoimmune component (e.g., atherosclerosis, inflammatory bowel disease, acquired immunodeficiency syndrome, etc.) [176-178]. In addition, at least three aspects of immune response are expected to shed more light on the mechanisms underlying epitope spreading. These include immune regulation, modulation of adaptive immunity by components of the innate immune response, and the host-environment interplay. At this time, there is far more information regarding pathogenic immune responses in epitope spreading than that for the regulatory aspects of the process. Bystander sup-

pression has been suggested to be the counterpart of pathogenic spreading [179]. Further integration of bystander suppression and the participation of CD4+CD25+ T cells and other regulatory T cells in the control of disease-propagating epitope spreading would significantly advance our understanding of the pathogenesis of autoimmune diseases. In regard to the role of components of innate immunity in autoimmunity, it has recently been shown that the complement and complement receptors [ 180] play a critical role in mediating effector arthritogenic response in KXBN transgenic mice [181], and in modulation of the induction of autoimmune myocarditis [182]. It is conceivable that the complement system and other mediators of innate immunity might have a significant effect on both pathogenic and regulatory epitope spreading. Finally, studies in animal models of autoimmunity [52-62] and those on the prevalence of autoimmune disorders in human populations living in different geographical regions of the world [183-188] have further validated the importance of the association between environment/infection and autoimmunity. Study of the interaction between immune response to subclinical/overt infections and autoimmune processes, and vice versa, would provide new insights into the influence of anti-microbial immunity on the induction and regulation of autoimmunity.

ACKNOWLEDGEMENTS We gratefully acknowledge the grant support from the National Institutes of Health (Bethesda, MD), the Arthritis Foundation (Atlanta, GA), the Maryland Chapter of Arthritis Foundation (Baltimore, MD), and the Maryland Arthritis Research Center (MARRC) (Baltimore, MD).

REFERENCES 1.

2.

Lehmann PV, Forsthuber T, Miller A, Sercarz EE. Spreading of T-cell autoimmunity to cryptic determinants of an autoantigen. Nature 1992;358(6382): 155-7. Sercarz EE. Immune focusing vs diversification and their connection to immune regulation. Immunol Rev 1998;164:5-10.

35

3.

Vanderlugt CL, Miller SD. Epitope spreading in immune-mediated diseases: implications for immunotherapy. Nature Rev Immunol 2002;2(2):85-95. 4. Moudgil KD, Chang TT, Eradat H, Chen AM, Gupta RS, Brahn E et al. Diversification of T cell responses to carboxy-terminal determinants within the 65-kD heatshock protein is involved in regulation of autoimmune arthritis. J Exp Med 1997;185(7):1307-16. 5. Swanborg RH. Experimental autoimmune encephalomyelitis in the rat: lessons in T-cell immunology and autoreactivity. Immunol Rev 2001;184:129-35. 6. Kuchroo VK, Anderson AC, Waldner H, Munder M, Bettelli E, Nicholson LB. T cell response in experimental autoimmune encephalomyelitis (EAE): role of self and cross-reactive antigens in shaping, tuning, and regulating the autopathogenic T cell repertoire. Annu Rev Immuno12002;20:101-23. 7. Lehmann PV, Sercarz EE, Forsthuber T, Dayan CM, Gammon G. Determinant spreading and the dynamics of the autoimmune T-cell repertoire. Immunol Today 1993;14(5):203-8. 8. McRae BL, Vanderlugt CL, Dal Canto MC, Miller SD. Functional evidence for epitope spreading in the relapsing pathology of experimental autoimmune encephalomyelitis. J Exp Med 1995;182(1):75-85. 9. Vanderlugt CL, Neville KL, Nikcevich KM, Eagar TN, Bluestone JA, Miller SD. Pathologic role and temporal appearance of newly emerging autoepitopes in relapsing experimental autoimmune encephalomyelitis. J Immuno12000; 164(2):670-8. 10. Miller SD, Vanderlugt CL, Lenschow DJ, Pope JG, Karandikar NJ, Dal Canto MC et al. Blockade of CD28/B7-1 interaction prevents epitope spreading and clinical relapses of murine EAE. Immunity 1995;3(6): 739-45. 11. Howard LM, Miga AJ, Vanderlugt CL, Dal Canto MC, Laman JD, Noelle RJ et al. Mechanisms of immunotherapeutic intervention by anti-CD40L (CD154) antibody in an animal model of multiple sclerosis. J Clin Invest 1999; 103(2):281-90. 12. Karandikar NJ, Eagar TN, Vanderlugt CL, Bluestone JA, Miller SD. CTLA-4 downregulates epitope spreading and mediates remission in relapsing experimental autoimmune encephalomyelitis. J Neuroimmunol 2000; 109(2): 173-80. 13. Yu M, Johnson JM, Tuohy VK. A predictable sequential determinant spreading cascade invariably accompanies progression of experimental autoimmune encephalomyelitis: a basis for peptide-specific therapy after onset of clinical disease. J Exp Med 1996; 183(4): 1777-88. 14. Tuohy VK, Yu M, Yin L, Mathisen PM, Johnson JM, Kawczak JA. Modulation of the IL-10/IL-12 cytokine circuit by interferon-beta inhibits the development of

36

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

epitope spreading and disease progression in murine autoimmune encephalomyelitis. J Neuroimmunol 2000; 111 (1-2):55-63. Yu M, Johnson JM, Tuohy VK. Generation of autonomously pathogenic neo-autoreactive Thl cells during the development of the determinant spreading cascade in murine autoimmune encephalomyelitis. J Neurosci Res 1996;45(4):463-70. Yin L, Yu M, Edling AE, Kawczak JA, Mathisen PM, Nanavati T et al. Pre-emptive targeting of the epitope spreading cascade with genetically modified regulatory T cells during autoimmune demyelinating disease. J Immuno12001;167(11):6105-12. Soos JM, Mujtaba MG, Schiffenbauer J, Torres BA, Johnson HM. Intramolecular epitope spreading induced by staphylococcal enterotoxin superantigen reactivation of experimental allergic encephalomyelitis. J Neuroimmuno12002; 123(1-2):30-4. Voskuhl RR, Farris RW, 2nd, Nagasato K, McFarland HE Dalcq MD. Epitope spreading occurs in active but not passive EAE induced by myelin basic protein. J Neuroimmunol 1996;70(2): 103-11. Jansson L, Diener P, Engstrom A, Olsson T, Holmdahl R. Spreading of the immune response to different myelin basic protein peptides in chronic experimental autoimmune encephalomyelitis in B 10.RIII mice. Eur J Immunol 1995;25(8):2195-200. Mor F, Cohen IR. Shifts in the epitopes of myelin basic protein recognized by Lewis rat T cells before, during, and after the induction of experimental autoimmune encephalomyelitis. J Clin Invest 1993;92(5):2199-206. Stevens DB, Gold DE Sercarz EE, Moudgil KD. The Wistar Kyoto (RTI(1)) rat is resistant to myelin basic protein-induced experimental autoimmune encephalomyelitis: comparison with the susceptible Lewis (RTI(1)) strain with regard to the MBP-directed CD4+ T cell repertoire and its regulation. J Neuroimmunol 2002; 126( 1-2):25-36. Robinson WH, Fontoura P, Lee BJ, De Vegvar HE, Tom J, Pedotti R et al. Protein microarrays guide tolerizing DNA vaccine treatment of autoimmune encephalomyelitis. Nat Biotechno12003;10:10. Karpus WJ, Pope JG, Peterson JD, Dal Canto MC, Miller SD. Inhibition of Theiler's virus-mediated demyelination by peripheral immune tolerance induction. J Immunol 1995;155(2):947-57. Miller SD, Vanderlugt CL, Begolka WS, Pao W, Yauch RL, Neville KL et al. Persistent infection with Theiler's virus leads to CNS autoimmunity via epitope spreading. Nat Med 1997;3(10):1133-6. Katz-Levy Y, Neville KL, Padilla J, Rahbe S, Begolka WS, Girvin AM et al. Temporal development of autoreactive Thl responses and endogenous presentation of

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

self myelin epitopes by central nervous system-resident APCs in Theiler's virus-infected mice. J Immunol 2000; 165(9):5304-14. Neville KL, Padilla J, Miller SD. Myelin-specific tolerance attenuates the progression of a virus-induced demyelinating disease: implications for the treatment of MS. J Neuroimmuno12002;123(1-2): 18-29. McFarland HI, Lobito AA, Johnson MM, Nyswaner JT, Frank JA, Palardy GR et al. Determinant spreading associated with demyelination in a nonhuman primate model of multiple sclerosis. J Immunol 1999; 162(4): 2384-90. Genain CP, Hauser SL. Experimental allergic encephalomyelitis in the New World monkey Callithrix jacchus. Immunol Rev 2001;183:159-72. Boon L, Brok HP, Bauer J, Ortiz-Buijsse A, Schellekens MM, Ramdien-Murli Set al. Prevention of experimental autoimmune encephalomyelitis in the common marmoset (Callithfix jacchus) using a chimeric antagonist monoclonal antibody against human CD40 is associated with altered B cell responses. J Immunol 2001 ;167(5):2942-9. Laman JD, 't Hart BA, Brok H, Meurs M, Schellekens MM, Kasran A et al. Protection of marmoset monkeys against EAE by treatment with a murine antibody blocking CD40 (mu5D 12). Eur J Immunol 2002;32(8): 2218-28. Tuohy VK, Yu M, Weinstock-Guttman B, Kinkel RP. Diversity and plasticity of self recognition during the development of multiple sclerosis. J Clin Invest 1997 ;99(7): 1682-90. Tuohy VK, Yu M, Yin L, Kawczak JA, Kinkel PR. Regression and spreading of self-recognition during the development of autoimmune demyelinating disease. J Autoimmun 1999; 13(1 ): 11-20. Tuohy VK, Yu M, Yin L, Kawczak JA, Kinkel RP. Spontaneous regression of primary autoreactivity during chronic progression of experimental autoimmune encephalomyelitis and multiple sclerosis. J Exp Med 1999;189(7): 1033--42. Ridgway WM, Fasso M, Fathman CG. A new look at MHC and autoimmune disease. Science 1999;284(5415):749,751. Kaufman DL, Clare-Salzler M, Tian J, Forsthuber T, Ting GS, Robinson P e t al. Spontaneous loss of T-cell tolerance to glutamic acid decarboxylase in murine insulin-dependent diabetes. Nature 1993;366(6450): 69-72. Tisch R, Yang XD, Singer SM, Liblau RS, Fugger L, McDevitt HO. Immune response to glutamic acid decarboxylase correlates with insulitis in non-obese diabetic mice. Nature 1993;366(6450):72-5. Tian J, Lehmann PV, Kaufman DL. Determinant spread-

38.

39.

40.

41.

42.

43.

44.

45.

46. 47.

48.

49.

50.

ing of T helper cell 2 (Th2) responses to pancreatic islet autoantigens. J Exp Med 1997;186(12):2039-43. Zechel MA, Elliott JF, Atkinson MA, Singh B. Characterization of novel T-cell epitopes on 65 kDa and 67 kDa glutamic acid decarboxylase relevant in autoimmune responses in NOD mice. J Autoimmun 1998;11(1):83-95. Bonifacio E, Scirpoli M, Kredel K, Fuchtenbusch M, Ziegler AG. Early autoantibody responses in prediabetes are IgG1 dominated and suggest antigen-specific regulation. J Immunol 1999;163(1):525-32. Bonifacio E, Lampasona V, Bernasconi L, Ziegler AG. Maturation of the humoral autoimmune response to epitopes of GAD in preclinical childhood type 1 diabetes. Diabetes 2000;49(2):202-8. Sohnlein P, Muller M, Syren K, Hartmann U, Bohm BO, Meinck HM et al. Epitope spreading and a varying but not disease-specific GAD65 antibody response in Type I diabetes. The Childhood Diabetes in Finland Study Group. Diabetologia 2000;43(2):210-7. Brooks-Worrell B, Gersuk VH, Greenbaum C, Palmer JP. Intermolecular antigen spreading occurs during the preclinical period of human type 1 diabetes. J Immunol 2001 ;166(8):5265-70. Viglietta V, Kent SC, Orban T, Hailer DA. GAD65reactive T cells are activated in patients with autoimmune type la diabetes. J Clin Invest 2002;109(7): 895-903. Wilson SB, Kent SC, Patton KT, Orban T, Jackson RA, Exley M e t al. Extreme Thl bias of invariant Valpha24JalphaQ T cells in type 1 diabetes. Nature 1998;391 (6663): 177-81. Pearson CM. Development of arthritis, periarthritis and periostitis in rats given adjuvants. Proc Soc Exp Biol Med 1956;91:95-101. Taurog JD, Argentieri DC, McReynolds RA. Adjuvant arthritis. Methods Enzymol 1988; 162:339-55. Holoshitz J, Matitiau A, Cohen IR. Arthritis induced in rats by cloned T lymphocytes responsive to mycobacteria but not to collagen type II. J Clin Invest 1984;73(1): 211-5. Van Eden W, Thole JE, van der Zee R, Noordzij A, van Embden JD, Hensen EJ et al. Cloning of the mycobacterial epitope recognized by T lymphocytes in adjuvant arthritis. Nature 1988;331(6152):171-3. Gaston JS, Life PF, Bailey LC, Bacon PA. In vitro responses to a 65-kilodalton mycobacterial protein by synovial T cells from inflammatory arthritis patients. J Immunol 1989; 143(8):2494-500. Quayle AJ, "Wilson KB, Li SG, Kjeldsen-Kragh J, Oftung F, Shinnick T et al. Peptide recognition, T cell receptor usage and HLA restriction elements of human heat-shock protein (hsp) 60 and mycobacterial 65-kDa

37

hsp- reactive T cell clones from rheumatoid synovial fluid. Eur J Immunol 1992;22(5):1315-22. 51. Moudgil KD. Diversification of response to hsp65 during the course of autoimmune arthritis is regulatory rather than pathogenic. Immunol Rev 1998;164: 175-84. 5 l a. Durai M, Gupta RS, Moudgil KD. The T cells specific for the carboxyl-terminal determinants of self (rat) heat-shock protein 65 escape tolerance induction and are involved in regulation of autoimmune arthritis. J Immunol 2004;172(5) (In press). 52. Moudgil KD, Kim E, Yun OJ, Chi HH, Brahn E, Sercarz EE. Environmental modulation of autoimmune arthritis involves the spontaneous microbial induction of T cell responses to regulatory determinants within heat shock protein 65. J Immunol 2001;166(6):4237-43. 53. Taurog JD, Richardson JA, Croft JT, Simmons WA, Zhou M, Fernandez-Sueiro JL et al. The germfree state prevents development of gut and joint inflammatory disease in HLA-B27 transgenic rats. J Exp Med 1994; 180(6):2359-64. 54. Rose NR, Saboori AM, Rasooly L, Burek CL. The role of iodine in autoimmune thyroiditis. Crit Rev Immunol 1997;17(5-6):511-7. 55. Penhale WJ, Young PR. The influence of the normal microbial flora on the susceptibility of rats to experimental autoimmune thyroiditis. Clin Exp Immunol 1988;72(2):288-92. 56. Murakami M, Nakajima K, Yamazaki K, Muraguchi T, Serikawa T, Honjo T. Effects of breeding environments on generation and activation of autoreactive B-1 cells in anti-red blood cell autoantibody transgenic mice. J Exp Med 1997;185(4):791-4. 57. Goverman J, Woods A, Larson L, Weiner LP, Hood L, Zaller DM. Transgenic mice that express a myelin basic protein-specific T cell receptor develop spontaneous autoimmunity. Cell 1993;72(4):551-60. 58. Thompson SJ, Elson CJ. Susceptibility to pristaneinduced arthritis is altered with changes in bowel flora. Immunol Lett 1993;36(2):227-31. 59. Kohashi O, Kuwata J, Umehara K, Uemura F, Takahashi T, Ozawa A. Susceptibility to adjuvant-induced arthritis among germfree, specific- pathogen-free, and conventional rats. Infect Immun 1979;26(3):791-4. 60. Benoist C, Mathis D. Autoimmunity. The pathogen connection. Nature 1998;394(6690):227-8. 61. Breban MA, Moreau MC, Fournier C, Ducluzeau R, Kahn ME Influence of the bacterial flora on collageninduced arthritis in susceptible and resistant strains of rats. Clin Exp Rheumatol 1993;11 ( 1):61-4. 62. Van den Broek MF, van Bruggen MC, Koopman JP, Hazenberg MP, van den Berg WB. Gut flora induces and maintains resistance against streptococcal cell

38

wall-induced arthritis in F344 rats. Clin Exp Immunol 1992;88(2):313-7. 62a. Singh B, Prange S, Iwnihar AM. Protective and destructive effects of microbial infection in insulin-dependent diabetes mellitus. Semin Immunol 1998; 10(1):79-86. 63. Ulmansky R, Cohen CJ, Szafer F, Moallem E, Fridlender ZG, Kashi Y e t al. Resistance to adjuvant arthritis is due to protective antibodies against heat shock protein surface epitopes and the induction of ILl0 secretion. J Immuno12002; 168(12):6463-9. 64. Prakken B J, van der Zee R, Anderton SM, van Kooten PJ, Kuis W, van Eden W. Peptide-induced nasal tolerance for a mycobacterial heat shock protein 60 T cell epitope in rats suppresses both adjuvant arthritis and nonmicrobially induced experimental arthritis. Proc Natl Acad Sci USA 1997;94(7):3284-9. 65. Taams LS, van Rensen AJ, Poelen MC, van Els CA, Besseling AC, Wagenaar JP et al. Anergic T cells actively suppress T cell responses via the antigen-presenting cell. Eur J Immunol 1998;28(9):2902-12. 66. Elson CJ, Barker RN, Thompson SJ, Williams NA. Immunologically ignorant autoreactive T cells, epitope spreading and repertoire limitation. Immunol Today 1995;16(2):71--6. 67. Alam A, Lambert N, Lule J, Coppin H, Mazieres B, de Preval C et al. Persistence of dominant T cell clones in synovial tissues during rheumatoid arthritis. J Immunol 1996; 156(9):3480-5. 68. Zandman-Goddard G, Shoenfeld Y. Novel approaches to therapy for SLE. Clin Rev Allergy Immunol 2003;25(1):105-12. 69. Via CS, Handwerger BS. B-cell and T-cell function in systemic lupus erythematosus. Curt Opin Rheumatol 1993;5(5):570-4. 70. Tsokos GC. Systemic lupus erythematosus. A disease with a complex pathogenesis. Lancet 2001;358(Suppl): $65. 70a. Datta SK. Major peptide autoepitopes for nucleosomecentered T and B cell interaction in human and murine lupus. Ann NY Acad Sci 2003;987:79-90. 70b. Hahn BH. Antibodies to DNA. N Engl J Med 1998;338: 1359-68. 71. James JA, Gross T, Scofield RH, Harley JB. Immunoglobulin epitope spreading and autoimmune disease after peptide immunization: Sm B/B'-derived PPPGMRPP and PPPGIRGP induce spliceosome autoimmunity. J Exp Med 1995;181(2):453-61. 72. Topfer F, Gordon T, McCluskey J. Intra- and intermolecular spreading of autoimmunity involving the nuclear self-antigens La (SS-B) and Ro (SS-A). Proc Natl Acad Sci USA 1995;92(3):875-9. 73. Deshmukh US, Lewis JE, Gaskin F, Kannapell CC, Waters ST, Lou YH et al. Immune responses to Ro60

and its peptides in mice. I. The nature of the irnmunogen and endogenous autoantigen determine the specificities of the induced autoantibodies. J Exp Med 1999;189(3): 531-40. 74. Deshmukh US, Lewis JE, Gaskin F, Dhakephalkar PK, Kannapell CC, Waters ST et al. Ro60 peptides induce antibodies to similar epitopes shared among lupus-related autoantigens. J Immunol 2000;164(12): 6655-61. 75. Paisansinsup T, Deshmukh US, Chowdhary VR, Luthra HS, Fu SM, David CS. HLA class II influences the immune response and antibody diversification to Ro60/ Sjrgren's syndrome-A: heightened antibody responses and epitope spreading in mice expressing HLA-DR molecules. J Immunol 2002; 168( 11):5876-84. 76. Scofield RH, Kaufman KM, Baber U, James JA, Harley JB, Kurien BT. Immunization of mice with human 60-kd Ro peptides results in epitope spreading if the peptides are highly homologous between human and mouse. Arthritis Rheum 1999;42(5): 1017-24. 77. Scofield RH, Pierce PG, James JA, Kaufman KM, Kurien BT. Immunization with peptides from 60 kDa Ro in diverse mouse strains. Scand J Immunol 2002;56(5):477-83. 77a. Kaliyaperumal A, Mohan C, Wu W, Datta SK. Nucleosomal peptide epitopes for nephritis-inducing T helper cells of murine lupus. J Exp Med 1996;183(6):245969. 77b. Datta SK, Kaliyaperumal A. Nucleosome-driven autoimmune response in lupus. Pathogenic T helper cell epitopes and costimulatory signals. Ann NY Acad Sci 1997;815:155-70. 77c. Eilat E, Dayan M, Zinger H, Mozes E. The mechanism by which a peptide based on complementarity-determining region-1 of a pathogenic anti-DNA auto-Ab ameliorates experimental systemic lupus erythematosis. Proc Natl Acad Sci USA 2001;98:1148-53. 78. Singh RR, Hahn BH. Reciprocal T-B determinant spreading develops spontaneously in murine lupus: implications for pathogenesis. Immunol Rev 1998; 164: 201-8. 79. Singh RR, Ebling FM, Sercarz EE, Hahn BH. Immune tolerance to autoantibody-derived peptides delays development of autoimmunity in murine lupus. J Clin Invest 1995;96(6):2990--6. 80. Kaliyaperumal A, Michaels MA, Datta SK. Antigenspecific therapy of murine lupus nephritis using nucleosomal peptides: tolerance spreading impairs pathogenic function of autoimmune T and B ceils. J Immunol 1999;162(10):5775-83. 81. Nakajima A, Azuma M, Kodera S, Nuriya S, Terashi A, Abe Met al. Preferential dependence of autoantibody production in murine lupus on CD86 costimulatory

molecule. Eur J Immunol 1995;25(11):3060-9. 82. Fan GC, Singh RR. Vaccination with minigenes encoding V(H)-derived major histocompatibility complex class I-binding epitopes activates cytotoxic T cells that ablate autoantibody-producing B cells and inhibit lupus. J Exp Med 2002;196(6):731--41. 83. Fisher DE, Reeves WH, Wisniewolski R, Lahita RG, Chiorazzi N. Temporal shifts from Sm to ribonucleoprotein reactivity in systemic lupus erythematosus. Arthritis Rheum 1985;28(12): 1348-55. 84. James JA, Mamula MJ, Harley JB. Sequential autoantigenic determinants of the small nuclear ribonucleoprotein Sm D shared by human lupus autoantibodies and MRL lpr/lpr antibodies. Clin Exp Immunol 1994;98(3): 419-26. 85. Neu E, Hemmerich PH, Peter HH, Krawinkel U, von Mikecz AH. Characteristic epitope recognition pattern of autoantibodies against eukaryotic ribosomal protein L7 in systemic autoimmune diseases. Arthritis Rheum 1997;40(4):661-71. 86. Arbuckle MR, Reichlin M, Harley JB, James JA. Shared early autoantibody recognition events in the development of anti-Sm B/B' in human lupus. Scand J Immunol 1999;50(5):447-55. 87. Ueki A, Isozaki Y, Tomokuni A, Hatayama T, Ueki H, Kusaka M e t al. Intramolecular epitope spreading among anti-caspase-8 autoantibodies in patients with silicosis, systemic sclerosis and systemic lupus erythematosus, as well as in healthy individuals. Clin Exp Immuno12002; 129(3):556-61. 88. Vincent A, Willcox N, Hill M, Curnow J, MacLennan C, Beeson D. Determinant spreading and immune responses to acetylcholine receptors in myasthenia gravis. Immunol Rev 1998;164:157-68. 89. Vincent A, Jacobson L, Shillito P. Response to human acetylcholine receptor alpha 138-199: determinant spreading initiates autoimmunity to self-antigen in rabbits. Immunol Lett 1994;39(3):269-75. 90. Shenoy M, Goluszko E, Christadoss P. The pathogenic role of acetylcholine receptor alpha chain epitope within alpha 146-162 in the development of experimental autoimmune myasthenia gravis in C57BL6 mice. Clin Immunol Immunopathol 1994;73(3):338-43. 91. Wang HB, Shi FD, Li H, Chambers BJ, Link H, Ljunggren HG. Anti-CTLA-4 antibody treatment triggers determinant spreading and enhances murine myasthenia gravis. J Immuno12001 ;166(10):6430-6. 92. Feferman T, Im SH, Fuchs S, Souroujon MC. Breakage of tolerance to hidden cytoplasmic epitopes of the acetylcholine receptor in experimental autoimmune myasthenia gravis. J Neuroimmunol 2003;140(1-2): 153-8. 93. Hill M, Beeson D, Moss P, Jacobson L, Bond A, Corlett

39

L e t al. Early-onset myasthenia gravis: a recurring Tcell epitope in the adult-specific acetylcholine receptor epsilon subunit presented by the susceptibility allele HLA-DR52a. Ann Neurol 1999;45(2):224-31. 94. Curnow J, Corlett L, Willcox N, Vincent A. Presentation by myoblasts of an epitope from endogenous acetylcholine receptor indicates a potential role in the spreading of the immune response. J Neuroimmunol 2001;115(1-2):127-34. 95. Chan LS, Vanderlugt CJ, Hashimoto T, Nishikawa T, Zone JJ, Black MM et al. Epitope spreading: lessons from autoimmune skin diseases. J Invest Dermatol 1998;110(2):103-9. 96. Zhu J, Pelidou SH, Deretzi G, Levi M, Mix E, van der Meide P et al. P0 glycoprotein peptides 56-71 and 180-199 dose-dependently induce acute and chronic experimental autoimmune neuritis in Lewis rats associated with epitope spreading. J Neuroimmunol 2001; 114( 1-2):99-106. 97. Deeg CA, Thurau SR, Gerhards H, Ehrenhofer M, Wildner G, Kaspers B. Uveitis in horses induced by interphotoreceptor retinoid-binding protein is similar to the spontaneous disease. Eur J Immunol 2002;32(9): 2598-606. 98. Alderuccio F, Toh BH, Tan SS, Gleeson PA, van Driel IR. An autoimmune disease with multiple molecular targets abrogated by the transgenic expression of a single autoantigen in the thymus. J Exp Med 1993;178(2): 419-26. 99. Lou Y, Tung KS. T cell peptide of a self-protein elicits autoantibody to the protein antigen. Implications for specificity and pathogenetic role of antibody in autoimmunity. J Immunol 1993;151(10):5790-9. 100. Lou YH, McElveen MF, Garza KM, Tung KS. Rapid induction of autoantibodies by endogenous ovarian antigens and activated T cells: implication in autoim. mune disease pathogenesis and B cell tolerance. J Immunol 1996; 156(9):3535--40. 101. Benichou G, Takizawa PA, Ho PT, Killion CC, Olson CA, McMillan M et al. Immunogenicity and tolerogenicity of self-major histocompatibility complex peptides. J Exp Med 1990;172(5):1341-6. 102. Ciubotariu R, Liu Z, Colovai AI, Ho E, Itescu S, Ravalli Set al. Persistent allopeptide reactivity and epitope spreading in chronic rejection of organ allografts. J Clin Invest 1998; 101 (2):398--405. 103. Liu Z, Colovai AI, Tugulea S, Reed EF, Fisher PE, Mancini D et al. Indirect recognition of donor HLADR peptides in organ allograft rejection. J Clin Invest 1996;98(5): 1150-7. 104. Molajoni ER, Cinti P, Orlandini A, Molajoni J, Tugulea S, Ho E et al. Mechanism of liver allograft rejection: the indirect recognition pathway. Hum Immunol

40

1997;53(1):57-63. 105. Tian J, Clare-Salzler M, Herschenfeld A, Middleton B, Newman D, Mueller R et al. Modulating autoimmune responses to GAD inhibits disease progression and prolongs islet graft survival in diabetes-prone mice. Nat Med 1996;2(12): 1348-53. 106. Yang L, DuTemple B, Gorczynski RM, Levy G, Zhang L. Evidence for epitope spreading and active suppression in skin graft tolerance after donor-specific transfusion. Transplantation 1999;67(11): 1404-10. 107. el-Shami K, Tirosh B, Bar-Haim E, Carmon L, Vadai E, Fridkin Met al. MHC class I-restricted epitope spreading in the context of tumor rejection following vaccination with a single irmnunodominant CTL epitope. Eur J Immunol 1999;29(10):3295-301. 108. Markiewicz MA, Fallarino F, Ashikari A, Gajewski TF. Epitope spreading upon P815 tumor rejection triggered by vaccination with the single class I MHC-restricted peptide P1A. Int Immunol 2001;13(5):625-32. 109. Disis ML, Grabstein KH, Sleath PR, Cheever MA. Generation of immunity to the HER-2/neu oncogenic protein in patients with breast and ovarian cancer using a peptide-based vaccine. Clin Cancer Res 1999;5(6): 1289-97. 110. Brossart P, Wirths S, Stuhler G, Reichardt VL, Kanz L, Brugger W. Induction of cytotoxic T-lymphocyte responses in vivo after vaccinations with peptide-pulsed dendritic cells. Blood 2000;96(9):3102-8. 111. van der Most RG, Sette A, Oseroff C, Alexander J, Murali-Krishna K, Lau LL et al. Analysis of cytotoxic T cell responses to dominant and subdominant epitopes during acute and chronic lymphocytic choriomeningitis virus infection. J Immunol 1996;157(12):5543-54. 112. Gallimore A, Hengartner H, Zinkernagel R. Hierarchies of antigen-specific cytotoxic T-cell responses. Immunol Rev 1998;164:29-36. 113. Butz EA, Bevan MJ. Massive expansion of antigenspecific CD8+ T cells during an acute virus infection. Immunity 1998;8(2):167-75. 114. Gross D, Huber BT, Steere AC. Molecular mimicry and Lyme arthritis. Curr Dir Autoimmun 2001 ;3:94-111. 115. Ehrenstein M, Isenberg D. Autoimmunity associated with infection: leprosy, acute rheumatic fever and Lyme disease. Curr Opin Immunol 1991 ;3(6):930--5. 116. Hinkula J, Svanholm C, Schwartz S, Lundholm P, Brytting M, Engstrom G e t al. Recognition of prominent viral epitopes induced by immunization with human immunodeficiency virus type 1 regulatory genes. J Virol 1997;71(7):5528-39. 117. Sercarz EE, Lehmann PV, Ametani A, Benichou G, Miller A, Moudgil K. Dominance and crypticity of T cell antigenic determinants. Annu Rev Immunol 1993;11:729-66.

118. Cibotti R, Kanellopoulos JM, Cabaniols JP, Halle-Panenko O, Kosmatopoulos K, Sercarz E et al. Tolerance to a self-protein involves its immunodominant but does not involve its subdominant determinants. Proc Natl Acad Sci USA 1992;89( 1):416-20. 119. Moudgil KD, Sercarz EE. Dominant determinants in hen eggwhite lysozyme correspond to the cryptic determinants within its self-homologue, mouse lysozyme: implications in shaping of the T cell repertoire and autoimmunity. J Exp Med 1993;178(6):2131-8. 120. Mamula MJ. The inability to process a self-peptide allows autoreactive T cells to escape tolerance. J Exp Med 1993;177(2):567-71. 121. Anderton SM, Viner NJ, Matharu P, Lowrey PA, Wraith DC. Influence of a dominant cryptic epitope on autoimmune T cell tolerance. Nat Immunol 2002;3(2): 175-81. 122. Lanzavecchia A. How can cryptic epitopes trigger autoimmunity? J Exp Med 1995;181(6):1945-8. 123. Di Rosa F, Bamaba V. Persisting viruses and chronic inflammation: understanding their relation to autoimmunity. Immunol Rev 1998;164:17-27. 124. Bottazzo GF, Pujol-Borrell R, Hanafusa T, Feldmann M. Role of aberrant HLA-DR expression and antigen presentation in induction of endocrine autoimmunity. Lancet 1983;2(8359):1115-9. 125. Sarvetnick N, Shizuru J, Liggitt D, Martin L, McIntyre B, Gregory A et al. Loss of pancreatic islet tolerance induced by beta-cell expression of interferon-gamma. Nature 1990;346(6287):844--7. 126. Oldstone MB. Molecular mimicry as a mechanism for the cause and a probe uncovering etiologic agent(s) of autoimmune disease. Curr Top Microbiol Immunol 1989;145:127-35. 127. Wucherpfennig KW, Strominger JL. Molecular mimicry in T cell-mediated autoimmunity: viral peptides activate human T cell clones specific for myelin basic protein. Cell 1995;80(5):695-705. 128. Zhao ZS, Granucci F, Yeh L, Schaffer PA, Cantor H. Molecular mimicry by herpes simplex virus-type 1: autoimmune disease after viral infection. Science 1998;279(5355): 1344-7. 129. Soloski MJ, Metcalf ES. The involvement of class Ib molecules in the host response to infection with Salmonella and its relevance to autoimmunity. Microbes Infect 2001;3(14-15):1249-59. 130. Rose NR. Viral damage or 'molecular mimicry'-placing the blame in myocarditis. Nat Med 2000;6(6):631-2. 131. Fairweather D, Rose NR. Type 1 diabetes: virus infection or autoimmune disease? Nat Immunol 2002;3(4): 338-40. 132. Rose NR, Hill SL. Autoimmune myocarditis. Int J Cardiol 1996;54(2):171-5.

133. Hill SL, Rose NR. The transition from viral to autoimmune myocarditis. Autoimmunity 2001 ;34(3): 169-76. 134. Horwitz MS, Bradley LM, Harbertson J, Krahl T, Lee J, Sarvetnick N. Diabetes induced by Coxsackie virus: initiation by bystander damage and not molecular mimicry. Nat Med 1998;4(7):781-5. 135. Voskuhl RR. Myelin protein expression in lymphoid tissues: implications for peripheral tolerance. Immunol Rev 1998;164:81-92. 136. Flodstrom M, Maday A, Balakrishna D, Cleary MM, Yoshimura A, Sarvetnick N. Target cell defense prevents the development of diabetes after viral infection. Nat Immunol 2002;3(4):373-82. 137. Casciola-Rosen LA, Anhalt G, Rosen A. Autoantigens targeted in systemic lupus erythematosus are clustered in two populations of surface structures on apoptotic keratinocytes. J Exp Med 1994; 179(4): 1317-30. 138. Casciola-Rosen L, Rosen A. Ultraviolet light-induced keratinocyte apoptosis: a potential mechanism for the induction of skin lesions and autoantibody production in LE. Lupus 1997;6(2):175-80. 139. Anderton SM, Wraith DC. Hierarchy in the ability of T cell epitopes to induce peripheral tolerance to antigens from myelin. Eur J Immunol 1998;28(4): 1251--61. 140. Harrington CJ, Paez A, Hunkapiller T, Mannikko V, Brabb T, Aheam M et al. Differential tolerance is induced in T cells recognizing distinct epitopes of myelin basic protein. Immunity 1998;8(5):571-80. 141. Tian J, Gregori S, Adorini L, Kaufman DL. The frequency of high avidity T cells determines the hierarchy of determinant spreading. J Immunol 2001; 166(12): 7144--50. 142. Mamula MJ, Janeway CA Jr. Do B cells drive the diversification of immune responses? Immunol Today 1993;14(4):151-2; discussion 153-4. 143. Mamula MJ, Fatenejad S, Craft J. B cells process and present lupus autoantigens that initiate autoimmune T cell responses. J Immunol 1994;152(3): 1453--61. 144. Shlomchik MJ, Craft JE, Mamula MJ. From T to B and back again: positive feedback in systemic autoimmune disease. Nat Rev Immunol 2001;1(2):147-53. 145. Simitsek PD, Campbell DG, Lanzavecchia A, Fairweather N, Watts C. Modulation of antigen processing by bound antibodies can boost or suppress class II major histocompatibility complex presentation of different T cell determinants. J Exp Med 1995;181(6): 1957-63. 146. Gor DO, Rose NR, Greenspan NS. TH1-TH2: a procrustean paradigm. Nat Immuno12003;4(6):503-5. 147. Jacob CO, Holoshitz J, Van der Meide P, Strober S, McDevitt HO. Heterogeneous effects of IFN-gamma in adjuvant arthritis. J Immunol 1989; 142(5): 1500-5. 148. Billiau A, Heremans H, Vandekerckhove F, Dijkmans

41

R, Sobis H, Meulepas E et al. Enhancement of experimental allergic encephalomyelitis in mice by antibodies against IFN-gamma. J Immunol 1988;140(5):1506-10. 149. Ferber IA, Brocke S, Taylor-Edwards C, Ridgway W, Dinisco C, Steinman L et al. Mice with a disrupted IFN-gamma gene are susceptible to the induction of experimental autoimmune encephalomyelitis (EAE). J Immunol 1996;156(1):5-7. 150. Willenborg DO, Fordham S, Bernard CC, Cowden WB, Ramshaw IA. IFN-gamma plays a critical down-regulatory role in the induction and effector phase of myelin oligodendrocyte glycoprotein-induced autoimmune encephalomyelitis. J Immunol 1996; 157(8):3223-7. 151. Krakowski M, Owens T. Interferon-gamma confers resistance to experimental allergic encephalomyelitis. Eur J Immunol 1996;26(7): 1641-6. 152. Kumar V, Sercarz E. Induction or protection from experimental autoimmune encephalomyelitis depends on the cytokine secretion profile of TCR peptide-specific regulatory CD4 T cells. J Immunol 1998;161(12): 6585-91. 153. Vermeire K, Heremans H, Vandeputte M, Huang S, Billiau A, Matthys P. Accelerated collagen-induced arthritis in IFN-gamma receptor-deficient mice. J Immunol 1997; 158( 11 ):5507-13. 154. Manoury-Schwartz B, Chiocchia G, Bessis N, Abehsira-Amar O, Batteux F, Muller Set al. High susceptibility to collagen-induced arthritis in mice lacking IFNgamma receptors. J Immunol 1997;158(11):5501-6. 155. Quinn A, Mclnerney B, Reich EP, Kim O, Jensen KP, Sercarz EE. Regulatory and effector CD4 T cells in nonobese diabetic mice recognize overlapping determinants on glutamic acid decarboxylase and use distinct V beta genes. J Immuno12001 ;166(5):2982-91. 156. Trembleau S, Penna G, Gregori S, Giarratana N, Adorini L. IL-12 Administration accelerates autoimmune diabetes in both wild-type and IFN-gammadeficient nonobese diabetic mice, revealing pathogenic and protective effects of IL-12-induced IFN-gamma. J Immunol 2003; 170(11):5491-501. 157. Eriksson U, Kurrer MO, Bingisser R, Eugster HP, Saremaslani P, Follath F et al. Lethal autoimmune myocarditis in interferon-gamma receptor-deficient mice: enhanced disease severity by impaired inducible nitric oxide synthase induction. Circulation 2001 ;103(1): 18-21. 158. Afanasyeva M, Wang Y, Kaya Z, Stafford EA, Dohmen KM, Sadighi Akha AA et al. Interleukin-12 receptor/STAT4 signaling is required for the development of autoimmune myocarditis in mice by an interferon-gamma-independent pathway. Circulation 2001; 104(25):3145-51. 159. Caspi RR, Chan CC, Grubbs BG, Silver PB, Wiggert B,

42

Parsa CF et al. Endogenous systemic IFN-gamma has a protective role against ocular autoimmunity in mice. J Immunol 1994;152(2):890-9. 160. Tarrant TK, Silver PB, Wahlsten JL, Rizzo LV, Chan CC, Wiggert B et al. Intefleukin 12 protects from a T helper type 1-mediated autoimmune disease, experimental autoimmune uveitis, through a mechanism involving interferon gamma, nitric oxide, and apoptosis. J Exp Med 1999;189(2):219-30. 161. Ring GH, Dai Z, Saleem S, Baddoura FK, Lakkis FG. Increased susceptibility to immunologically mediated glomerulonephritis in IFN-gamma-deficient mice. J Immunol 1999; 163(4):2243-8. 162. Furlan R, Brambilla E, Ruffini F, Poliani PL, Bergami A, Marconi PC et al. Intrathecal delivery of IFNgamma protects C57BL/6 mice from chronic-progressive experimental autoimmune encephalomyelitis by increasing apoptosis of central nervous system-infiltrating lymphocytes. J Immuno12001;167(3):1821-9. 163. Madakamutil LT, Maricic I, Sercarz E, Kumar V. Regulatory T cells control autoirranunity in vivo by inducing apoptotic depletion of activated pathogenic lymphocytes. J Immunol 2003; 170(6):2985-92. 164. Veys EM, Mielants H, Verbruggen G, Grosclaude JP, Meyer W, Galazka A e t al. Interferon gamma in rheumatoid arthritis - a double blind study comparing human recombinant interferon gamma with placebo. J Rheumatol 1988; 15(4):570-4. 165. Lemmel EM, Brackertz D, Franke M, Gaus W, Hartl PW, Machalke K et al. Results of a multicenter placebocontrolled double-blind randomized phase HI clinical study of treatment of rheumatoid arthritis with recombinant interferon-gamma. Rheumatol Int 1988;8(2): 87-93. 166. Cannon GW, Pincus SH, Emkey RD, Denes A, Cohen SA, Wolfe F et al. Double-blind trial of recombinant gamma-interferon versus placebo in the treatment of rheumatoid arthritis. Arthritis Rheum 1989;32(8): 964-73. 167. Peng SL, Moslehi J, Craft J. Roles of interferongamma and interleukin-4 in murine lupus. J Clin Invest 1997;99(8): 1936--46. 168. Kumar V. Determinant spreading during experimental autoimmune encephalomyelitis: is it potentiating, protecting or participating in the disease? Immunol Rev 1998;164:73-80. 169. Takacs K, Altmann DM. The case against epitope spread in experimental allergic encephalomyelitis. Immunol Rev 1998;164:101-10. 170. Jones RE, Bourdette D, Moes N, Vandenbark A, Zamora A, Offner H. Epitope spreading is not required for relapses in experimental autoimmune encephalomyelitis. J Immuno12003;170(4):1690-8.

171. Kotzin BL, Falta MT, Crawford F, Rosloniec EE Bill J, Marrack Pet al. Use of soluble peptide-DR4 tetramers to detect synovial T cells specific for cartilage antigens in patients with rheumatoid arthritis. Proc Natl Acad Sci USA 2000;97(1):291--6. 172. Trollmo C, Meyer AL, Steere AC, Hailer DA, Huber BT. Molecular mimicry in Lyme arthritis demonstrated at the single cell level: LFA-1 alpha L is a partial agonist for outer surface protein A-reactive T cellS. J Immunol 2001 ;166(8):5286-91. 173. Fahmy TM, Bieler JG, Schneck JP. Probing T cell membrane organization using dimeric MHC-Ig complexes. J Immunol Methods 2002;268(1):93-106. 174. Hueber W, Utz PJ, Steinman L, Robinson WH. Autoantibody profiling for the study and treatment of autoimmune disease. Arthritis Res 2002;4(5):290-5. 175. Steinman L. Despite epitope spreading in the pathogenesis of autoimmune disease, highly restricted approaches to immune therapy may still succeed (with a hedge on this bet). J Autoimmun 2000; 14(4):278-82. 176. Gordon PA, George J, Khamashta MA, Harats D, Hughes G, Shoenfeld Y. Atherosclerosis and autoimmunity. Lupus 2001 ;10(4):249-52. 177. Bouma G, Strober W. The immunological and genetic basis of inflammatory bowel disease. Nat Rev Immunol 2003;3(7):521-33. 178. Zandman-Goddard G, Shoenfeld Y. HIV and autoimmunity. Autoimmun Rev 2002; 1(6):329-37. 179. Bach JF, Koutouzov S, van Endert PM. Are there unique autoantigens triggering autoimmune diseases? Immunol Rev 1998;164:139-55.

180. Carroll MC. The role of complement and complement receptors in induction and regulation of immunity. Annu Rev Immunol 1998;16:545-68. 181. Ji H, Ohmura K, Mahmood U, Lee DM, Hofhuis FM, Boackle SA et al. Arthritis critically dependent on innate immune system players. Immunity 2002;16(2): 157-68. 182. Kaya Z, Afanasyeva M, Wang Y, Dohmen KM, Schlichting J, Tretter T et al. Contribution of the innate immune system to autoimmune myocarditis: a role for complement. Nat Immuno12001;2(8):739-45. 183. Bulman DE, Sadovnick AD, Ebers GC. Age of onset in siblings concordant for multiple sclerosis. Brain 1991;114(Pt 2):937-50. 184. Ewing C, Bernard CC. Insights into the aetiology and pathogenesis of multiple sclerosis. Immunol Cell Biol 1998;76(1):47-54. 185. Rouse BT. Virus-induced immunopathology. Adv Virus Res 1996;47:353-76. 186. Wilson C, Tiwana H, Ebringer A. Molecular mimicry between HLA-DR alleles associated with rheumatoid arthritis and Proteus mirabilis as the Aetiological basis for autoimmunity. Microbes Infect 2000;2(12): 1489-96. 187. Singh B, Rabinovitch A. Influence of microbial agents on the development and prevention of autoimmune diabetes. Autoimmunity 1993; 15(3):209-13. 188. Rose NR. The role of infection in the pathogenesis of autoimmune disease. Semin Immunol 1998;10(1): 5-13.

43

9 2004 Elsevier B. V All rights resen,ed. Infection and Autoimmunity Y. Shoenfeld and N.R. Rose, editors

Molecular Mimicry in Multiple Sclerosis: Role of MHC-Altered Peptide Ligands (MAPL) Dong-Gyun Lim and David A. Hater

Laboratory of Molecular Immunology, Centerfor Neurologic Diseases, Brigham and Women's Hospital and Harvard Medical School Boston, MA, USA

Multiple sclerosis (MS) is a chronic inflammatory illness affecting the CNS white matter that can lead to progressive neurologic dysfunction. Together with other organ-specific autoimmune diseases such as type 1 diabetes mellitus and rheumatoid arthritis, MS is thought to be mediated by autoreactive T cells that recognize CNS self-antigens. Supporting evidence for the autoimmune basis of MS includes the inflammatory nature of the CNS lesions, the genetic linkage to the MHC region, similarities to the animal model experimental autoimmune encephalomyelitis (EAE), and the therapeutic effects of immunomodulatory drugs [1]. Data from animal studies in the EAE model established that CD4 +Thl cells specific to myelin antigens can play a central role in the induction and progression of autoimmune demyelinating disease [2, 3]. This finding has fueled efforts to dissect the antigenic specificity and functional characteristics of myelin-reactive CD4 + T cells in the peripheral blood and cerebrospinal fluid of patients with MS. Several peptide epitopes derived from myelin proteins have been found to activate CD4 + T cells in the circulation of patients with MS. Among them, MBP (myelin basic protein) 84-102 and MBP148-162 have been classified as immunodominant epitopes in the context of HLA-DR2 haplotype [4-6]. However, CD4 + T cells reactive to these self-peptides have also been detected in healthy persons, suggesting that the presence of myelinreactive T cells is not sufficient for the development of MS. While resting autoreactive T cells are a part of normal T cell repertoire, the activation status of these cells is different in patients than in healthy

individuals. Consistent with the properties of preactivated T cells, myelin-reactive T cells recovered from patients are less dependent on costimulation for activation and express activation markers, such as IL-2R, on their cell surfaces [7-9]. These findings lead to the question of what induces the activation of myelin-reactive T lymphocytes in patients with MS. One attractive hypothesis is based on the role of microbial infections in the activation of selfreactive T cells. If invading microorganisms contain protein antigens with sufficient structural homology to human proteins, infection itself may be sufficient to activate pre-existing self-reactive T cells. This is the basic concept of molecular mimicry. Several clinical and epidemiological studies support the hypothesis of mimicry as a mechanism of autoimmunity. According to the original observation by Fujinami and Oldstone [10], rabbits immunized with a hepatitis B polymerase peptide that shared six amino acids with the MBP sequence developed CNS lesions reminiscent of EAE. In humans, upper respiratory infections often precede MS exacerbations. The original concept of molecular mimicry evolved from the recent understanding of the degenerative recognition of antigens by the T cell receptor (TCR). In examining the immune response to MBP, we found that complementary mutations in an antigenic peptide allow for cross-reactivity of autoreacfive T cell clones that may be related to shifts of the TCR structure itself [11]. Using combinatorial libraries, Hemmer and co-workers [12] demonstrated that totally unrelated peptides also activate autoreactive T cells. Furthermore, recent data suggest that even different MHC molecules complexed

45

with diverse peptides can activate MBP-reactive T lymphocytes in an agonistic or antagonistic fashion. In the following sections, we will describe examples of this broader concept of molecular mimicry and its clinical relevance in MS.

reveal signs of superantigenic activation of myelinreactive T cells in MS. Thus, molecular mimicry is still a very attractive theory to explain the frequent association of microbial infection with the development or exacerbation of MS.

1. RELATIONSHIP BETWEEN MICROBIAL INFECTIONS AND MS

2. M O L E C U L A R MIMICRY ASSOCIATED WITH SEQUENCE H O M O L O G Y

Since 1894, when Pierre Marie proposed that infection is the cause of MS, many reports have supported the possible involvement of infectious pathogens. Epidemiological studies have formulated a list of suspicious microbes, including Borrelia burgdorferi, Chlamydia pneumoniae, measles virus, rabies virus, paramyxovirus, coronavirus, EpsteinBarr virus, cytomegalovirus, varicella-zoster virus, herpes simplex virus, human herpes virus 6, rubella virus, mumps virus, Marek's disease virus, Semiliki Forest virus, human retroviruses, and human lymphoma virus type I [13]. So far, none of these infectious agents have been found to be specific for MS, although an MS-specific agent may yet be discovered. The apparent absence of an MS-specific infectious agent and the autoimmune nature of MS suggest another role of infection in the development of this disease. Three mechanisms have been proposed to explain the association of microbial infection and MS. One is the molecular mimicry theory, which has received a great deal of attention and will be discussed in detail in the following sections. Another is bystander activation, including epitope spreading [ 14]. Infections can activate autoreactive T cells through the release of sequestered myelin proteins as a result of infection-related tissue damage, activation of antigen-presenting cells (APCs), and induction of secretion of inflammatory cytokines and chemokines, irrespective of the particular microbial determinants. The third is superantigenic T cell activation. Several bacterial and viral products are able to cross-link TCR and MHC molecules independent of specific antigen recognition through the TCR. Cross-linking leads to activation of T cells with particular V[3 families of TCR. Myelin-reactive T cells with a particular TCR VI3 chain can be activated after infection with microbes whose superantigen recognizes this specific VI3 chain. Again, many studies have failed to

Molecular mimicry describes a situation whereby a foreign antigen can initiate an immune response in which a T or B cell component cross-recognizes self. The previous concept for the antigen specificity of T cells predicted that the presence of strict sequence homology between the microbial antigens and self-peptide was necessary to induce autoimmunity (Fig. 1, top A). There are two approaches to determining the foreign microbial antigens that are cross-reactive to self-antigens. One is the initial identification of causative microbes and their major antigenic epitopes and subsequent demonstration of their cross-reactivity to self-antigens. This approach was applied successfully to reveal that two autoimmune diseases are caused by a molecular mimicry mechanism. Autoimmune Lyme arthritis is preceded by infection with B. burgdorferi and is associated with hLFA-1-reactive T cells that were initially primed and expanded by the OspA(165-173) peptide of this spirochete [15]. Herpetic stromal keratitis (HSK) follows infection of the eye with herpes simplex virus 1 and is caused by the activation of autoreactive cells by viral UL6(299-314) peptide [16]. However, it is difficult to apply this strategy for the study of molecular mimicry in MS because, as mentioned previously, no microbial agent has been directly associated with the disease. The other approach is to first identify T cell determinants capable of inducing autoimmunity and then search for the microbial antigens with sequences homologous to those determinants. An initial study adapting this latter strategy successfully demonstrated that a peptide from hepatitis B virus polymerase (HBVP) with six consecutive amino acids in common with an encephalitogenic determinant of MBP (ICGYGSLPQE in HBVP vs TTHYGSLPQK in MBP) could induce subclinical EAE in rabbits [10]. However, proteins sharing a sequence of more than six amino acids are not

46

Molecular mimicry

Cognate MHC/peptide ligand

A. Extensive sequence homology

I

L

l

1

i

!

TCR

1,

I

B. Minimal sequence homology

I

L

1 L,I

C. Entirely unrelated peptide

J..........

[

I,,

Y

Variant molecular mimicry A. Different MHC/same peptide

B. Different MHC/different peptide

L

I Tc R

!

[

I

!

!

I

Figure 1. Schematic representation of molecular mimicry. Top: Classical molecular mimicry occurs by the sequence homology between self and microbial peptide (A). Expanded molecular mimicry can occur either by peptides with minimal sequence homology (B) or even by entirely unrelated peptides complexed with cognate MHC molecule (C). Bottom: Variant molecular mimicry means the cross-recognition of a different MHC/same (A) or different peptide (B) by the identical TCR. common in nature, and this type of homologous foreign peptides has not yet been identified in human MS. Extensive biochemical and structural characterization of the recognition of MHC/peptide ligand by TCR in humans has greatly influenced

the identification of cross-reactive epitopes for the given TCR. While a few amino acid residues of a peptide are important for binding to the MHC molecule, the other one or two amino acid residues serve as critical residues for the recognition by the TCR (Fig. 1, top B). Specifically, in the immunodo-

47

minant peptide ligand (MBP85-99) presented by DRB* 1501, two hydrophobic residues (Val-89 and Phe-92) serve as the primary anchors to the HLADR2 molecule, while Phe-91 and Lys-93 have been defined as the primary TCR contact residues for a MBP85-99-specific T cell clone. Mutations in other amino acid residues are tolerated with respect to recognition by the TCR [17]. On the basis of this information, Wucherpfennig and Strominger [18] developed an efficient strategy to search for peptides that share these critical contact motives rather than sequence homology. This search yielded seven viral and one bacterial peptides derived from herpes simplex virus, adenovirus, human papillomavirus, Epstein-Barr virus, influenza type A virus, reovirus type 3, and Pseudomonas that efficiently activated T cell clones. Interestingly, only one of them has been identified as a molecular mimic by sequence alignment. These data clearly indicated that more foreign peptides may act as molecular mimics than was previously thought.

3. M O L E C U L A R MIMICRY W I T H O U T SEQUENCE H O M O L O G Y Conservation of critical MHC and TCR contact residues appears to be a general rule for activation by the peptide of a specific T cell. However, several peptide sequences with no sequence homology have been shown to be able to activate identical T cells (Fig. 1, top C) [19, 20]. In addition, the use of synthetic peptide combinatorial libraries has clearly illustrated the extreme degeneracy of TCR recognition of antigen. Hemmer et al [12] examined the response of CD4 + T cell clones specific for MBP86-96 to a set of 220 11-mer peptide sublibraries, each containing 10 degenerated amino acids and one defined amino acid in the positional scanning format. The new knowledge obtained from this study was 1) that of highly degenerative recognition of peptides by autoreactive CD4 § T cells, including identification of stimulatory ligands not sharing a single amino acid in corresponding positions with the antigen used to establish the T cell clone and 2) the identification of more potent agonistic peptides than cognate self-peptide. From the database search for natural proteins with deduced peptide sequences, they found four self and three microbial proteins

48

as stimulatory ligands to this T cell clone. Most interestingly, one self-peptide (protein-glutamine gamma-glutamyltransferase, 675-685) and two microbial peptides (human CMV UL71, 166-176; UDP-N-acetylenolpyruvoyl-glucosamine reductase of Salmonella typhimurium, 227-237) were defined as even more potent agonists than MBP86-96. This strategy opened the way for the identification of molecular mimics if relevant autoreactive T cells are defined in MS.

4. VARIANT M O L E C U L A R MIMICRY: MHC-ALTERED PEPTIDE LIGAND The studies discussed above focused on the molecular mimics with respect to the peptide portion of TCR ligand. Since TCRs recognize MHC/peptide as a single unit and exhibit greatly degenerative recognition of their ligands, they might recognize a different MHC combined with a cognate or even a different peptide as an agonist (Fig. 1, bottom). Since most people are heterozygous for the HLADR locus, this type of molecular mimicry is likely to occur in physiological situations. It has been shown that the human MBP peptide 84-102 binds to several different DR molecules and that T cells recognize this peptide presented by these different HLADR2 molecules [21]. We systematically exarpdned this type of cross-reaction using a panel of CD8 § T cell clones specific to Taxi 1-19 peptide of HTLV-1 in the context of HLA-A*0201 [22]. When CD8 § T cells were stimulated with cognate Taxi 1-19 peptide presented by different HLA-A2 subtype alleles, which have one to four amino acid differences at specific positions (Table 1), they showed a diverse pattern of T cell function, encompassing agonistic, weak agonistic, or partial agonistic, depending on the individual T cell clones (Fig. 2). This is similar to the effects induced by antigenic altered peptide ligands (APL). In addition, atypical partial agonistic T cell function was observed; i.e., a number of CD8 + T cell clones proliferated in response to Taxi 1-19 ~ presented by the HLA-A*0205 subtype even though they did not exhibit any cytotoxic activity. The analysis of the structural interaction between the TCR and MHC/peptide complex indicated that polymorphic amino acids in the HLA-A2 peptide-binding groove, especially the D-pocket, rather than the dif-

Table 1. Summary of amino acid sequences at polymor-

phic HLA-a2 positionsa HLA-A2 subtype

Amino acid sequence at specific position .

o~1 domain

.

.

.

.

.

t~2 domain

9

43

95

149

152

156

A*0201

F

Q

V

A

V

L

A*0202

F

R

L

A

V

W

A*0203

F

Q

V

T

E

W

A*0205

Y

R

L

A

V

W

A*0206

Y

Q

V

A

V

L

aReproduced from Ref. [22] by copyright permission of the Rockefeller University Press. ferences on the MHC residues in direct contact with the TCR, were responsible for this partial agonism (Fig. 3A). This was supported by the finding that reciprocal mutations of the Tax peptide side chain that engaged the D-pocket restored the agonist functions of the MHC/peptide complex (Fig. 3B). Thus, our study clearly demonstrated that MHC molecules play an important role in T cell activation not only by restricting antigen element but also by altering the antigenic nature of peptide. We termed this type of TCR ligand MHC-altered peptide ligand (MAPL). CD4 § T cell can also exhibit MAPL effects on the recognition of variant MHC class II molecules. Germain and colleagues [23] previously suggested that a peptide presented on a mutant MHC class II molecule induces a response different from the response induced with the native MHC molecule. We also recently examined the modulation of T cell responses induced by stimulation with different MHC class II/peptide complexes, demonstrating the existence of significant cross-reactivity of autoreactive CD4 + T cells in the context of the distinct MHC class II molecules. The Ob 1A12 T cell clone is reactive to the immunodominant MBP peptide, MBP85-99, presented by HLA-DRA/DRB*1501 and was generated from peripheral blood mononuclear cells of a patient with MS [24]. Characterization of this T cell clone showed that it recognizes several single amino acid substitutes of MBP85-99 presented by DRB 1" 1501 in a degenerative fashion.

Moreover, when the responses of this T cell to APLs in the context of the other self-DR allele (DRM DRBI*0401) were examined, the OblA12 T cell clone responded to the MBP85-99 88V--->K APL, even though it did not recognize the MBP85-99 presented by this allele (Fig. 4, [25]). These data indicate that APL-induced cross-reactivity provides a further degree of T cell receptor degeneracy among different DR molecules. It should be noted that the functional outcome of this type of crossreaction could be agonistic, partial agonistic, or even antagonistic depending on the specific combination of MHC molecules and peptides. The cross-reaction induced by MAPLs containing cognate, altered, or even irrelevant peptides has not been well appreciated as a potential mechanism of molecular mimicry. However, our data generated from the systemic analyses in vitro, as mentioned above, and the increasing awareness of the highly degenerative nature of TCR recognition strongly predict that this type of molecular mimicry might effect the activation of autoreactive T cells. In fact, a recent report provides a good example for this type of molecular mimicry by showing that T cells derived from a patients with MS recognized an Epstein-Barr virus DNA polymerase peptide in the context of DRB5*0101 as well as the immunodominant MBP85-99 epitope in the context of DRB 1" 1501 [26]. The structural similarity in TCR contact surfaces between these two MHC/peptide complexes explains this cross-reactivity. If this type of molecular mimicry is frequent, the chances for autoreactive T cell activation would be higher than previously thought, while the identification of mimicry peptides would be more complicated. This means that we need to consider both the peptide epitopes and MHC class II alleles to carry out an exhaustive study of the molecular mimics.

5. THE RELEVANCE OF M O L E C U L A R MIMICRY IN MS It is important to define whether molecular mimicry induces the initiation of autoimmune responses or contributes only to the exacerbation of existing autoimmune responses. Unfortunately, no convincing evidence is currently available for a role of molecular mimicry in the initiation of any autoim-

49

A

KS2E 11.7

TP60

TP7

-~ _~ _~ -~ _g -z~

- ; -8 -7 - ; -5 _z~

-9 -8 -7 -6 -5 -4

TP41

TP27

TP49

8O 60 40 20 .....

0

~8o .....

--

A*0201 A*0202

& "

A*0203

."

"

r

A*0205 " A*0206

60 40 20

CO

o~

__ I

-9 -8 -7 -6 -5 -4

=IP

"I"

I-

9

,,

=I"

"IP

-9 -8 -7 -6 -5 -4

, I

I

I

I

,, I

I

-9 -8 -7 -6 -5 -4

,,

Log molar peptide

v

B Cytotoxicity

~

% specific lysis 9 >50

Proliferation

S.I. " > 2 0

IFN-T secretion

pg/ml 9 > 1,000

.=,

,,,

~

% specific lysis 9 10-50

S.I. " 2 - 2 0

pg/ml" 1 0 0 - 1,000

.

% specific lysis 9 ~

60000

40000 20000 ,

,

0 e-

,

,

,

,

,

,

,

i

,

,

,

,

co

B

ooooi_/.a _/amm=Hi DRB1*1501

120000

E=. 80000 u 40000

0 .

.

.

m

.

,""

~

OO

GO

OO

GO

OO

o

OO

OO

= = '0'0 OO GO

OO OO

GO

, OO OO

'

00 00

,'---', OO OO

=o GO

, OO OO

GO

OO

DRBI*~01

100000

~

,

. . . . . . .

:

60000 40000 20000 0 ,

m

m

m

m

m

,

~

m

= ~

m

m

,

r a m , =

=

m =

=

m =

~ =

~

~ m , = =

o c

Figure 4. Cross-reaction of human autoreactive T-cell receptor with APLs in the absence of reaction to cognate peptide presented by different MHC class II molecules. (A) The MBP85-99 peptide was substituted in the core recognition region and presented by either DRBI*1501 or DRBI*0401. The responses of OblA12.TCR hybridoma to these ligands were analyzed by HT.2 assay measuring the interleukin-2 production. (B) As the MBP 88K peptide was recognized in the context of DRBI*0401, a series of MBP85-99 peptide with substitutions at position 88 were synthesized and tested for their ability to activate Ob 1A12.TCR hybridoma.

mune disease. It should be emphasized that all the cross-reactive foreign peptides have been defined by their stimulatory capacity and selected from protein databases or screening based on the role of individual amino acid residues. Therefore, it is not clear whether these peptides can be processed by and presented on APCs after microbial infection. If they can be, it still remains to be defined whether or not cross-reactive T cells can be successfully activated and expanded sufficiently to attack self. Even though epiderniological studies suggested that sev-

52

eral viral and bacterial infections often precede the exacerbation of MS [27], it is difficult to discriminate between two potential mechanisms for this effect: the molecular mimicry described here and the bystander activation of autoreactive T cells. The latter includes the effect of inflammation, epitope spreading, or a superantigenic effect of invading microbes. In addition, a chronologic relationship is difficult to establish because the starting point of the disease is usually unknown and most suspicious microbes induce persistent infection. The lack of an

animal model of spontaneous development of MS is another obstacle to finding answers to these questions. Thus, although a great deal of fragmentary evidence is available, a definitive role for molecular mimicry in the pathogenesis of MS has not been proven. Potential evidence for the role of molecular mimicry in human MS has recently emerged from a clinical trial of altered peptide ligand therapy [28, 29]. In vitro and animal studies found that APLs can inhibit the response of T cells to agonistic autoantigen and can block EAE [30, 31]. Furthermore, APLs could induce a novel, APL-reactive T cell population that had a Th2 phenotype and cross-reacted with native MBP antigens [32]. These data supported the use of APLs as a therapeutic agent for MS. Unfortunately, unexpected detrimental side effects, including a hypersensitivity reaction to APLs and exacerbation of disease, necessitated the premature termination of all the trials before completion. An immunological study suggested that the latter side effect came from the expansion of APL-specific T cells with pro-inflammatory phenotype, which cross-reacted with MBP [30]. This finding is perhaps the strongest argument that, in some situations, molecular mimicry indeed can play a role in the pathogenesis of MS. Considering the highly degenerative nature of T cell recognition, it can be proposed that APLs may be presented by any of the HLA class II molecules and activate the pro-inflammatory T cells, which cross-react with myelin antigens presented either by APL-restricting or another class II molecule. However, the previous studies employed a single HLADR molecule for the selection of the APLs. Since the majority of patients are heterozygous for HLADR locus, the effect of APLs should be considered in conjunction with the individual's entire HLA class II haplotype. Actually, we recently observed that one APL (88V---)K) of MBP85-99 could stimulate the OblA12 TCR in the context of both DRB*1501 and DRBI*0401, whereas the original MBP85-99 peptide could not be recognized in the context of DRB*0401 [25]. Therefore, stricter criteria, considering all the self-MHC class II molecules as potential restriction elements, should be applied for the selection of APLs.

6. CONCLUDING R E M A R K S Molecular mimicry still remains an attractive hypothesis to explain the initiation and maintenance of MS lesions. Epidemiological and immunological studies support this theory. Indeed, several microbial peptides that cross-react with self myelin proteins have been identified. In addition, more microbial peptides will be identified in the near future if the MAPL concept is considered during screening for cross-reacting antigens. However, as mentioned previously, not all the available data can be taken as direct evidence that this molecular mechanism is at work in the pathogenesis of MS. Considering the highly degenerative nature of TCR recognition and the presence of self-reactive T cells in a normal repertoire, why do only a small percentage of people suffer from MS after microbial infections? One possible explanation is the multifactorial origin of MS. A recent twin study convincingly showed the importance of genetic traits as a risk factor [33]. If genetically predisposed individuals suffer from infections, they may develop MS due to molecular mimicry. Recently, there has been a conceptual movement in the MS field that MS is not a single disease entity but rather a syndrome composed of different disorders with different causes [34]. In addition, major pathogenic mechanisms might differ depending on the disease stage. Therefore, the molecular mimicry hypothesis should be re-evaluated according to the new concept of the complexity of T cell cross-reactivity as well as the disease entity.

REFERENCES 1.

2.

3.

4.

MartinR, McFarland HF, McFarlin DE. Immunological aspects of demyelinating Diseases. Annu Rev Immunol 1992;10:153-187. Mokhtarian F, McFarlin DE, Raine CS. Adoptive transfer of myelin-basic protein-sensitized T cells produces chronic relapsing demyelinating disease in mice. Nature 1984;309:356--358. ZamvilS, Nelson P, Trotter J, Mitchell D, Knobler R, Fritz R, Steinman L. T-cell clones specific for myelin basic protein induce chronic relapsing paralysis and demyelination. Nature 1985;317:355-358. Ota K, Matsui M, Milford EL, Mackin GA, Weiner HL, Hater DA. T-cell recognition of an immunodominant

53

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

54

myelin basic protein epitope in multiple sclerosis. Nature 1990;346:183-187. Pette M, Fujita K, Wilkinson D, Altman DM, Trowsdale J, Giegerich G, Hinkkanen A, Epplen JT, Kappos L, Wekede H. Myelin autoreactivity in multiple sclerosis: recognition of myelin basic protein in the context of HLA-DR2 products by T lymphocytes of multiple sclerosis patients and healthy donors. Proc Natl Acad Sci USA 1990;87:7968-7972. Martin R, Jaraquemada D, Flerlage M, Richert J, Whitaker J, Long EO, McFarlin DE, McFarland HF. Fine specificity and HLA restriction of myelin basic protein-specific cytotoxic T cell lines from multiple sclerosis patients and healthy individuals. J Immunol 1990;145:540-548. Lovett-Racke AE, Trotter JL, Lauber J, Perrin PJ, June CH, Racke MK. Decreased dependence of myelin basic protein-reactive T cells on CD28-mediated costimulation in multiple sclerosis patients. A marker of activated/memory T cells. J Clin Invest 1998;101: 725-730. Scholz C, Patton KT, Anderson DE, Freeman GJ, Hafler DA. Expansion of autoreactive T cells in multiple sclerosis is independent of exogenous B7 costimulation. J Immunol 1998;160:1532-1538. Zhang J, Markovic-Plese S, Lacet B, Raus J, Weiner HL, Hafler DA. Increased frequency of intedeukin 2responsive T cell specific for myelin basic protein and proteolipid protein in peripheral blood and cerebrospinal fluid of patients with multiple sclerosis. J Exp Med 1994;179:973-984. Fujinami RS, Oldstone MBA. Amino acid homology between the encephalitogenic site of myelin basic protein and virus: mechanism for autoimmunity. Science 1985;230:1043-1045. Ausubel LJ, Kwan CK, Sette A, Kuchroo VK, Hafler DA. Complementary mutations in an antigenic peptide allow for crossreactivity of autoreactive T-cell clones. Proc Natl Acad Sci USA 1996:93;15317-15322. Hemmer B, Fleckenstein BT, Vergelli M, Jung G, McFarland H, Martin R, Wiesmuller K-H. Identification of high potency microbial and self ligands for a human autoreactive class H-restricted T cell clone. J Exp Med 1997;185:1651-1659. Kastrukoff LS, Rice GPA. Virology. In: Paty DW, Ebers GC, eds. Multiple Sclerosis. Philadelphia: FA Davis, 1998;370-402. Johnson RT, Griffin DE. Virus-induced autoimmune demyelinating disease of the central nervous system. Chapter 24. In: Notkins AL, Oldstone MBA, eds. Concepts in Viral Pathogenesis II. New York: Springer, 1986;203-209. Gross DM, Forsthuber T, Tary-Lehmann M, Etling

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

c, Ito K, Nagy ZA, Field JA, Steere AC, Huber BT. Identification of LFA-1 as a candidate autoantigen in treatment-resistant Lyme arthritis. Science 1998;281: 703-706. Zhao Z-S, Granucci F, Yeh L, Schaffer PA, Cantor H. Molecular mimicry by herpes simplex virus-type I: autoimmune disease after viral infection. Science 1998;279:1344-1347. Wucherpfennig KW, Sette A, Southwood S, Oseroff C, Matsui M, Strominger JL, Hailer DA. Structural requirements for binding of an immunodominant myelin basic protein peptide to DR2 isotypes and for its recognition by human T cell clones. J Exp Med 1994;179:279-290. Wucherpfennig KW, Strominger JL. Molecular mimicry in T cell-mediated autoimmunity: Viral peptides activate human T cell clones specific for myelin basic protein. Cell 1995;80:695-705. Bhardwaj V, Kumar V, Geysen HM, Sercarz EE. Degenerate recognition of a dissimilar antigenic peptide by myelin basic protein-reactive T cells. Implications for thymic education and autoimmunity. J Immunol 1993;151:5000-5010. Hagerty DT, Allen PM. Intramolecular mimicry. Identification and analysis of two cross-reactive T cell epitopes within a single protein. J Immunol 1995;155: 2993-3001. Martin R, Howell MD, Jaraquemada D, Flerlage M, Richert J, Brostoff S, Long EO, MaFarlin DE, McFarland HE A myelin basic protein peptide is recognized by cytotoxic T cells in the context of four HLA-DR types associated with multiple sclerosis. J Exp Med 1991;173:19-24. Lim D-G, Slavik JM, Bourcier K, Smith ICJ, Hailer DA. Allelic variation of MHC structure alters peptide ligands to induce atypical partial agonistic CD8 § T cell function. J Exp Med 2003;198:99-109. Racit~ppi L, Ronchese F, Matis LA, Germain RN. Peptide-major histocompatibility complex class II complexes with mixed agonist/antagonist properties provide evidence for ligand-related differences in T cell receptor-dependent intracellular signaling. J Exp Med 1993:177; 1047-1060. Ota K, Matsui M, Milford EL, Mackin GA, Weiner HL, Hailer DA. T-cell recognition of an immunodominant myelin basic protein epitope in multiple sclerosis. Nature 1990:346;183-187. Mycko MP, Waldner H, Bourcier KD, Wucherpfennig K, Kuchroo VK, Hailer DA. Cross-reactive human autoreactive T-ceU receptor responses to altered peptide ligands presented by different MHC class II molecules. (Submitted). Lang HLE, Jacobsen H, Ikemizu S, Andersson C,

Harlos K, Madsen L, Hjorth P, Sondergaard L, Svejgaard A, Wucherpfennig K, Stuart DI, Bell JI, Jones EY, Fugger L. A functional and structural basis for TCR cross-reactivity in multiple sclerosis. Nat Immuno12002;3:940-943. 27. Sibley WA, Bamford CR, Clark K. Clinical viral infections and multiple sclerosis. Lancet 1985;1: 1313-1315. 28. Bielekova B, Goodwin B, Richert N, Cortese I, Kondo T, Afshar G, Gran B, Eaton J, Antel J, Frank JA, McFarland HF, Martin R. Encephalitogenic potential of the myelin basic protein peptide (amino acids 83-99) in multiple sclerosis: results of a phase II clinical trial with an altered peptide ligand. Nat Med 2000;6:1167-1175. 29. Kappos L, Comi G, Panitch H, Oger J, Antel J, Conlon P, Steinman L. Induction of a non-encephalitogenic type 2 T helper-cell autoimmune response in multiple sclerosis after administration of an altered peptide ligand in a placebo-controlled, randomized phase II trial. The altered peptide ligand in relapsing MS study group. Nat Med 2000;6:1176-1182.

30. De Magistris MT, Alexander J, Coggeshall M, Altman A, Gaeta FC, Grey HM, Sette A. Antigen analog-major histocompatibility complexes act as antagonists of the T cell receptor. Cell 1992;68:625-634. 31. Karin N, Mitchell DJ, Brocke S, Ling N, Steinman L. Reversal of experimental autoimmune encephalomyelitis by a soluble peptide variant of a myelin basic protein epitope: T cell receptor antagonism and reduction of interferon ~/and tumor necrosis factor ct production. J Exp Med 1994;180:2227-2237. 32. Nicholson LB, Greer JM, Sobel RA, Lees MB, Kuchroo VK. An altered peptide ligand mediates immune deviation and prevents autoimmune encephalomyelitis. Immunity 1995;3:397-405. 33. Ebers GC, Sadovnick AD, Risch NJ. A genetic basis for familial aggregation in multiple sclerosis. Canadian collaborative study group. Nature 1995;377:150-151. 34. Lucchinetti C, Bruck W, Parisi J, Scheithauer B, Rodriguez M, Lassmann H. Heterogeneity of multiple sclerosis lesions: Implications for the pathogenesis of demyelination. Ann Neurol 2000;47:707-717.

55

9 2004 Elsevier B. V All rights reserved. Infection and Autoimmunity Y. Shoenfeld and N.R. Rose, editors

Adverse Events of Desirable Gain in Immunocompetence: The Immune Restoration Inflammatory Syndromes (IRIS) Matthias Stoll and Reinhold E. Schmidt

Department Clinical Immunology, Medical School Hannover, Hannover, Germany

1. INTRODUCTION Sumilar signs of inflammation are common and correspondingly in a broad spectrum of heterogeneous infectious diseases. In infectious diseases inflammation reflects the ability of the adaptive immune system to actively compete with "nonself'. Inflammatory symptoms develop for the first time after an incubation period, in which usually (a) the infectious agent spreads inapparently in the host and (b) primary or secondary specific adaptive immune responses develop. Beginning from that point declining extent of inflammation reflects increasing immunological control of infection in an immunocompetent host. Therefore intensity of inflammation and extent of infection are sometimes equated improper. In contrast, in the immunodeficient host inflammation will not occur or to less extent - even in presence of latent infection. Therefore in case of (re-)emerging immunocompetence in an immunodeficienct state an increase of "paradoxical" inflammatory response may develop (Fig. 1). The delayed acute onset of "paradoxical" inflammation resembles a recently acquired intercurrent or acute disease. Therefore IRIS rather reflects an uncommon pathophysiologic mechanism or special kind of inflammation than the (retrospective) recognition of a - previously incompetent - reconstituted immune system. This may justify the assumption of a "paradoxical" inflammatory response. However, both underlying possible scenarios require specific clinical management: Spread of infection which escapes immunosurveillance demands an intensification or change of antinfective

treatment. But prolonged inflammation as sign of improving immunologic control might need predominantly antiinflammatory treatment [ 1]. A consent for the definition of IRIS has still to be found. Until than it could be characterized: (i) by the description of predispopsing risks for certain forms of IRIS; (ii) by its particular characteristics and clinical appearance; (iii) by its specific therapeutical requirements; (iv) and by the identification of distinct immunopathologic features in IRIS.

2. STATE OF IMMUNOCOMPETENCE The term IRIS might suggest restriction to one particular state of immunity. Indeed it may develop under different clinical conditions. IRIS may occur in individuals with defects in the specific or the innate immune system respectively. Beyond immunodeficiencies IRIS has been described as well in cases with extensive infections presenting without immunodeficiency.

2.1. IRIS Without Apparent Immunodeficiency Any whitespread infection by itself may overwhelm the capacity of the immune control in a non-immunocompromized host. In that case effective antimicrobial treatment allows the host's immune system to generate more potent inflammatory response at the site of remaining infection by lowering the high burden of infectious antigen. The historically classical example for IRIS is called "reversal reaction"

57

Figure 1. Model illustrating balance of driving forces in inflammatory reactions during immune restoration. and occurs commonly during the treatment of multibacillary "lepromatous" leprosy. Multibacillary forms of leprosy reflect a lower extent of specific immunity with predominant Th2-immune response than Thl dominated paucibacillary "tuberculoid" leprosy [2]. Therefore certain antiinflammatory or

58

immunosuppressive drugs are added to polychemotherapy of lepromatous leprosy. Correspondingly in disseminated tuberculosis IRIS with development of cerebral tuberculoma after initiation of tuberculostatic treatment may cause fatal outcome [3]. Reversion from negative tuberculin skin test in

disseminated tuberculosis to positive reaction after chemotherapy is of additional evidence for specific immune reconstitution in such scenario [4]. The addition of immunosuppressive corticosteroids is established and improves the outcome of certain forms of tuberculosis [5] or bacterial meningitis [6] by controlling the overshoot of inflammation under antimicrobial chemotherapy, indicating clinical significance of IRIS.

frequently. Commonly these cases suffer previously from fever of 'unknown' origin (FUO) during neutropenia. FUO is characterized by the lack of any infectious focus. The subsequent development of such foci - often after complete resolution of fever - c a n be explained by reconstitution of elements of the innate immune system in neutropenia [10, 13].

3. CLINICAL CHARACTERISTICS

2.2. Cellular Immunodeficiency Until the pandemic spread of the human immunodeficiency virus (HIV) severe defects of the cellular immune system were uncommon in adults and up to the introduction of highly active antiretroviral treatment (HAART) there was hardly any option to enhance immunity in immunodeficient patients [7]. I ~ S as an apparent new entity of disease have been described in cases of virologic and immunologic successful HAART [8-10]. IRIS occurs predominantly in patients who are severely immunocompromised in a narrow timely correlation to the initiation of HAART. The synchronous and widespread use of HAART in a large number of HIV infected patients with severe immunodeficiencies focusses in a considerable number of cases presenting with IRIS within a short period. Because of the individually proven immunological effectiveness of HAART in these cases IRIS imposes as unexpected and atypical event of inflammation. In most cases it is due to HIV-associated or AIDS-defining infectious disease, but IRIS presents often with unusual clinical features as compared to the corresponding classical opportunistic diseases. A couple of studies give evidence for the hypothesis that IRIS can be explained as increased inflammatory response during the restoration of the previously incompetent specific immunity [ 11, 12].

2.3. Defects of Innate Immunity Paradoxical inflammatory responses are not restricted to the reconstitution of specific adaptive immunity. Extensive and acutely developing pulmonary infiltrates or spread of focal hepatic lesions during the hematopoietic reconstitution after high dose cytostatic chemotherapy in patients with hematological malignancies have been described

IRIS imposes as unexpected event or paradoxical deterioration, when increasing inflammatory potency in a reconvalescent host either unmasks subclinical disease or worsens intensity of an already apparent inflammation. By certain clinical characteristics progress of classical overt disease can be distinguished from complementary IRIS. 3.1. Mycobacteria Tuberculous IRIS occurs commonly in (a) a narrow timely correlation to the initiation of antimycobacterial treatment or the initiation of HAART, (b) conversion of specific skin tests in some cases, and (c) the lack of increase of acid fast stained rods [8], which reflects increase of inflammation without increase of the underlying pathogen. IRIS reveals emerging and long lasting pulmonary infiltrates, lymphadenitis or tuberculoma [3, 14]. Focal lymphadenitis as a typical presentation of IRIS due to nontuberculous mycobacteria [15] clearly differs from the classical disseminated multiorgan disease in advanced immunodeficiency.

3.2. Herpesviridae Relapses of Cytomegalovirus (CMV)-retinitis under maintenance CMV-therapy were seen in up to half of all HIV-infected individuals at risk after the initiation of HAART [8, 9]. In contrast to AIDSdefining CMV-retinitis, IRIS may occur in patients with CD4+ T cells within normal range [8], without markers of CMV replication, and with atypical clinical manifestations including uveitis, vitritis, macular edema, epiretinal membranes and cataract [16]. A more pronounced increase of specific anti-CMV-IgG antibodies in these patients might serve as evidence for a specific pathogenetic role of

59

immune reconstitution in CMV-IRIS [ 11 ]. Repeated reactivation of dermatomal zoster developed up to more than half a year after HAART has been started [9, 17]. VZV-IRIS has been described in cases with higher increase of CD4+ T cells during the first weeks of HAART [9] and with more elevated CD8+ T cell proportions [ 17]. Herpes simplex (HSV) reactivation with unusually localized erosive lesions has been reported as IRIS in cases, who shared certain HLA Class I and HLA Class II antigens [18]. Transient reactivation of EBV replication occurs after initiation of HAART more often in patients with good immunological responses, especially in those, who respond immunologically well, present with increased immunoglobulin levels but do not reach complete control of retroviral replication [121. IL-6, a growth factor for HHV-8 was found to be elevated in patients with IRIS [19]. In this context two HHV-8 associated diseases presented atypically during immune restoration: Castleman's disease and Kaposi's sarcoma [20]. 3.3. Noninfectious

Disease

Increase of inflammatory response during immune restoration in a couple of noninfectious diseases provides additional evidence for the pathophysiological concept of IRIS: Descriptions include cases of autoimmune diseases, like systemic lupus erythematosous [21], rheumatoid arthritis [22], polymyositis [231, autoimmune thyreopathy [24], alopecia universalis [25], allergic reaction e.g. against prexisiting tattoos [26], disease of unknown etiology, like sarcoidosis [27], and the induction of atherogenic chronic inflammation respectively [28-30]. Additional clinical aspects of IRIS are summarized in detail elsewhere [ 1, 8-10, 31 ]

4. I M M U N O P A T H O L O G I C F E A T U R E S Certain correlations could be demonstrated between immunological markers or genetic characteristics and IRIS. As different opportunistic diseases require particular mechanisms of immunologic control disease specific "risk factors" could be

60

Table 1. Putative risk factors for development of IRIS after initiation of HAART in HIV-infection with severe immunodeficiency 9 9 9 9

Duration of immunodeficiency Extent of immunodeficiency Velocity and (relative) extent of immune reconstitution Specific pattern of immune reconstitution under HAART - immune reconstitution without complete suppression of HIV replication - high levels of CD8+ T lymphocytes - high levels of IL-6 and soluble IL-6 receptor - increase in (CMV-) specific IgG antibodies - high levels of soluble CD30 and soluble CD26 (dipeptidyl peptidase IV) activity - high levels of IFN gamma producing cells - Increased expression of CCR3 and CCR5 on monocytes and/or granulocytes - persisting polyclonal hypergammaglobulinemia - development of specific delayed type hypersensitivity 9 Genetic susceptibility - Distinct HLA haplotypes (e.g.: HLA B72, Cw0202, DRB4; HLA A2, B44 and HLA A1, B8, DR3 in conjunction with TNF-alpha polymorphism) - Polymorphisms in cytokine genes 9 TNF-alpha (in conjunction with certain HLAhaplotypes) 9 IL-6 9 IL-12

identified for IRIS (Table 1). The risk increases with speed and strength of immune restoration. In the majority of mycobacterial IRIS increase of CD4+ cells remained suboptimal whereas a more pronounced rise of CD4+ and/or CD8+ T cells were found in Herpesvirus IRIS [8, 9, 11, 17]. In CMVand EBV-IRIS predominance of Th2- over T h l immune response develops [ 11, 12, 32, 33] and correspondingly IL-6 and soluble IL-6 receptor were found elevated [19]. Increased chemokine receptor expression may result in persistence of irranunostimulation as one additional risk factor for the onset of IRIS [33] and usually succesful HAART leads to a decrease of immunostimulation [34]. Evidence for the individually genetic predisposition for certain manifestations of IRIS is based on the association of IRIS with certain histocompafibility antigens [18, 35] and with gene polymorphisms of cytokines like IL-6 and TNF-alpha [19, 36].

Table 2. Proposal for diagnostic criteria for IRIS within a setting of immune reconstitution Clinical criteria

Immunological criteria Major

Minor

Unexpected onset or paradoxical deterioration

Demonstrable immune reconstitution

Specific and predisposing pattern of immune stimulation or immune restoration a

Specific symptoms, which should allow distinction from "regular" opportunistic disease

Specific immune reconstitution against the presumed pathogen a

Predisposing genetic factors (e.g. histocompatibility antigens, cytokine gene polymorphisms)a

Rapid onset and close correlation to restoration of the immune system Proof of underlying pathogen a Routine tests for these criteria are either not available or these criteria are preliminary and therefore restricted to scientific investigation.

5. DIAGNOSIS

Consensus for diagnostic criteria and treatment is not yet defined. One proposal for diagnostic criteria for IRIS within a setting of immune reconstitution is given [ 1] in Table 2.

6. TREATMENT Until treatment guidelines will be defined and evaluated in clinical studies an empirical proposal for treatment of IRIS should consider [ 1]: 9 Treatment of underlying disease by antiinfective chemotherapy. 9 Temporarily suppression of inflammation by suitable interventions, e.g. with antiphlogistic or immunesuppressive drugs. 9 Surgical or symptomatic treatment of certain manifestations of I ~ S (e.g.: iritis bombata, necrotizing lymphadenitis).

and can mimic opportonistic disease, allergic reaction or features of autoimmunity respectively. 9 Timely correlation to immune restoration, distinct clinical appearance, particular immunologic and genetic characteristics, and different therapeutic requirements justify the assumption of an indepent entity for IRIS. 9 Differential therapeutic approaches require to separate IRIS from corresponding opportunistic disease: Antiinflammatory or immunosuppressive drugs are cornerstones of treatment in IRIS. 9 Certain patterns of immune response, different levels and pathways of immune stimulation and genetic factors have been identified in IRIS. In future these factors may allow an individual risk assessment and development of rational therapeutic approaches.

REFERENCES 1.

7. TAKE HOME MESSAGES

9 Immune restoration inflammatory syndromes (IRIS) present as unexpected event or paradoxical deterioration of inflammation. 9 IRIS is correlated to a variety of infectious diseases or noninfectious immunogenic antigens

2.

3.

Stoll M, Schmidt RE. Immune restoration inflammatory syndromes: The dark side of successful antiretroviral treatment. Curr Infect Dis Rep 2003;5:266-276. Moubasher AD, Kamel NA, Zedan H, Raheem DD. Cytokines in leprosy, II. Effect of treatment on serum cytokines in leprosy. Int J Dermatol 1998;37(10): 741-746. Lees AJ, MacLeod AF, Marshall J. Cerebral tuberculo-

61

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

62

mas developing during treatment of tuberculous meningitis. Lancet 1980; 1(8180): 1208-1211. Rooney JJ, Crocco JA, Kramer S, Lyons HA. Further observations on tuberculin reactions in active tuberculosis. Am J Med 1976;60(4):517-522. Dooley DE Carpenter JL, Rademacher S. Adjunctive corticosteroid therapy for tuberculosis: a critical reappraisal of the literature. Clin Infect Dis 1997;25(4): 872-887. De Gans J, van de Beek D, European Dexamethasone in Adulthood Bacterial Meningitis Study Group. Dexamethasone in adults with bacterial meningitis. N Engl J Med 2002;347(20): 1549-1556. Egger M, May M, Chene G, Phillips AN, Ledergerber B, Dabis F et al. Prognosis of HIV-l-infected patients starting highly active antiretroviral therapy: a collaborative analysis of prospective studies. Lancet 2002;360(9327): 119-129. Behrens G, Meyer D, Stoll M, Schmidt RE. Immune reconstitution syndromes in human immunodeficiency virus infection following effective antiretroviral therapy. Immunobio12000;202:186-193. French MA, Lenzo N, John M, Mallal SA, McKinnon EJ, James IR et al. Immune restoration disease after the treatment of immunodeficient HIV-infected patients with highly active antiretroviral therapy. HIV Med 2000; 1(2): 107-115. Cheng VC, Yuen KY, Chan WM, Wong SS, Ma ES, Chan RM. Immunorestitution disease involving the innate and adaptive response. Clin Infect Dis 2000;30( 6):882-892. Stone SE Price E Tay K, French MA. Cytomegalovirus (CMV) retinitis immune restoration disease occurs during highly active antiretroviral therapy-induced restoration of CMV-specific immune responses within a predominant Th2 cytokine environment. J Infect Dis 2002; 185(12): 1813-1817. Righetti E, Ballon G, Ometto L, Cattelan AM, Menin C, Zanchetta Met al. Dynamics of Epstein-Barr virus in HIV-l-infected subjects on highly active antiretroviral therapy. AIDS 2002; 16(1):63-73. Heussel CP, Kauczor HU, Heussel G, Fischer B, Mildenberger E Thelen M. Early detection of pneumonia in febrile neutropenic patients: use of thin-section CT. AJR Am J Roentgenol 1997;169(5):1347-1353. Narita M, Ashkin D, Hollender ES, Pitchenik AE. Paradoxical worsening of tuberculosis following antiretroviral therapy in patients with AIDS. Am J Respir Crit Care Med 1998;158(1):157-161. Race EM, Adelson-Mitty J, Kriegel GR, Barlam TF, Reimann KA, Letvin NL et al. Focal mycobacterial lymphadenitis following initiation of protease-inhibitot therapy in patients with advanced HIV-1 disease.

Lancet 1998;351:252-255. 16. Whitcup SM. Cytomegalovirus retinitis in the era of highly active antiretroviral therapy. JAMA 2000;283(5): 653-657. 17. Domingo P, Torres OH, Ris J, Vazquez G. Herpes zoster as an immune reconstitution disease after initiation of combination antiretroviral therapy in patients with human immunodeficiency virus type-1 infection. Am J Med 2001;110(8):605-609. 18. Fox PA, Barton SE, Francis N, Youle M, Henderson DC, Pillay D et al. Chronic erosive herpes simplex virus infection of the penis, a possible immune reconstitution disease. HIV Med 1999;1(1):10-18. 19. Stone SF, Price P, Keane NM, Murray RJ, French MA. Levels of IL-6 and soluble IL-6 receptor are increased in HIV patients with a history of immune restoration disease after HAART. HIV Med 2002;3(1):21-27. 20. Zietz C, Bogner JR, Goebel FD, Lohrs U. An unusual cluster of cases of Castleman's disease during highly active antiretroviral therapy for AIDS. N Engl J Med 1999;340(24): 1923-1924. 21. Behrens G, Knuth C, Schedel I, Mendila M, Schmidt RE. Highly active antiretroviral therapy. Lancet 1998;351:1057-1058. 22. Bell C, Nelson M, Kaye S. A case of immune reconstitution rheumatoid arthritis. Int J STD AIDS 2002; 13(8): 580-581. 23. Sellier P, Monsuez JJ, Evans J, Minozzi C, Passeron J, Vittecoq D et al. Human immunodeficiency virusassociated polymyositis during immune restoration with combination antiretroviral therapy. Am J Med 2000; 109(6):510-512. 24. Jubault V, Penfornis A, Schillo F, Hoen B, Izembart M, Timsit J et al. Sequential occurrence of thyroid autoantibodies and Graves' disease after immune restoration in severely immunocompromised human immunodeficiency virus-l-infected patients. J Clin Endocrinol Metab 2000;85( 11):4254--4257. 25. Sereti I, Sarlis NJ, Arioglu E, Turner ML, Mican JM. Alopecia universalis and Graves' disease in the setting of immune restoration after highly active antiretroviral therapy. AIDS 2001;15(1):138-140. 26. Silvestre JF, Albares MP, Ramon R, Botella R. Cutaneous intolerance to tattoos in a patient with human immunodeficiency virus: a manifestation of the immune restoration syndrome. Arch Dermatol 2001;137(5): 669-670. 27. Wittram C, Fogg J, Farber H. Immune restoration syndrome manifested by pulmonary sarcoidosis. AJR/san J Roentgenol 2001;177(6):1427. 28. Behrens G, Stoll M, Schmidt RE. Lipodystrophy syndrome with protease inhibitors: what is it, what causes it and how can it be managed? Drug Saf 2000;23:

57-76. 29. Grahame C, Alber DG, Lucas SB, Miller R, Vallance P. Association between Kaposi's sarcoma and atherosclerosis: implications for gammaherpesviruses and vascular disease. AIDS 2001; 15(14): 1902-1904. 30. Lewis W. Atherosclerosis in AIDS: potential pathogenetic roles of antiretroviral therapy and HIV. J Mol Cell Cardio12000;32(12):2115-2129. 31. Shelburne SA, Hamill RJ. The immune reconstitution inflammatory syndrome. AIDS Rev 5(2):67-79. 32. Johnson SC, Benson CA, Johnson DW, Weinberg A. Recurrences of cytomegalovirus retinitis in a human immunodeficiency virus-infected patient, despite potent antiretroviral therapy and apparent immune reconstitution. Clin Infect Dis 2001;32(5):815-819. 33. Keane NM, Price P, Lee S, Stone SF, French MA. An evaluation of serum soluble CD30 levels and serum CD26 (DPPIV) enzyme activity as markers of type

2 and type 1 cytokines in HIV patients receiving highly active antiretroviral therapy. Clin Exp Immunol 2001 ;126(1): 111-116. 34. Behbahani H, Landay A, Patterson BK, Jones P, Pottage J, Agnoli M et al. Normalization of immune activation in lymphoid tissue following highly active antiretroviral therapy. J Acquir Immune Defic Syndr 2000;25(2): 150-156. 35. Price P, Mathiot N, Krueger R, Stone S, Keane NM, French MA. Immune dysfunction and immune restoration disease in HIV patients given highly active antiretroviral therapy. J Clin Virol 2001 ;22(3):279-287. 36. Price P, Morahan G, Huang D, Stone E, Cheong KY, Castley A et al. Polymorphisms in cytokine genes define subpopulations of HIV-1 patients who experienced immune restoration diseases. AIDS 2002; 16(15): 2043-2047.

63


CD5-Expressing B Cells and Infection Y. Renaudineau, J.O. Pers and P. Youinou

Laboratory of Immunology, Brest University Medical School, Brest, France

The 67 kDa T cell marker CD5 was originally identified on malignant human B cells [1], and subsequently shown to act as a coreceptor on a proportion of normal B lymphocytes in humans [2] and mice [3]. These have been classified into B-2, representing the conventional cells, and B-l, predominating in serous cavities [4]. The latter population comprises B-la which express CD5 (Fig. 1), and B-lb subpopulations which do not, but share all the other attributes of B-1 cells [5], such as the presence of mRNA for CD5, the expression of the myelomonoytic marker Mac-1, and the reduced density of the high molecular weight isoform of the common leukocyte antigen (Ag) CD45RA. Over the past decade, evidence has been accumulating to suggest that such B-1 lymphocytes are key in the defense from infectious agents. For example, they produce much of the immunoglobulin (Ig), almost all natural antibody (Ab) reactive with lipopolysaccharide [6], and most of the innate Ab in serum [7], although constituting only a minor fraction of the B compartment. Furthermore, they contribute significantly to the IgA-producing plasma cells in the lamina propria of the gut [8]. Indeed, in germ-free conditions, few peritoneal B-1 cells are detected in the mouse, while a number of these exist in specific pathogen-free conditions, indicating that bacterial infections are necessary for their generation [9]. The finding that numerous autoimmune conditions are associated with elevated levels of circulating B-1 cells, and the demonstration that, due to their resistance to apoptosis, such lymphocytes accumulate in chronic lymphocytic leukemia (CLL) and other B cell malignancies have sparked off a great deal of interest in the possibility of a role

of B-1 cells in the pathophysiology of a number of diseases, including infectious conditions. The general conclusions regarding B-1 lymphocytes are that only in some cases has this subset been demonstrated to be responsible for the production of Ab to infectious agents as well as autoAb. In general, however, the function of these cells in different disease states remains unclear.

1. C H A R A C T E R I Z A T I O N OF CD5+ B CELLS

1.1. Origins of the Cells Still it is unclear whether CD5 signifies a different lineage or is induced by activation. On the one hand, there are a number of evidence in support of a different lineage. Early experiments showed that irradiated mice could be reconstituted with CD5+ B cells if the graft contained bone marrow stem cells together with peritoneal cells [10]. Additional support came from studies on the severe combined immunodeficiency mouse (SCID). The animals fail to develop either T or B cells due to a genetic deficiency in the enzyme required for rearrangements of Ag receptor genes. Immunological reconstitution with CD5+ B cells can be achieved by injection of fetal liver but not adult bone marrow. In similar experiments with severe common immune deficiency (SC1D) mice, repopulation of CD5+ B cells, but not conventional B cells, B2, was obtained by injection of fetal omental cells. Levels of serum IgM and IgG3 Ab became detectable in these mice. Although the SCID mice experiments seem definitive, it has been argued that these animals might provide an alien environment for the develop-

65

Figure 1. On the left, double-staining of peripheral blood lymphocyte using a B cell marker which is CD19 and a T cell marker which is CD5 permits identification of B cells which express CD5, i.e., B-1 cells (arrows), as opposed to conventional B-2 cells. On the fight, polymerase chain reaction of the transcripts for CD5 followed by electrophoresis and blots show the message in T cells, and, to a lesser degree, in B cells.

ment of B cells along the conventional pathway from a common B cell precursor. Also consistent with the lineage paradigm, is the striking similarity in the relative proportion of circulating CD5+ B cells in monozygotic twins [11 ] and family members [12] of patients with rheumatoid arthritis (RA). On the other hand, arguments against the notion that CD5+ B cells are a separate lineage support the general view that such cells are just an activated subpopulation. One argument is that human CD5B cells can be induced to become CD5+ in vitro by activation with phorbol myristic acetate (PMA) [13] and several cytokines modulate the expression of this molecule [ 14]. Similar data is now available in the mouse showing that anti-IgM and interleukin (IL)-6 treatment increased CD5 expression on splenic B cells. To account for these opposite viewpoints, we

66

have proposed a reconciliation of the divergent theses [15] by postulating two different classes of CD5+ B cells (Fig. 2): those where CD5 expression is constitutive ("classical" CD5+B cells) particularly in the fetal liver and in the cord blood (CB), and conventional B-2 cells induced to express CD5 on appropriate activation ("induced" CD5+B cells) usually located in the germinal center of any secondary lymphoid organ.

1.2. Functions of CD5+ B Cells This question has been approached using immortalized CB clones. There appeared that B-1 lymphocytes have a propensity to produce low-affinity polyreactive Ab binding to self, as well as to exogenous Ag including several bacteria [16]. Some positive clones were specific for the carbohydrate

Figure 2. Innate or "classical" CD5+B cells in the cord blood are distinct from acquired or "induced" CD5+ B cells in the germinal center of a secondary lymphoid organ. B-1a refers to CD5-positive and B-lb to CD5-negative B-1 cells, as opposed to conventional B-2 cells.

erythrocyte iI [ 17], which is accessible to the B lymphocytes in the CB. This observation suggests that the CD5+ clones have been driven by such autoAg and, therefore, selected in vivo, and agrees with the previous finding that, at least in the Hemophilus influenza model, innate (B-1 cells) and acquired (B2 cells) humoral immunities are mediated by distinct arms of the system [7]. Most of these B-1 cells, however, use germ-line genes [18] and express cross-reactive idiotypes [ 16].

1.3. Control of the CD5+ B Cell Population As suggested by the data, already mentioned, on monozygotic twins [11] and family members of RA patients [12], the size of B cell subsets seems to be under the control of the major histocompatibility complex, as established in the mouse [19]. The distribution of B cells into B-1 a, B-lb and B-2 would thus be genetically regulated, while IL-10

rescues B-1 a cells from apoptosis and encourages B-lb cell proliferation [20]. Interestingly, an Igindependent regulating feedback mechanism of the B cell compartments has also been described in the mouse [21 ].

2. CD5+ B CELLS AND DISEASE

2.1. Connective Tissue Diseases Several groups have reported that the CD5+ B cell subset may be expanded in patients with RA. Using double-fluorochrome ultraviolet light rvficroscopy, this was originally found to comprise an average of 20% of the B cells of 16 RA patients, compared to a maximum of 3% in eight normal controls, though the average absolute numbers of circulating B cells were comparable in these two groups of subjects [12]. Although CD5 molecules are present at low

67

Autoimmune disease

Control

3% I

Chronic lymphocytic leukemia

22%

91%

I

CD 19-FITC F i g u r e 3. The level of CD5+ B cells is increased in some nonorgan-specific diseases as well as infectious states. These cells accumulate in chronic lymphocytic leukemia.

density on B lymphocytes, this is increased [13] following treatment of B-lymphocyte-enriched cell suspensions with PMA. These results were subsequently confirmed by flow cytometry analysis (Fig. 3). It was thus possible to detect coexpression of CD5 on a larger population of B cells from RA patients and controls than earlier studies, but the mean proportions of B cells that express CD5 were still greater in patients than in controls. These data conflict with some reports showing no significant differences in percentages of circulating CD5+ B cells in RA patients compared with normal individuals. In fact, they fall into two categories, twothirds with CD5+B cell levels within the normal range, and a third with elevated levels. Clearly, the elevation do CD5+ B cells p e r se is insufficient to give rise to RA. The corollary is that high levels of this B-cell subpopulation is not a prerequisite for developing the disease. The number of circulating CD5+ B cells does, however, correlate with the titer of rheumatoid factor (RF). Whilst there were reports of significantly elevated frequencies of these cells in RA patients with extremely high titers of RF, other studies have claimed that increased levels of CD5+B cells were associated with RF and antinuclear antibody in such patients. The level of CD5+B cells was also elevated in patients with primary Sjrgren's

68

syndrome [22] particularly in those patients with associated monoclonal Ig. Surprisingly, in most of the cases of systemic lupus erythematosus (SLE), there are not elevated numbers of CD5+B cells. Increased numbers have, however, been described in some of them [23], suggesting that polyclonal activation might also affect this B cell subset in a proportion of SLE patients.

2.2. Lymphoid Malignancies CLL comprise a heterogeneous group of disorders, in which three main conditions have emerged: CLL, pro-lymphocytic leukaemia (PLL) and hairy cell leukaemia (HCL). The malignant cells from approximately 95% of the CLL patients co-express CD5 and other B-cell surface markers [24]. Thus, in most cases of CLL there is a proliferation of a B cell clone characterized by low amounts of surface Ig and increased expression of the CD5 Ag. Different subtypes of CLL have, however, been delineated, and CD5 shown to be expressed on the leukemic cells, not only from patients with genuine CLL, but also from those with cleaved cell lymphocytic leukaemia, and with lymphoplasmocytoid leukaemia. The B cells of CLL with a particularly indolent character have been claimed to be more frequently CD5+ than those B cells in patients with more

aggressive CLL [25], but no correlation was definitively established between the expression of CD5 and surface Ig class or type, clinical stage, disease activity or age at diagnosis by other investigators. B cell markers are present on the leukemic cells in all cases of PLL, but in some cases they do not express the CD5 marker, and only weakly to moderately in others. In contrast, PLL have been found to carry molecules recognized by anti-CD5 monoclonal Ab. Typically, malignant cells forming HCL do not express CD5, although is expression has been reported occasionally [26]. Expression of CD5 is not a trait shared by immature B cell malignancies, such as pre-B acute lymphoblastic leukaemia, or by endstage differentiated B cell malignancies, e.g., multiple myeloma or Waldenstrrm's macroglobulinemia. Nonetheless, there are no clear differences between the latter disease and an otherwise typical CLL associated with monoclonal Ig in the serum and urine or cases terminating in lymphoplasmocytoid leukaemia or lymphoma. Less than half of the B cell-derived non-Hodgkin's lymphomas are CD5+ [27]. The marker is mainly found on B cell lymphomas composed predominantly of small lymphocytes, such as diffuse well-differentiated lymphocytic lymphomas, or intermediate lymphocytic lymphomas. Malignant cells are prone to express CD5, when solid tumors are associated with lymphocytosis. In fact, the immunophenotype of these lymphoma or leukaemia cells is reminiscent of that of lymphocytes in normal primary follicles and the mantle zone of secondary follicles of secondary lymphoid organs. It is well documented that surface Ig receptors on malignant B cells exhibit specificity for a variety of self Ag. This concept has been extended [28] by studies of IgM Ab secreted by leukemic B cells, after stimulation with PMA, which results in the production of low-affinity polyreactive autoAb. 2.3. Infections States

Elevated levels of circulating CD5+ B cells have also been reported in a great number of infectious diseases, especially those from viral origin. Surprisingly, this has been described in infectious mononucleosis [29], but never confirmed because Epstein-Barr virus reduces the expression of CD5. On the other hand, CD5+ B cells have repeatedly

been found enhanced in chronic hepatitis C virus infection (HCV), as compared with the patients with resolved infection [30]. In fact, chronic infection with HCV is associated with disturbance of B lymphocyte activation and function, leading to serological abnormalities, such as autoAb production, mixed cryoglobulinemia and B cell lymphomas [31]. The possibility exists that they reflect chronic Ag stimulation or aberrant signaling through the B cell CD81 glycoprotein. Alternatively, CD81 which is a putative HCV receptor is upregulated in CD5+ cells, compared with conventional B cells [32]. Are these unique cells involved in the defense from more common infections? The major target Ag of B-1 lymphocyte-derived IgA are normal intestinal bacteria [33]. Their coating with IgA results in immune exclusion, as established for pathogenic microbes, although the bacterial microflora of the gut is an extremely stable ecosystem. Furthermore, several parasites have been associated with an expansion of CD5+ B cells. Included are Toxoplasma gondii [34], Trypanosoma evansi [35] and Schistosiamis mansoni [36]. In the latest parasitemia, whether or not B-1 a cells are responsible for Ab against egg Ag polylactosamine sugars, as described for a mouse model previously, has not yet been determined.

3. FUNCTION OF THE CD5 M O L E C U L E CD5 is physically [37] and functionally [38] associated with the B cell receptor (BCR). Increased numbers of CD5+ B cells might thus reflect defective regulation of B cell function through CD5 itself (Fig. 4). There is now a growing body of evidence that CD5 is essential in modulating signals downstream of the BCR. In this respect, we have shown that ligation of CD5 or IgM on tonsillar B but not blood T cells resulted in apoptosis [39]. This observation has since been confirmed in a group of patients with CLL [40], and shown to take the BCR pathway [41]. In addition, anti-CD5 sustains the proliferation of tonsillar B cells pre-activated with anti-IgM Ab and IL-2 [42]. This was in contrast to CB CD5+ B cells which do not apoptose in response to anti-CD5, but might rather reflect the fact these CB B cells are continuously exposed to autoAg in vivo. It is important to note that the src-homolgy

69

(~ P-Tyrosine @ Serine

Figure 4. The B cell receptor (BCR) comprises membrane IgM with Igct/Ig[3as transducing molecules. CD5 is made up of three extra-cytoplasmic domains (D l-D3) and associated with the BCR and brings about the src-homology 2 domaincontaining phosphatase (SHP-1) to dampen down the transducing cascade. The tyrosine residues are phorphorylated (P) by phosphorylases, but not the serine(s) residues.

2 domain-containing protein tyrosine phosphatase (SHP-1) is constitutively linked with the Ig~/Ig]] chains of the BCR through the immunoregulatory tyrosine based inhibitory motif of CD5 [43]. The tyorisine residues are phosphorylated by phosphorylases (P), but not the serine(s) residues. It has thus been suggested that such an interaction with CD5 "sequesters" the SHP-1, and limits its role with important molecules in positive signaling through the BCR [44]. Furthermore, the role of CD5 in the maintenance of clonal anergy has recently been addressed by the elegant experiments of Hippen et al using the hen egg lysosyme (HEL)-Ig transgenic (Tg) mouse [45]. In this model, mice Tg for HEL-Ig and the membrane-bound form of the self Ag HEL produce apoptosis of anti-HEL B cells, while those Tg for HEL-Ig and the soluble form of HEL initiate anergy through the SHP-1. Breeding of the latter Tg mice onto a CD5-/- background results in loss of tolerance. These data indicate that the presence of CD5 raises the threshold required for activation of self-reactive B cells, in such a way that it determines their ultimate face. Consistent with this role for CD5 is a more recent model in which CD5- spleen cells from mice made Tg for anti-ribonucleo-protein

70

(RNP), a common autoAb in SLE and other connective tissue diseases, were injected into irradiated naive mice. They migrated to the peritoneal cavity where most of the CD5+ B cells are found, and began to express CD5 which prevented their production of anti-RNP autoAb [46]. As well as CD5 being important in this negative regulation of autoreactive B cells, other molecules have been shown to play a role. For example, CD 19 amplifies BCR signaling by favoring the activity of phosphorylases, such that a modest 10-20% increase in CD 19 expression may be sufficient to shift the balance between tolerance and immunity to autoimmunity [47]. In contrast, CD22, dampens down the signals by recruiting SHP-1, so that deficiency in CD22 encourages the development of autoimmunity [48]. Furthermore, defective signaling through the BCR has already been demonstrated for B cells from patients with SLE [49].

4. REGULATION OF THE CD5 E X P R E S S I O N Several lines of evidence indicate that the expression of CD5 is tightly regulated. Thus, membrane density of CD5 is --30-folds higher in T than in B cells, and the expression of CD5 developmentally regulated, since the membrane density of CD5 is higher on mature T cells than on thymocytes. Moreover, CD5-expressing B cells represent the majority of B lineage cells during fetal and neonatal fife, but the number of CD5+ B cells declines in relative number with age [50]. Further evidence in support of tight CD5 regulation comes from experiments showing that ex vivo B cells downregulate their membrane CD5 expression when cultured in the presence of IL-4 [14], but upregulate CD5 following activation with PMA [ 13, 51 ], or when their membrane IgM is cross-linked in the presence of IL-6 [52]. Consistent with this view are findings that, despite their loss of membrane expression of CD5-, B-lb cells retain CD5 mRNA, albeit at lower levels than B-la cells [5]. Finally, a feedback regulation of murine B-la cells has also been advocated [21]. In an apparent contrast to these findings, a recent report suggested that all B cells constitutively express CD5, but the level of expression varies considerably, from B l a cells at one end of the spectrum to B2 cells at the other [53]. All in all, these observations imply that

Figure 5. The CD5 gene is made up of 12 exons. Exon 1 associates the classical exon 1, termed 1A and the alternative exon 1, designated lB. The former is expressed in B and T cells, and the latter present exclusively in B cells. When exon 1A splices to exon 2, the initiation site AUG is located within the exon 1 and the resulting CD5 molecule is full-length, whereas when exon 1B splices out exon 1A and binds to AUG-free exon 2, the first initiation site AUG is located within exon 3 and the resulting CD5 molecule truncated, because the 5' segment of exon 3 is not transcribed into mRNA.

multiple regulatory mechanisms for CD5 expression exist.

4.1. Shedding of CD5 At the posttranslational level, shedding of the molecule has been described, and suggested to be exaggerated in nonorgan-specific autoimmune and infectious diseases [54]. Cell-free CD5 could even bind to cells endowed with the related receptors, leading to an over-estimation of CD5+ B cells.

4.2. Internalization of the Membrane Molecule Interplay between several mechanisms is likely to be involved in the accurate regulation of CD5 expression at the membrane. CD5 internalization which is enhanced in T cells, but inhibited in B cells, upon Ag receptor crosslinking is another mechanism known to be involved in CD5 regulation [55], even though the spontaneous turn-over of this molecule is rather low.

pre-mRNA splicing, message stability and translation in different lymphocyte populations, are totally unknown. These 11 exons are, however, conserved in size and number in the mouse, as well as most of their transcription regulatory elements. A novel exon 1 that is exclusively transcribed in B cells has just been discovered [57]. Intriguingly, the existence of this new exon is due to a defective human endogenous retrovirus (HERV). The data also provides attractive evidence for a reciprocal expression of this alternative exon 1, designated exon 1B, with the conventional exon 1, hereafter referred to as exon 1A. Exon 1B-type transcripts are translated into a truncated variant of the CD5 molecule devoid of leader peptide. Consequently, whereas exon 1A promotes expression of membrane CD5 protein in T and a subset of B cells (Fig. 5), exon 1B acts to reduce CD5 protein expression in BL and, therefore, possibly, reduce the signaling functions of CD5, such as the production of Ab against infectious agents and auto-Ab. This balance between the two exons 1 might be important in the regulation of membrane expression of CD5.

4.3. Alternative Splicing of the Gene The CD5 protein is encoded by a single gene in both T and B cells, mapping to chromosome 11q12.2, and consisting of at least 11 exons [56]. The precise stages of regulation, i.e., transcription, alternative

To conclude, the Bar Mitzvah is being celebrated for B-1 cells [58]. We are indeed close to understand the way they operate in autoimmunity and infection. Paradoxically, in the light of recent findings of the modulation of B cell signaling by CD5, this

71

and other molecules play a crucial role in preventing autoimmunity. Aberrations of the transduction through CD5 are thought to exist. They could lead to autoimmune disorders. Hence, the present views on the potential functions of CD5+ B cells in autoimmunity are quite different from the previous and rather naive interpretation that the increased levels of CD5 B cells in patients with nonorganspecific autoimmune diseases represented a direct source of autoAb leading to pathogenesis.

ACKNOWLEDGEMENTS

8.

9.

10.

11.

Studies mentioned in this review were supported by the Acad6mie Nationale Franqaise de M6decine, the Conseil R6gional de Bretagne and the C o m m u naut6 Urbaine de Brest. The secretarial assistance of S imone Forest is appreciated.

12.

REFERENCES

13.

1.

2.

3.

4. 5.

6.

7.

72

Boumsell L, Bernard A, Lepage V, Degos L, Lemerle J, Dausset J. Some chronic lymphocytic leukemia cells beating surface immunoglobulins share determinants with T cells. Eur J Immunol 1978;8:900-904. Caligaris-Cappio F, Gobbi M, Bofill M, Janossy G. Infrequent normal B lymphocytes express features of B-chronic lymphocytic leukemia. J Exp Med 1982;155: 623--628. Ledbetter JA, Evans RL, Lipinski M, CunninghamRundles C, Good RA, Herzenberg LA. Evolutionary conservation of surface molecules that distinguish T lymphocyte helper/inducer and cytotoxic/suppressor subpopulations in mouse and man. J Exp Med 1981;153:310-323. Kantor A. A new nomenclature for B cells. Immunol Today 1991;12:388-391. Kasaian MT, Ikematsu H, Casali P. Identification and analysis of a novel human surface CD5-B lymphocyte subset producing natural antibodies. J Immunol 1992; 148:2690-2702. Su SD, Ward MM, Apicella MA, Ward RE. The primary B cell response to the O/core region of bacterial lipopolysaccharide is restricted to the Ly-1 lineage. J Immunol 1991;146:327-331. Baumgarth N, Herman OC, Jager GC, Brown L, Herzenberg LA, Herzenberg LA. Innate and acquired humoral immunities to influenza virus are mediated by distinct arms of the immune system. Proc Natl Acad Sci

14.

15.

16.

17.

18.

19.

20.

USA 1999;96:2250-2255. Murakami M, Honjo T. Involvement of B-1 cells in mucosal immunity and autoimmunity. Immunol Today 1995;16:534-539. Murakami M, Nakajima K, Yamazaki K, Muraguchi T, Sorikawa T, Honjo T. Effects of breeding environments on generation and activation of autoreactive B-1 cells in anti-red blood cell autoantibody transgenic mice. J Exp Med 1997;185:791-794. Hayakawa K, Hardy RR, Herzenberg LA, Herzenberg LA. Progenitors for Ly-1 B cells are distinct from progenitors for other B cells. J Exp Med 1985;161: 1554-1558. Kipps TJ, Vaughan JH. Genetic influence on the levels of circulating CD5+B lymphocytes. J Immunol 1987;139:1060-1064. Youinou P, MacKenzie L, Katsikis P, Merdrignac G, Isenberg DA, Tuaillon N, Lamour A, Le Goff P, Jouquan J, Drogou A, Muller S, Genetet B, Moutsopoulos HM, Lydyard PM. The relationship between CD5expressing B lymphocytes and serologic abnormalities in rheumatoid arthritis patients and their relatives. Arthritis Rheum 1990;33:339-348. Youinou P, MacKenzie L, Jouquan J, Le Goff P, Lydyard PM. CD5-positive B cells in patients with rheumatoid arthritis: phorbol ester-mediated enhancement of detection. Ann Rheum Dis 1987;46:17-22. Defrance T, Vanbervliet B, Durand I, Banchereau J. Human interleukin 4 down-regulates the surface expression of CD5 on normal and leukemic cells. Eur J Immunol 1989; 19:293-299. Youinou P, Jamin C, Lydyard PM. CD5 expression in human B-cell populations. Immunol Today 1999;20: 312-316. Lydyard PM, MacKenzie LE, Youinou P, Deane N, Jefferis R, Mageed RA. Specificity and idiotype expression of IgM produced by CD5+ and CD5- cord blood B cells. Ann NY Acad Sci 1992;651:527-535. Vencovsk3~J, Shokri F, Salpekar A, Kahan M, Isenberg DA, Youinou P, Lydyard PM, Mageed RA. Autoreactivity of antibodies encoded by the human V4-34 gene promotes over-representation in cord blood B 1 lymphocytes. Submitted. Deane M, MacKenzie LE, Stevenson FK, Youinou PY, Lydyard PM, Mageed RA. The genetic basis of human VH4 gene family-associated cross-reactive idotype expression in CD5+ and CD5- cord blood B-lymphocyte clones. Scand J Immunol 1993;38:348-358. Pers JO, Jamin C, Lydyard PM, Charreire J, Youinou P. The H-2 haplotype regulates the distributioin of B cells into B-la, B-lb and B-2 subsets. Immunogenetics 2002;54:208-211. Pers JO, Jamin C, Youinou P, Charreire J. Role of IL- 10

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

in the distribution of B-la and B-lb cells in the B-1 cell population. Eur Cytokine Netw 2003; 14:178-185. Lalor PA, Herzenberg LA, Adams S, Stall AM. Feedback regulation of murine Ly- 1 B cell development. Eur J Immunol 1989; 19:507-513. Youinou P, MacKenzie LE, Le Masson G, Papadopoulos NM, Jouquan J, Pennec YL, Angelidis P, Katsikis PD, Moutsopoulos HM, Lydyard PM. CD5-expressing B lymphocytes in the blood and salivary glands of patients with primary Sjrgren's syndrome. J Auotimmun 1988;1:185-194. Becker H, Weber C, Storch S, Federlin K. Relationship between CD5+ lymphocytes and the activity of systemic autoimmunity. Clin Immunol Immunopathol 1990;56:219-225. Martin PJ, Hansen JA, Siadak AW, Nowinski RC. Monoclonal antibodies recognizing normal human T lymphocytes and malignant human B lymphocytes: a comparative study. J Immunol 1981;127:1920-1923. Caligaris-Cappio F, Gobbi M, Bergui L, Campana D, Lauria F, Fierro MT. B-chronic lymphocytic leukemia patients with stable benign disease show a distinctive membrane phenotype. Br J Haematol 1984;56: 655-660. Den Ottolander GL, Schuit HRE, Waayer JLM, Huibregtsen L, Hijmans W, Jansen J. Chronic Bcell leukemias: relation between morphological and immunological features. Clin Immunol Immunopathol 1985;35:92-102. Borowitz MJ, Bousvaros A, Brynes RK, Cousar JB, Crissman JD, Whitcomb CC, Kerns BJ, Byrne GE Jr. Monoclonal antibody phenotyping of B-cell nonHodgkin's lymphomas. The Southeastern Center Study Group experience. Am J Pathol 1985;121:514-521. Brrker BM, Klejman A, Youinou P, Jouquan J, Worman CP, Murphy J, MacKenzie L, Quartey-Papafio R, Blaschek M, Collins P, Lal S, Lydyard PM. Chronic lymphocytic leukemic cells secrete multispecific autoantibodies. J Autoimmun 1988;1:469-481. Hassan J, Feighery C, Bresnihan B, Whelan A. Increased CD5+ B cells in infectious mononucleosis. Br J Haematol 1990;74:375-376. Curry MP, Golden-Mason L, Nolan N, Parfrey NA, Hegarty JE, O'Farrelly C. Expansion of peripheral blood CD5+ B cells is associated with mild disease in chronic hepatitis C infection. J Hepatol 2000;32: 121-125. Curry MP, Golden-Mason L, Doherty DG, Deignan T, Norris S, Duffy M, Nolan N, Hall W, Hegarty JE, O'Farrelly C. Expansion of innate CD5 (pos) B cells expressing high levels of CD81 in hepatitis C virus infected liver. J Hepato12003;38:642-650. Dutra WO, Martins-Filho OA, Cancado JR, Pinto-Dias

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

JC, Brener Z, Freeman Junior GL, Colley DG, Gazzinelli G, Parra JC. Activated T and B lymphocytes in peripheral blood of patients with Chagas' disease. Int Immunol 1994;6:499-506. Kroese FG, de Waard R, Bos NA. B-1 ceils and their reactivity with the murine intestinal microflora. Semin Immunol 1996;8:11-18. Chen M, Aosai F, Norose K, Mun HS, Yano A. The role of anti-HSP70 autoantibody-forming V(H) 1-J(H) 1 B- 1 cells in Toxoplasma gondii-infected mice. Int Immunol 2003;15:39-47. Onah DN, Hopkins J, Luckins AG. Increase in CD5+ B cells and depression of immune responses in sheep infected with Trypanosoma evansi. Vet Immunol Immunopathol 1998;63:209-222. el-Cheikh MC, Bonomo AC, Rossi MI, Pinho M de F, Borojevic R. Experimental murine Schistosiamis mansoni: modulation of the B-1 lymphocyte distribution and phenotype expression. Immunology 1998;199: 51-62. Lankester AC, van Scjrdel GM, Cordell J, van Noessel CJ, van Lier RA. CD5 is associated with the human B cell antigen receptor complex. Eur J Immunol 1994;24: 812-816. Jamin C, Le Corre R, Lydyard PM, Youinou P. CD5+ B cells: differential capping and modulation of IgM and CD5. Scand J Immunol 1996;43:73-80. Pers JO, Jamin C, Le Corre R, Lydyard PM, Youinou P. Ligation of CD5 on resting B cells, but not on resting T cells, results in apoptosis. Eur J Immunol 1998;28: 4170-4176. Pers JO, Berthou C, Porakishvili N, Burdjanadze M, Le Calvez G, Lydyard PM, Jamin C. CD5-induced apoptosis of B cells in some patients with chronic lymphocytic leukemia. Leukemia 2002; 16:44--52. Pers JO, N6dellec S, Renaudineau Y, Berthou C, Lydyard PM, Youinou P. Role of B cell antigen receptorassociated molecules and lipid rafts in CD5-induced apoptosis of B chronic lymphocytic leukemia. Submitted. Jamin C, Le Corre R, Lydyard PM, Youinou P. AntiCD5 extends the proliferative response of human CD5+ B cells activated with anti-IgM and interleukin-2. Eur J Immunol 1996;26:57-62. Sen G, Bikah G, Venkataraman C, Bondada S. Negative regulation of antigen receptor-mediated signafing by constitutive association of CD5 with the SHP-1 protein tyrosine phosphatase in B-1 B cells. Eur J Immunol 1999;29:3319-3328. Bikah G, Carey J, Ciallella JR, Tarakhovsky A, Bondada S. CD5-mediated negative regulation of antigen receptor-induced growth signals in B-1B cells. Science 1996;274:1906-1909.

73

45. Hippen K, Tze LE, Behrens TW. CD5 maintains tolerance in anergic B cells. J Exp Med 2000;191:883-889. 46. Qiou Y, Sartiago C, Borrero M, Tedder TF, Clarke SH. Lupus-specific antiribonucleoprotein B cell tolerance in nonautoimmune mice is maintained by differentiation of B-1 and governed by B cell receptor signaling thresholds. J Immuno12001 ;166:2412-2419. 47. Sato S, Hasegawa M, Fujimoto M, Tedder TF, Takehara K. Quantitative genetic variation in CD19 expression correlates with autoimmunity. J Immunol 2000;165: 6635-6643. 48. Smith KG, Tarlington DM, Doody GM, Hibbs ML, Fearon DT. Inhibition of the B cell by CD22: a requirement for Lyn. J Exp Med 1998;187:807-811. 49. Liossis SN, Kovacs B, Dennis G, Kammer GM, Tsokos GC. B cells from patients with systemic lupus erythematosus display abnormal antigen receptor-mediated early signal transduction events. J Clin Invest 1996;98: 2549-2557. 50. Bergler W, Adam S, Gross HJ, HiSrmann K, SchwartzAlbiez R. Age-dependent altered populations of tonsillar lymphocytes. Clin Exp Immunol 1999; 116:9-18. 51. Miller RA, Gralow J. The induction of Leu-1 antigen expression in human malignant and normal B cells by phorbol myristis acetate. J Immunol 1984;133: 3408-3414.

74

52. Ying-zi C, Rabin E, Wortis HH. Treatment of murine CD5-B cells with anti-Ig, but not LPS, induces surface CD5: two B-cell activation pathways. Int Immunol 1991;3:467-476. 53. Kaplan D, Smith D, Meyerson H, Pecora N, Lewandowska K. CD5 expression by B lymphocytes and its regulation upon Epstein-Barr virus transformation. Proc Natl Acad Sci USA 2001;98:13850-13853. 54. Jamin C, Magadur G, Lamour A, MacKenzie LE, Lydyard PM, Katsikis PD, Youinou P. Cell-free CD5 in patients with rheumatoid diseases. Immunol Letter 1991;31:79-84. 55. Lu X, Axtell RC, Collawn JF, Gibson A, Justement LB, Raman C. AP2 adaptator complex-dependent internalization of CD5: differential regulation in T and B ceils. J Immuno12002; 168:5612-5620. 56. Padilla O, Calvo J, Vilh JM, Arman M, Gimferrer I, Places L, Arias MT, Pujana MA, Vives J, Lozano E Genomic organization on the human CD5 gene. Immunogenetics 2000;51:993-1001. 57. Renaudineau Y, Mageed RA, Youinou P. An alternative exon 1 of human CD5 gene regulates CD5 expression. Submitted. 58. Tarakhovsky A. Bar Mitzvah for B-1 cells: how will they grow up ? J Exp Med 1997; 185:981-984.


Endothelial Cell Autoreactivity and Infection C. Dugu6, Y. Renaudineau and E Youinou

Laboratory of Immunology, Brest University Medical School Brest, France

Owing to their permanent contact with circulating immune effectors, endothelial cells (EC) have long been suspected of being a target for antibody (Ab)-mediated assaut. In spite of incredibly wide variations in the results [ 1], it is, therefore, not surprising that anti-EC Ab (AECA) have been reported in a variety of clinical settings having in common to be accompanied by vascular changes [2]. These include not only most of the nonorgan-specific auto-immune diseases, but also numerous infectious states. AECA were first detected by indirect immunofluorescence (IIF) analysis [3, 4], and subsequently characterized using purified IgG and F(ab') 2 fragments [5]. The disorders associated with AECA has become impressively diverse [6], and sera apparently negative for this autoAb on a given cell type may turn positive if, instead, appropriate substrate cells are used [7]. Thus, AECA certainly represent a heterogeneous family of autoAb. As a corollary, the antigens (Ag) recognized by AECA may be infered to be multiple, though we have hitherto been unable to identify any of these EC Ag [8]. Furthermore, the presence of such autoAb does not imply causation, since it may follow, rather than precede EC damage. There is, nonetheless, compelling evidence that they are pathogenic. At this time, the most persuasive argument for this interpretation has come from the development of an idiotypic experimental model of systemic vasculitis [9]. A number of recent findings have indeed kindled a new debate on their pathogenicity. Concomitantly, the interest for further analysis of infection-induced AECA has been revived by the finding that infectious agents, such as Mycobacterium leprae, cytomegalovirus (CMV) and

dengue virus [10-12] colonize EC and contribute to the pathophysiology of vasculitis. Evidence has also been presented that some AECA recognize the membrane, while others react with components of their cytosol [13]. The former may be involved in the pathogenesis, but the latter would merely constitute a disease marker. Such findings support the view that AECA are also functionally heterogeneous, depending on their specificity.

1. PITFALLS IN AECA DETECTION The detection methods of AECA can be classified into those requiring fixation of the cells, and those using suspensions of EC. The group with fixation includes IIF on tissue sections, cytotoxicity of EC labelled with Cr51 or I111, radio-immunoassay and cell enzyme-linked immunosorbent assay (ELISA). There are, in fact, some major pitfalls in this method developed by Hashemi et al [14]. We have indeed reported that heterophile Ab against fetal calf serum (FCS) may be mistaken for AECA [15]. Such an interference can be eliminated simply by absorption with FCS-containing dilution buffer. Given that antinuclear Ab and rheumatoid factor may interfere with AECA, these may also be detected by an ELISA using Ab from EC lysate [16]. In addition, since false-negative AECA may serult from lack of expression of specific Ag, analysis of infectious or other AECA [ 17] dictates the use of several EC types, including microvascular EC. However, because fixation with glutaraldehyde permeabilizes the cells, autoAb to non-EC-specific cytosolic components can been detected in these assays, as reported in malaria [13].

75

Table 1. Prevalence of antiendothelial cell antibodies in infectious disease Infectious diseases

Antiendothelial cell antibodies [No positive/No tested (% positive)]

Reference

Year

13/16 (81)

[24]

1986

3/17 (18)

[28]

1992

13/72 (18)

[27] [30] [29]

1993 1997 1999

35/68 (51)

[33] [131

2001 2003

30/34 (88)

[ 13 ]

2003

Viral infections 9 Kawasaki disease 9 Puumala virus o Behqet's disease 9 Cytomegalovirus

10/23 (43)

9 Hepatitis C virus

28/69 (41)

Bacterial infections 9 Infective endocarditis 9 Leprosy

7/15 (47)

Parasitic infections 9 Malaria 9 Toxocara infection

3/5 (60)

unpublished

2003

9 Amebiasis

4/5 (80)

unpublished

2003

9 Echinococcis

4/6 (67)

unpublished

2003

9 Schistosomiasis

5/5 (100)

unpublished

2003

EC may rather be used as a suspension. Fluorescence-activated cell sorting (FACS) analysis, immunoprecipitation (IP) and Western blotting (WB) have thus been developed [18]. Human umbilical vein endothelial cells (HUVEC) remain the most widely used substrate cells. Hybrid cell lines, such as EAhy-926 and various cell lines, are occasionally employed [19]. However, needless to stress that HUVEC have a very limited use on a routine basis, due to the fact that their number is so limited (approximately 106 cells are eluted from a cord) that the procedure becomes extremely tedious. In addition, the phenotype of the cells is not stable, and EC die at the third or fourth passage. EA.hy926 cells have the unvaluable advantage of consisting of a non-limited number of cells, of which the phenotype remains the same until 50 passages or more. It may be necessary to absorb the sera with the mother epithelial cells A-549 before use. However, though "AECA" has long been coined as a designation for these autoAb, this does not necessarily mean that they recognize exquisitely EC. Numerous other human, bovine, and murine cell lines are available. With regard to the practical development of the AECA binding, the most accurate method is the

76

cell-ELISA [ 18]. It is quantitative since EC are used only once confluent. It is therefore possible to test a large number of sera at the same time. l i e a semiquantitative test, is not recommended, given that it does not permit adequate follow-up of the AECA levels. Cytotoxicity tests and radioimmunoassay are not in use any more. FACS is a technique where the cells do not adhere to a matrix, but are suspended in a buffer [ 18, 19]. It has been claimed to be the most suitable method to measure membrane-specific AECA [20], and we have shown that such autoAb detected that way predominate in leprosy [13]. Yet, its standardization has not been achieved and creating suspension of adherent cells may underestimate the expression of certain Ag [21]. The results of the ongoing programs could standardize the test and perhaps bring about more insights into the understanding of these autoAb. At present, they have to be compared with those obtained by ELISA. EC display different Ag distributions on their surface, depending on whether or not they are in contact with the solid support (subendothelial matrix or plastic), or flooding in the supernatant. AutoAb against the extracellular matrix can maskarade AECA [22], and have patho-

logical significance in vascular damage. IP and WB are too complicated to be applied on a routine basis. An additional problem is that cytoplasmic proteins can be precipitated, along with surface glycoproteins. Due to the uncertainty of such results, we have evaluated the same sera using in-house ceI1-ELISA, FACS analysis and WB completed by densitometric quantitation, and identified AECA using these three different methods [18]. We came to the conclusion that, ideally, the three methods should be applied. It is equally important to agree on the mean to express of results [23]. Given the variations in the number of cells in a well, they should not be expressed in optical densities, but in binding index, using the formula: 100 x (S - A) / (B - A), where S is the result of the disease, A the negative value and B the positive value.

2. DETECTION OF AECA

2.1. Viral Infections Despite obvious differences in their pathophysiology, similar AECA have been reported in a vast array of diseases (Table 1). For example, IgG and IgM AECA have long been described [24] in Kawasaki syndrome (KS), and claimed to be involved in the development of its acute phases [25]. Such a statement has been subsequently challenged [26], so that still very little is known about the etiology of the syndrome. In Behqet disease which is also of unknown origin but possibly triggered by a virus, AECA have been detected in 18% of the cases, found to be associated with thrombotic events, and therefore suspected to be pathogenic [27]. There is increasing evidence that AECA are associated with various viral infections. The autoAb have thus been reported in nephropathia epidemica, one of the milder forms of haemorrhagic fever caused by Puumala virus, and in three of 17 patients, as well as in four of nine and two of 19 sera from patients with influenza A and influenza B, respectively [28]. Interestingly, AECA is a common finding in hepatitis C virus (HCV) infection, but not in non-HCV chronic liver diseases [29]. It might be a risk factor for vascular rejection in CMV-infected recipients of cardiac or renal [30] or liver [31 ] allograft. In this respect, the dengue virus is another

particular case, because EC are permissive to this virus and AECA generated in the majority of the patients [32].

2.2. Other Infections AECA have also been found in under half of the sera from infective endocarditis [33], and over half of those from leprosy [13]. In the latter disease, more of the patients with the lepromatous and the borderline lepromatous that of the patients with the tuberculoid and the borderline tuberculoid forms scored positive, and the truly specific autoAbs to the membrane of EC were preferentially associated with multibacillary than with paucibacillary leprosy. Other bacteria, including Chlamydia pneumoniae and Helicobacter pylori activate EC, though, for unknown reasons, they do not promote the synthesis of AECA [34]. Finally, the cell-ELISA for these autoAb was recorded positive in 30 of 34 patients with malaria, compared with 17 of 50 local controls [13]. This baseline production in healthy African individuals is relatively high, as would be expected in an area where parasitic infection is endemic. Consistent with this view is the high level of AECA in toxocara infection (60%), amebiasis (80%), echinococas (67%) and schistosomiasis (100%).

2.3. Pathogenic Effects of AECA The pathogenicity of AECA remains uncertain. The likelihood of such an effect was first suggested by the observation that the autoAb levels fluctuate with disease activity in patients with systemic lupus erythematosus (SLE), Wegener granulomatosis (WG) and KS. The AECA test can even identify subsets of systemic sclerosis (SSc) [35], vasculitides [36] or inflammatory myopathies [37] with differing prognoses. Interestingly, the production of AECA is complicated by renal failure in SLE [38], vasculitis in rheumatoid arthritis (RA) [39] and lung fibrosis in dermatomyositis [40]. Some of them cause complement-mediated killing of EC in SLE [5], KS [25] and hemolytic-uremic syndrome [41], or induce Ab-dependent cellular cytotoxicity in WG [42]. Thrombomodulin, an EC-specific glycoprotein, is released by damage to these cells in WG and other systemic vasculitides [43]. Plasma from patients with thrombotic thrombo-

77

Figure 1. On the left, endothelial cells (EC) stained with May-Griinwald-Giemsa exhibit a typical morphological aspect of apoptosis (arrows). On the right, agarose gel electrophoresis shows EC incubated with control IgG or with apoptosis-inducing autoantibodies. These generate fragmentation of DNA. cytopenic purpura and sporadic hemolytic-uremic syndrome induces apoptosis in restricted lineages of human microsvascular EC, although the agents responsible for initiating EC injury and the exact role played by AECA is elusive [44]. Also supporting an apoptotic process in these thrombotic microangiopathies are the EC detachment from affected vessels, their appearance in the periphery [45], and the clear absence of inflammatory changes. The recent finding that one of the earliest events of a chicken model of SSc is EC apoptosis [46] may be highly relevant to this problem. Similar endothelial changes are also found in the initial phase of human generalized and local scleroderma. In addition, speculation about the mechanism of vascular conditions associated with AECA has focused on raised expression of adhesion molecules, such as E-selectin, intercellular adhesion molecule 1 and vascular cell adhesion molecule 1, by EC [47]. This enhancement, together with the production of chemotactic cytokines, e.g., intefleukin (IL)- 1~, IL6, IL-8 and monocyte chemotatic protein 1, would facilitate adhesion of leukocytes to the inflamed vessel walls, followed by their extravascular migra-

78

tion and granuloma formation. A different group of AECA has been shown to induce tissue factor in EC [48]. An other appealing possibility is that EC apoptosis is initiated by AECA. In this study, incubation of EC with AECA derived from patients with vasculitis or mouse monoclonal AECA resulted in the expression of phosphatidylserine (PS) on the surface of the cells, as established through the binding of cationic annexin V [49]. Hypoploid cell enumeration, DNA fragmentation study and optical, immunofluorescence, confocal and electron microscopy analysis confirmed apoptosis of EC (Fig. 1). In some but not all sera, a subgroup of AECA may thus be pathogenic that way. Such a complication has been described in leprosy [13] and denguevirus infection [50]. We have, therefore, addressed the issue of whether activation is a prerequisite for AECA-mediated apoptosis of EC [51 ], shown that the ability of some AECA to activate the cells was irrelevant to the nature of the underlying disorder, and established that activation does not play a role in the advent of apoptosis. We have since extended these studies and found

that AECA binding to HUVEC makes anionic phospholipids (PL) accessible to anti-[32 glycoprotein I (I3zGPI) Ab (52]. A mechanism by which some antiPL Ab (aPL) bind to EC has thus been proposed. Should PS become available, following the binding of AECA, circulating ~2GPI would attach to EC and thereby, allow the recognition of the 132GPI~L complex by autoimmune aPL. It is not yet known whether anti- I3zGPI Ab from patients with primary aPL syndrome recognize new epitopes formed after binding of the molecule to anionic structures displayed by native ~2GPI when available at increased density, as one would expect for low-affinity Ab [53]. In line with the first interpretation is the recent report by Pittoni et al that a monoclonal from an SLE patient reacts with a cryptic epitope on I]2GPI, following binding to apoptotic cells [54]. Thus, not only AECA encourage the binding of pre-existing aPL to apoptotic EC, but the PS exposure might result in de novo production of aPL. It may be argued that if AECA were essential to the production of aPL, they should be present in all patients with aPL. However, a proportion of sera contain aPL but not AECA. One possibility is that, by the time a patient is investigated for autoAb, AECA may have already disappeared, so that the serum, while aPL-positive, has become AECAnegative. As suggested by Shoenfeld [55], aPL may be infectiously-induced. Inasmuch as the infectious diseases to be associated with AECA are plenty, it is tempting to speculate on a role for these infectionrelated autoAb in the production of aPL.

4. MECHANISMS OF AECA PRODUCTION IN INFECTIOUS DISEASE 4.1. Direct Involvement of EC

Three mechanisms deserve to be considered. The first refers to molecular mimicry, as described in dengue virus infection where Ab cross-react with EC, and their binding inhibited by pretreatment of the cells with nonstructural protein 1 from the virus [32]. Human heat shock protein (HSP)-70, which is a chaperone molecule, is recognized by some autoAb from some lepromatous sera [13]. This target Ag of AECA reproduces the C-terminal half of the M. leprae HSP-70 [56]. Accordingly, it

may initiate cross-reactive autoAb, either singularly or through interaction with any chaperoned autoAg. Alternatively, AECA may represent one of many serological hallmarks of polyclonal B cell activation. This second mechanism, well established in SLE, has been demonstrated in the production of malaria AECA [13, 57]. The third possibility is the induction of cell proliferation and morphological changes through colonization of EC. One example is M. leprae, most notably in those lining epineurial and perineurial blood vessels [10]. Dengue virus-EC interaction as also been studied in depth by using differential display reverse transcriptionpolymerase chain reaction (PCR), real time PCR, and Affymetrix oligonucleotide microarrays [58]. Stricking changes in gene expression were seen after infection of HUVEC with the virus. 4.2. Indirect Involvement of EC

Several indirect mechanisms have been suspected to initiate the production of AECA. These include activation of EC by IL-lct released by epithelial cells infected with respiratory syncytial virus [59], and upregulation of CD40 expression on EC infected with CMV [60].

5. THE TARGET AG OF AECA 5.1. Cell Membrane Specificity

Early studies have excluded anti-ABO and antiHLA Ab from the AECA [61]. Specificity for membrane, that cannot be absorbed with cytosolic lysates, predominates over that for cytosolic components in leprosy [13]. WB analysis, expression bank evaluation and two-dimensional electrophoresis have revealed that calreticulin, vimentin, tubulin and HSP-70 are recognized by AECA from patients with leprosy, while numerous proteins remain unidentified. (Table 2). In SLE, which is the prototype vasculitis-associated disease, 19 bands ranging from 15 to 200 kDa were identified by van der Zee et al [62] using WB, and Ab against 38, 41 and 150 kDa proteins shown to be tightly associated with lupus nephritis. Li et al [63] reported, however, that SLE patients with nephritis, vasculitis and hypocomplement raise IgG-AECA against a 66 kDa membrane

79

Table 2. Membrane components recognized by antiendo-

thelial cell antibodies Disease Leprosy

Calreticulin, vimentin, tubulin, Heat-Shock Protein (HSP)-70

Systemic lupus erythematosus

Ribosomal P-protein

Systemic vasculitides

Triose phosphate isomerase

Systemic sclerosis

Heparan sulfate

Wegener granulomatosis 70 kDa protein (HSP-70?)

Ag, whereas a 55 kDa Ag would be the specific target for AECA in patients with thrombocytopenia, and another 18 kDa component in those with pleuritis. Other groups have demonstrated that ribosomal P protein is an EC target for autoAb. It may be involved in the pathogenesis of lupus nephritis [64]. Although the AECA epitopes vary from a given patient to another, a subgroup can be IP only by SLE sera, suggesting that the way AECA react might be specific for each disease [65]. It has also been established that in renal diseases and kidney transplantation, AECA from patients with systemic vasculitis recognize 30-35 kDa Ag. In contrast, a 28 kDa Ag has been claimed to be specific for vasculitis and to share 93% amino acid with triose phosphate isomerase [66]. Interestingly, Wheeler et al [67] have demonstrated the association of these anti-triose phosphate isomerase Ab with IgM anti-vimentin Ab in human transplantassociated coronary artery disease. The latter observation might be a clue to the concept that a fraction of AECAs enter the cells. In RA, 12 proteins, ranging from 16 to 48 kDa, have been identified by WB and IP. In those patients with RA vasculitis, IgG-AECA were as directed towards a 44 kDa EC membrane Ag [68]. This is reminiscent of the intriguing finding that a 44 kDa protein was targeted by AECA from patients with Behqet's disease when human dermal microvascular EC were used [69]. Del Papa et al [65] found that any one of five proteins in WG reacts with AECA (180, 155, 125, 38 and 25 kDa). Ab binding to a

80

43 kDa as yet unknown component in the cytosol and the nucleus of human microvascular renal EC have also been identified in hemolytic uremic syndrome and thrombotic thrombocytopenic purpura [70]. In heparin-induced thrombocytopenia, some circulating Ab react with platelet factor 4 complexed with heparin, and others with heparan sulfate incorporated into the membrane of EC. Thus, it is not unreasonable to assume that they may play an active role in the development of thrombosis [71]. Similar reactions have been reported to occur [72] in connective tissue diseases associated with vasculitis. Finally, a 18 kDa EC membrane antigen was shown important for autoAb from patients with SSc. It should be stressed that the related AECA are associated with the CREST variant of this disorder. Finally, the possibility exists that murine monoclonal AECA produced by idiotypic manipulation with human Ab recognizes HSP-70 [73].

6. OTHER AG POSSIBLY R E C O G N I Z E D BY AECA

There appears that most of the AECA-positive malarian sera react with the cytosol but not with the membrane of EC [13]. This is substantiated (Fig. 2) by our finding that sera negative in the FACS analysis turn positive, once EC have been permeabilized with saponin. Clearly the target Ag of these pseudoAECA are specific neither for malaria nor for EC. The majority of AECA that bind to Ag seem to be EC membrane proteins. Yet the demonstration that extensive washes of radiolabeled preparations with high molar buffers result in the reduction of AECA from SLE sera indicates that some of these Ab are also able to recognize non-constitutive proteins. This is further supported by the description of monoclonal and polyclonal anti-DNA Ab binding in vitro to EC through DNA or DNA/histone complexes attached to the cell membrane [74]. To conclude, there is compelling evidence that more and more specific proteins and epitopes recognized by those apoptosis-inducing AECA are identified. Still, clarification of the function of target Ag is required to achieve a better understanding of the effects of AECA on associated infectious diseases.

Leprosy After

}, |l

i

s

Malaria '

il

After

Figure 2. Detection of antiendothelial cell (EC) antibodies (Ab) from lepromatous (top) and malarian (bottom) sera using flow cytometry analysis. The serum from the patient with leprosy is already positive before permeabilization of the cells since Ab bind to the membrane, whereas that from the patient with malaria needs incubation of EC with saponin to encounter cytosolic antigens.

ACKNOWLEDGEMENTS The studies m e n t i o n e d in this review were supported by the Conseil R6gional de Bretagne and the C o m m u n a u t 6 Urbaine de Brest. The secretarial assistance of S imone Forest is appreciated.

5.

6.

7.

REFERENCES 1.

2. 3.

4.

Youinou P, Meroni PL, Khamashta MA, Shoenfeld Y. A need for standardization of the anti-endothelial cell antibody test. Immunol Today 1995; 16:363-364. Youinou P. Antiendothelial cell autoantibodies and disease. Intern Med Clin Lab 1995;3:7-10. Lindqvist KJ, Osterland CK. Human antibodies to vascular endothelium. Clin Exp Immunol 1970;9: 753-760. Tan EM, Pearson CM. Rheumatic disease sera reactive with capillaries in the mouse kidney. J Clin Invest

8.

9.

1972;15:23-28. Cines DB, Lyss AP, Reeber M, Bina M, DeHoratius RJ. Presence of complement-fixing anti-endothelial cell antibodies in systemic lupus erythematosus. J Clin Invest 1984;73:611-625. Meroni PL, Youinou P. Endothelial cell antibodies. In: Peter JB, Shoenfeld Y, eds. Autoantibodies. Amsterdam: Elsevier, 1996;245-252. Praprotnik S, Blank M, Meroni PL, Rozman B, Eldor A, Shoenfeld Y. Classification of anti-endothelial cell antibodies into antibodies against microvascular and macrovascular endothelial cells. Arthritis Rheum 2001 ;44:1484-1494. Castillo S, Revelen R, Bordron A, Renaudineau Y, Dueymes M, Youinou P. Antiendothelial cell reactivity, the unresolved enigma... Int J Immunopathol Pharmacol 2001;14:109-118. Damianovich M, Gilburd B, Georges J, Del Papa N, Afek A, Goldberg L, Kopolovic Y, Roth D, Barkai G, Meroni PL, Shoenfeld Y. Pathogenic role of antiendothelial cell antibodies in vasculitis: an idiotypic experi-

81

10. 11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

82

mental model. J Immunol 1996;156:4946-4951. Fite GL. The vascular lesions of leprosy. Int J Lep 1941 ;9:193-202. Ho DD, Rota TR, Andrews CA, Hirsch MS. Replication of human cytomegalovirus in endothelial cells. J Infect Dis 1984; 150:956-957. Funahara Y, Ogawa K, Fujita N, Okuno Y. Three possible triggers to induce thrombocytopenia in dengue virus infection. Southeast Asian J Trop Med Public Health 1987;18:351-355. Dugu6 C, Perraut R, Youinou P, Renaudineau Y. Effects of antiendothelial cell antibodies in leprosy and malaria. Infect Immun 2003; in press. Hashemi S, Smith CD, Izaguirre CA. Anti-endothelial cell antibodies: detection and characterization using a cellular enzyme-linked immunosorbent assay. J Lab Clin Med 1987;109:434-440. R6v61en R, Bordron A, Dueymes M, Youinou P, Arvieux J. False positivity in a cyto-ELISA for antiendothelial cell antibodies caused by heterophile antibodies to bovine serum proteins. Clin Chem 2000;46: 273-278. Drouet C, Nissou MF, Ponard D, Arvieux J, DumestrePerand C, Gaudin P, Imbert B, Massot C, Sarrot-Reynauld E Detection of antiendothelial cell antibodies by an enzyme-linked immunosorbent assay using antigens from cell lysate: minimal interference with antinuclear antibodies and rheumatoid factors. Clin Diagn Lab Immuno12003; 10:934-939. Renaudineau Y, R6v61en R, Levy Y, Salojin K, Gilburd B, Shoenfeld Y, Youinou P. Anti-endothelial cell antibodies in systemic sclerosis. Clin Diagn Lab Immunol 1999;6:156-160. REv6len R, d'Arbonneau E Guillevin L, Bordron A, Youinou P, Dueymes M. Comparaison of cell-ELISA, flow cytometry and Western blotting for the detection of antiendothelial cell antibodies. Clin Exp Rheumtol 2002;20:19-26. Edgell CJS, McDonald CC, Graham JB. Permanent cell line expressing human factor VIII-related antigen established by hybridization. Proc Natl Acad Sci USA 1983 ;80:3734-3737. Westphal JR, Boerbooms AMT, Schalkwijk CJM, Kwast H, de Weijert M, Jacobs C, Vierwinden G, Ruiter DH, van de Putte LBA, de Waal RMW. Anti-EC antibodies in sera of patients with autoimmune diseases: comparison between ELISA and FACS analysis. Clin Exp Immunol 1994;96:444-449. Forsyth KD, Talbot V. Assessment of endothelial immunophenotype: limitation of flow cytometric analysis. J Immunol Methods 1991;144:93-99. Direskeneli H, D'Cruz D, Khamashta MA, Hughes GRV. Autoantibodies agains endothelial cells, extra-

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

cellular matrix and humn collagen type IV in patients with systemic vasculitis. Clin Immunol Immunopathol 1994;70:206-210. Rosenbaum J, Pottinger BE, Woo P, Black CM, Loizou S, Byron MA, Pearson JD. Measurement and characterization of circulating anti-endothelial cell IgG in connective tissue diseases. Clin Exp Immunol 1988;72: 450--456. Leung DYM, Collins T, Lapierre LA, Geha RS, Pober JS. Immunoglobulin M antibodies in the acute phase of Kawasaki syndrome lyse cultured vascular endothelial cells stimulated by gamma interferon. J Clin Invest 1986;77:1428-1435. Leung DYM, Geha RS, Newburger JW, Bums JC, Fiers W, Lapierre LA, Pober JA. Two monokines, intefleukin 1 and tumor necrosis factor, render cultured vascular endothelial cells susceptible to lysis by antibodies circulating during Kawasaki syndrome. J Exp Med 1986;164:1958-1972. Nash MC, Shah V, Reader JA, Dillon MJ. Anti-neutrophil cytoplasmic antibodies and anti-endothelial cell antibodies are not increased in Kawasaki disease. Brit J Rheumatol 1995;34:882-887. Aydintug AO, Tokg6z G, D'Cruz DP, Giirler A, Cervera R, Diizgiin N, Atmaca LS, Khamashta MA, Hughes GRV. Antibodies to endothelial cells in patients with Behqet's disease. Clin Immunol Immunopathol 1993;67:157-162. Wangel AG, Temonen M, Brummer-Korvenkontio M, Vaheri A. Anti-endothelial cell antibodies in nephropathia epidemica and other viral diseases. Clin Exp Immunol 1992;90:13-17. Cacoub P, Ghillani P, Revelen R, Thibault V, Calvez V, Charlotte F, Musset L, Youinou P, Piette JC. Anti-endothelial cell autoantibodies in hepatitits C virus mixed cryoglobulinemia. J Hepatol 1999;31:598-603. Toyoda M, Galfayan K, Galera OA, Petrosian A, Czer LS, Jordan SC. Cytomegalovirus infection induces antiendothelial cell antibodies in cardiac and renal allograft recipients. Transplant Immunol 1997;5:104-111. Varani S, Muratori L, De Rewo N, Vivarelli M, Lazzarotto T, Gabrielli L, B ianchi FB, Bellusci R, Landini MP. Autoantibody appearance in cytomegalovirusinfected liver transplant recipients: correlation with antigenema. J Med Vir 2002;66:56-62. Lin CF, Lei HY, Shiau AL, Liu CC, Liu HS, Yeh TM, Chen SM, Lin YS. Antibodies from dengue patient sera cross-react with endothelial cells and induce damage. J Med Vir 2003;69:82-90. Portig I, Beck V, Pankuweit S, Maisch B. Antiendothelial antibodies in sera of patients with infective endocarditis. Basis Res Cardiol 2001;96:75-81. Schumacher A, Seljeflot I, Lerkerod AB, Sommervoll

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

L, Ottersdat JE, Arnesen H. Does infection with Chlamydia pneumoniae and/or Helicobacter pylori increases the expression of endothelial cell adhesion molecules in humans? Clin Microbiol Infect 2002;8:564-661. Salojin KV, Le Tonqu~ze M, Saraux A, Nassonov EL, Dueymes M, Piette JC, Youinou P. Antiendothelial cell antibodies: useful markers of systemic sclerosis. Am J Med 1997;102:178-185. Salojin KV, Le Tonqu~ze M, Nassonov EL, B louch MT, Baranov AA, Saraux A, Guilevin L, Fiessinger JN, Piette JC, Youinou P. Anti-endothelial cell antibodies in patients with various forms of vasculitis. Clin Exp Rheumatol 1996;14:163-169. Salojin KV, Bordron A, Nassonov EL, Shtutman VZ, Guseva NG, Baranov AA, Targoff IN, Youinou P. Antiendothelial cell antibody, thrombomodulin, and von Willebrand factor in idiopathic inflammatory myopathies. Clin Diagn Lab Immuno 1997;4:519-521. D'Cruz DP, Haussiau FA, Ramirez G, Baguley E, McCutcheon J, Vianna J, Haga HJ, Swana GT, Khamashta MA, Taylor JC, Davies DR, Hughes GRV. Antibodies to endothelial cells in systemic lupus erythematosus: a potential marker for nephritis and vasculitis. Clin Exp Immunol 1991;85:254-261. Heurkens AHM, Hiemstra PS, Lafeber GJM, Daha MR, Breedveld FC. Anti-endothelial cell antibodies in patients with rheumatoid arthritis complicated by vasculitis. Clin Exp Immunol 1989;78:7-12. Cervera R, Ramirez G, Fernandez-Sola J, D'Cruz D, Casademont J, Grau JM, Asherson RA, Khamashta MA, Urbano-Marquez A, Hughes GRV. Antibodies to endothelial cells in dermatomyositis: association with interstitial lung disease. Br Med J 1991 ;302:880-881. Leung DYM, Moake JL, Havens PL, Kim M, Pober JS. Lytic anti-endothelial cell antibodies in haemolyticuraemic syndrome. Lancet 1998;333:183-186. Del Papa N, Meroni PL, BarceUini W, Sinico A, Radice A, Tincani A, D'Cruz D, Nicoletti F, Borghi MO, Khamashta MA, Hughes GRV, Balestrieri G. Antibodies to endothelial cells in primary vasculitides mediate in vitro endothelial cytotoxicity in the presence of normal peripheral blood mononuclear cells. Clin Immunol Immunopathol 1992;63:267-274. Boehme MWJ, Schmitt WH, Youinou P, Stremmel WR, Gross WL. Clinical relevance of elevated serum thrombomodulin and soluble E-selectin in patients with Wegener's granulomatosis and other systemic vasculitides. Am J Med 1996;101:387-394. Raife TJ, Atkison B, Aster RH, McFarland JG. Gottschall JL. Minimal evidence of platelet and endothelial cell reactive antibodies in thrombotic thrombocytopenic purpura. Am J Hematol 1999;63:267-274. George F, Poncelet P, Laurent JC, Massot O, Arnoux

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

D, Lequeux N, Ambrosi P, Chichepotiche C, Sampol J. Cytofluorometric detection of human endothelial cells in whole blood using S-Endo 1 monoclonal antibody. J Immunol Methods 1991;139:65-75. Sgonc R, Gruschwitz MS, Dietrich H, Recheis H, Gershwin ME, Wick G. Endothelial cell apoptosis is a primary pathogenetic even underlying skin lesions in avian and human scleroderma. J Clin Invest 1996;98: 785-792. Del Papa N, Guidali L, Sironi M, Shoenfeld Y, Mantovani A, Tincani A, Balestrieri G, Radice A, Sinico RA, Meroni PL. Anti-endothelial cell IgG antibodies form patients with Wegener's granulomatosis bind to human endothelial cell in vitro and induce adhesion molecule expression and cytokine secretion. Arthritis Rheum 1996;39:758-792. Kornberg A, Renaudineau Y, Blank M, Youinou E Shoenfeld Y. Anti-beta 2-glycoprotein I antibodies and anti-endothelial cell antibodies induce tissue factor in endothelial cells. IMA 2000;2 (supplement):27-31. Bordron A, Dueymes M, Levy Y, Jamin C, Leroy JP, Piette JC, Shoenfeld Y, Youinou E The binding of some antiendothelial cell antibodies induces endothelial cell apoptosis. J Clin Invest 1998; 101:2029-2035. Avirutnan P, Malasit P, Seliger B, Bhakdi S, Hussmann M. Dengue virus infection of human endothelial cell leads to chemokine production complement activation and apoptosis. J Immunol 1998;161:6338-6346. Bordron A, R6v61en R, d'Arbonneau E Dueymes M, Renaudineau Y, Jamin C, Youinou E Functional heterogenetity of anti-endothelial antibodies. Clin Exp Immuno12001;124"492-501. Bordron A, Dueymes M, Levy Y, Jamin C, Leroy JP, Piette JC, Shoenfeld Y, Youinou E Antiendothelial cell antibody binding makes negatively-charged phospholipids accessible to antiphospholipid antibodies. Arthritis Rheum 1998;41:1738-1747. Ronda N, Haury M, Nobrega A, Kaveri SV, Coutinho A, Kazatchkine MD. Analysis of natural and diseaseassociated autoantibody repertoires: anti-endothelial cell IgG autoantibody activity in the serum of healthy individuals and patients with systemic lupus erythematosus. In Immunol 1994;6:1651-1660. Pittoni V, Ravirajan CT, Donohoe S, Machin SJ, Mackie IJ, Lydyard PM, Isenberg DA. Human monoclonal antiphospholipid antibodies bind to membrane phospholipids and a cryptic dpitope of ~2-glycoprotein I on apoptotic cells [abstract]. Arthritis Rheum 1998;41 suppl 12:$166. Shoenfeld Y. Etiology and pathogenetic mechanisms of the antiphospholipid syndrome unraveled. Trends Immuno12003;24:2-4. Janson AA, Klatser PR, van der Zee R, Cornelisse

83

57.

58.

59.

60.

61.

62.

63.

64.

65.

84

YE, de Vries F, Thole JE, Ottenholf TH. A systematic molecular analysis of the T cell-stimulating antigens from Mycobacterium leprae with T cell clones of leprosy patients. Identification of a novel M. leprae HSP70 fragment by M. leprae-specific T cells. J Immunol 1991;147:3530-3537. Adebajo AO, Charles P, Maini RN, Hazleman BL. Autoantibodies in malaria, tubercuolosis and hepatitis B in a West African population. Clin Exp Immunol 1993;92:73-76. Warke RV, Xhaja K, Martin KJ, Fournier MF, Shaw SK, Brizuela N, de Bosch N, Lapointe D, Ennis FA, Rothman AL, Bosch I. Dengue virus induces novel changes in gene expression of human umbilical vein endothelial cells. J Viro12003;77:11822-11832. Chang CH, Huang Y, Anderson R. Activation of vascular endothelial cells by IL-1 alpha released by epithelial cells infected with respiratory syncitial virus. Cell Immuno12003;221:37-41. Maisch T, Kropff B, Sinzger C, Mach M. Upregulation of CD40 expression on endothelial cells infected with human cytomegalovirus. J Virol 2002;76:1280312812. Ferraro G, Meroni PL, Tincani A, Sinico A, Barcellini W, Radice A, Gregorini G, Froldi M, Borghi MO, Balestrieri. Anti-endothelial cell antibodies in patients with Wegener's granulomatosis and micropolyarteritis. Clin Exp Immunol 1990;79:47-53. Van der Zee JM, Heurkens AH, van der Voort EA, Daha MR, Breedveld FC. Characteriztion of anti-endothelial antibodies in patients with rheumatoid arthritis complicated by vasculitis. Clin Exp Rheumatol 1991;9: 589-594. Li JS, Liu MF, Lei HY. Characterization anti-endothelial cell antibodies in patients with systemic lupus erythematosus: a potential marker for disease activity. Clin Immunol Immunopathol 1996;79:211-216. Reichlin M, Wolfson-Reichlin M. Evidence for the participation of antiribosomal P antibodies in lupus nephritis. Arthritis Rheum 1999;42:2728-2729. Del Papa N, Conforti G, Gambini D, La Rosa J, Tincani A, D'Cruz D, Khamashta M, Hughes GR, Balestrieri G, Meroni PL. Characterization of the endothelial surface proteins recognized by anti-endothelial antibodies in

66.

67.

68.

69.

70.

71.

72.

73.

74.

primary and secondary autoimmune vasculitis. Clin Immunol Immunopathol 1994;70:211-216. Moriya S, Khouri NA, Cameron JS, Frampton G. Preliminary report of a vasculitis associated 28 kD endothelial cell autoantigen which shares a 93% identity with triose phosphate isomerase [abstract]. Lupus 1995;4:99. Wheeler CH, Collins A, Dunn MJ, Crisp SJ, Yacoub MIJ, Rose MJ. Characterization of endothelial antigens associated with transplant-associate coronary artery disease. J Heart Lung Transplant 1995;14:188-197. Heurkens AH, Hiemstra PS, Lafber GJ, Daha MR, Breedveld FC. Antiendothelial cell antibodies in patients with rheumatoid arthritis complicated by vasculitis. Clin Exp Immunol 1989;78:7-12. Lee KH, Bang D, Choi ES, Chun WH, Lee ES, Lee E. Presence of circulating antibodies to a disease-specific antiben on cultured human dermal microvascular endothelial cells in patients with Behqet's disease. Arch Dermatol Res 1999;291:374-381. Koenig DWL, Daniel TO. A western blot assay detects autoantibodies to cryptic endothelial antigens in thrombotic microangiopathies. J Clin Immunol 1993;13: 204-211. Visenti GP, Ford SE, Scott JP, Aster RH. Antibodies from patients with heparin-induced thrombocytopenia/ thrombosis are specific for platelet factor 4 complexed with heparin or bound to endothelial cells. J Clin Invest 1994;93:31-38. Renaudineau Y, R6v61en R, Bordron A, Mottier D, Youinou P, Le Corre R. Two populations of endothelial cell antibodies cross-react with heparin. Lupus 1998;7: 56-94. Levy Y, Gilburd B, George J, Del Papa N, Mallone R, Damianovich M, Blank M, Radice A, Renaudineau Y, Youinou P, Wiik A, Malavasi F, Meroni PL, Shoenfeld Y. Characterization of murine monoclonal anti-endothelial cell antibodies (AECA) produced by idiotypic manipulation with human AECA. Int Immunol 1998; 10:861-868. Chan TM, Frampton G, Staines NA, Hobby P, Perry GJ, Cameron JS. Different mechanisms by which anti-DNA MoAbs bind to human EC and glomerular mesangial cells. Clin Exp Immunol 1992;88:68-74.


Induction of Autoimmunity by Adjuvant Hydrocarbons Kindra M. Kelly, Yoshiki Kuroda, Dina C. Nacionales, Jun Akaogi, Minoru Satoh and Westley H. Reeves

Division of Rheumatology & Clinical Immunology, Center for Systemic Autoimmune Diseases, University of Florida, Gainesville, FL, USA

Although susceptibility to systemic lupus erythematosus (SLE) is genetically determined [1 ], environmental factors such as ultraviolet radiation, chemicals, or infections are likely to play an important role in triggering the disease in susceptible individuals. In view of the diversity of environmental exposures in human populations, animal models afford the best opportunity to identify exogenous triggers of lupus. We have reported that a lupus-like disease with disease-specific autoantibodies and nephritis develops in non-autoimmune prone mice treated with pristane, a hydrocarbon derived from the metabolism of chlorophyll [2, 3]. Some strains also develop an erosive and destructive arthritis reminiscent of rheumatoid arthritis [4]. BALB/c, C57BL/6, and nearly all other immunocompetent mice are susceptible to pristane-induced lupus, but autoantibody production, the severity of renal disease, and the development of arthritis exhibit strain-to-strain variation [5]. More recently, it has become clear that other hydrocarbons, notably the mineral oil Bayol F and the endogenous hydrocarbon squalene, also can induce lupus-like disease in mice [6]. These substances share the capacity to serve as immunological adjuvants, defined as "substances used in combination with a specific antigen that produce more immunity than the antigen alone" [7]. The induction of murine lupus by immunological adjuvants is significant for two reasons. First, it provides a model for the interaction of environmental triggers with the genetic background in systemic autoimmunity and secondly, it raises the possibility that adjuvant hydrocarbons might trigger autoimmune disease in susceptible humans. Consistent with that possibility, mineral oil and hydrocarbon adjuvants are known to

induce inflammatory disease in humans, including lipoid pneumonia, granulomas, and synovitis [8-10]. This review summarizes the current state of understanding about the immunological effects of adjuvants and our recent work on the induction of autoimmunity by these materials.

1. IMMUNOLOGICAL BASIS OF ADJUVANTICITY

Although many early vaccines consisted of live, attenuated intact microorganisms or heat-killed intact organisms, the use of these vaccines is limited by difficulties culturing certain organisms and by adverse effects, such as the induction of disease in immunocompromised hosts or unacceptable inflammatory reactions. This has led to the development of more antigenically restricted vaccines, such as viral subunit vaccines. However, it has been long recognized that many protein antigens are, by themselves, poorly immunogenic in comparison with intact microorganisms, e.g. viruses or bacteria. Accordingly, adjuvants have been added to these antigens to boost the immune responses. Adjuvants used in humans must be selected carefully so as to enhance immune responsiveness sufficiently without causing undue toxicity. At present, few effective adjuvants are considered safe for use in humans. 1.1. Mineral Salts

Alum, the only adjuvant currently approved in the United States for human vaccines, is a relatively weak adjuvant and a poor inducer of cell-mediated

87

Table 1. Some adjuvants used in human or veterinary vaccines Adjuvant

Potency

Status

Advantages/disadvantages

Aluminum-based mineral salts (alum)~

Weak

FDA approved for human use.

Safe; poor enhancement of cell-mediated immunity; can induce IgE responses and allergic reactions

MF59 (Squalene)b

Moderate Licensed in Europe. In clinical trials for influenza, HIV, herpes simplex, CMV, and hepatitis B vaccines.

More potent than alum in enhancing cell mediated immunity. Safe and well tolerated in humans.

Incomplete Freund' s adjuvant (mineral oil)b

Moderate Used in human influenza vaccine in the 1940s. No longer used in human vaccines, but used commonly in veterinary vaccines,

More potent than alum. Human use discontinued due to formation of granulomas and abscesses and because of induction of plasmacytomas in mice.

Complete Freund' s High adjuvant (mineral oil plus heat killed mycobacteria)b'~

Limited use in animals. Not approved in humans

Too toxic to use in humans (causes severe granulomatous inflammation, sterile abscesses, pain and fever)

Muramyl dipeptide~

Moderate Limited use in animals

Good inducer of both humoral and cellmediated immunity. Unacceptable for human use due to fever.

CpG DNA~

Moderate In clinical trials

Good inducer of both humoral and cellmediated immunity. So far appears to be safe and well tolerated.

Lipopolysaccharide c

Moderate Not approved for human use (but whole organism vaccines for typhoid, cholera, and pertussis contain substantial amounts of LPS)

Good inducer of both humoral and cellmediated immunity. Too toxic for human use (fever).

aMineral salt. bOil emulsion. Clmmunostimulatory microbial product (Toll receptor ligand).

immunity (Table 1). Alum and other mineral salts are thought to enhance immune responses mainly by binding antigen and acting as a slow-release depot for the antigen, thus prolonging antigen presentation [ 11].

1.2. Oil Emulsions In contrast to the mineral salts, oil adjuvants stimulate cellular as well as humoral immunity. A variety of hydrocarbon oils exhibit adjuvant properties [ 11, 12] (Table 1). Le Moignic and Pinoy discovered the adjuvant effect of mineral oil in 1916. Freund subsequently discovered that the potency could be enhanced significantly by adding heat-killed mycobacteria. Incomplete Freund's adjuvant (IFA) consists of the mineral oil Bayol F plus an emulsifier

88

(Arlacel A), whereas complete Freund's adjuvant (CFA) contains heat-killed mycobacteria in IFA [13]. IFA was used in human influenza vaccines up until the early 1960s when its use was discontinued due to the occurrence of granulomatous inflammatory reactions and sterile abscesses in some individuals, and in light of evidence that mineral oil can induce plasmacytomas in mice [14]. Despite these concerns, there generally were few side effects and in one 16-18-year follow-up study of 18,000 military recruits who received influenza vaccine in the mineral oil Draceol, there was no evidence of an increased incidence of either neoplasia or systemic autoimmune disease [ 15]. Besides mineral oil, a number of other hydrocarbon oils, including squalene (MF59) and pristane (2,6,10,14-tetramethylpentadecane), have adjuvant

effects (Table 1). Straight chain hydrocarbons containing 15-20 carbons (C15-20) can substitute effectively for standard mineral oil (a complex mixture of hydrocarbons) in Freund's adjuvants, at least for inducing experimental autoimmune encephalomyelitis, whereas longer and shorter carbon chains are ineffective [ 16]. The threshold for adjuvanticity is a chain length of 12 carbons (C 12) [ 17]. The adjuvant effect appears to be related in some manner to inflammation. Inflammatory responses to hydrocarbon oils have been studied in experimental models employing turpentine, alkanes, and a variety of other hydrocarbons [ 16, 18, 19]. Nevertheless, although all commonly used adjuvant oils cause inflammation, short-chain alkanes are intensely inflammatory but are poor adjuvants, suggesting that there is not a simple relationship between adjuvanticity and the strength of the inflammatory response. Recent studies have shed some light on the question of how hydrocarbon oils exert their adjuvant effects [20]. Squalene (MF59) does not appear to facilitate the transport of antigen nor does it have a depot effect, as previously supposed [21]. Rather, the oil appears to be internalized by macrophages, which are transported to regional lymph nodes where they undergo apoptotic cell death and are taken up by dendritic cells [20, 22]. These or other antigen-presenting cells exposed to the oil probably are responsible for the subsequent T and B cell activation seen in squalene-treated mice. Despite these conceptual advances, it remains uncertain precisely how hydrocarbon oils cause inflammation.

1.3. lmmunostimulatory Adjuvants Innate immunity mediated by antigen presenting cells tightly controls the activation of antigen-specific T and B-lymphocytes. A number of immunostimulatory molecules of microbial origin have been used as vaccine adjuvants (Table 1). Some of these substances, such as lipopolysaccharide (LPS), bacterial DNA containing unmethylated CpG motifs, and muramyl dipeptide, are ligands for Toll-like receptors (TLRs) (Table 2), whereas others are the cytokines (e.g. IL-12, GM-CSF) produced upon engagement of TLRs. Like oil emulsion adjuvants, immunostimulatory adjuvants effectively stimulate both cellular and humoral immune responses. Most of these substances are too toxic to use as human

Table 2. Stimulation of IFNot/~ by Toll-like receptors Toll-like Ligands receptor

MyD88 MyD88 dependent independent pathway pathway

TLR3

Doublestranded RNA

Yes

Yes

TLR4

Lipopolysaccharide, Yes Lipoteichoic acid, Taxol, RSV F protein, HSP60

Yes

TLR7

Imidazoquinoline Yes compounds (imiquimod; R-848)

No

TLR9

CpG DNA

No

Yes

vaccine adjuvants, though it should be noted that heat-killed intact organisms are replete with TLR ligands. CpG DNA is perhaps the most promising of the immunostimulatory adjuvants, and currently is undergoing human trials [23, 24]. Antigen presenting cells, natural killer cells, complement, and type I interferons (IFNs) constitute the innate immune system [25, 26]. Dendritic cells (DCs) link innate immunity with the adaptive immune response [25, 27]. Antigens picked up by DCs from apoptotic cells are transported to the lymph nodes [25, 28, 29]. When antigens are presented to T cells by immature DCs, the outcome is usually tolerance [30, 31 ]. In contrast, when DCs are confronted with a "dangerous" foreign antigen [32], they mature and express higher levels of costimulatory molecules (e.g. CD86) and MHC class II [25]. Antigen presenting cells sense "danger" via pattern receptors such as the TLRs. The immunostimulatory adjuvants LPS, lipoarabinomannan (muramyl dipeptide), and CpG DNA signal via TLR 4, TLR2, and TLR9, respectively [33]. Moreover, whole organism vaccines for typhoid, cholera, and pertussis contain substantial amounts of the TLR4 ligand LPS. The adapter protein MyD88 is a key element of the signaling pathway for many of the TLRs, including TLR4 and TLR9 [33, 34]. Signal transduction through MyD88 recruits the IL-1 receptor associated kinase (IRAK) and leads to activation of the transcription factors NFw_B and MAP kinase. TLR4 also signals via a MyD88 independ-

89

ent pathway, which activates the transcription factor interferon regulatory factor 3 (IRF-3) [35]. NFrd3 and IRF-3, in turn, activate several "primary viral response genes" including IFN~I, IP-10, RANTES and others [36]. The interaction of IFNI~ with the type I IFN receptor activates the transcription of an additional set of "secondary viral response genes", further enhancing cellular defenses. TLRs that efficiently stimulate type I IFN production include TLR3, TLR4, TLR7, and TLR9 [37]. Recent studies strongly suggest that type I IFNs (IFNo~, ~, to) are important mediators of the "adjuvant effect" [38].

2. IMPORTANCE OF TYPE I INTERFERONS FOR ADJUVANTICITY

Type I IFNs were found in 1957 to possess potent anti-viral activity. Since then, much has been leamed about their regulation and mode of action. Type I IFNs include up to 18 IFNct genes and pseudogenes, one IFN~ and one IFNto gene, all of which are located in the IFN gene cluster on the short arm of human chromosome 9 [39, 40]. The Type I IFNs are active as monomers and bind a specific receptor complex composed of two subunits, IFNAR1 and IFNAR2 [41]. The Jak non-receptor tyrosine kinases Tyk2 and Jakl associate with IFNAR1 and IFNAR2. IFN-receptor interactions cause reciprocal transphosphorylation of Jaks leading to receptor phosphorylation and recruitment and phosphorylation of STAT1 and STAT2. This causes activation of IFN regulatory factors (IRFs), transcription factors that induce the expression of IFN regulated proteins. Type I IFNs are a key component of the cellular response to viral infection. Working in part through the extracellular TLR3 molecule, viral double stranded (ds) RNA activates several intracellular kinases including the dsRNA-dependent protein kinase (PKR), culminating in IFN expression [42]. IFNs can activate PKR in a positive feedback loop, and induce other antiviral proteins that amplify the antiviral response. A variety of other stimuli are now known to induce IFN~I5 production including bacterial lipopolysaccharide (LPS), and bacterial CpG DNA working through TLR4 and TLR9 respectively. IFN inducible genes mediate various effector

90

pathways, including PKR, 2'5' oligoadenylate synthase (OAS), the Mx proteins, TRAIL, caspases, IRFs, and other proteins [43]. PKR can inhibit translation by phosphorylating Eif-2t~ and activates inflammation by causing nuclear translocation of NFrd3. OAS induces mRNA degradation, the Mx proteins inhibit viral replication, and TRAIL inhibits viral infectivity. IRFs and IRSs are involved in transduction of IFN signals. 2.1. Adjuvanticity is Associated with IFN Production

Type I IFNs have immunomodulatory effects, including enhancement of class II MHC expression on APCs and promotion of DC maturation and survival [38, 44]. IFNt~ stimulation increases the expression of BLyS and APRIL, promoting CD40-independent immunoglobulin class switching as well as plasma cell differentiation [45]. IFNtx also promotes TH1 responses [44] and the survival of activated T cells [46]. Its effects on B cells are more diverse. B cell lymphopoiesis is inhibited by Type I IFNs through a mechanism involving down-regulation of Bcl-2 and apoptosis [47], whereas in other situations, B cell survival is prolonged [48] and the maturation of B cells into plasmablasts facilitated [49]. The production of polyclonal IgG in response to influenza virus is abrogated when plasmacytoid dendritic cells, the primary Type I IFN producing cells, are depleted. Plasmacytoid dendritic cells drive CD40L activated B cells to mature into plasmablasts, which subsequently undergo IL-6-mediated differentiation into antibody secreting cells (see below). Moreover, Type I IFNs influence the balance of immunoglobulin isotypes produced in response to polyclonal [50] or antigen-specific [38] immune stimulation. Production of Type I IFNs is stimulated by several TLR ligands, including double stranded RNA, LPS, and CpG DNA [36, 51, 52]. Since these molecules also are potent adjuvants, there is considerable interest in the immunomodulatory effects of Type I IFNs. It has been shown that Type I IFNs augment antigen-specific immunoglobulin production following immunization with soluble antigen and are required for memory B cell responses [38]. IFNtx is as potent an adjuvant as IFA, an effect mediated by its action on dendritic cell maturation. In conclusion, there is increasing evidence that Type I IFNs are critical

mediators of adjuvanticity, a fact that may have considerable implications for the induction of autoimmunity by hydrocarbon adjuvants.

Table 3. Susceptibility to pristane-induced lupus BALB/c DBA/1 C57BL/6 B10.S Anti-RNP/Sm

55%

83%

24%

5%

3. INDUCTION OF LUPUS AUTOANTIBODIES BY HYDROCARBON ADJUVANTS

Anti-ribosomal P

0

0

16%

62%

Anti-dsDNA

38%

N/A

0

0

Anti-NF90/NF45

0

0

26%

33%

During the course of generating monoclonal antibody-enriched ascitic fluid, we found unexpectedly that the intraperitoneal injection of pristane (2,6,10,14-tetramethylpentadecane, 0.5 ml i.p.) in BALB/c and other non-autoimmune strains of mice results in the production of autoantibodies characteristic of SLE as well as immune complex-mediated glomerulonephritis resembling lupus nephritis [2, 3]. More recently, it has become apparent that certain other hydrocarbons have the same effect, and that the ability to induce lupus autoantibodies correlates to some degree with adjuvanticity and cytokine production [6]. Remarkably, although various hydrocarbons may be more or less likely to induce lupus like disease, they all induce a similar spectrum of lupus-associated autoantibodies suggesting that their mechanisms of action are similar. Following intraperitoneal injection of pristane, IFA, squalene, or hexadecane, BALB/c and most other immunocompetent mice develop high levels of antinuclear antibodies [2, 3, 6] (Y Kuroda et al, unpublished data). The specificities include autoantibodies thought to be pathognomonic of SLE, such as anti-Sm, anti-dsDNA, and anti-ribosomal P [5, 53, 54] (Table 3). Other specificities include antinRNP, anti-Su, anti-chromatin, anti-ssDNA, and anti-NF90/NF45 [55] as well as myositis-specific anti-OJ autoantibodies [56]. The pattern of autoantibody production is remarkably similar to that seen in SLE (Fig. 1), and titers are comparable to those found in lupus-prone strains, such as MRL, or in humans with SLE. A puzzling aspect of this and other murine models is the complete absence of responses to the Ro (SS-A) and La (SS-B) antigens. We have examined many strains bearing different H-2 haplotypes, and although there is significant inter-strain variability in the frequency of standard autoantibodies induced by pristane [5], anti-Ro and La are never seen. Following intraperitoneal injection of pristane

Anti-OJ

0

0

0

5%

Anti-Ro (SSA) or La (SSB)

0

0

0

0

Anti-Scl70, fibrillarin, RNA polymerase I/III

0

0

0

0

Glomerular IC

S

S

S

S

Proteinuria

S

S

R

N/A

Arthritis

S

S

R

R

S, susceptible; R, resistant; N/A, not available.

or other active hydrocarbons, the earliest autoantibodies to appear are IgM anti-ssDNA mad IgM anti-chromatin antibodies, which appear after about 2 weeks [3]. Anti-Su antibodies are detected at 2-3 months followed by anti-nRNP/Sm at 3-4 months. Anti-dsDNA antibodies appear much later (6-10 months). Interestingly, this is well after the onset of nephritis. Unlike the early IgM anti-ssDNA response, these late-appearing autoantibodies are primarily of T cell dependent isotypes: IgG2a and IgG2b in the case of anti-Su, and IgG2a in the case of anti-nRNP/Sm and anti-dsDNA. Autoantibody production varies somewhat from strain to strain, but only within the rather limited repertoire mentioned above. Thus, pristane treatment causes BALB/c mice to produce anti-nRNP/ Sm (55%) but not anti-ribosomal P (Table 3). In contrast, SJL and B 10.S mice produce anti-ribosomal P frequently but not anti-Sm [5, 54]. Other strains fall somewhere between these extremes. There also are some differences depending on the hydrocarbon used to induce peritoneal inflammation. Pristane more potently induces anti-nRNP/Sm, whereas medicinal mineral oils induce mainly antissDNA and anti-chromatin. It is important to note that other than the major specificities (anti-nRNP/

91

Figure 1. Spectrum of autoantibodies induced in mice by pristane is similar to that seen spontaneously in SLE. K562 cells were labeled with [35S]methionine and cell extracts were immunoprecipitated using sera from patients with SLE, serum from a healthy control subject (NHS), or with sera from pristane-treated or medicinal mineral oil-treated BALB/c mice. Immunoprecipitated proteins characteristic of SLE include the Sm (U5-200, Sm-B, Sm-D, Sm-E/F, and Sm-G) and RNP (U1-A, U1-C) proteins, the 100 kDa Su antigen, and the ribosomal P (rP) proteins P0, P1, and P2.

Sm, Su, ribosomal P, dsDNA, ssDNA, chromatin, and to a lesser degree NF90/NF45 and anti-OJ), additional specificities are unusual regardless of which hydrocarbon is used to induce autoantibody production. Following intraperitoneal injection of pristane, large amounts of IL-6, IL-12, and/or TNFa are produced locally in BALB/c and many other strains of mice [6]. There is inter-strain variability, however. For instance, C57BL/6 mice produce little IL-6 [57]. We investigated the effects of these cytokines on pristane-induced lupus in cytokine knockout mice. In view of the absence of autoantibodies in T cell deficient nude mice [58], we also examined the effects of T cell cytokines (IL-4 and IFN7) in cytokine knockout mice (Table 4).

92

Table 4. Cytokine dependence of pristane-induced lupus in BALB/c mice a

Manifestation

IL-6 -/-

IL-12 -/-

IFN)'-/-

IL-4 -/-

Anfi-dsDNA

,H,

No A

,l,,I,

No A

Anti-chromatin

,1,,I,

$,1,

,1,$

No A

Anti-nRNP/Sm

No a b

S J,

$,1,

No a

Nephritis

,H,

$,1,

$,],

No h

aData from Refs. [53, 58, 75, 76]. bDecreased levels but comparable prevalence.

3.1. IL-6 Deficient Mice IL-6 is a pleiotropic cytokine thought to play a pivotal role in immune regulation through its effects on B and T cells [59]. It acts primarily on the late phase of B cell differentiation, consistent with the observation that IL-6R is expressed on mitogenactivated, but not resting, B cells [60]. Although Type I IFN is a key mediator of B cell differentiation into plasmablasts, their further maturation into antibody secreting cells requires IL-6 [49]. Moreover, IL-6 promotes the development of plasmacytomas in pristane and mineral oil treated mice [61-63]. Polyclonal activation and hypergammaglobulinemia in lupus may reflect both the over-expression of IL-6 receptors on B cells [64, 65] and increased IL-6 levels [66]. IL-6 overproduction in disorders such as atrial myxomas and Castleman's disease is associated with hypergammaglobulinemia and autoimmune phenomena [59, 67, 68]. In view of the strong indirect evidence linking IL-6 with autoantibody formation, we examined whether IL-6 deficient mice were susceptible to pristane-induced lupus [53]. Pristane induces high levels of IgG anti-ssDNA and chromatin antibodies as well as anti-dsDNA in wild type BALB/c mice, but IL-6 deficient BALB/c mice do not produce these autoantibodies [53]. Interestingly, wild type mice spontaneously develop low levels of anti-chromatin autoantibodies in an age-dependent manner, and effect abrogated in IL-6 -/- mice [53]. In contrast to anti-DNA, the frequencies of anti-nRNP/Sm an anti-Su antibodies are similar in pristane-treated IL-6-/- and IL-6 +/+ mice, although levels were lower in the knockout mice, consistent with the known effects of this cytokine on B cell maturation. We concluded from analyzing IL-6 deficient mice that anti-DNA and chromatin antibodies in pristane-treated mice are strictly IL-6dependent, whereas induction of anti-nRNP/Sm and Su autoantibodies is relatively IL-6-independent. 3.2. IL-12 and IFNy Deficient Mice The potent adjuvant CFA is derived from IFA by adding heat-killed mycobacteria, which stimulate TLRs and enhance the production of IL-12 by maturing DCs [29, 69, 70]. For example, CpG oligonucleotides stimulate IL-12 production through

binding to TLR9, polarizing the immune response to TH1 [71, 72]. Analysis of the isotypes of polyclonal IgG and specific autoantibodies induced by pristane, especially anti-nRNP/Sm and anti-dsDNA, indicated a strong IgG2a response [73]. Since IgG2a is an IFNy [74] and IFNt~ [50] dependent isotype, we hypothesized that pristane-induced lupus may be a TH1 mediated disease. This possibility was examined using IL-12 and IFN 7 knockout mice. IgG anti-chromatin autoantibodies are absent following pristane treatment of IFNy deficient BALB/c mice, whereas their frequency and level are similar in IL-4 deficient mice vs. wild type controls [75]. The frequency of IgG anti-nRNP/Sm autoantibodies was reduced markedly in pristane-treated IFNy 4- mice compared with +/+ controls (22% vs. 77%), but in the few IFNy-/- mice producing these autoantibodies, levels were comparable to those in wild type controls. Likewise, the frequency of anti-Su antibodies was reduced (55% vs. 17%), but not the level. IL-12 and IL-18 are key cytokines produced by antigen presenting cells (APCs) that regulate the production of IFNy. As noted above, local (intraperitoneal) IL-12 production is greatly enhanced by pristane treatment. IL-12 deficient mice exhibit a similar, though not identical, autoantibody phenotype to that of IFNy-/- mice [76]. The major difference is that anti-dsDNA autoantibody production following pristane treatment is not substantially reduced in the IL-12-/- mice vs. controls, despite the fact that glomerulonephritis is nearly completely abrogated [76].

4. INDUCTION OF POLYCLONAL HYPERGAMMAGLOBULINEMIA BY HYDROCARBON ADJUVANTS One of the first changes noted in BALB/c mice given an intraperitoneal injection of pristane is a striking increase in total serum IgM [3]. This is followed by a rise in IgG1, IgG2a and IgG2b [77]. In most strains, IgG2a increases out of proportion to IgG1. Despite the role of IL-6 in hypergammaglobulinemia in other situations, an increased total IgM level is apparent as early as 2 weeks after pristane treatment in BALB/c IL-6 +/+ as well as IL-6 -/- mice, but not in PBS-treated controls. IgG2a, IgG2b, and

93

IgG3 levels also increase markedly from 1 to 3 months after pristane treatment both in IL-6 +/+ and -/- mice, suggesting that other cytokines contribute. Not surprisingly, a substantial reduction in polyclonal IgG2a (an IFN~, dependent isotype) inducible by pristane is seen in BALB/c IFNT 4- mice [75]. Likewise, IgG1 (IL-4 dependent isotype) polyclonal hypergammaglobulinemia following pristane treatment is substantially reduced in IL-4 deficient mice [75]. Different hydrocarbons can yield quite different effects on the total levels of IgM vs. IgG isotypes. Thus, whereas pristane and to a lesser extent squalene and IFA promote IgG2a hypergammaglobulinemia, medicinal mineral oils tend to stimulate the T cell independent isotypes IgM and IgG3 out of proportion to IgG2a. These data suggest that the polyclonal hypergammaglobulinemia induced by pristane, mineral oil, squalene [6, 75], and silicone oil [78] is mediated at least in part through the stimulation of cytokine production. Consistent with that notion, intraperitoneal injection of lupus-inducing hydrocarbons (pristane, squalene, and IFA) causes the production of IL-6, IL-12, and sometimes TNFcz [6]. The production of these cytokines is substantially lower when nonlupus inducing hydrocarbons, such as medicinal mineral oils, are injected. IL-6 production also is likely to contribute to the growth of plasma cell neoplasms arising in the peritoneal cavity of BALB/ cPtAn mice treated with pristane [14, 61]. The cytokine-dependent induction of specific autoantibodies thus occurs in the setting of polyclonal hypergammaglobulinemia, the nature of which is influenced by the type of hydrocarbon injected and the cytokine response to it.

5. INDUCTION OF GLOMERULONEPHRITIS AND ARTHRITIS BY HYDROCARBON ADJUVANTS

complexes, but not renal disease [80]. The role of inflammatory cells in nephritis depends largely on the site of immune complex deposition: subendothelial and mesangial deposits are associated with inflammatory lesions, whereas subepithelial lesions (characteristic of membranous nephropathy) are not [81 ]. The predominantly mesangial and subendothelial immune deposits characteristic of human lupus nephritis and pristane-induced lupus are consistent with the inflammatory nature of the renal lesion in SLE. Renal lesions in pristane treated mice are initially mesangial, but subendothelial lesions reminiscent of diffuse proliferative lupus nephritis develop later on [3]. The mesangial expansion in pristane-induced lupus [3, 54] is consistent with the appearance of lupus nephritis [82]. IL-6 stimulates mesangial cell proliferation [83, 84] and may exacerbate lupus nephritis [85]. IFN~/also has been implicated in lupus nephritis [86; 87]. The light microscopic changes and severe proteinuria in pristane treated BALB/c mice are nearly abrogated by IFNT or IL- 12 deficiency and there is a significant reduction in immune complex deposition [75, 76]. In contrast, IL-4 deficiency has little effect or actually increases the severity of nephritis. IL-6 deficient BALB/cAn mice also are highly resistant to the induction of lupus nephritis by pristane. Light microscopic changes and proteinuria are eliminated and glomerular immune complex deposition is reduced dramatically [53]. Taken together, these data strongly suggest that the proinflammatory cytokines IFN),, IL-12, and IL-6 are critical mediators of pristane-induced lupus nephritis. It is possible that the striking reduction in glomerular immune complex deposits and the lower levels of immune complexes containing IgG2a in particular decreases the inflammatory response mediated by FcTRI and III. Alternatively, cytokine deficiency could decrease the recruitment of phagocytes and/or T cells into the glomerular lesions [88].

5.1. Nephritis 5.2. Arthritis

The interaction of immune complexes with Fc receptors (FqRI or Fc~,RIII) on phagocytes causes production of proinflammatory cytokines and other mediators of glomerulonephritis [79]. Interestingly, NZB/W mice lacking the common T chain shared by FcTR I and III exhibit glomerular immune

94

Several hydrocarbon oils induce arthritis in rodents [4]. BALB/cJ, DBA/1 and several other strains of mice develop synovial hyperplasia, periostitis, and marginal erosions reminiscent of rheumatoid arthritis following intraperitoneal pristane treatment

[4, 89]. Serological abnormalities consistent with rheumatoid arthritis also develop, including autoantibodies against type II collagen and rheumatoid factor [4]. Susceptibility to pristane-induced arthritis is associated with the IF-1 locus, which regulates circulating IFNct/~, TNFo~, and IL-6 levels induced by Newcastle disease virus [4]. BALB/c and DBA/1 mice (susceptible to arthritis) have the IF-1 ~allele, whereas most resistant strains, such as C57BL/6 and DBA/2, have the IF-lh allele and express high levels of I F N ~ , TNFct, and IL-6 when infected [90]. This is unexpected in view of the importance of TNFot and IL-6 in the pathogenesis of rheumatoid arthritis and collagen-induced arthritis [91]. It is all the more surprising in view of a report indicating that pristane-induced arthritis is ameliorated by TNFt~ inhibition [92]. Other hydrocarbon oils also can induce arthritis in rats. The adjuvanticity of a hydrocarbon corresponds to its arthritogenicity. Normal (unbranched) alkanes with 12 or more carbons are arthritogenic, whereas C12-C16 alkenes (olefins) consistently induce less severe disease than corresponding saturated hydrocarbons [17]. Squalene is arthritogenic in arthritis-prone DA rats following intradermal injection, raising the possibility that this endogenous cholesterol precursor has the potential to trigger autoimmune disease [93]. The arthritis induced by squalene, like pristane-induced arthritis in mice, is erosive and T cell-mediated, but is not accompanied by anti-collagen autoantibodies. The pathogenesis of erosive arthritis in rodents exposed to hydrocarbons is not well understood. TNFt~ has been implicated as well as monocyte/macrophages, suggesting that the pathogenesis may be similar in some respects to that of rheumatoid arthritis.

6. ACCELERATION OF SPONTANEOUS LUPUS BY PRISTANE Besides triggering the onset of autoantibody production and lupus-like autoimmune disease in non-lupus prone mice, pristane can accelerate spontaneous (genetically mediated) lupus-like disease in (NZB X NZW) F1 and MRL +/+ mice (Table 5). This is a useful model for studying the interaction between the environment and the genetic background in lupus.

6.1. (NZB X NZW) F1 Mice

In addition to anti-chromatin/DNA responses, NZB/W mice spontaneously produce autoantibodies against the double-stranded RNA binding protein RNA helicase A (RHA). In contrast, this strain does not produce autoantibodies against the nRNP, Sm, Ro, and La antigens. Pristane exposure greatly accelerates the production of anti-chromatin and anti-DNA antibodies and dramatically accelerates renal disease. Production of anti-nRNP/Sm and Su autoantibodies also is induced, indicating that the unresponsiveness of NZB/W mice to these antigens can be overcome. Unexpectedly, pristane treatment does not enhance the production of anti-RHA and may actually inhibit it, suggesting that these autoantibodies are regulated differently. 6.2. MRL Mice and Effect of the

lpr Defect

Similarly, the onset of spontaneous anti-nRNP/Sm and anti-Su autoantibody production and the development of lupus nephritis were accelerated greatly in MRL +/+ mice by pristane treatment (A Mizutani et al, submitted for publication) (Table 5). However, a major surprise is that the Ipr and gld mutations eliminate susceptibility to pristane-induced lupus nearly completely [57]. The TNF/TNFR family members Fas and FasL signal apoptosis and act as lupus susceptibility genes [94]. In lupus-prone mice, deficiency of Fas or FasL greatly accelerates the onset and severity of lupus. Even in non-autoimmune B6 mice, Fas Jp~promotes the development of mild autoimmunity, with anti-chromatin autoantibody production. How Fas deficiency promotes autoimmunity is unclear, but it has been suggested that autoreactive B and/or T cells are deleted in the periphery by Fas-FasL signaling [95, 96]. In view of the consistent acceleration of spontaneous lupus-like disease in mice by lpr (Fas mutation) or gld (FasL mutation) the finding that these mutations have just the opposite effect on autoantibody formation in B6 mice with pristaneinduced lupus was unexpected [57]. B6/lpr and B6/gld mice are highly resistant to the induction of autoantibodies by pristane. Pristane induces IgM anti-ssDNA at 2 weeks and IgG anti-nRNP/Sm/Su/ribosomal P autoantibodies at 6 months in wild type B6 mice, but this is abrogated

95

Table 5. Effects of pristane on spontaneous autoimmunity Autoimmune strain

Manifestation

Effect of pristane

Reference

(NZB X NZW)F1

Anti-dsDNA Anti-nRNP/Sm Anti-Su Anti-RNA helicase A Glomerulonephritis

Acceleration Induction Induction Inhibition Acceleration

[ 103]

MRL +/+

Anti-dsDNA Anti-nRNP/S m Anti-Su Glomerulonephritis

Induction/Acceleration Acceleration No effect Acceleration

Unpublished data

MRL/Ipr

Anti-dsDNA Anti-nRNP/Sm Anti-Su Glomerulonephritis

No effect No effect No effect Inhibition

Unpublished data

B6/lpr

Anti-chromatin

No effect

[57]

CBA/N

Anti-RNA helicase A

Inhibition

[ 101 ]

in lpr or gld mice, suggesting that intact Fas signaling is necessary for autoantibody induction. Pristane also does not enhance IgG anti-chromatin antibody production in B6/lpr or B6/gld mice, suggesting that it does not influence spontaneous autoantibody production in Fas deficient mice. Similarly, although autoantibody production and nephritis in MRL +/+ mice are accelerated by pristane treatment, MRL lpr/lpr mice are completely refractory and the onset of renal disease is delayed by pristane treatment. This paradoxical effect of Fas deficiency might be explained in several ways. Cells undergoing Fas-mediated cell death may provide a source of antigens driving autoantibody formation in pristaneinduced lupus, consistent with suggestions that the abnormal clearance of apoptotic material induces autoantibody formation [97, 98]. Alternatively, pristane might cause Fas-mediated cell death of a population of cells that normally prevents autoimmunity. Another possibility is that pristane and Fas stimulate autoimmunity through mutually antagonistic pathways and that pre-existing Fas or FasL deficiency precludes the induction of autoimmunity by pristane. Recent studies in our laboratory raise the possibility that pristane exposure may promote the maturation of myeloid DCs, an important producer of IL-12. It has been proposed that whereas immature DCs tolerize autoreactive T cells, mature DCs

96

are activators [30, 99]. Since DCs capture apoptotic cells via interactions with surface receptors such as the integrin avl] 5 [99, 100], perhaps the uptake of apoptotic cells by a subset of pristane-activated DCs could help trigger autoimmunity. 6.3. CBA/N (x/d) Mice Although generally considered to be an "immunocompromised" strain, CBA/N mice, which have a genetic defect in the Bruton's tyrosine kinase (btk) gene, have recently be shown to produce autoantibodies against RNA helicase A (RHA) [ 101 ]. AntiRHA autoantibodies are associated with spontaneous lupus in humans [ 102] and mice [ 103]. Pristane treatment antagonizes the spontaneous production of anti-RHA autoantibodies in both CBA/N and NZB/W mice (Table 6) [101,103]. In contrast to the striking predominance of IgG2a class anti-nRNP/ Sm autoantibodies in pristane treated mice and in MRL mice with spontaneous lupus, IgG1 antiRHA autoantibodies are produced at high levels. We hypothesize that by enhancing the production of IFNy and IFNr pristane may inhibit the production of IgG1 anti-RHA antibodies. However, it remains to be determined why anti-nRNP/Sm is so strongly skewed toward IgG2a whereas anti-RHA responses are skewed more toward IgG1.

7. HOW DO HYDROCARBONS CAUSE AUTOANTIBODY PRODUCTION?

Just as the mechanisms responsible for adjuvanticity remain incompletely understood, we do not know precisely why certain hydrocarbons promote autoimmunity while others do not. Interestingly, there is a rough correlation between the two phenomena: hydrocarbons that are good adjuvants tend to promote lupus whereas those that are weak adjuvants, such as medicinal mineral oils, do not [6]. Lupus-inducing hydrocarbons tend to be more potent than inactive hydrocarbons at stimulating early IL-12 production [6], consistent with its role in linking innate with adaptive immunity. However, there are many other unanswered questions that are only now being addressed. 7.1. Is There a Receptor for Pristane?

Hydrocarbons could bind to a specific receptor or receptors that transmit a "danger signal" [32] analogous to the binding of LPS to TLR4 or glycolipids to CD1 [104]. The fact that CDld deficient mice remain sensitive to pristane-induced lupus argues that CD1 is not a receptor for pristane or other hydrocarbons [56]. The role of Toll-like receptors in the recognition of hydrocarbons is not known and under investigation. An alternative possibility is that pristane becomes incorporated into the plasma membrane [105-107], altering inflammatory signaling pathways or that pristane gains access to the inside of the cell by dissolving in the plasma membrane followed either by binding to an intracellular receptor or modification of a subset of intracellular proteins, as has been proposed for urushiol, the inflammatory oil that causes poison ivy [108, 109]. 7.2. Is the Site of Pristane Exposure Important?

Thus far, studies of hydrocarbon-induced lupus all have employed intraperitoneal injection of the oil. There is little information about whether intraperitoneal injection is necessary or if other sites work equally well. Experiments to address this question are underway in our laboratory. There are significant differences in secondary lymphoid tissues located at different sites, such as lymph nodes, Peyer's patches, or the peritoneal cavity. The peritoneal

cavity of mice, in particular, is highly enriched in the B-I subset of B-lymphocytes [110]. However, we have shown that these cells are rapidly depleted upon injection of pristane [111]. We do not know at present whether the B-1 cells undergo apoptosis or become sequestered at some other site, such as in "granulomas" forming in response to intraperitoneal pristane injection. 7.3. What is the Role of "Oil Granulomas" Induced by Pristane?

Potter et al were the first to record the development of inflammatory nodules in the peritoneal cavities of mice following the injection of mineral oil or pristane [112-114]. They termed these structures "granulomas" in view of the fact that they contained inflammatory cells and numerous oil droplets. These later become organized into polypoid structures. Plasmacytomas develop within the granulomas of certain strains of mice, notably BALB/c, starting -10 months after pristane injection. The evolution of granulomas may begin with so-called "milky spots" [115]. More recently, we have found that the term "granuloma" is not entirely accurate. These structures closely resemble ectopic lymphoid tissue (tertiary lymphoid tissue, lymphoid neogenesis), since they contain collections of B and T lymphocytes, dendritic cells, and macrophages and have a variety of other features consistent with lymphoid tissue (D Nacionales et al, manuscript in preparation). The role of this ectopic lymphoid tissue in the pathogenesis of autoantibodies and lupus-like disease is incompletely defined. However, ectopic lymphoid tissue in a variety of other situations is associated with autoantibody-mediated autoimmune diseases [116]. The thyroid gland in Hashimoto's thyroiditis, the thymus of some patients with myasthenia gravis, CNS lesions in multiple sclerosis, the salivary glands in Sj6gren's syndrome, and the synovium in rheumatoid arthritis all have morphological and functional features of lymphoid neogenesis, including the presence of high endothelial venules, DCs and follicular dendritic cells, antigen-driven clonal proliferation of B-cells, and lymphoid tbllicles with clonally expanded lymphocytes [ 117-120]. Ectopic lymphoid tissue may provide a focal milieu where interactions between lymphocytes and APCs occur in the partial absence of normal censoring mecha-

97

nisms. 7.4. What is the Role of Microbial Exposure as a Co-factor?

Autoantibody production induced by pristane is attenuated in mice housed under specific pathogen free conditions in comparison with mice housed under standard conditions [77]. In contrast, autoimmune diabetes is milder in NOD mice housed under standard conditions than in SPF mice [121]. These and other observations suggest that the microbial environment can modulate autoimmune disease. Studies of pristane-induced lupus in SPF mouse raised the possibility that pristane might increase the exposure to microbial substances such as LPS, which stimulate innate immunity. For instance, inflammation of the bowel has been shown to increase bowel permeability to bacteria [122]. To address this question, we recently completed studies of pristane-induced lupus in germfree mice (A Mizutani et al, submitted for publication). Germfree mice were susceptible to pristane-induced lupus, and made autoantibodies at frequencies comparable to those in SPF mice. Thus, it seems likely that pristane has other actions besides just increasing microbial stimulation. We have found that the peritoneal exudate cells from pristane treated mice are hyper-responsive to stimulation by LPS, suggesting that pristane may act synergistically with certain TLR ligands. 7.5. What is the Role of Cytokines in the Pathogenesis of Pristane Induced Lupus?

As described above, IL-6, IL-12, and IFNy appear to be directly involved in the pathogenesis of pristane-induced lupus. Anti-nRNP,-Sm, and-Su, and -dsDNA autoantibody production is greatly reduced in IL-12 or IFNy deficient mice, and glomerulonephritis is much milder. In contrast, whereas antidsDNA antibody production is nearly eliminated in IL-6 deficient mice, the prevalence of anti-nRNE Sm, and Su autoantibodies is not altered substantially. More recently, we have shown that "granulomas" induced by pristane have high type I IFN activity in comparison with those induced by medicinal mineral oil (DC Nacionales et al, unpublished data), suggesting that type I IFNs are involved in the

98

pathogenesis of pristane induced lupus. IL-6, IL-12, IFNT, and I F N ~ have important effects on the generation of autoreactive T and B cells that may help explain their role in the pathogenesis of hydrocarbon-induced lupus in mice (Fig. 2). DCs, which undergo maturation in response to IFN 7, IFN~, or LPS, regulate T cell activation [25, 30]. Immature dendritic cells (iDC) are tolerogenic whereas mature DCs are sfimulatory. IFNot promotes the differentiation of monocytes into DC-like cells, which can capture antigens from dying cells and present them to CD4+ T cells and also enhances both DC and T cell survival [38, 44, 46, 123]. Mature DCs produce IL-12, which along with IFN 7 polarizes the CD4+ T cells toward the TH1 phenotype. B cell activation and isotype switching are influenced by signals delivered by CD4+ T cells, including CD40L and cytokines. Switching to IgG2a, the predominant isotype of autoantibodies induced by pristane, is enhanced by IFNy and IFNc~ [50, 74]. The maturation of B cells into plasmablasts and plasma cells is driven by IFN~ and IL-6 [49]. Thus, the major cytokines implicated in pristane-induced lupus may promote autoimmunity at multiple levels, including DC maturation, autoreactive T cell activation and survival, and activation, isotype switching, and maturation of B cells. 7.6. Are the Cytokine Abnormalities in PristaneInduced Lupus Relevant to Human SLE?

The cytokines involved in pristane-induced lupus, IL-6, IFN~I3, and IFN 7, have been implicated in the pathogenesis of human SLE. For instance, IL6 levels have been reported to be elevated in SLE [124] and both IL-6 and IFNy are thought to play a role in the pathogenesis of lupus nephritis in humans [ 125, 126]. Both Castleman's disease [ 127] and atrial myxoma [59] are associated with the production of high levels of IL-6 as well as the production of lupus autoantibodies and the development of autoimmune disease. Therapeutic use of IFNy has been associated with the development of lupuslike disease [ 128]. The same is true of IFNc~, which when used to treat hepatitis C infection, malignant carcinoid syndrome, or chronic myelogenous leukemia is sometimes associated with autoimmune phenomena, including sarcoidosis [129], autoimmune

Figure 2. Cytokine effects on autoreactive T and B cells. Cytokines implicated in the pathogenesis of hydrocarbon-induced lupus and their possible mechanisms of action. Dendritic cell (DC) maturation is promoted by IFNtx and IFN~,, as well as by TLR ligands, such as lipopolysaccharide (LPS). Whereas immature dendritic cells (iDC) promote T cell tolerance, mature DCs promote T cell activation. IL-12 and IFN~,drive the differentiation of type I (Tal) T cells, and IFNtx permits activated T cells to survive. B cell development also is cytokine dependent. IFN~,produced by TH1 cells promotes isotype switching to IgG1, the predominant isotype of autoantibodies produced in hydrocarbon-induced lupus. IFNct and IL-6 stimulate B cell differentiation into plasmablasts and plasma cells, respectively, and type I IFNs are necessary for the generation of memory B cells.

thyroiditis, and autoimmune hepatitis [130]. The induction of antinuclear antibodies and anti-dsDNA antibodies as well as overt lupus has been reported, as well [131-133]. Moreover, the serum level of IFNct correlates with anti-dsDNA antibody levels and disease activity in SLE [134-136]. Finally, recent studies suggest the existence of a type I IFN gene expression "signature" that is associated with active SLE [137, 138]. Together, these data suggest that the cytokine abnormalities identified in pristane-induced lupus are relevant to human SLE. A major challenge for the future will be to understand how non-specific inflammation and cytokine production resulting from hydrocarbon exposure leads to the production of a highly restricted subset of autoantibodies that is pathognomonic of SLE. A second challenge will be to determine whether environmental triggers of lupus, such as viral infections or exposure to chemicals, promote disease by stimulating increased production of IL-6, IFN~,, and type I IFNs.

ACKNOWLEDGEMENTS We gratefully acknowledge the technical assistance of Ms. Minna Honkanen-Scott. This work was supported by research grants R01-AR44731 and R01-AI44074 from the United States Public Health Service, by training grant T32-AR07603, and by research support from Lupus Link.

REFERENCES 1.

2.

3.

4.

Morel L, Mohan C, Yu Y, Croker BE Tian N, Deng A et al. Functional dissection of systemic lupus erythematosus using congenic mouse strains. J Immunol 1997;158( 12):6019-6028. Satoh M, Reeves WH. Induction of lupus-associated autoantibodies in BALB/c mice by intraperitoneal injection of pristane. J Exp Med 1994;180:2341-2346. Satoh M, Kumar A, Kanwar YS, Reeves WH. Antinuclear antibody production and immune complex glomerulonephritis in BALB/c mice treated with pristane. Proc Natl Acad Sci USA 1995;92:10934-10938. Wooley PH, Seibold JR, Whalen JD, Chapdelaine

99

5.

6.

7. 8. 9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

100

JM. Pristane-induced arthritis. The immunologic and genetic features of an experimental murine model of autoimmune disease. Arthritis Rheum 1989;32:10221030. Satoh M, Richards HB, Shaheen VM, Yoshida H, Shaw M, Naim JO et al. Widespread susceptibility among inbred mouse strains to the induction of lupus autoantibodies by pristane. Clin Exp Immunol 2000;121: 399-405. Satoh M, Kuroda Y, Yoshida H, Behney KM, Mizutani A, Akaogi Jet al. Induction of lupus autoantibodies by adjuvants. J Autoimm 2003 ;21:1-9. Ramon G. Sur la toxine et sur l'anatoxine diphtheriques. Ann Inst Pasteur 1924;38:1-10. Spickard A. Exogenous lipoid pneumonia. Arch Intern Med 1994;154:686-692. Wanless IR, Geddie WR. Mineral oil lipogranulomata in liver and spleen: a study of 465 autopsies. Arch Pathol Lab Med 1985; 109:283-286. Gwynne-Jones DP. Accidental self inoculation with oil based veterinary vaccines. NZ Med J 1996;109: 363-365. Gupta RK, Relyveld EB, Bizzini B, Ben-Efraim S, Gupta CK. Adjuvants: a balance between toxicity and adjuvanticity. Vaccine 1993; 11:293-306. Audibert FM, Lise LD. Adjuvants: current status, clinical perspectives and future prospects. Immunol Today 1993;14:281-284. Kohashi O, Tanaka S, Kotani S, Shiba T, Kusumoto S, Yokogawa K et al. Arthritis-inducing ability of a synthetic adjuvant, N-acetylmuramyl peptides, and bacterial disaccharide peptides related to different oil vehicles and their composition. Infect Immun 1980;29: 70-75. Potter M, Boyce CR. Induction of plasma-cell neoplasms in strain BALB/c mice with mineral oil and mineral oil adjuvants. Nature (Lond) 1962; 193:10861087. Beebe GW, Simin AH, Vivona S. Long-term mortality follow-up of army recruits who received adjuvant influenza virus vaccine in 1951-1953. Am J Epidemiol 1972;95:337-346. Shaw CM, Alvord EC, Kies MW. Straight chain hydrocarbons as substitutes for the oil in Freund's adjuvants in the production of experimental "allergic" encephalomyelitis in guinea pig. J Immunol 1964;92:24-27. Whitehouse MW, Orr KJ, Beck FWJ, Pearson CM. Freund's adjuvants: relationship of arthritogenicity and adjuvanticity in rats to vehicle composition. Immunol 1974;27:311-330. Fattori E, Cappelletti M, Costa P, Sellitto C, Cantoni L, Carelli M e t al. Defective inflammatory response in interleukin 6-deficient mice. J Exp Med 1994;180:

1243-1250. 19. Crowle AJ, Hu CC. A comparison on n-hexadecane and mineral oil emulsions in induction of hypersensitivity in mice. Proc Soc Exp Biol Med 1966;123:94-97. 20. Dupuis M, Denis-Mize K, LaBarbara A, Peters W, Charo IF, McDonald DM et al. Immunization with the adjuvant MF59 induces macrophage trafficking and apoptosis. Eur J Immunol 2001 ;31 (10):2910-2918. 21. Herbert WJ. The mode of action of mineral-oil emulsion adjuvants on antibody production in mice. Immunol 1968;14(3):301-318. 22. Dupuis M, Murphy TJ, Higgins D, Ugozzoli M, Van Nest G, Ott G e t al. Dendritic cells internalize vaccine adjuvant after intramuscular injection. Cell Immunol 1998; 186( 1): 18-27. 23. Krieg AM, Yi AK, Matson S, Waldschmidt TJ, Bishop GA, Teasdale R et al. CpG motifs in bacterial DNA trigger direct B-cell activation. Nature (Lond) 1995;374: 546-549. 24. Krieg AM. CpG motifs in bacterial DNA and their immune effects. Annu Rev Immunol 2002;20:709760. 25. Banchereau J, Steinman RM. Dendritic cells and the control of immunity. Nature (Lond) 1998;392:245252. 26. Palucka K, Banchereau J. Linking innate and adaptive immunity. Nat Med 1999;5(8):868-870. 27. Cyster JG. Chemokines and the homing of dendritic cells to the T cell areas of lymphoid organs. J Exp Med 1999; 189(3): 447-450. 28. Inaba K, Turley S, Yamaide F, Iyoda T, Mahnke K, Inaba M et al. Efficient presentation of phagocytosed Cellular fragments on the major histocompatibility complex class 1I products of dendritic cells. J Exp Med 1998; 188(11):2163-2173. 29. Reis e Sousa C, Hieny S, Scharton-Kersten T, Jankovic D, Charest H, Germain RN et al. In vivo microbial stimulation induces rapid CD40 ligand- independent production of interleukin 12 by dendritic cells and their redistribution to T cell areas. J Exp Med 1997;186(11): 18189-11829. 30. Steinman RM, Turley S, Mellman I, Inaba K. The induction of tolerance by dendritic cells that have captured apoptotic cells. J Exp Med 2000;191(3): 411-416. 31. Hawiger D, Inaba K, Dorsett Y, Guo M, Mahnke K, Rivera M et al. Dendritic ceils induce peripheral T cell unresponsiveness under steady state conditions in vivo. J Exp Med 2001;194(6):769-779. 32. Matzinger P. Tolerance, danger, and the extended family. Annu Rev Immunol 1994;12:991-1045. 33. Kaisho T, Akira S. Toll-like receptors as adjuvant receptors. Biochim Biophys Acta 2002; 1589(1): 1-13.

34. Akira S, Yamamoto M, Takeda K. Role of adapters in Toll-like receptor signalling. Biochem Soc Trans 2003;3 l(Pt 3):637-642. 35. Kawai T, Takeuchi O, Fujita T, Inoue J, Muhlradt PF, Sato S et al. Lipopolysaccharide stimulates the MyD88independent pathway and results in activation of WNregulatory factor 3 and the expression of a subset of lipopolysaccharide-inducible genes. J Immunol 2001;167(10):5887-5894. 36. Doyle S, Vaidya S, O'Connell R, Dadgostar H, Dempsey P, Wu T et al. IRF3 mediates a TLR3/TLR4specific antiviral gene program. Immunity 2002;17(3): 251-263. 37. Hemmi H, Kaisho T, Takeuchi O, Sato S, Sanjo H, Hoshino K et al. Small anti-viral compounds activate immune cells via the TLR7 MyD88-dependent signaling pathway. Nat Immuno12002;3(2):196--200. 38. Le Bon A, Schiavoni G, D'Agostino G, Gresser I, Belardelli F, Tough DE Type I interferons potently enhance humoral immunity and can promote isotype switching by stimulating dendritic cells in vivo. Immunity 2001;14(4):461--470. 39. Sen GC, Lengyel E The interferon system. J Biol Chem 1992;267(8):5017-5020. 40. Diaz MO, Pomykala HM, Bohlander SK, Maltepe E, Malik K, Brownstein B e t al. Structure of the human type-I interferon gene cluster determined from a YAC clone contig. Genomics 1994;22(3):540-552. 41. Taniguchi T, Takaoka A. A weak signal for strong responses: interferon-alpha/beta revisited. Nat Rev Mol Cell Bio12001;2(5):378-386. 42. Jiang Z, Zamanian-Daryoush M, Nie H, Silva AM, Williams BR, Li X. Poly(dI x dC)-induced Toll-like receptor 3 (TLR3)-mediated activation of NFkappa B and MAP kinase is through an interleukin-1 receptor-associated kinase (IRAK)-independent pathway employing the signaling components TLR3-TRAF6-TAK1-TAB2PKR. J Biol Chem 2003;278(19):16713-16719. 43. Brierley MM, Fish EN. Review: IFN-alpha/beta receptor interactions to biologic outcomes: understanding the circuitry. J Interferon Cytokine Res 2002;22(8): 835-845. 44. Kadowaki N, Antonenko S, Lau JY, Liu YJ. Natural interferon alpha/beta-producing cells link innate and adaptive immunity. J Exp Med 2000; 192(2):219-226. 45. Litinskiy MB, Nardelli B, Hilbert DM, He B, Schaffer A, Casali P et al. DCs induce CD40-independent immunoglobulin class switching through BLyS and APRIL. Nat Immuno12002;3(9):822-829. 46. Marrack P, Kappler J, Mitchell T. Type I interferons keep activated T cells alive. J Exp Med 1999;189(3): 521-529. 47. Zuckerman E, Zuckerman T, Sahar D, Streichman S,

48. 49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

Attias D, Sabo E et al. The effect of antiviral therapy on t(14;18) translocation and immunoglobulin gene rearrangement in patients with chronic hepatitis C virus infection. Blood 2001;97(6):1555-1559. Su L, David M. Inhibition of B cell receptor-mediated apoptosis by IFN. J Immunol 1999;162:6317--6321. Jego G, Palucka AK, Blanck JP, Chalouni C, Pascual V, Banchereau J. Plasmacytoid dendritic cells induce plasma cell differentiation through type I interferon and interleukin 6. Immunity 2003; 19(2):225-234. Finkelman FD, Svetic A, Gresser I, Snapper C, Holmes J, Trotta PP et al. Regulation by interferon alpha of immunoglobulin isotype selection and lymphokine production in mice. J Exp Meal 1991;174(5):1179-1188. Sun S, Zhang X, Tough DE Sprent J. Type I interferonmediated stimulation of T cells by CpG DNA. J Exp Med 1998;188(12):2335-2342. Krug A, Rothenfusser S, Homung V, Jahrsdoffer B, Blackwell S, Ballas ZK et al. Identification of CpG oligonucleotide sequences with high induction of IFNalpha/beta in plasmacytoid dendritic cells. Eur J Immuno12001;31(7):2154-2163. Richards HB, Satoh M, Shaw M, Libert C, Poli V, Reeves WH. IL-6 dependence of anti-DNA antibody production: evidence for two pathways of autoantibody formation in pristane-induced lupus. J Exp Med 1998; 188:985-990. Satoh M, Hamilton KJ, Ajmani AK, Dong X, Wang J, Kanwar YS et al. Autoantibodies to ribosomal P antigens with immune complex glomerulonephritis in SJL mice treated with pristane. J Immunol 1996;157: 3200-3206. Satoh M, Shaheen VM, Kao PN, Okano T, Shaw M, Yoshida H et al. Autoantibodies define a family of proteins with conserved double-stranded RNA binding domains as well as DNA-binding activity. J Biol Chem 1999;274:34598-34604. Yang JQ, Singh AK, Wilson MT, Satoh M, Stanic AK, Park JJ et al. Immunoregulatory role of CDld in the hydrocarbon oil-induced model of lupus nephritis. J Immuno12003;171(4):2142-2153. Satoh M, Weintraub JP, Yoshida H, Shaheen VM, Richards HB, Shaw M et al. Fas and Fas ligand mutations inhibit autoantibody production in pristane-induced lupus. J Immunol 2000;165:1036-1043. Richards HB, Satoh M, Jennette JC, Okano T, Kanwar YS, Reeves WH. Disparate T cell requirements of two subsets of lupus-specific autoantibodies in pristanetreated mice. Clin Exp Immunol 1999;115:547-553. Hirano T, Taga T, Yasukawa K, Nakajima K, Nakano N, Takatsuld F et al. Human B-cell differentiation factor defined by an anti-peptide antibody and its possible role in autoantibody production. Proc Natl Acad Sci USA

101

1987 ;84:228-231. 60. Taga T, Kawanishi u Hardy RR, Hirano T, Kishimoto T. Receptors for B cell stimulatory factor 2. Quantitation, specificity, distribution, and regulation of their expression. J Exp Med 1987;166:967-981. 61. Nordan RP, Potter M. A macrophage-derived factor required by plasmacytomas for survival and proliferation in vitro. Science (Wash, DC) 1986;233:566-569. 62. Hilbert DM, Kopf M, Mock BA, Kohler G, Rudikoff S. Interleukin 6 is essential for in vivo development of B lineage neoplasms. J Exp Med 1995;182:243-248. 63. Dedera DA, Urashima M, Chauhan D, LeBrun DP, Bronson RT, Anderson KC. Interleukin-6 is required for pristane-induced plasma cell hyperplasia in mice. Br J Hematol 1996;94:53-61. 64. Nagafuchi H, Suzuki N, Mizushima Y, Sakane T. Constitutive expression of IL-6 receptors and their role in the excessive B cell function in patients with systemic lupus erythematosus. J Immunol 1993; 151:6525-6534. 65. Kobayashi I, Matsuda T, Saito T, Yasukawa K, Kikutani H, Hirano T et al. Abnormal distribution of IL-6 receptor in aged MRL/lpr mice: elevated expression on B cells and absence on CD4+ cells. Int Immunol 1992;4: 1407-1412. 66. Tang B, Matsuda T, Akira S, Nagata N, Ikehara S, Hirano T et al. Age-associated increase in interleukin 6 in MRL/Ipr mice. Int Immunol 1991;3:273-278. 67. Jourdan M, Bataille R, Seguin J, Zhang XG, Chaptal PA, Klein B. Constitutive production of interleukin-6 and immunologic features in cardiac myxomas. Arthritis Rheum 1990;33:398-402. 68. Yoshizaki K, Matsuda T, Nishimoto N, Kuritani T, Taeho L, Aozasa K et al. Pathogenic significance of interleukin-6 (IL-6/BSF-2) in Castleman's disease. Blood 1989;74:1360-1367. 69. Trinchieri G. Interleukin- 12 and the regulation of innate resistance and adaptive immunity. Nat Rev Immunol 2003 ;3(2): 133-146. 70. Yip HC, Karulin AY, Tary-Lehmann M, Hesse MD, Radeke H, Heeger PS et al. Adjuvant-guided type-1 and type-2 immunity: infectious/noninfectious dichtomy defines the class of response. J Immunol 1999;162: 3942-3949. 71. Chu RS, Targoni OS, Krieg AM, Lehmann PV, Harding CV. CpG oligodeoxynucleotides act as adjuvants that switch on T helper 1 (Thl) immunity. J Exp Med 1997;186:1623-1631. 72. Krug A, Towarowski A, Britsch S, Rothenfusser S, Hornung V, Bals R et al. Toll-like receptor expression reveals CpG DNA as a unique microbial stimulus for plasmacytoid dendritic cells which synergizes with CD40 ligand to induce high amounts of IL-12. Eur J Immunol 2001 ;31 (10):3026-3037.

102

73. Satoh M, Treadwell EL, Reeves WH. Pristane induces high titers of anti-Su and anti-nRNP/Sm autoantibodies in BALB/c mice. Quantitation by antigen capture ELISAs based on monospecific human autoimmune sera. J Immunol Methods 1995;182:51-62. 74. Snapper CM, Perchel C, Paul WE. IFN-gamma stimulated IgG2a secretion by murine B cells stimulated with lipopolysaccharide. J Irmnunol 1988; 140:2121-2127. 75. Richards HB, Satoh M, Jennette JC, Croker BP, Reeves WH. Interferon-gamma promotes lupus nephritis in mice treated with the hydrocarbon oil pristane. Kidney Int 2001 ;60:2173-2180. 76. Calvani N, Satoh M, Croker BP, Reeves WH, Richards HB. Nephritogenic autoantibodies but absence of nephritis in II-12p35-deficient mice with pristaneinduced lupus. Kidney Int 2003;64:897-905. 77. Hamilton KJ, Satoh M, Swartz J, Richards HB, Reeves WH. Influence of microbial stimulation on hypergammaglobulinemia and autoantibody production in pristane-induced lupus. Clin Immunol Immunopathol 1998;86:271-279. 78. Naim JO, Satoh M, Buehner NA, Ippolito KM, Yoshida H, Nusz D et al. Induction of hypergammaglobulinemia and macrophage activation by silicone gels and oils in female A.SW mice. Clin Diagn Lab lmmunol 2000;7(3):366-370. 79. Ravetch JV, Bolland S. IgG Fc receptors. Annu Rev Immuno12001 ;19:275-290. 80. Clynes R, Dumitru C, Ravetch JV. Uncoupling of immune complex formation and kidney damage in autoimmune glomerulonephritis. Science (Wash, DC) 1998;279(5353): 1052-1054. 81. Fries JWU, Mendrick DL, Rennke HG. Determinants of immune complex-mediated glomerulonephritis. Kidney Int 1988;34:333-345. 82. Couser WG. Pathogenesis of glomerulonephritis. Kidney Int 1993;44 (Suppl. 42):19-26. 83. Horii Y, Muraguchi A, Iwano M, Matsuda T, Hirayama T, Yamada H et al. Involvement of IL-6 in mesangial proliferative glomerulonephritis. J Immunol 1989;143: 3949-3955. 84. Ruef C, Budde K, Lacy J, Northemann W, Baumann M, Sterzel RB et al. Interleukin 6 is an autocrine growth factor for mesangial cells. Kidney Int 1990;38: 249-257. 85. Kiberd BA. Interleukin-6 receptor blockage ameliorates murine lupus nephritis. J Am Soc Nephrol 1993;4: 58-61. 86. Haas C, Le Hir M. IFN-gamma is essential for the development of autoimmune glomerulonephritis in MRL/lpr mice. J Immunol 1997;158:5484-5491. 87. Schwarting A, Wada T, Kinoshita K, Tesch G, Kelley VR. IFN-gamma receptor signaling is essential for

88.

89.

90.

91.

92.

the initiation, acceleration, and destruction of autoimmune kidney disease in MRL-Fas ~pr mice. J Immunol 1998;161:494-503. Naito T, Yokoyama H, Moore KJ, Dranoff G, Mulligan RC, Kelley VR. Macrophage growth factors introduced into the kidney initiate renal injury. Mol Med 1996;2(3): 297-312. Potter M, Wax JS. Genetics of susceptibility to pristane-induced plasmacytomas in BALB/cAn: Reduced susceptibility in BALB/cJ with a brief description of pristane-induced arthritis. J Immunol 1981 ;127:15911595. Raj NB, Cheung SC, Rosztoczy I, Pitha PM. Mouse genotype affects inducible expression of cytokine genes. J Immunol 1992; 148:1934-1940. Feldmann M, Brennan FM, Maini RN. Role of cytokines in rheumatoid arthritis. Annu Rev Immunol 1996;14:397-440. Beech JT, Thompson SJ. Anti-tumour necrosis factor therapy ameliorates joint disease in a chronic model of inflammatory arthritis. Br J Rheumatol 1997;36(10): 1129.

93. Carlson BC, Jansson AM, Larsson A, Bucht A, Lorentzen JC. The endogenous adjuvant squalene can induce a chronic T-cell-mediated arthritis in rats. Am J Pathol 2000; 156( 6):2057-2065. 94. Cohen PL, Eisenberg RA. Lpr and gld: single gene models of systemic autoimmunity and lymphoproliferative disease. Annu Rev Imrnunol 1991 ;9:243-269. 95. Russell JH, Rush B, Weaver C, Wang R. Mature T cells of autoimmune lprllpr mice have a defect in antigenstimulated suicide. Proc Natl Acad Sci USA 1993;90: 4409-4413. 96. Stohl W, Lynch DH, Starling GC, Kiener PA. Superantigen-driven, CD8 § T cell-mediated down-regulation: CD95 (Fas)-dependent down-regulation of human Ig responses despite CD95-independent killing of activated B cells. J Immunol 1998;161:3292-3298. 97. Casciola-Rosen L, Andrade F, Ulanet D, Wong WB, Rosen A. Cleavage by granzyme B is strongly predictive of autoantigen status: implications for initiation of autoimmunity. J Exp Med 1999;190(6):815-826. 98. Botto M, Dell'Agnola C, Bygrave AE, Thompson EM, Cook HT, Petry F et al. Homozygous C lq deficiency causes glomerulonephritis associated with multiple apoptotic bodies. Nat Genet 1998; 19(1):56-59. 99. Albert ML, Pearce SF, Francisco LM, Sauter B, Roy P, Silverstein RL et al. Immature dendritic cells phagocytose apoptotic cells via alphavbeta5 and CD36, and cross-present antigens to cytotoxic T lymphocytes. J Exp Med 1998;188(7): 1359-1368. 100. Fadok VA, Bratton DL, Henson PM. Phagocyte receptors for apoptotic cells: recognition, uptake, and conse-

quences. J Clin Invest 2001;108(7):957-962. 101. Satoh M, Mizutani A, Behney KM, Kuroda Y, Akaogi J, Yoshida H et al. X-linked immunodeficient mice spontaneously produce lupus-related anti-RNA helicase A autoantibodies, but are resistant to pristane-induced lupus. Int Immuno12003;15(9):1117-1124. 102. Takeda Y, Caudell P, Grady G, Wang G, Suwa A, Sharp GC et al. Human RNA helicase A is a lupus autoantigen that is cleaved during apoptosis. J Immunol 1999;163(11):6269--6274. 103. Yoshida H, Satoh M, Behney KM, Lee CG, Richards HB, Shaheen VM et al. Effect of an exogenous trigger on the pathogenesis of lupus in NZB X NZW (F1) mice. Arthritis Rheum 2002;46:2235-2244. 104. Beckman EM, Porcelli SA, Morita CT, Behar SM, Furlong ST, Brenner MB. Recognition of a lipid antigen by CDl-restricted alpha beta+ T cells. Nature (Lond) 1994 ;372(6507):691-694. 105. B ly JE, Garrett LR, Cuchens MA. Pristane induced changes in rat lymphocyte membrane fluidity. Cancer Biochem Biophys 1990;11:145-154. 106. Janz S, Shacter E. A new method for delivering alkanes to mammalian cells: preparation and preliminary characterization of an inclusion complex between 13-cyclodextrin and pristane (2,6,10,14-tetramethylpentadecane). Toxicology 1991 ;69:301-315. 107. Gawrisch K, Janz S. The uptake of pristane (2,6,10,14tetramethylpentadecane) into phospholipid bilayers as assessed by NMR, DSC, and tritium labeling methods. Biochim Biophys Acta 1991" 1070(2):409--418. 108. Kalergis AM, Lopez CB, Becket MI, Diaz MI, Sein J, Garbarino JA et al. Modulation of fatty acid oxidation alters contact hypersensitivity to urushiols: role of aliphatic chain beta-oxidation in processing and activation of urushiols. J Invest Dermatol 1997;108(1):57-61. 109. Kalish RS, Wood JA. Induction of hapten-specific tolerance of human CD8+ urushiol (poison ivy)-reactive T lymphocytes. J Invest Dermatol 1997;108(3):253-257. 110. Hardy RR, Hayakawa K. B cell development pathways. Annu Rev Immuno12001;19:595-621. 111. Richards HB, Reap EA, Shaw M, Satoh M, Yoshida H, Reeves WH. B cell subsets in pristane-induced autoimmunity. Curr Top Microbiol Immuno12000;252: 201-207. 112. Leak LV, Potter M, Mayfield WJ. Response of the peritoneal mesothelium to the mineral oil, pristane. Curr Top Microbiol Immunol 1985;122:221-233. 113. Anderson AO, Wax J S, Potter M. Differences in the peritoneal response to pristane in BALB/cAnPt and BALB/cJ mice. Curr Top Microbiol Immunol 1985;122:242-253. 114. Potter M, MacCardle RC. Histology of developing plasma cell neoplasia induced by mineral oil in BALB/c

103

mice. J Natl Cancer Inst 1964;33:497-515. 115. Garosi G, Di Paolo N. Recent advances in peritoneal morphology: the milky spots in peritoneal dialysis. Adv Perit Dial 2001;17:25-28. 116. Hjelmstrom P. Lymphoid neogenesis: de novo formation of lymphoid tissue in chronic inflammation through expression of homing chemokines. J Leukoc Bio12001;69(3):331-339. 117. Stott DI, Hiepe F, Hummel M, Steinhauser G, Berek C. Antigen-driven clonal proliferation of B cells within the target tissue of an autoimmune disease. The salivary glands of patients with Sjtigren's syndrome. J Clin Invest 1998;102(5):938-946. 118. Xanthou G, Polihronis M, Tzioufas AG, Paikos S, Sideras P, Moutsopoulos HM. "Lymphoid" chemokine messenger RNA expression by epithelial cells in the chronic inflammatory lesion of the salivary glands of SjSgren's syndrome patients: possible participation in lymphoid structure formation. Arthritis Rheum 2001 ;44(2):408-418. 119. Schroder AE, Greiner A, Seyfert C, Berek C. Differentiation of B cells in the nonlymphoid tissue of the synovial membrane of patients with rheumatoid arthritis. Proc Natl Acad Sci USA 1996;93(1):221-225. 120. Randen I, Mellbye OJ, Forre O, Natvig JB. The identification of germinal centres and follicular dendritic cell networks in rheumatoid synovial tissue. Scand J Immunol 1995;41 (5):481-486. 121. Singh B, Rabinovitch A. Influence of microbial agents on the development and prevention of autoimmune diabetes. Autoimmunity 1993;15:209-213. 122. Adams RB, Planchon SM, Roche JK. IFN-gamma modulation of epithelial barrier function: time course, reversibility, and site of cytokine binding. J Immunol 1993;150:2356-2363. 123. Blanco P, Palucka AK, Gill M, Pascual V, Banchereau J. Induction of Dendritic Cell Differentiation by IFNalpha in Systemic Lupus Erythematosus. Science (Wash, DC) 2001 ;294(5546): 1540-1543. 124. Linker-Israeli M, Deans RJ, Wallace DJ, Prehn J, OzeriChen T, Klinenberg JR. Elevated levels of endogenous IL-6 in systemic lupus erythematosus. A putative role in pathogenesis. J Immunol 1991;147(1):117-123. 125. Malide D, Russo P, Bendayan M. Presence of tumor necrosis factor alpha and interleukin-6 in renal mesangial cells of lupus nephritis patients. Hum Pathol 1995;26:558-564. 126. Martin M, Schwinzer R, Schellekens H, Resch K. Glomerular mesangial cells in local inflammation: induction of the expression of MHC class II antigens by IFN-gamma. J Immunol 1989; 142(6): 1887-1894.

104

127. Nanki T, Tomiyama J, Arai S. Mixed connective tissue disease associated with multicentric Castleman's disease. Scand J Rheumatol 1994;23:215-217. 128. Graninger WB, Hassfeld W, Pesau BB, Machold KP, Zielinski CC, Smolen JS. Induction of systemic lupus erythematosus by interferon-gamma in a patient with rheumatoid arthritis. J Rheumatol 1991;18(10): 16211622.

129. Fiorani C, Sacchi S, Bonacorsi G, Cosenza M. Systemic sarcoidosis associated with interferon-alpha treatment for chronic myelogenous leukemia. Haematologica 2000;85(9): 1006-1007. 130. Dumoulin FL, Leifeld L, Sauerbruch T, Spengler U. Autoimmunity induced by interferon-alpha therapy for chronic viral hepatitis. B iomed Pharmacother 1999;53(5-6):242-254. 131. Ioannou Y, Isenberg DA. Current evidence for the induction of autoimmune rheumatic manifestations by cytokine therapy. Arthritis Rheum 2000;43:14311442. 132. Kalkner KM, Ronnblom L, Karlsson Parra AK, Bengtsson M, Olsson Y, Oberg K. Antibodies against doublestranded DNA and development of polymyositis during treatment with interferon. Q J Med 1998;91(6): 393-399. 133. Ronnblom LE, Aim GV, Oberg KE. Autoimmunity after alpha-interferon therapy for malignant carcinoid tumors (see comments). Ann Intern Med 1991;115(3): 178-183. 134. Kim T, Kanayama Y, Negoro N, Okamura M, Takeda T, Inoue T. Serum levels of interferons in patients with systemic lupus erythematosus. Clin Exp Immunol 1987 ;70(3 ):562-569. 135. Hooks JJ, Moutsopoulos HM, Geis SA, Stahl NI, Decker JL, Notkins AL. Immune interferon in the circulation of patients with autoimmune disease. N Engl J Med 1979;301 (1):5-8. 136. Preble OT, Black RJ, Friedman RM, Klippel JH, Vilcek J. Systemic lupus erythematosus: presence in human serum of an unusual acid-labile leukocyte interferon. Science (Wash, DC) 1982;216(4544):429--431. 137. Baechler EC, Batliwalla FM, Karypis G, Gaffney PM, Ortmann WA, Espe KJ et al. Interferon-inducible gene expression signature in peripheral blood cells of patients with severe lupus. Proc Natl Acad Sci USA 2003; 100(5):2610-2615. 138. Bennett L, Palucka AK, Arce E, Cantrell V, Borvak J, Banchereau J e t al. Interferon and granulopoiesis signatures in systemic lupus erythematosus blood. J Exp Med 2003;197(6):711-723.


Vaccination and Autoimmunity Anabel Aron-Maor ~and Yehuda Shoenfeld ~,2

;Department of Medicine 'B' & Center for Autoimmune Diseases, Chaim Sheba Medical Center (affiliated to Tel-Aviv University), Tel-Hashomer 52621, Israel; 2Incumbent of the Laura Schwarz-Kipp Chair for Research of Autoimmune Diseases, Tel-Aviv University

Starting in the third decade of the 20th century vaccination against some of the most common infectious diseases (measles, mumps, diphtheria, rubella, polio) was introduced, reducing the morbidity of these diseases by close to 100% by the end of the century [ 1]. Thus, the benefits of immunization are irrefutable. However, there have been over the last 15 years or so several reports of adverse autoimmune reactions to various vaccines. Mostly the connection between the vaccination and the autoimmune reaction was temporal and not causal. This nevertheless did not prevent such concern (to put it mildly) among the medical and general community that certain childhood vaccinations all but stopped being administered to large populations (such as the MMR vaccine in Britain in the 80's). There are two kinds of vaccination: 9 Active vaccination: when a live, generally attenuated infectious agent (microbe or virus) is used, or an inactivated infectious agent (or constituents thereof), or products obtained by genetic recombination. Active vaccination may also be achieved when injecting a toxoid. 9 Passive vaccination: usually provides temporary immunity and consists of immune globulin preparations or antitoxins. The vaccines usually contain not just the specific antigens but also adjutants (such as aluminum salts or carder proteins). These are introduced in order to potentiate, or boost, the immune response to some antigens. Their action is non-specific. The purpose of vaccination is to induce immunization a reaction of the immune system that will provide the organism with protection against -

disease. As any other medical treatment, vaccines also have side effects, from local reactions, to systemic side effects such as flu-like or hypersensitivity reactions. For more than 15 years reports have been accumulating of autoimmune reactions to various vaccines. Mostly case reports but also some case series of patients who developed autoimmune signs and syndromes. The aim of this chapter is to surmnarize the autoimmune manifestations that have been reported in connection with various vaccines, as well as the possible explanations to these occurrences. As already mentioned, it is imperative to emphasize that so far no causal connection has been demonstrated between any one vaccine and an autoimmune syndrome (even though strong evidence exists to suggest such a connection in regard to reactive arthritis following rubella vaccine [2, 3]). All reported cases of autoimmune manifestations have been only temporally related to the respective vaccines.

1. VACCINES AND ARTHRITIS The occurrence of arthritis has been described following administration of several vaccines (Table 1) and can be divided into isolated or reactive arthritis (poly or monoarticular) and arthritis as a symptom of a systemic autoimmune disease (such as Systemic Lupus Erythematosus (SLE) or rheumatoid arthritis (RA)). Some vaccines have been implicated more often than others.

105

Table 1. Vaccines associated with post-vaccination reactive arthritis and RA Disease

Vaccine

Reactive arthritis

HBV

Rheumatoid arthritis

SLE

Rubella Mumps and measles Influenza DPT Typhoid HBV

Tetanus HBV Polyvaccine: mumps, measles, tetanus, meningococcal, hepatitis A, polio

1.1. Arthritis and Hepatitis B Vaccine Over the past 10-15 years more than 30 cases of arthritis following vaccination with the HBV recombinant vaccine have been reported. Some were cases of isolated inflammation of the joints, others turned out to be harbingers of frank RA. In 1990 two cases of arthritis were reported shortly after the patients had received HBV vaccines [4, 5]. One of the patients developed poly-arthritis and also erythema nodosum and the other patient reactive arthritis only. The symptoms receded in both patients and there was no evidence of a systemic autoimmune illness later on. During the next nine years more reports were published of people developing arthritis after HBV vaccination [6-12]. Some of the patients described were found to have high titers of rheumatoid factor (RF) in their sera without fulfilling other ACR criteria for RA [7]. Others were carders of genetic markers predisposing to autoimmune disease [12]. In 1998 eleven patients

106

No. of cases

1 4 3 1 >10 >100 1 1 2 2 15 1 6 5 13 7 5

Additional symptoms Erythema nodosum Migratory arthritis, urticaria, oedema of the glottis Hypercalcemia, lytic bone lesions Myalgia Vasculitis Adult-onset Still's disease Reactive arthritis alone

Cutaneous, renal, hematological Cutaneous, renal, hematological

were reported who developed arthritis after receiving HBV recombinant vaccine [12]. Ten of these patients fulfilled the ACR criteria for RA, nine of those required disease modifying drugs. Five of the subjects were carriers of the HLA-DR4 haplotype. Nine of the eleven patients genotyped for HLA-DR and DQ expressed the RA shared motif in their HLA class II genes. The findings from this report suggested that HBV recombinant vaccine may trigger RA in genetically prone individuals. An additional case report supports to a certain extent this hypothesis [9]. This is the case of a 44 year old man who had had myasthenia gravis 20 years earlier and had developed arthritis shortly after administration of HBV vaccine. Overall the occurrence of arthritis, and especially RA after anti-HBV immunization is a rare one [13]. Moreover, in studies that examined the response of known RA patients to HBV vaccination it was shown that the administration of the vaccine was not associated with an appreciable deterioration of any laboratory or clinical parameters of

the disease [ 14], and that 68% of patients produced antibodies. Older age and higher scores of daytime pain were associated with a lower rate of antibody production after vaccination. To summarize there are three explanations suggested [15] to the apparent association between immunization and arthritis: - it represents the chance occurrence of two common phenomena. - immunization precipitates a specific form of arthritis that is distinct from RA and that is usually self-limited (post-immunization arthritis). - immunization is one of the factors that can trigger the manifestation of R A - as can infection.

1.2. Arthritis and Rubella Vaccine Joint manifestations have been documented often in connection with rubella vaccine, as well as with the wild virus itself [16, 17]. In 1991 the Institute of Medicine released a report in JAMA examining adverse effects of the DPT (diphtheria-pertussis-tetanus) vaccine and the rubella vaccine (strain 27/3). The report concluded that the evidence suggests a causal relation between rubella vaccine and acute arthritis in adult women [18, 19]. No animal studies were available to support or disprove this conclusion. In a large study [20] that included 2658 immunized and 2359 non-immunized children, the incidence of joint manifestations was assessed six weeks after immunization with the MMR (measlesmumps-rubella) vaccine. There was an increased risk of arthralgia or arthritis in the immunized children six weeks after immunization. The risk for frank arthritis was less than after wild rubella infection. As with the HBV vaccine a genetic predisposition to develop autoimmune disease may play a role in the manifestation of acute reactive arthritis in proximity to the administration of the rubella vaccine. In 1998 [21] a group of scientists examined the frequency of HLA-DR in relation to the incidence of acute joint manifestations in 283 white women who had received rubella vaccination post-partum. The conclusion, based on statistical analysis (after adjustment for age, treatment and time post-partum) was that the risk for developing arthritis was 1.9 times greater after rubella vaccination that after placebo. The risk for arthropathy was also influenced by DR interactions - odds to develop post-vaccination arthropathies were 8 times greater in individu-

als with both DR1 and DR4, and 7.6 times greater with both DR4 and DR6 present. An additional risk factor to the development of post-vaccine arthropathy [22] apparently is the titer of pre-vaccine rubella antibodies: the lower the titer the higher the risk of post-vaccination arthropathy. In conclusion, there seems to be a causal relation between the rubella vaccine and post-vaccination arthropathies (arthralgia and arthritis) with an increased risk for individuals with genetic predisposition (HLA-DR4) and for individuals with low pre-vaccination titers of anti-rubella antibodies.

1.3. Arthritis and BCG Vaccine (see separate chapter) Oligo- and poly-articular arthritis has been reported in approximately 3% of patients treated with intravesicular BCG (for bladder carcinoma) 1-3 months after start of treatment [23]. The arthritis is sterile and HLA-B27 has been demonstrated in several of these patients suggesting a resemblance to reactive arthritis [24].

2. VACCINES AND SLE Systemic lupus erythematosus (SLE) is an autoimmune illness involving multiple organs. Its etiology is believed to be multifactorial since presentation (as well as flare-up) of the disease has been observed after exposure to infectious agents(see chapter), ultra-violet light, drugs and various chemicals. Genetic factors also, inevitably, play a significant role in determining who will develop SLE and when. Viral infections have specifically been causally associated with SLE [25-31] and there are documented cases of SLE presenting after vaccinations.

2.1. SLE and HBV Vaccine The HBV vaccine has been relatively frequently associated with manifestations of SLE in both sexes and all age groups [32-37]. It is interesting to note familial "clustering" of cases of post-vaccination lupus. Such as the case of a 24 year old woman and her 7 year old daughter who both developed autoimmune disease (the mother SLE and the daughter

107

ITP-idiopathic thrombocytopenic purpura) 4 and 10 months (respectively) after vaccination against HBV [33]. The issue of safety of HBV immunization of SLE patients has also been addressed so far only in retrospective studies. The safety of such immunization has yet to be definitely determined and prospective studies are needed. So far it seems that most patients mount an adequate immune response to vaccination, even though it may be quantitatively and qualitatively less than in healthy controls [37]. It is recommended that individuals at risk of exposure to hepatitis B, be immunized.

3. VACCINATION AND N E U R O L O G I C A L AUTOIMMUNE MANIFESTATIONS

Table 2. Vaccines associated with GBS Tetanus toxoid [48, 49] Bacille Calmette-Guerin [48] Rabies [48] Smallpox [48] Mumps [48] Rubella [48] Hepatitis B [48] Diphtheria [49] Poliovirus [48, 49]

Three major neurological autoimmune manifestations have been addressed in conjunction with vaccination: the Guillain- Barre syndrome (GBS), multiple sclerosis and autism.

lion adults received influenza ("swine-flu") virus vaccine [48] the incidence of GBS increased by a factor of four to eight. Additional vaccines have been associated with the occurrence of GBS and Table 2 summarizes the vaccines that have been related with GBS.

3.1. Vaccination and GBS

3.2. Vaccination and Multiple Sclerosis

GBS is a transient neurological disorder characterized by areflexic motor paralysis with mild sensory disturbances. In the patients' cerebrospinal fluid there is an acellular rise of total protein associated with inflammatory demyelination of the peripheral nerves [38]. The exact etiology of the syndrome remains unclear, however there is increasing evidence to suggest an autoimmune etiology [39]. Autoantibodies to various myelin-associated glycoconjugates are described in GBS patients [40] and prior viral infections are often associated with the onset of GBS [41-45]. Approximately 30% of GBS cases are preceded by Campylobacterjejuni infections [46] as detected by serologic tests. It can be concluded that GBS is probably both a humoral and a cellular autoimmune disease induced by infection with multiple microorganisms. The presence of microbe-specific antibodies and T-cells with cross-reactivity to various nerve-sheath components initiates inflammatory demyelination and shedding of peripheral nerve auto-antigens [47]. Presumably by a similar mechanism vaccines can induce an autoimmune reaction. In the autumn of 1976 after a government-sponsored mass-inoculation program in which 45 mil-

Multiple sclerosis (MS) is a disease characterized by central nervous system demyelination and progressive paralysis. It is considered an autoimmune disease of unknown etiology in which the pathologic process is caused by a cell-mediated autoimmune process directed against nerve-sheath myelin. Autoantibodies specific for the central nervous system/oligodendrocyte glycoprotein were identified [50]. These autoantibodies were specifically bound to disintegrating myelin around axons in lesions of acute MS. MS has been connected to hepatitis B vaccine in one of the largest and most heated debates and law suites in France and the US [51]. More than 600 cases of illnesses, many with MS-like symptoms have been reported in France among people who have received recombinant HBV vaccine. The temporal association between MS and HBV vaccination has been reported on few occasions [52-53]: neurological symptoms and signs as well as magnetic resonance imaging documenting CNS demyelinization have been documented days to weeks after BV vaccination. On the other hand, a French government sponsored study in 1997 revealed that vaccinated individuals were less likely to have MS [52].

108

The measles vaccine also has been investigated in conjunction with MS. There are several lines of evidence that support the possibility that MS may be an age dependent host-response to measles. MS patients have higher titers of measles antibodies than do healthy controls and paramyxovirus inclusions have been found in brain cells of MS patients [54]. A hypothesis has been raised that measles might be one cause of MS, based on the finding that in areas where measles occurs at a later age in the majority of population (such as in Scandinavia) the incidence of MS is higher. Whereas in areas where measles occurs at a generally younger age the incidence of MS is significantly lower [54]. However, even though the incidence of measles dropped precipitously after measles vaccination began in the US in 1963, no effect was seen approximately 30 years later on the incidence of MS [55].

3.3. Vaccination and Autism The behavioural syndrome of autism in children is considered to be a neuro-developmental disorder identified by neuro-psychiatric manifestations that include few or no imaginative and language skills, repetitive rocking and self-injurious behavior, and abnormal responses to sensations, people, events and objects. The cause of the syndrome is not known but the etiology may be multifactorial, including environmental, genetic, immunological and as yet undiscovered biochemical and neuropathological factors. An immune hypothesis involving autoimmunity as one possible pathogenetic mechanism in autism has been suggested [56] based on a family study of infantile autism in the presence of autoimmune disease. The vaccine most commonly associated with autism was the measles vaccine. The hypothesis suggested was that were an immunological assault to occur prenatally or post-nataly (during infancy or early childhood) it could possibly result in poor myelination or abnormal function of the axon myelin. The hypothesis was supported by an association found in autistic children, between anti-viral and brain autoantibodies [57]. Slightly higher titers of measles IgG were found in autistic children compared to normal controis and the higher the measles antibodies' titer the greater the likelihood of brain autoantibodies [58]. An additional theory has been suggested connect-

ing autism, the measles-mumps-rubella vaccine and a new pathological entity - a form of chronic inflammatory bowel syndrome - lymphoid nodular hyperplasia (LNH) [59]. It is a different entity from Crohn's disease or ulcerative colitis - this is a reactive swelling of the lymphoid (immune) tissue of the ileal and colonic lining. Symptoms include abdominal pain and change of bowel habits and it may transient or persistent. Autistic children with this finding were also found to be immune deficient [60] lacking in one or more lymphocyte subsets or in immunoglobulin IgG subclasses - findings consistent with an acquired immunodeficiency. The measles component of the MMR vaccine has been implicated in the etiology of this syndrome. The connection between the gastrointestinal findings and autism was hypothetically explained by the possibility that the primary site of damage in autism is outside the brain causing an arrest in the normal development of the brain and its function. Other studies, however, failed to demonstrate the presence of measles virus in lesions of inflamed bowel from these children. So far data is conflicting and there has been no consistent scientific support to the alleged connection between the measles vaccine (or wild virus) and autism [61 ]. An epidemiological study reassessing the association between measles and autism was published in 1999 [62]. It transpired from this study that since 1979 there was a steady increase in the number of cases of autistic children by year of birth with no sudden increase after the introduction of the MMR vaccine. Also there was no difference in age of diagnosis between children vaccinated before or after 18 months of age and those who were never vaccinated. An increased potential risk for neurodevelopmental disorders might related to increased doses of thimerosal- an organic mercury compound that is metabolized to ethyl-mercury and thiosalycylate and that has been used as a preservative in some vaccines since the 1930's [63]. Other, systemic, autoimmune phenomena have been sometimes described in relation to several vaccines. Especially renal involvement has been documented on several occasions [64-66]. The renal disorder developed after receiving a variety of vaccines, among which polio vaccine [67], smallpox, tetanus toxoid and influenza [68]. Rarely vasculitis has been described in conjunction with vaccination

109

with HBV recombinant vaccine [69].

4. VACCINATION OF PATIENTS W I T H KNOWN A U T O I M M U N E DISEASE The question was addressed in a number of studies over the years and several vaccines have been examined in this context. The swine-flu influenza vaccination was well tolerated by patients with MS [70--71] without any influence on the course of the disease (more exacerbations). Likewise, the widely used influenza vaccine has been well tolerated by MS patients [72] with similar rates of systemic reactions as the general population. Influenza vaccine did not cause any worsening of the disease in SLE patients either [73-74]. Response to immunization however (as measured by antibody titers), was lower in patients with lupus than in healthy controls [74-75]. A similar kind of reduced response to vaccination was observed in RA patients as well [76]. Nevertheless the vaccine does not have any noxious effects on RA Patients and does not cause an increased frequency of flares of the disease [76-77]. Since the danger to these patients with systemic autoimmune diseases is great in case of influenza infection they should receive the vaccine. The same is true for other vaccines where the infection itself poses a real and significant danger to these patients.

5. POSSIBLE M E C H A N I S M S OF VACCINERELATED A U T O I M M U N I T Y Many common infections can induce a transient rise in autoantibody production. A similar rise in autoantibody production has been observed after various vaccinations. Such autoantibodies usually resolve within a period of two months [78] but can persist in rare cases. Several studies indicate that stimulation of autoantibody production has become one of the criteria of establishing the safety of vaccines. It is to be remembered however that although autoantibodies are a characteristic of autoimmune disease it is often unclear whether they are an epiphenomena or represent the causal agents of the illness. The human immune system is highly complex, it displays both specificity and memory and

110

is designed to provide protection against almost all infections. The drawback of such a complex and broadly responding immune system is that in responding to infection, the immune system of a few individuals will "turn against" the self and cause autoimmunity [79]. An infection (or vaccine for that matter) can induce autoimmunity via two mechanisms: antigen-specific or antigen-non-specific. An autoimmune condition will arise, however, only if the individual is genetically predisposed to that condition. A common explanation of how an infectious agent can cause autoimmunity via an antigen-specific mechanism is the molecular mimicry theory. Antigenic determinants of the infectious can thus be recognized by the host immune system as being similar to antigenic determinants of the host itself [79]. The situation is more complex for molecular mimicry that involves T lymphocytes. These cells recognize their antigen as short peptides bound to MHC molecules. To serve as a molecular mimic an infectious agent's antigen must copy the shape of a self-antigentic epitope bound to the appropriate MHC molecule. Experimental findings have shown that a single T-cell receptor can recognize a broad range of sequences [80-81 ] including peptides with totally different sequences. It has been calculated that each individual T-cell should be able to recognize more than one million distinct peptide epitopes [81]. Thus, the probability of T-cell cross-reactivity is so high that one wonders why all infection do not induce severe autoimmune disease?! Another mechanism whereby microorganisms (or vaccines) may cause autoimmunity involves bystander activation. This is an antigen-non-specific mechanism. In this instance the infecting agent causes release of previously sequestered self-antigens or stimulates the innate immune response, resulting in activation of self-antigen-expressing antigen presenting cells. Evidence for this mechanism has come from several studies on transgenic mice [82-83]. Autoimmune disease is most likely to be induced in the infected organ. For example mice that harbour high numbers of islet antigen-specific T cells developed diabetes only when infected with an islet-cell tropic virus [84]. The effect of this virus has been reproduced by an islet-cell damaging drug but not by non-specific T-cell activation [85]. These findings imply that viruses can precipitate disease by damaging tissue and causing the release and presentation of previ-

ously sequestered self-antigens. So far we can conclude that there is a high probability for infectiousagents' antigen to cross-react with self-antigens and also that autoimmune disorders can be triggered by the innate immune response to microorganisms. The fact that autoimmune disease does not occur more frequently is probably due to a "fail-safe" mechanism that the immune system has evolved to prevent extensive tissue damage in response to infection. The immune system is controlled by homeostatic mechanisms [86]. Lymphocytes have to compete with each other for antigen and growth factors. Furthrmore, T cell reaction to antigen is limited by activation-induced cell-death [87]. These mechanisms are designed to keep the lymphocyte population at an optimal predetermined level thus limiting the expansion of self-reactive lymphocytes [86]. In a lymphopenic setting, self-reactive lymphocytes undergo homeostatic proliferation and are released from peripheral tolerance' thus causing autoimmune disease [88-91]. The immune system is equipped with a wide variety of lymphocytes bearing receptors with varying affinity to antigen [79]. The immune response to a given antigen selects only a strictly limited set of these lymphocytes. The selection depends on several mechanisms" a) the role of antigen processing and MHC-peptide complex formation; b) selective binding of antigenic epitopes to specific MHC molecules; and c) selective depletion of specific lymphocytes by overstimulation (clonal exhaustion or deletion) [92]. Also, the fact that the threshold for activation of T cells is close to the threshold for activation-induced cell death results in a highly focused reaction of the immune system to any antigen [93, 94]. These mechanisms limit the immune response to antigen and prevent activation of cells beating high-affinity receptors. Since high-affinity receptors are more likely to be crossreactive it is likely this mechanism has evolved to prevent collateral tissue damage, which occurs during the immune response to infection, and to limit the likelihood of self-reactive lymphocyte activation during infection [79]. An additional control mechanism is imposed on the immune system by "regulatory T cells". The best characterized subset pf T cells is the CD4+CD25+ cell [95, 96]. These cells arise in the thymus where they are positively selected by recognition of self antigen [97]. Unlike the majority of the T cell population which leave the

thymus as naive lymphocytes, the CD25+ cells emigrate the thymus but do not proliferate in response to antigen. They are capable of suppressing the response to self antigens. These cells were first described by Sakaguchi et al [98], who noted that thymectomy of young mice prevented their generation and resulted in widespread autoimmune disease in adult animals. Additional examples exist to the role and importance of these cells [99]. The physiological role of T-regulatory type I cells is probably to moderate the immune response to infection and thereby limit the collateral damage that results from the immune response to an infectious agent [100]. These combined homeostatic and regulatory mechanisms have evolved to ensure that the immune response is focused and controlled, and they prevent the individual from developing autoimmune disease during the course of infection [79]. These mechanisms also apply to the host response to vaccination. It is probable that a killed vaccine would be less likely to activate the innate response to infection and to cause tissue disruption, that a live-attenuated one, thereby reducing the risk of autoimmune disease. Nevertheless the degree of activation achieved by an attenuated organism will be much less than that induced by the wild strain. Every new vaccine should therefore be assessed on a case-by-case basis giving extreme consideration to the potential benefit, in terms of public health provision.

6. VACCINATION AND DIABETES Over the past few decades there has been a steady increase in the incidence of type I diabetes in most countries in the world. It is not surprising therefore that childhood vaccinations have been deemed suspect as a potential trigger for this disease. This possibility has been assessed in several epidemiological studies. Results of a case-control study done in Sweden in the 1980's has shown no significant influence of several vaccines (anti tuberculosis, smallpox, tetanus, pertussis and rubella) on the incidence of type I diabetes [101 ]. One vaccine in particular has been suggested to be related with an increased risk for diabetes - the Haemophilus influenza type b (Hib) vaccine [102, 103] especially if given at age two months or older. This theory however, was not confirmed in a 10-year follow-

111

up study that included more than 100,000 Finnish children [104]. Results of this study showed no increased risk of diabetes when children who had received four doses of vaccine at ages 3, 4, 6 and 14-18 months were compared with those who had received only one dose at age 2 years. Additionally, findings of a study undertaken in four large health-maintenance organizations in the USA did not suggest an association between administration of routine childhood vaccines and an increased risk for diabetes, irrespective of the timing of Hib or hepatitis B vaccination [105]. Therefore, at the present time there is no conclusive evidence of any great effect of childhood vaccines on the occurrence of diabetes type I.

7. CONCLUSION Vaccination has been perhaps the greatest medical discovery of the 20th century with the greatest impact on public health. Thanks to this treatment some infectious diseases have been virtually eradicated since the beginning of the century. Over the last two decades there has been increasing concern about possible effects of vaccines on the occurrence of autoimmune diseases, based mainly on case reports connecting the onset of autoimmune phenomena to vaccination. In this chapter we have reviewed the main data available on vaccines and autoimmune diseases. There exist no criteria for diagnosing vaccine-related autoimmune disease. Epidemiological studies so far have not shown conclusively that there exists a causal relation between any one vaccine and any autoimmune disease. Appropriate epidemiological studies should be done before a particular autoimmune clinical condition is associated with a given vaccination. A possible increased risk for the development of autoimmune conditions has been suggested by findings in several case reports and series, where familial or genetic risk factors for autoimmune conditions has been found in many of the patients who had developed autoimmune disease (or phenomena) shortly after vaccinations. It is interesting to note that autoimmune phenomena related to vaccination occur equally in males and females, unlike "regular" autoimmune diseases which are prevalent mainly in women). Based on the above, vaccination of any person with known

112

such risk factors should be carefully considered. However, the degree of vaccine-related risk should always be compared with that associated with the corresponding natural infection, either for the whole population or for a specific subgroup. It is important to mention that vaccination of patients with known autoimmune diseases (such as RA or SLE) has not caused exacerbation of their condition and that most of these patients have mounted a good antibody response to the vaccine (even though attenuated as compared to healthy controls). Criteria for the assessment of adverse effects of vaccines have been established by the World Health Organization (WHO). The four basic principles that apply to autoimmune disease are: the consistency, strength and specificity of the association between the administration of a vaccine and an adverse event, and the temporal association. 1) Consistency and strength- the findings should be the same if the vaccine is given to a different group of people, by different investigators and irrespective of the method of investigation. 2) Specificity - the association should be distinctive and the adverse event linked uniquely or specifically to the vaccine concerned. An adverse event could be caused by a vaccine adjuvant or additive, rather than by its active component. 3) Temporal relation - receipt of the vaccine should precede the earliest manifestation of the event or a clear exacerbation of a continuing condition. A clear distinction should be made between autoimmunity and autoimmune disease. Autoimmunity is a feature of the healthy immune system. Laboratory measurable signs of autoimmunity can associate with infection and occasionally with vaccination. Fortunately, the immune system has evolved sufficient fail-safe mechanisms to prevent these sign from developing into clinical autoimmune disease in the majority of instances.

REFERENCES 1. 2. 3.

NassM, Safety of the smallpox vaccine among military recipients. JAMA 2003;290:2123--4. NossalGJV. Vaccination and autoimmunity. J Autoimmunity 2000;14:15-22. ShoenfeldY, Aron-Maor A, Sherer Y. Vaccination as an additional player in the mosaic of autoimmunity. Clin

4.

5.

6.

7.

8.

9.

10.

11.

12.

13. 14.

15.

16.

17.

18.

Exp Rheumato12000; 18. Rogerson SJ, Nye FJ. Hepatitis B vaccine associated with erythema nodosum and polyarthritis. BMJ 1990;301:345. Haschulla E, Houvenagel E, Mingui A, Vincent G, Laine A. Reactive arthritis after hepatitis B vaccination. J Rheumatol 1990; 17:1250-1251. Biasi D, De Sandre G, Bambara LM, Carletto A, Caramaschi P, Zanoni G, Tridente G. A new case of reactive arthritis after hepatitis B vaccination. Clin Exp Rheumatol 1993; 11:215. Biasi D, Carletto A, Caramaschi P, Frigo A, Pacor ML, Bezzi D, Bambara LM. Rheumatological manifestations following hepatitis B vaccination. A report of two clinical cases. Recenti Prog Med 1994;85:438-440. Gross K, Combe C, Kruger K, Schattenkirschner M. Arthritis after hepatitis B vaccination. Report of three cases. Scand J Rheumatol 1995;24:50-52. Cathebras P, Cartry O, Lafage-Proust MH, Lauwers A, Acquart S, Thomas T, Rousset H. Arthritis, hypercalcemia and lytic bone lesions after hepatitis B vaccination. J Rheumatol 1996;23:558-560. Maillefert JF, Sibilia J, Toussirot E, Vignon E, Eschard JP, Lorcerie B, Juvin R, Parchin-Geneste N, Piroth C, Wendling D, Kuntz JL, Tavernier C, Gaudin P. Rheumatic disorders developed after hepatitis B vaccination. Rheumatology (Oxf) 199;38:978-983. Grasland A, Le Maitre F, Pouchot J, Hazera P, Bazin C, Vinceneux P. Adult-onset Still's disease after hepatitis A and B vaccination. Rev Med Interne 1998;91: 134-136. Pope JE, Stevens A, Howson W, Bell DA. The development of rheumatoid arthritis after recombinant hepatitis B vaccination. J Rheumatol 1998;25:1687-1693. Sibilia J, Maillefert JF. Vaccination and rheumatoid arthritis. An Rheum Dis 2002;61:575-576. Elkayam O, Yaron M, Caspi D. Safety and efficacy of vaccination against hepatitis B in patients with rheumatoid arthritis. Ann Rheum Dis 2002;61:623-625. Symmons DPM, Chakravarty K. Can immunization trigger rheumatoid arthritis. Ann Rheum Dis 1993;52: 843-844. Weibel RE, Bemor DE. Chronic arthropathy and musculoskeletal symptoms associated with rubella vaccine. A review of 124 claims submitted to the National Vaccine Injury Compensation Program. Arthritis Rheum 1996;39:1529-1534. Ray P, Black S, Shinefield H, Dillon A, Schwalbe J, Holmes S, Hadler S, Chen R, Cochi S, Wassilak S. Risk of chronic arthropathy among women after rubella vaccination. Vaccien Safety Datalink Team. JAMA 1997;278:551-556. Howson CP, Fineberg HV. Adverse events following

19.

20.

21.

22.

23. 24.

25.

26. 27.

28.

29.

30. 31. 32.

33.

pertussis and rubella vaccines. Summary of a report of the Institute of Medicine. JAMA 1992;267:392-396. Howson CP, Katz M, Johnston RB Jr, Fineberg HV. Chronic arthritis after rubella vaccination. CUn Infect Dis 1992;15:307-312. Benjamin CM, Chew CG, Silman AJ. Joint and limb symptoms in children after immunization with measles, mumps and rubella vaccine. BMJ 1992;304:10751077. Mitchell LA, Tingle AJ, MacWilliam L, Home C, Keown P, Gaur LK, Nepom GT. HLA-DR class II associations with rubella vaccine-induced joint manifestations. J Infect Dis 1998;177:5-12. Mitchell LA, Tingle AJ, Grace M, Middleton P, Chalmers AC. Rubella virus vaccine associated arthropathy in postpartum immunized women: influence of preimmunization immunologic status on development of joint manifestations. J Rheumatol 2000;27:418--423. Tischler M, Shoenfeld Y. Tuberculose et immunite: ou en sommes nous? Ann Inst Pasteur 1996;7:133-136. Belmatong E, Levy Ojebbour, Appelboom T, De Gennes C, Peltier AP, Meyer O, Kahn MF, Carbon C. Polyarthritis in four patients treated with intravesical BCG for bladder carcinoma. Rev Rhum 1993;60: 130--131. James JA, Kaufman KM, Farris AD, Taylor-Albert E, Lehman TJ, Harle JB. An increased prevalence of Epstein-Barr virus infection in young patients suggests a possible etiology for systemic lupus erythematosus. J Clin Invest 1997;100:3019-3026. Vaughan JH. Epstein-Barr virus and systemic lupus erythematosus. J Clin Invest 1997; 100:2939-2940. Dror Y, Blachar Y, Cohen P, Livni M, Rosenmann E, Ashkenazi A. Systemic lupus erythematosus associated with acute Epstein-Barr virus infection. Am J Kidney Dis 1998;32:825-828. Incaprera M, Rindi L, Bazzichi A, Garzelli C. Potential role of Epstein-Barr virus in systemic lupus erythematosus autoimmunity. Clin Exp Rheumatol 1998; 16: 289-294. Dostal C, Newkirk MM, Duffy KN, Paleckova A, Cerna M, Zd'arsky M, Zvarova J. Herpes viruses in multicase families with rheumatoid arthritis and systemic lupus erythematosus. Ann 1Wt Acad Sci 1997;815:334-337. Keen GA, Yeats J. Epstein-Barr virus-induced systemic lupus erythematosus. S Afr Med J 1996;86:850. Matsukawa Y. Epstein-Barr virus-induced systemic lupus erythematosus. S Afr Med J 1996;86:850-851. Tudela P, Marti S, Bonal J. Systemic lupus erythematosus and vaccination against hepatitis B. Nephron 1992;62:236. Finielz P, Lam-Kam-Sang LF, Guiserix J. Systemic lupus erythematosus and thrombocytopenic purpura in

113

34. 35.

36.

37.

38. 39. 40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

114

two members of the same family following hepatitis B vaccine. Nephrol Dial Transplant 1998; 13:2420-2421. Guiserix J. Systemic lupus erythematosus following hepatitis B vaccine. Nephron 1996;74:441. Mamoux V, Dumont C. Lupus erythemateux dissemine et vaccination contre l'hepatite B. Arch Pediatr 1994;1: 307-308. Grezard P, Chaff M, Philippot V, Perrot H, Faisnat M. Cutaneous lupus erythematosus and buccal aphthosis after hepatitis B vaccination in a &year old child. Ann Dermatol Venereol 1996;123:657-659. Ioannou Y, Isenberg DA. Immunization of patients with systemic lupus erythematosus: the current state of play. Lupus 1999;8:497-501. Ropper AH. The Guillain-Barre syndrome. New Engl J Med 1992;326:1130-1136. Rostami AM. Pathogenesis of immune-mediated neuropathies. Pediatr Res 1993;33:590-594. Fredman P, Vedeler CA, Nyland H, Aarli JA, Svennerholm L. Antibodies in sera from patients with inflammatory demyelinating polyradiculoneuropathy reactive with ganglioside LM1 and sulfatide of peripheral nerve myelin. J Neurol 1991;238:75-79. Melnick SC. Role of infection in the Guillain-Barre syndrome. Neurol Neurosurg Psychiatry 1964;27: 395-407. Dowling PC, Cook SD. Role of infection in GuillainBarre syndrome: laboratiry confirmationof herpes virus in 41 cases. Ann Neurol 1981;9(suppl):44-55. Van Doom PA, Brand A, Vermeulen N. Anti-neuroblastoma cell-line antibodies in inflammatory demyelinating polyneuropathy: inhibition in vitro by intravenous immunoglobulins. Neurology 1988;38:1592-1595. Schmitz H, Enders G. Cytomegalovirus as a frequent cause of Guillaine-Barre syndrome. J Med Virol 1977;1:21-27. Grose C, Spizland I. Guillain-Barre syndrome following administration of live measles vaccine. Am J Med 1976;60:441-443. Greunewald R, Ropper AH, Lior H, Chan J, Lee R, Molinaro VS. Serologic evidence of Campylobacter jejuni and E. coli enteritis in patients with GuillainBarre syndrome. Arch Neurol 1991 ;48:1080-1082. Abu-Shakra M, Shoenfeld Y. Natural Autoantibodies. In: Shoenfeld Y, Isenberg D, eds. Bocaa-Raton: CRC Press, 15-33. Ropper AH, Victor M. Influenza vaccination and the Guillain-Barre syndrome. New Engl J Med 1998;339: 1845-1846. Stratton KR, Howe CJ, Johnston RB. Adverse events associated with childhood vaccines other than pertussis and rubella. JAIMA 1994;271:1602-1605. Genain CP, Canella B, Hauser SL, Raine CS. Identifica-

51. 52.

53. 54. 55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

65.

tion of autoantibodies associated with myelin damage in multiple sclerosis. Nature Med 1999;5:170-175. Marshall E. A shadow falls on hepatitis B vaccination effort. Science 1998;281:630-631. Herroelen L, de Keyser J, Ebinger G. Central nervous system demyelination after immunization with recombinant hepatitis B vaccine. Lancet 1991;338: 1174-1175. Nadler JP. Multiple sclerosis and hepatitis B vaccination. Clin Infect Dis 1993; 17:928-929. Alter M. Is multiple sclerosis an age-dependent host response to measles? Lancet 1976;28:456--457. Currier RRD. Measles vaccination has had no effect on the occurrence of multiple sclerosis. Neurol 1996;53: 1216. Weizman A, Weizman R, Szekely GA, Wijsenbeek H, Livni E. Abnormal immune response to brain tissue antigen in the syndrome of autism. Am J Psychiat 1982;7:1462-1465. Singh VK, Warren RP, Odell JD, Warren WL, Cole P. Antibodies to myelin basic protein in children with autistic behaviour. Brain Behav Immun 1993;7: 97-103. Singh VK, Sheren XL, Yang VC. Serological association of measles virus and human herpesvirus-6 with brain autoantibodies in autism. Clin Immunol Immunopathol 1998;89:105-108. Lee JW, Melgaard B, Clementa CJ. Autism, inflammatory bowel disease and MMR vaccine. Lancet 1998 ;351:905-909. Gupta S, Aggarnal S, Heads C. Brief report: Dysregulated immune system in children with autism: beneficial effects of intravenous immunoglobulins on autistic characteristics. J Autism Develop 1996;26:439-450. Peltola J, Patja A, Leinikki P, Valle M, Davidkin L, Paunio M. No evidence for measles, mumps and rubella vaccine-associated inflammatory bowel disease or autism in a 14 year prospective study. Lancet 1998;351: 1327-1328. Taylor B, Miller E, Farrington CP, Petropoulos MC, Favot-Mayaud I, Li J, Waight PA. Autism and measles, mumps and rubella vaccine: no epidemiological evidence for a causal association. Lancet 1999;353: 2026--2029. Geier MR. Neurodevelopmental disorders following thimerosal-containing vaccines. Exp Biol Med 2003;228:660--4. Freud P, Maffia AJ, Hosbach RE, Valicenti PW, Smallpox vaccination followed by acute renal failure. Amer J Dis Child 1960;99:98-100. Stefanini M, Piomelli S, Ostrowski JF, Colpoys WP. Acute vascular purpura following immunization with asiatic influenza vaccine. New Engl J Med 1958;259:

9-12. 66. Bishop WB, Carlton RF, Fanders LL. Diffuse vasculitis and death after hyperimmunization with pertussis vaccine. Report of a case. New Engl J Med 1966;274: 661-669. 67. Izumi AK, Matsunaga J. BCG vaccine-induced lupus vulgaris. Arch Dermatol 1982; 118:171-172. 68. Baxter AG, Horsfall AC, Healey D, Ozegbe P, Day S, Wiliams DJ, Cooke A. Mycobacteria precipitate an SLlike syndrome in diabetes-prone NOD mice. Immunology 1994;83:227-231. 69. Le Hello C, Cohen P. Boousser MG, Letellier P. Suspected hepatitis B vaccination-related vasculitis. J Rheumatol 1999;26:191-192. 70. Bamford CR, Sibley WA, Laguna JE Swine influenza vaccination in patients with multiple sclerosis. Arch Neurol 1978;35:242-243. 71. Kurland LT, Molgaard CA, Kurland EM, Wiederholt WC, Kirkpatrick JW. Swine flu influenza vaccine and multiple sclerosis. JAMA 1984;251:2672-2675. 72. Salvetti M, Pisani A, Bastianello S, Millefiorini E, Buttinelli C, Pozzilli C. Clinical and MRI assessment of disease activity in patients with multiple sclerosis after influenza vacciniation. J Neurol 1995;242:143-146. 73. Brodman R, Gilfillan R, Glass D, Schur PH. Influenzal vaccine response in systemic lupus erythematosus. Ann Intern Med 1976;88:735-740. 74. Williams GW, Steinberg AD, Reinertsen JL, Klassen LW, Decker JL, Dolin R. Influenza immunization in systemic lupus erythematosus. A double blind trial. Ann Intern Med 1978;88:729-734. 75. Ristow SC, Douglas RG, Condemi JJ. Influenza vaccination of patients with systemic lupus erythematosus. Ann Intern Med 1978;88:786-789. 76. Cimmino MA, Seriolo B, Accardo S. Influenza vaccination in rheumatoid arthritis. J Rheumatol 1995;22: 1802. 77. Heron A, Dettleff G, Hixon B, Brandwin L, Ortbals D, Hornick R, Hahn B. Influenza vaccination in patients with rheumatic diseases. Safety and efficacy. JAMA 1979;242:53-56. 78. Borchers AT, Keen CL, Shoenfeld Y, Silva J, Gershwin ME. Vaccines, viruses and voodoo. J Invest Allergol Clin Immuno12002;12:155-168. 79. Wraith DC, Goldman M, Lambert PH. Vaccination and autoimmune disease: what is the evidence? Lancet (published online on June 3, 2003 on http: //image.thelancet.corn/extras/02art9340web.pdf). 80. Hemmer B, Jacobsen M, Somner M. Degeneracy in T-cell antigen recognition: implications for the pathogenesis of autoimmune diseases. J Neuroimmunol 2000; 107:148-153. 81. Mason D. A very high level of crossreactivity is an

82.

83.

84.

85.

86.

87.

88.

89.

90.

91.

92.

93.

94.

95. 96.

essential feature of the T-cell receptor. Immunol Today 1998;19:395-404. Kissler S, Anderton SM, Wraith DC. Antigen-presenting cell activation: a link between infection and autoimmunity. J Autoimmun 2001; 16:303-308. Kissler S, Anderton SM, Wraith DC. Cross-reactivity and T-cell receptor antagonism of myelin basic proteinreactive T cells is modulated by the activation state of the antigen-presenting cell. J Autoimmun 2002;19: 183-193. Horwitz MS, Bradley LM, Harbertson J. Diabetes induced by coxsackie virus: initiation by bystander damage and not molecular mimicry. Nat Med 1998;4: 781-785. Horwitz MS, Ilic A, Fine C, Rodriguez E, Sarvetnick N. Presented antigen from damaged pancreatic beta cells activates autoreactive T-cells in virus mediated autoimmune diabetes. J Clin Invest 2002; 109:79-87. Theophilopoulos AN, Durraner W, Kono DW. T cell homeostasis of systemic autoimmunity. J Clin Invest 2001; 108:335-340. Walker LS, Abbas AK. The enemy within: keeping self-reactive T cells at bay in the periphery. Nat Rev Immuno12002;2:11-19. Bucy RP, Xu XY, Li J, Huang G. Cyclosporin Ainduced autoimmune disease in mice. J Immunol 1993; 151:1039-1050. Sakaguchi N, miyai K, Sakaguchi S. Ionizing radiation and autoimmunity: induction of autoimmune disease in mice by high dose fractionated total lymphoid irradiation and its prevention by inoculating normal T cells. J Immunol 1994;152:2586-2595. Morse SS, Sakaguchi N, Sakaguchi S. Virus and autoimmunity: induction of autoimmune disease in mice by mouse T lymphotropic vLrus (MTLV) destroying CD4+ T cells. J Immunol 1999;162:5309-5316. Kono DH, Balomenos D, Pearson DL. The prototypic Th2 autoimmunity induced by mercury is dependent on WN-gamma and not Thl/Th2 imbalance. J Immunol 1998; 161:234-240. Sercarz EE, Lehmann PV, Ametani A. Dominance and crypticity of T cell antigenic determinants. Annu Rev Immunol 1993; 11:729-766. Anderton SM, Radu CG, Lowrey PA, Ward ES, Wraith DC. Negative selection during the peripheral immune response to antigen. J Exp Med 2001;193:1-11. Anderton SM, Wraith DC. Selection and fine-tuning of the autoimmuune T cell repertoire. Nat Rev Immunol 2002;2:487-498. Maloy KJ, Powrie F. Regulatory T cells in the control of immune pathology. Nat Imuno12001 ;2:816-822. Certified Professionals SEM. CD4+ CD25+ suppressor T cells. J Exp Med 2001;193:41-45.

115

97. Itoh M, Takahashi T, Sakaguchi N. Thymus and autoimmunity: production of CD25+CD4+ naturally anergic and suppressive T cells as a key function of the thymus in maintaining immunologic self tolerance. J Immunol 1999;162:5317-5326. 98. Sakaguchi S, Sakaguchi N, Assano M, Itoh M, Toda M. Immunologic self tolerance maintained by activated T cells expressing IL-2 receptor alpha-chains (CD25): breakdown of a single mechanism of self-tolerance causes various autoimmune diseases. J Immunol 1995;155:1151-1164. 99. Olivares-Villagomez D, Wang Y, Lafaille JJ. Regulatory CD4+ cells expressing endogenous T cell receptor chains protect myelin basic protein-specific transgenic mice from spontaneous autoimmune encephalomyelitis. J Exp Med 1998;188:1883-1894. 100. Trinchieri G. Regulatory role of T cells producing both interferon gamma and interleukin 10 in persistent infection. J Exp Med 2001;194:53-57.

116

101. Blom L, NystromL, Dahlquist G. The Swedish childhood diabetes study: vacinations and infections as risk determinants for diabetes in childhood. Diabetologia 1991;34:176--181. 102. Classen JB, Classen DC. Immunization in the forst month of life may explain decline in incidence of IDDM in the netherlands. Autoimmunity 1999;31: 43-45. 103. Classen B, Classen DC. Association between type I diabetes and Hib vaccine: causal relation is likely. BMJ 1999;319:1133. 104. Karvonen M, Cepaitis Z, Tuomilehto J. Association between type I diabetes and Haemophilus Influenza type b vaccination: birth cohort study. BMJ 1999;318: 1169-1172. 105. DeStefano F, Mullooly JP, Okoro CA. Childhood vaccinations, vaccination timing and risk of type I diabetes mellitus. Paediatrics 2001;108:E112.


BCG Vaccination Moshe Tishler

Department of Medicine 'B', Assaf Harofe Medical Center, Zerifin, Israel; Tel Aviv University Sackler School of Medicine, Israel

1. INTRODUCTION

2. BACKGROUND PATHOPHYSIOLOGY

The issue of autoimmune manifestations or diseases following administration of various vaccines has been discussed extensively in the literature, both in case reports and reviews. Even though vaccinations have been found to be one of the greatest achievements of modem medicine, a growing number of reports have raised the question of the causal relationship between them and autoimmune phenomena. The relationship between vaccinations and autoimmunity is bi-directional. On one hand, vaccinations prevent infectious diseases and thus can prevent autoimmune diseases that might be triggered by infectious agents. On the other hand, case reports and series that describe post-vaccination autoimmune phenomena give rise to the suspicion that they can also trigger some autoimmune diseases in a similar way to the infectious agents from which we try to protect. Bacillus Calmette-Gu6rin (BCG) vaccine was derived from an attenuated strain of M. bovis and was first administered to humans in 1921. This vaccine is recommended for routine use at birth in countries with high tuberculosis prevalence and for health care employees in high risk areas. Another use for this vaccine is as an adjuvant treatment for superficial bladder cancer where it is given intravesically. In this chapter the link between mycobacterial infection, BCG immunotherapy and autoimmunity will be discussed.

The BCG vaccine acts mainly by an immunological mechanism. Mycobacteria have been found to be immunogenic [ 1] and antibodies to phosphoglycolipids extracted from mycobacterial cell wall have been suggested as a diagnostic tool for detecting active tuberculosis. Furthermore, many autoantibodies can be detected with high frequency in infected patients with mycobacteria [2]. It has been postulated that mycobacteria share antigens with human tissue. Monoclonal anti-DNA antibodies derived from patients and mice with systemic lupus erythematosus (SLE) were found to bind to three glycoproteins derived from mycobacterial cell wall [3]. This binding could be inhibited by prior incubation of the antibodies with glycolipid antigens and with anti ss-DNA, thus indicating that mycobacterial and human tissue share common antigens. On the other hand, autoantibodies detected in chromic relapsing experimental autoimmune encephalomyelitis have been shown to react as well with mycobacteria [5]. The similarity between human and mycobacterial antigens is not limited to the humoral levels and extends to cellular mechanisms also. Studies done in 1984/5 in a model or of rat with adjuvant arthritis succeeded in establishing a Tcell clone specific for M. tuberculosis which was strongly arthritogenic [6]. This clone has been able to recognize, in addition to M. tuberculosis antigens present in human synovial fluid, progeoglycans purified from human cartilage and chondrocyte culture medium [7]. The molecular mimicry between mycobacteria and human tissues has been based on studies of

117

adjuvant arthritis in a rat model. In this model cross antigenicity between M. tuberculosis and proteoglycans purified from human cartilage has been demonstrated [7]. Furthermore, T-lymphocytes taken from patient with rheumatoid arthritis (RA) have shown augmented reactivity to a fraction of mycobacteria cross-reactive with human cartilage [8]. The systemic immune effect of mycobacteria is not limited to this model and has also been assessed in autoimmune-prone mice. In an animal SLE model a single dose of M. bovis injected to prediabetic NOD mice resulted in prevention of type I diabetes and onset of a systemic autoimmune disease similar to SLE [9]. This disease was characterized by hemolytic anemia antinuclear antibodies (ANA), severe sialadenitis and glomerulonephritis. Characterization of the B-cell responses in these mice have shown that they were directed against ds DNA and the Sm ribonucleoprotein complex [10]. The antids DNA and anti-Sm antibodies were not a direct result of polyclonal stimulation although it was most likely occurring [11]. This molecular similarity between mycobacterial and host antigens, which has repeatedly been determined, was the conceptual basis for the usage of mycobacteria as a therapeutic agent.

only as case reports with no more than 30 cases reported thus far [14-17]. Usually it is manifested by symmetric polyarthritis affecting the large joints of the lower limbs, associated also with low back pain. Arthritis can spread to other joints after recurrent instillations, and severe polyarthritis has been described even after the eighth BCG administration [17]. Radiographic evaluation usually reveals no abnormal findings and laboratory tests show only non-specific signs of inflammation. In all cases reported the synovial fluid culture was negative for mycobacterial and negative PCR tests argue against the possibility of active mycobacterial infection. Review of the literature discloses that in approximately 50% of the cases HLA B27 was positive and in fewer cases HLA DR4 was demonstrated [18]. Most cases resolve completely with NSAID treatment without further joint sequelae. Another very rare autoimmune phenomenon induced by BCG vaccination given as TB prophylaxis is dermatomyositis. Only three case reports could be detected in the literature, describing three adolescents developing dermatomyositis [ 19-21 ].

4. BCG AND ARTHRITIS- MECHANISM OF ACTION 3. BCG IN CLINICAL PRACTICE Treatment of urinary bladder carcinoma by intravesical installation of Bacillus Calmette-Gu6rin has been used successfully since 1976. The BCG given for superficial bladder carcinoma does not destroy the tumor cells directly but rather increases the local immune response, eventually eliminating the tumor cells [ 12]. The great majority of patients (- 95%) tolerate treatment without any serious side effects. The most common side effects are malaise, low-grade fever, cystitis and hematuria, all of which are short term and resolve spontaneously. Intravesical BCG can also result in more severe side effects with systemic phenomena such as rash (0.5%), renal abscess (0.1%), epididymitis (8.4%), sepsis (8.4%), pneumonitis and hepatitis (0.7%), cytopenia (0.7%) and arthritis or arthralgia (0.5%) [13]. Arthritis secondary to intravesical BCG administration is a rare systemic side effect poorly documented in the literature. BCG induced arthritis has been reported

118

The mechanism of action of BcG induced arthritis is not clear. Animal studies have shown that an intact host immune system is required for the antitumoral effect of BCG [15], which hints us that the probable mechanism involved in the therapeutic activity of the vaccine is immunologic. Furthermore, clinical and laboratory evidence suggests that the anti-tumor activity is concentrated at the site of BCG administration, thus supporting the local immune mechanism as an important factor for its therapeutic effect [22]. It has been shown that following repeated instillations of BCG organisms into the bladder, large quantities of various cytokines can be detected in the urine. Although the responses among patients were heterogeneous, concentration of all cytokines detected (IL-1, IL-2, IL-6, IL-8, IL-10, TNF alpha, IFN gamma and soluble ICAM-1) were increased following BCG intravesical therapy [24]. The most likely explanation is based on the concept of molecular mimicry which has developed following the

model of adjuvant arthritis. In this model performed in rats cross antigenicity was demonstrated between M. tuberculosis and proteoglycans purified from cartilage [7]. Moreover, T-cell fines grown from patients shortly after intravesical BCG treatment showed a strong expression of HLA-DR on their surface which persisted several months afterwards [24]. The results of these studies suggest that proliferation of a CD4 T-cell clone might take place. This clone might have specificity to a common antigen shared by both cartilage proteoglycans and the mycobacterial cell wall [25]. Penetration of the bacterial or bacterial antigens through the wall of the vascularized tumor in the bladder to the circulation can generate a systemic immune response. Therefore, the T-cell clone that has been suggested to proliferate can now attack the joints. Such attack can take place in genetically susceptible individuals such as those having specific HLA antigens. Indeed, more than 50% of patients with BCG induced arthritis were carrying the HLA B27 antigen. Genetic susceptibility has been demonstrated by in vitro studies as well. It has been shown that peripheral blood lymphocytes from patients with RA responded vigorously to specifically to PPD but not to other polyclonal mitogens [26]. Enhanced T-cell responses to a fraction of mycobacteria have also been shown in RA patients. This T-cell response was more pronounced in these patients compared to patients with degenerative joint disease and to healthy controls. This effect of generating an attack on a host organism from within has been given the name "Trojan Horse" [27] and is not unique just to BCG vaccination but rather to numerous autoimmune phenomena following vaccination.

important role as well.

5. C O N C L U S I O N

9.

BCG immunization, especially by intravesical instillation, can act as "a double edged sword". Although it proved to be a powerful tool in the treatment of superficial bladder cancer, it can trigger autoimmune phenomena and even full blown autoimmune disease, which are fortunately rare. The link between mycobacteria and autoimmunity is probably a consequence of molecular mimicry, although genetic and environmental factors play an

REFERENCES 1.

2.

3.

4.

5.

6.

7.

8.

10.

11. 12.

TeplizkiH, Buskila D, Alkan M, Coates A, Baumgarten A, Shoenfeld Teplizki H, Buskila D, Alkan M, Coates A, Baumgarten A, Shoenfeld. ELISA measurement of antibody titer to purified protein derivative and mycobacteria derived phosphoglycolipids: tools for diagnosing active pulmonary tuberculosis. Isr J Med Sci 1987;23:1121-1124. Lindqvist KJ, Coleman RE, Osterland CK. Autoantibodies in chronic, pulmonary tuberculosis. J Chron Dis 1997;22:717-725. ShoenfeldY, Volmar Y, Coates ARM, Rauch J, Shaul D, Pinhas J. Monoclonal anti-tuberculosis antibodies react with DNA, and monoclonal anti-DNA and autoantibodies react with Mycobacterium tuberculosis. Clin Exp Immunol 1986;66:255-261. Thorns CJ, Morris JA. Common epitopes between mycobacteria and certain host tissue antigens. Clin Exp Immunol 1985;61:323-328. Glynn P, Weeda D, Edwards J, Suchliny AJ, Cuzner ML. Humoral immunity in chronic relapsing experimental encephalomyelitis. The major oligoclonal IgG bands are antibodies to mycobacteria. Neurol Sci 1982;57:369-364. HoloshitzJ, Matitiahu A, Cohen IR. Arthritis induced in rats by cloned T-lymphocytes responsive to mycobacteria but not to collagen type II. J Clin Invest 1984;73:211-215. Van Eden W, Holoshitz J, Nevo Z, Frenkel A, Klujman A, Cohen IR. Arthritis induced by a T-lymphocyte clone that responds to M. tuberculosis and to cartilage proteoglycans. Proc Natl Acad Sci 1985;82:5113-5120. Holoshitz J, Irdujman A, Drucker I, Lapidot Z, Yaretzky A. Frenkel A et al. T-lymphocytes of rheumatoid arthritis patients show augmented reactivity to a fraction of mycobacteria cross-reactive with cartilage. Lancet 1986;2:305-309. Baxter AG, Horsfall AC, Healy D, Ozegbe P, Day S, William DG et al. Mycobacterial precipitate an SLElike syndrome in diabetes-prone NOD mice. Immunol 1994;83:235-248. Horsfall AC, Howson R, Silveira RA, Williams DG, Baxter G. Characterization and specificity of B cell responses in lupus induced by M. bovis in NOD/Lt. mice. Immunology 1998;95:8. Baxter AG, Cooke A, Peptide therapy in diabetes. Lancet 1974;343:1169. Kurth KR, Bouffloux C, Sylvester R, van der Meijden

119

13.

14.

15.

16.

17.

18.

19. 20.

120

AP, Oosterlink W, Brause M. Treatment of superficial bladder tumors: achievements and needs. The EORTC Genitourinary Group. Eur Urol 2000;27(Suppl 3): 1-9. Lamm DL, van der Meijden AP, Morales A, Brosman SA, Catalona WJ, Herr HW et al. Incidence and treatment of BCG intravesical therapy in superficial bladder cancer. J Urol 1992;147:596-600. Mas AJ, Romera M, Valverde Garcia JM. Articular manifestations after the administration of intravesical BCG. Joint Bone Spine 2002;69:92-93. Shoenfeld Y, Aron-Maor A, Tanai A, Ehrenfeld M. BCG and autoimmunity: Another two-edged sword. J Autoimmun 2001 ;16:235-240. Mouly S, Berenbaum F, Kaplan G. Remitting seronegative synovitis with pitting edema following intravesical BCG instillation. J Rheumato12001;28:1699-1701. Pardalidis NP, Papatsoris AG, Kosmaoglou EV, Georganas C. Two cases of acute polyarthritis secondary to intravesical BCG adjuvant therapy for superficial bladder cancer. Clin Rheumato12002;21:536--537. Clavel G, Grados F, Cayrolle G, Bellony R, Leduc I, Lafont Bet al. Polyarthritis following intravesical BCG immunotherapy. Report of a case and review of 20 cases in the literature. Rev Rhum Engl Ed 1999;66:115-118. Kass E, Straume S, Munthe E. Dermatomyositis after BCG vaccination. Lancet 1978;8067:772 (Letter). Ehrengut W. Dermatomyositis and vaccination. Lancet 1978;8072:1040-1041.

21. Kass E, Straume S, Mellbye OJ, Munthe E, Salheim BG. Dermatomyositis associated with BCG vaccination. Scand J Rheum 1979;8:187-191. 22. Prescott S, Jackson AM, Hawkyard SJ, Alexandroff AB, James K. Mechanisms of action of intravesical BCG: local immune mechanisms. Clin Infect Dis 2000;31:591-593. 23. Jackson AM, Alexandroff AB, Kelly RW, Skibinska A, Esuvaranathan K, Prescott S et al. Changes in urinary cytokines and soluble intracellular adhesion molecule-1 (ICAM-1) in bladder cancer patients after BCG immunotherapy. Clin Exp Immunol 1995;99:369-375. 24. Prescott S, James K, Busuttil A, Hargreave TB, Chisholm GD, Smyth JF. HLA-DR expression by high grade superficial bladder cancer treated with BCG. Br J Urol 1989;63:264-269. 25. Holoshitz J, Naparstek Y, Ben Nun A, Cohen IR. Lines of T lymphocytes induce or vaccinate against autoimmune arthritis. Science 1983;219:56-58. 26. Abrahamson G, Froland SS, Matvig JB. In vitro mitogen stimulation of synovial fluid lymphocytes from rheumatoid arthritis and juvenile rheumatoid arthritis patients: dissociation between the response to antigens and polyclonal mitogens. Scand J Immunol 1978;7: 81-90. 27. Aron-Maor A, Shoenfeld Y. BCG immunisation and the 'qu Horse" phenomenon of vaccination. Clin Rheumato12003;22:6-7.


Viruses: The Culprits of Autoimmune Diseases? A.M. Denman ~and B. Rager-Zisman 2

Worthwick Park Hospital Harrow, Harrow, UK; 2Dept of Microbiology and Immunology, The University Center for Cancer Research, Faculty of Health Sciences, Ben Gurion University of the Negev, Beer Sheva, Israel

1. INTRODUCTION One of the first principles in the organisation of the immune system is the need to avoid autoimmune diseases resulting from immune reactions to self antigens (Ehrlich's "horror autotoxicus"). The classical explanation for the prevention of autoimmunity was given by Burnet in his clonal selection theory which postulated that clones with the potential to produce autoantibodies are selectively eliminated during the development of immune responses in normal individuals. Breakdown of tolerance to serf-antigens was viewed as a general mechanism leading to autoimmune diseases [1]. Subsequent discoveries have revealed the complexities of the immune system and its regulation. Consequently this theory is now seen by many observers as an over simplification since autoimmune responses may have a physiological role in tissue repair and maintenance [2]. Furthermore it is increasingly evident that autoantibodies in human autoimmune disease are directed at a very limited number of self-proteins, perhaps 2% of the estimated 20,000-60,000 of the potential total [3]. Indeed the same limited repertoire of auto-antigens is the target in mouse strains developing spontaneous autoimmune disease. Although this observation does not negate classical theories based on broken tolerance, it suggests that changes in the milieu in which auto-antigens are presented to the immune system must be at least as important as universal defects in tolerance maintenance. In recent years the idea has been prominent that autoimmune diseases result from a failure of regulatory mechanisms. Particular emphasis has been placed on the regulatory

role of the T cell suppressor population identifiable by their specific CD25 + CD4 § phenotype [4]. The basic issue of tolerance maintenance has important implications for those seeking an infectious aetiology for autoimmune diseases. If the immune system has evolved to meet pathogens, it is likely that these disorders arise as part of the anti-microbial response and not as the result of the random emergence of auto-reactive clones unrelated to host defence [5]. Furthermore, the pathogenesis of human autoimmune diseases is unlikely to have a single cause since experimental models point to aberrations in a wide range of immune responses which may lead to autoimmune diseases [6]. Tissue damage, irrespective of the cause, may induce ephemeral autoimmune reactions but these rarely persist as autoimmune diseases. Viral infections are an attractive explanation for human autoimmune diseases because of the many potential ways in which these may perpetuate local inflammation and also subvert local and systemic immune responses. In classical terminology, proponents of the viral theory of autoimmune diseases must find persuasive evidence to back the many explanations based on the theoretical ability of these agents to break tolerance. This task has at least been eased by the recognition that auto-reactive T and B cell clones are part of the normal immune repertoire. The issue can be more precisely defined as ways in which virus infection may subvert the mechanisms which normally limit this reactivity. There are many excellent recent reviews summarising the evidence for a viral aetiology of autoimmune diseases [7, 8]. However fresh insights into

123

the complex mechanisms of anti-viral host defence and its interactions with different viruses continue to provide new insights and experimental evidence for this hypothesis. The object of this review is to examine the validity of the viral hypothesis in the light of this evidence. In a broader perspective there is increasing acceptance that infections contribute to chronic diseases of unknown aetiology [9]. Given the wealth of information on this issue, our approach is largely dictated by the need to devise acceptable criteria for linking infection by a given virus to a specific autoimmune disease. For this purpose we have attempted to adapt Koch's classical postulates [10] for a bacterial aetiology of disease in a way which takes account of the complex interactions between viruses and the host immune response. However we recognise that almost certainly autoimmune diseases do not arise through a single mechanism. Furthermore autoimmune diseases can be induced experimentally which are clearly independent of viral or indeed any obvious antigenic stimulation. A cogent example is myopathy with autoantibodies to histidyl-transfer RNA synthetase (anti-Jo-1) in mice with transgenic overexpression of MHC class 1 molecules [ 11 ].

2. HISTORY AND EVOLUTION OF K O C H ' S POSTULATES There were many doubts in Koch's times about the relevance of bacterial infection to the pathogenesis of disease even in pulmonary tuberculosis when the characteristic symptoms and signs of infection formed a reasonably consistent picture [10]. His postulates provided a satisfactory basis for experimental validation because isolating tubercle bacilli correlated reasonably well with disease activity. Furthermore there was an available laboratory model of tuberculosis induced by the bacillus. Yet even in this situation there are recognised difficulties. The presence of m. tuberculosis in tissues does not necessarily imply tuberculous disease. The outcome of infection is largely determined by host immunity. Perhaps the greatest benefit derived from the postulates was to establish verifiable rules in contrast with the unbridled, often philosophical speculation which characterised so much contemporary discussion on the aetiology of tuberculosis

124

and indeed most diseases. Any attempt to link viral infections with human autoimmune diseases must depend on comparable rules adapted to test this hypothesis. However the difficulties are formidable. Viruses have more complex life histories with a wider range of possible outcomes than bacterial infections. The range of anti-viral host responses is immense and limiting or eradicating viral infection necessitates an attack on the cells in which viruses complete their replication cycle or persist in defective or latent form. Furthermore, viruses have evolved strategies for circumventing or subverting these responses in order to enhance their survival and replication. Thus there is an almost unlimited number of plausible mechanisms by which viral infection could initiate autoimmune diseases. A basic problem is our ignorance of many fundamental issues concerning the natural history of bacterial and viral infections alike and their interactions with host defence mechanisms. To compound these difficulties, there is still controversy over the extent to which autoimmunity contributes to many chronic inflammatory human diseases even when these are associated with autoimmune features such as circulating autoantibodies. There is a vast body of information about the immunopathological features of these diseases but these have thrown rather limited light on their origin. Equally, there is no lack of contrived or spontaneous experimental models of diseases whose features resemble human autoimmune disease. A major difficulty is to determine the relevance of these models to the aetiology operating in human autoimmune disease. Perhaps the most reassuring feature of these models is the almost identical yet limited repertoire of auto-antigens recognized by the immune system in human a autoimmune diseases models [3]. This review attempts to expand and adapt Koch's criteria to the problem of linking viral infection with autoimmune disease.

3. GENERAL T H E O R I E S OF AUTOIMMUNE DISEASES There are many excellent general reviews on the general subject of autoimmune diseases. However our ideas on the origin of these diseases general are rapidly changing. Even the traditional distinction between conventional and autoimmune reactions

appears less absolute [12]. The molecular mechanisms of organ-specific autoimmune disease such as myasthenia gravis [ 13] have been elucidated in fine detail. The precise role of autoimmunity is more difficult to unravel in multi-system diseases such as the connective tissue diseases and rheumatoid arthritis where autoantibody may be secondary to other proinflammatory mechanisms. Its contribution to other diseases of unknown aetiology such as multiple sclerosis is even more controversial. We consider only those aspects which are crucial to the issue of viral infection and autoimmune disease. Auto-immune reactivity has been shown to form part of the normal T and B cell repertoire. Indeed self recognition is likely to prove an essential step in the generating an effective response to foreign antigens including microbial infection [14]. For example T cells from normal donors proliferate in response to in vitro stimulation with histones and nucleosomes and clonally expanded T cells reactive to these auto-antigens circulate in similar numbers in normal control subjects and SLE patients [15]. Even persistent, circulating autoantibodies are not necessarily associated with disease; organ-specific autoantibodies, rheumatoid factor, and anti-nuclear antibodies are present in some 2-5% of females over the age of 40 without any apparent immunopathological consequences. Classical immunological theory rigidly distinguished between non-specific, innate immunity and specific immunity mediated by T and B cells. The origin of autoimmune diseases has long been considered to arise from anomalous T or B lymphocyte function. It is now evident that this rigid distinction is incorrect; since all modalities of nonspecific and specific immunity have been variously implicated even immediate type hypersensitivity, an area not usually considered relevant in the context of autoimmune diseases, can not be ignored. Histamine release in immediate type reactions enhances Tn 1 responses through its activating effect on their type 1 histamine receptors [16]. As a result histamine secretion by mast cells contributes to the inflammatory reaction in experimental autoimmune diseases such as experimental allergic encephalomyelitis and to human autoimmune diseases such as rheumatoid arthritis and bullous pemphigus [ 17]. In keeping with these observations, mice strains which are genetically deficient in mast cells are resistant

to an experimental model of inflammatory arthritis

[~8]. The wider spectrum of responses potentially contributing to autoimmune disease is highly relevant to theories of initiation by viral or indeed other forms of infection because the many it increases the potential points at which excessive inflammation could initiate autoimmune disease. Nevertheless the dogma that autoimmune diseases result from a breakdown of tolerance to self antigens remains the key issue [19] even if classical theory needs extensive revision. There is still a consensus view that central tolerance depends on the elimination in the thymus of potentially autoreactive T cells with high affinity receptors for self-antigens. This form of tolerance is difficult to break. Peripheral tolerance is defined as the process by which T cells migrating from the thymus to the periphery express low affinity receptors for autoantigens and normally remain tolerant. However peripheral tolerance is readily broken, usually as the result of inflammation or tissue damage. Indeed, peripheral autoreactivity is not necessarily pathological but can reasonably be considered a physiological process. Furthermore, the ability to mount reactions against self-components is an integral part of the process of counteracting the destructive effects of inflammatory reactions. This role is amplified in the host response to infections associated with tissue damage to an extent dependent on the duration and extent of the infection. It is a situation especially inherent in viral infections which are commonly characterised by indefinite viral persistence in a wide range of host cells. Transient autoimmune phenomena commonly accompany infectious diseases but rarely lead to overt autoimmune disease. Peripheral events affecting tolerance are complex because of the vast range of interactions between different cell populations and their products. For example experimental manipulation indicates that a given cytokine for example will break or enhance tolerance depending on the experimental conditions. Transgenic models show that constitutive over production of selected cytokines induces autoimmune diseases and this over-expression is reproduced in the resulting inflammatory lesions. Nevertheless some general principles have emerged. Disease induction depends critically on

125

the nature of auto-antigen presentation and abrogation of the many regulatory steps which limit the proliferation of T cells with autoreactive receptors [6]. Transgenic models emphasise the many safeguards against the prolonged breakdown of peripheral tolerance and the emergence of autoimmune disease. There is good experimental evidence that the induction of tolerance in T cells exposed to novel virus-related antigens is also regulated during the maturation of these cells [20]. There are other mechanisms for controlling potential aggressiveness by autoreactive T cells which have escaped elimination in the thymus, a process which has been termed "fine tuning" [21]. It is not surprising therefore that persistent autoimmunity represents a rare escape from a network of restraining factors. The factors governing B cell tolerance are similarly elaborate. Germ line encoded autoreactivity is a normal feature of B cell development and maturation before eventual IgV gene diversification through hypermutation [22]. Experiments involving autoreactivity to snRNPs and ss DNA indicate that tolerance induction in B cells partly depends on encountering auto-antigens during the transition from immature to mature B cells [23]. B cell maturation depends on interactions with other cell populations and cytokines. In addition it is possible that an accumulation of immune complexes interferes with B cell tolerance to auto-antigens; an inability to clear immune complexes and autoantibody production are characteristic findings in experimentally induced complement deficiency and genetically determined complement deficient in man. There are also intriguing observations that malignant lymphoproliferation, an unequivocal example of failed B cell regulation, is associated with a wide spectrum of organ specific and systemic autoimmune diseases [24].

4. GENETIC SUSCEPTIBILITY TO AUTOIMMUNE DISEASE Any hypothesis invoking viral infection in the pathogenesis of autoimmune disease must satisfactorily explain the overwhelming evidence for genetic factors in determining susceptibility to these disorders. Indeed there are animal models of autoimmune disease in which at first sight it appears

126

superfluous to invoke any factors other than single inherited errors in the immune system. In support of classical; theories, genetic abnormalities in T and B cell regulation lead to autoimmune disorders with features resembling human disease. For example mice lacking protein kinase C lambda develop hypergammaglobulinaemia and glomerulonephritis which are characteristic of systemic lupus erythematosus (SLE) [25]. Regulatory defects confined to B cell activation such as kinase deficiency also lead to autoimmune disease [26]. Inherited defects in apoptosis resulting from a single gene defect also induces autoimmune disease as, for example, in mice lacking the membrane tyrosine kinase c-mer [27]. However even autoimmune diseases induced by single gene manipulation result in secondary abnormalities and are thereby realistic models of what may occur in diseases where there are no such clear leads to the underlying error or errors in immune regulation [6]. These abnormalities could well lead to an unusual outcome of infections. Indeed as yet unidentified environmental factors are important in determining the outcome in individuals with familial susceptibility to organ-specific immune diseases who inherit a mutated immune regulatory gene (AIRE) on chromosome 21q22.3 and also HLA class II genes linked to susceptibility to these diseases [28, 29]. A mouse model of this mutation suggests that the proliferation of antigen responsive T cells may also be exaggerated [30] and this could also affect the response to infection. Models in which a combination of genetic factors unequivocally determines the onset of autoimmune disease more closely reproduce the multigenic susceptibility characteristic of human autoimmune disorders such as type I diabetes and SLE. Thus Fas (CD95) deficient mice develop more severe autoimmune haemolytic anaemia and thrombocytopenia when IL-10 levels are constitutively high [31]. We can postulate that human autoimmune diseases arise when infection perturbs the cytokine network and acts in combination with other genetic determined events. Certainly studies of disease incidence in identical twins who are genetically susceptible to autoimmune diseases emphasise the importance of as yet undefined environmental factors. Particular attention has been given to the well documented association between susceptibility or resistance to autoimmune diseases and the inherit-

ance of certain class II HLA genes in man and their MHC counterpart in mice. There is good evidence that these genes control antigen presentation and hence the character of the resulting anti-viral T cell reposnse. This response largely determines the outcome of viral infections. For example HLA antigens determine susceptibility to HIV infection [32] and a vigorous response to this virus anti-viral is associated with the expression of HLA A*02 [33]. HIV infected individuals inheriting HLA B'5701 or 5703 are more resistant to HIV infection because they develop a broad range of CD8 T cell responses variants of the conserved viral p24 epitope [34]. Indeed the evolution of HIV-1 infection is largely determined by HLA restricted anti-viral immune responses [35]. Similarly the HLA class II haplotype DQB 1"0301 protects against hepatitis C virus infection because it enables infected individuals to maintain an effective CD4 T cell response [36]. It is also apparent that more subtle genetic variation determines the risk of post-infectious autoimmune disease. Chagas disease is caused by Trypanosoma cruzi infection and the most serious disease feature is cardiomyopathy. In a mouse model of the disease the myocarditis is directly related to the genetically controlled levels of anti-myosin cell mediated and humoral response [37]. In the human disease there is good evidence that genetically controlled T cell recognition of myosin in an inflammatory milieu contributes to cardiomyopathy [38]. In other situations selective experimental manipulation of the cytokine network affects the pattern of postinfectious immunopathology. Borrelia burgdorferei infection in man usually causes acute disease manifestations including arthritis and myocarditis (Lyme disease). Chronic arthritis with a persistent acute phase response and rheumatoid factor production has been linked to autoimmunity. In experimental Borrelia burgdorferei infection of mice, passively administered antibody to IL-12 increases the severity of the arthritis [39]. This observation suggests that genetically determined differences in cytokine production determine the outcome of human infection by this agent. Linkage studies in human autoimmune diseases emphasise the polygenic risk factors. It is conceivable that these could operate through host resistance to infections including viral infection. Genes predisposing to autoimmune disease control many other features of innate and specific

immunity including cytokine production, complement, and apoptosis. Increasing the complexities, genetic mapping shows that susceptibility genes increase the risk of developing several, different autoimmune diseases and not just a single disease. For example there is an association within the same families between increased susceptibility to organ specific autoimmune diseases and SLE. Genetic predisposition to organ specific autoimmune disease within the same family extends to type I diabetes and thyroiditis. Disease severity is also genetically controlled. The increased susceptibility could also be explained by polygenic influence on the outcome of infections. However the polygenic control of disease manifestation and severity further complicates the exploration of a viral aetiology. The perturbations induced by viral infections could influence the immune response might be secondary effects stages which are not confined to the initial interactions between viral antigens, specialised antigen presenting cells, and responding T cells. In support of this view, studies in a wide range of animal models and clinical situations indicate that genetically determined susceptibility to autoimmune diseases involves many different mechanisms [40, 41].

5. IDENTIFICATION OF HUMAN AUTOIMMUNE DISEASES Autoimmune diseases are defined as diseases arising from autoimmune attack on target organs, tissues, or cells. Autoimmune diseases affect 3-5% of most populations in developed countries although there is some indication that this figure may be lower in undeveloped regions of Africa and Asia. The prevalence of these diseases has been estimated to be as high as 20% [42] but this is probably an exaggeration. Nevertheless the number of autoimmune diseases continues to increase with investigative advances and diagnostic ascertainment is also improving. Conventionally, autoimmune diseases are classified as organ specific or systemic. Organ specific autoimmune diseases are associated with autoantibodies to antigens usually specific for the target organ (Table 1) and the distribution of multiorgan autoimmune disease correlates with the distribution of these antigens. It is reasonable to conclude that these anti-

127

Table 1. Classification of autoimmune diseases Autoimmune pathogenesis

Disease

Established Organ specific

Endocrine

Thyroiditis Type 1 (insulin dependent) diabetes Adrenal insufficiency (Addison's disease) (uncommon) Hypoparathyroid Ovarian disease

Gastro-intestinal

Gastritis and pernicious anemia Hepatitis (some forms)

Haematological

Haemolytic anemia Trombocytopenia Neutropenia

Nervous

Myasthenia gravis Peripheral neuropathy (some forms) Para-neoplastic neuro-muscular diseases Rare disorders (e.g. Stiff-man syndrome)

Renal

Goodpasture's syndrome Nephritis (some forms)

Skin

Pemphigus Pemphigoid Chronic urticaria

Ocular

Sympathetic ophthalmia

Possible Organ specific

Coeliac disease Ulcerative colitis Fibrosing alveolitis Uveitis (some forms) Multiple sclerosis Myocarditis Rheumatic fever

Established Systemic

Systemic lupus erythematosus SjiSgren's disease Vasculitis and polyarthritis (some forms)

Possible Systemic

Rheumatoid arthritis Polymyositis; dermatomyositis Scleroderma

gens are the targets for the immunopathological events which eventually destroy the target organs. However it is more difficult to be sure that these antigens are the same as those which initiated the autoimmune reaction. It should also be admitted that the true contribution of autoimmunity to many of the diseases listed in Table 1 remains controversial. Longitudinal studies of first degree relatives of patients with autoimmune diseases who develop autoantibodies before the advent of clini-

128

cally detectable disease suggest that the number of initiating antigens is very limited. Subsequently self antigens may become the target for autoimmune reactions because the autoirranune response becomes more widespread as the result of epitope spreading or bystander effects. The delayed appearance of antibodies to other auto-antigens may have important clinical consequences even though these appear as secondary events. Thus autoantibodies to the beta 1-adrenoreceptor contribute to the circula-

tory problems of patients with idiopathic dilated cardiomyopathy [43] but this potential consequence of myocardial infection is not considered in experimental models of virus induced myocarditis. This issue is relevant to the viral hypothesis because analysis of the initiating auto-antigens is likely to provide the best clues to the responsible viruses. As an additional misleading factor, an initiating infection in the mother may be responsible for autoimmune disease in her children. Maternal antibodies may well stimulate the production of autoreactive T cells [44, 45]. In support of this idea, eliminating the transfer of maternal autoantibodies to pancreatic islet cells prevents diabetes in NOD mice [46]. The attribution of pathogenic effects to autoantibodies is well attested in myaesthenia gravis and other neurological diseases in which autoantibodies to acetylcholine receptors, voltage-gated calcium channels, and voltage-gated potassium channels account for muscle weakness and related symptoms. Reasonably selective removal of these autoantibodies leads to clinical improvement [13]. Nevertheless, while autoantibodies are often good predictors of organ specific autoimmune disease, circulating autoantibodies are not necessarily associated with clinically significant disease [47]. Nevertheless, these may be provoked by the same factors which initiate autoimmune disease. For example rheumatoid factors, so long regarded as harmful contributors to the immunopathogenesis of rheumatoid arthritis, will probably prove to be part of the normal host defence against infection [48]. Autoantibodies in general may only become harmful when they achieve high affinity. There are other difficulties in defining the contribution of autoimmunity in experimental models and in human disease. T cells isolated from target organs and clones derived from these cells may be demonstrably autoreactive for target organ cells or their antigens judged by in vitro proliferative or cytotoxic assays. However the interpretation of these findings is not straightforward. In vitro experiments do not necessarily reflect the in vivo behaviour of effector populations subject to regulatory constraints. Furthermore tests selected to demonstrate destructive autoimmunity may overlook T cell populations with protective properties; thus autoreactive T cells protect vulnerable neurones from damage in rats with experimentally damaged optic nerves [49].

Similarly, a population of CD4 alpha/beta cells in non-obese diabetic (NOD) mice limits the damage to pancreatic islet cells induced by autoreactive T cells [50]. Such observations have been incorporated in a more general scheme which assigns a possibly regulatory role to autoreactive T cells in response to infection [51 ].

6. PREVALENCE OF AUTOIMMUNE DISEASES The epidemiology of autoimmune diseases could in principle give clues to an infectious aetiology for these disorders although to date such hopes have not materialised. There is much speculation that the incidence of allergic and autoimmune diseases is increasing in developed societies because innate and specific immunity have been blunted by immunisation strategies, improved hygiene, and antibiotics. A decline in the incidence of infectious diseases has been considered a factor in the seeming increasing in the incidence of allergic and autoimmune diseases possibly because of resulting distortions in the relative numbers of T cell populations; decreased infection might lead to a reduction in Th2 cell numbers and of immunoregulatory IL-10 [52]. There is little evidence that immunisation directly induces autoimmune disease [53, 54]. However a longer period of observation and detailed analysis of immunological memory for the immunising virus may be needed, to take full account of immunisation's effects on the long term accuracy of immunological memory for the immunising viruses. T cell memory may be directed at unrelated viruses and may extend to inappropriate reactivity against self antigens. It is also possible that low level virus persistence in target organs following immunization may induce autoimmune rather than effective anti-viral responses [44]. Although there is little doubt that the prevalence of autoimmune diseases is higher in developed than in third world countries, there is no firm evidence that this is related to immunisation.

129

7. ANTI-VIRAL HOST DEFENCES AND THEIR SUBVERSION

Viruses have evolved many strategies to avoid elimination by host defences [55, 56]. LCM virus is a striking example of a virus which can persist in low levels despite a seemingly efficient immune response [57]. Viral persistence increases the risks of autoimmune disease in two general ways. Firstly, a continuing anti-viral response creates an inflammatory environment in which peripheral tolerance is more likely to be broken. Secondly, the strategies by which virus infections persist are likely to disrupt other regulatory mechanisms which normally prevent autoimmunity. A related issue is the duration of virus infection's effects on the regulation of immunity. If disruption continues after the initiating virus has been eliminated or it can no longer replicate, it will be more difficult to link remote infection with persistent immunopathology. Indeed long term disruption of regulatory mechanisms could theoretically act in the same way as inherited abnormalities conferring susceptibility to autoimmune disease. Traditionally, anti-viral immunity has been analysed in rigid compartments, namely innate immunity, and specific humoral and cell mediated immunity. However it is now evident that there are multiple pathways whereby these systems interact in order to mount effective inflammatory and immune reactions to viruses and other pathogens whilst limiting the damage these responses may inflict on infected tissues [58]. The complexities of virus-host interactions are further increased by genetically determined variations in the control of each component of the host response. These topics are the subject of comprehensive reviews and we only consider the implications for theories of virus-induced autoimmunity (Table 2). It is also important to emphasise an emerging theme that defects in innate or specific immunity predispose to inappropriate B cell activation. Defects in the acute phase protein SAP, the complement system, and NK cells predispose to this problem [59]. Viral infection leads to apoptotic or immune mediated host cell death associated with non-specific local inflammation and often a specific autoreactive response to the antigens released by cell death. Delayed clearance of cell debris secondary to viral infection and the resulting immune complexes accentuate the risk.

130

Table 2. Major contributors to innate anti-viral immunity

Reference NK cells

[60]

Complement

[64]

Dendritic cells

[65]

Defensins

[68]

Chemokines

[66, 67]

Cytokines

[69]

Innate T and B lymphocytes

[70]

Gamma/delta T cells

[63]

Inherited defects are extreme examples of situations which may also be acquired as the result of viral infection. For example herpeseviruses in general and CMV in particular down regulate the expression of adhesion molecules and class I MHC molecules as a strategies for evading the host immune response [60]. Similarly some viruses have evolved a strategy for blocking antigen presentation by antigen presenting cells to naive CD8 T cells by encoding "viral proteins interfering with antigen presentation ("VIPRs") [61]. Indeed the range of strategies for viral persistence continues to grow. For example human and murine CMV encodes proteins which have been termed immunoevasins. These interfere with viral antigen processing and presentation by class II MHC antigens [62]. 7.1. Toll Receptors and the Defensin System

In common with other pathogens, viruses are recognised by germline encoded receptors termed pattern recognition receptors which include Tolllike receptors [71, 72] and the defensin system. In mice with defective expression of the Toll receptor TLR-4 deficiency respiratory syncytial virus infection (RSV) induces severe lung disease with immunopathological features resembling fibrosing alveolitis in man [73]. Toll-like receptors expressed in high levels on microglia and astrocytes in the central nervous system present auto-antigens to the immune system and this may contribute to chronic immunopathology [74].

7.2. Cytokines and Chemokines Interactions with inflammatory and specific immune cell populations are mediated mainly by chemokines and cytokines. Genetically determined polymorphism in the chemokine and cytokine systems largely determines the pattern and intensity of innate and specific responses to microbial infection. Inherited patterns of chemokine receptors govern cell activation in response to external stimuli including microbial products [75]. In common with other complex mediator systems, there are also regulatory proteins. The IRAK family of kinases down regulate the pro-inflammatory signals transmitted to monocytes and macrophages thereby reducing cytokine release [72]. It is now apparent that these systems contribute to autoimmune disease. Rheumatoid arthritis is an interesting example of the likely interplay of innate and specific immunity in the pathogenesis of a suspected autoimmune disease. Innate immunity is activated in the rheumatoid synovial membrane. Inappropriate macrophage stimulation could lead to excessive cytokine production including IL-12 initiated by CpG sequences in DNA [76]. Chemokines belong to a superfamily of chemoattractants which attract leukocytes to inflammatory lesions. This superfamily comprises at least 50 structurally related peptide agonists and 20 G protein-coupled receptors. So far more than 30 virally encoded mimics of chemokines and chemokine receptors have been identified [67]. These viruses are mainly herpesviruses, poxviruses, and lentiviruses. The net effect of any given viral infection in vivo is unpredictable and could in theory enhance or suppress local inflammation. The potential of this mechanism to influence the outcome of infection is increased by the discovery that chemokines also influence lymphocyte traffic [77]. Furthermore, T cell subsets with the chemokine receptors CXCR3 and CCR5 are specifically associated with inflammatory reactions [78]. Recent evidence suggests that in a transgenic model of type I diabetes, T cells may be attracted to pancreatic islet beta cells because the latter express chemokine CXC ligand 9 [79]. The distinction between innate and specific immunity has become increasingly blurred in terms of host defence against microbial infections and the risk of developing autoimmune diseases. The vast field of cytokine physiology began with

the discovery of the anti-viral effects of the interferons. The complexities even of the interferon system in isolation are daunting. While there are only fourteen Interferon-alpha, one interferon-beta, and one interferon-gamma gene, DNA microarray technology has revealed that there are hundreds of genes whose expression is influenced by the interferon family [69]. The regulation of other cytokine is also complex. There are many ways in which virus infection could initiate imbalances in the cytokine network leading to autoimmune disease. However the pleomorphic effects of individual cytokines and the complexities of the cytokine network make it difficult to predict the outcome of such infections [80]. The relevance of this field to theories of autoirnmune disease is well illustrated by the variable outcome of viral infections in hosts with genetically determined differences in their interferon response. A crucial phase in the susceptibility of mice to diabetes following coxsackie B4 virus infection is the extent to which the virus grows in host cells. Interferon alpha and beta production in pancreatic islet beta cells is genetically controlled. Low levels of production predispose to cell mediated destruction of infected cells initially mediated by NK cells [81]. However the interferon system illustrates a major dilemma; it is very difficult to disentangle viral virulence enhanced by genetically determined deficiencies from suppression mediated by viral genes [69]. For example mutations in the non-structural genes of the highly virulent influenza virus strain H5N1 confer resistance to interferons and TNF alpha [82].

7.3. Non-Specific Inflammatory Cell Populations The activation of B cells and non-specific inflammatory cell populations also depends on Fc gamma receptors. Deficient expression of the neutrophil specific Fc gamma receptor Fc gamma RIIIB is a risk factor for developing SLE and deletions in the promoter region for an inhibitory Fc gamma receptor are found in some lupus prone mouse strains [83]. Viral infection of dendritic cells is a potent strategy for suppressing specific anti-viral immunity or even inducing long term tolerance. Human cytomegalovirus is especially adept at influencing dendritic cells to reduce viral antigen presentation by class II

131

MHC molecules and deleting potential anti-CMV T cells [84]. Measles virus has similar capacities [85]. Dendritic cells also determine whether immunity or tolerance follows antigen presentation to T cells. This issue is complicated by the heterogeneity of dendritic cells and the nature of cytokine exposure during the process of antigen presentation [65]. Furthermore some dendritic cells have the unusual capacity to perform a process termed cross presentation. Usually MHC class II antigens present peptides derived from microbial antigens which enter the cell by endocytosis. These peptides are recognised by helper CD4 T cells. In contrast MHC class I molecules present peptides primarily derived from endogenously synthesised self or microbial proteins which have been degraded by the proteasome and then transported by TAP molecules of antigen presenting cells. These peptides are presented to cytotoxic CD8 T cells. This distinction ensures that CD8 cells only attack virus infected host cells thereby reducing reduces the risk that uninfected cells will be the target of an autoimmune response. Crosspresentation permits peptides derived from exogenous viral antigens processed by dendritic cells to be presented by class I MHC antigens to cytotoxic T cells. [86]. Probably many viral antigens including influenza, vaccinia, polio, and LCMV are processed and presented in this manner. Cross-presentation implies that self antigens associated with viral peptides can potentially be presented to cytotoxic T cells thereby inducing an autoimmune response. Indeed there is experimental evidence which supports this conjecture. Mice expressing influenza haemagglutinin as a transgenically coded develop diabetes after subsequent infection with influenza virus because anti-viral T cells destroy the islet cells [20]. However it is impossible to generalise as the effects of viral infection on dendritic cells depends on the infecting agent; many viruses including herpesvirus type I, measles, retroviruses, vaccinia, and LCM suppress antigen presentation by these cells. There are other mechanisms by which inflammatory responses to virus infections are expedited with the potential risk of provoking autoimmune reactions. Innate T and B lymphocytes are especially interesting because they express a restricted set of germ-line encoded receptors which recognise conserved structures including self-antigens that are

132

encountered in inflammatory situations including infections [70]. Similarly the activation of NK cells depends on a balance between stimulating signals transmitted by activating receptors and inhibitory signals dependent on contact with histocompatibility antigens. NK cells are activated by contact with virus infected cells and are especially important in host defence against herpesvirus infections. The outcome is determined partly by the nature of the viral infection and partly by host genetic factors. Herpesviruses in general and cytomegalovirus (CMV) in particular depress NK cell activation by simulating the normal inhibitory signal transmitted by histocompatibility antigens [60]. Genetic control of NK cell activation has been shown to affect virus growth. For example C3H/HeN mice are relatively resistant and CBA/J mice are relatively susceptible to Polyomavirus. The virus grows to high titres in the latter strain because the inhibitory molecule CD94-NKG2A is related to weak T cell cytotoxicity for Polyoma virus infected cells [87, 88]. Memour T cells in CBA/J mice are also impaired. In fact viruses have evolved a wide range of mechanisms for suppressing anti-viral defence by NK cells. These involve modulating the expression of receptors on virus infected cells which normally activate NK cells or on NK cells themselves. Furthermore viruses can increase the production of cytokines which inhibit NK cell responses [89]. Other virusinduced proteins inhibit NK destruction of HIV and EBV infected cells. Conversely, NK cells may be non-specifically activated through recognition of viral antigens on infected cells by V gamma 9 / V lambda 2 receptors on NK cells respond to antigens expressed by virus infected cells. These receptors also engage cell surface lipids on infected cells undergoing apoptosis [70]. These interactions are potent routes for activating auto-reactive T cells.

7.4. Specific Immunity The interactions between innate and specific immunity and the many viral factors which influence innate immunity complicate former, relatively simple assumptions about the maintenance of tolerance in the face of acute or persistent viral infections. Clearly, acquired or induced defects in innate

immunity predispose to loss of both T and B cell tolerance. In addition defective T cell function encourages virus virulence and persistence. In particular, infection is aided by major genetic defects which negate normal CD8 T cell function or genetic variation which reduces the efficiency of the response to certain infections. In addition some viruses affect T and B cell function as a survival strategy. Conversely, the host response is a balance between efficiently killing infected cells and the risks of excessive cell death mediated in part at least by autoimmunity. CD8 T cell mediated killing is regulated in large part by CD4 § CD25 § T cells. Appropriate regulation is crucial to the outcome of persistent infections by all pathogens; excessive regulation favours persistence and re-infection while deficient regulation encourages cell death [90]. Another problem arises from the nature of T cell anti-viral memory. This must be sufficiently specific for the host to respond to further infections by the same agent yet sufficiently diverse to respond appropriately to other pathogens [91]. However increasing diverse memory increases the risk of potentially harmful autoreactivity. T cell anti-viral memory becomes less virus specific with the passing of time. For example CD8 memory T cells sensitised by an initial LCM virus infection produce interferon gamma after in vivo exposure to vaccinia virus infection [92]. Furthermore the cross-reactivity with epitopes expressed by a different virus has immunopathological consequences since it induces different forms of lung inflammation including bronchiolitis and destructive changes. The issue of anti-viral memory specificity may be important if T cell memory remains latent after an initial infection but can be reactivated by subsequent infections by the same or different viruses against residual viral antigens from the first infection. This possibility is foreshadowed by experiments in which T cell tolerance to influenza haemagglutin in the islet cells of transgenic mice is broken by subsequent influenza virus infection [20]. There is limited information concerning the duration and nature of anti-viral T cell memory. Antibody levels may be a poor guide to previous infection. Thus specific CD4 and CD8 T cell responses to hepatitis C virus infection are often detectable when anti-viral antibodies are no longer detectable [93]. The duration of memory in different T cell popu-

lations also varies. CD8 T cell memory for LCM virus infection is more stable than CD4 cell memory [94]. The phenotype of memory T cells is not the same in different herpesvirus infections [95]. Viral persistence in different lymphocyte populations has long been recognised and has been explored using both in vitro and in vivo systems. The usual outcome is immunosuppression. Recently many mechanisms have been described by which virus infection aids its persistence by inhibiting specific immunity to the infecting agent. Thus many viruses including respiratory syncytial virus infection suppress CD8 T cell memory [96]. Direct infection is not a prerequisite for selective immunosuppression since there are inhibitory receptors on T cells which can be exploited by viruses such as cytomegalovims [87]. One issue of potential importance is the possibility that viral infections initiate autoimmune reactions by their effects on the repertoire of receptors expressed by T cells and antibodies secreted by B cells. The T cell immunodeficiency characteristic of HIV infection is associated with autoimmune thrombocytopenia and connective tissue diseases resembling SjiSgren's syndrome. However it is not clear whether these complications are related to the accompanying polyclonal proliferation of B cells or to more subtle defects involving altered antigen recognition by surviving T cells. Autoantibody production in vitro by EBV infected B cells is a well documented consequence of the accompanying polyclonal stimulation and ephemeral autoantibody production is a recognised feature of infectious mononucleosis. However the relevance of these observations to the pathogenesis of autoimmune disease remains doubtful. Indeed the short duration and usually inconsequential nature of in vivo autoimmune phenomena after primary EBV infection is more a tribute to the robust nature of regulatory mechanisms than evidence of EBV aetiology for autoimmune diseases. Nevertheless the concept that viruses may stimulate autoantibody production through a mitogenic effect is interesting because other ubiquitously encountered viral antigens such as influenza haemagglutinin have this property, at least in vitro [97].

133

8. PROPOSED MECHANISMS FOR VIRAL INDUCTION OF AUTOIMMUNE DISEASES 8.1. General Theories Many general theories have been advanced to account for virus-induced autoimmune diseases (Table 3). However this classification is artificial since many mechanisms may operate simultaneously. In particular T and B cells may be activated relatively specifically because of defects in the immune system or as a secondary event in an inflammatory environment. It is especially difficult to determine what determines the transition form post-viral immunopathology to a sustained autoimmune response which is more than a passing phenomenon [98]. It is also necessary to consider ways in which viruses might cause organ specific or systemic autoimmune diseases in different circumstances [99]. If theories for a viral aetiology of autoimmune diseases are to prove useful, they must satisfy two sets of conditions. Firstly, they must fit what is already known about the disease in question. Above all, any postulated mechanism must account for an autoimmune attack on a dominant auto-antigen even if this is not confined to a single organ [19]. Indeed the theory must account for the limited number of autoantigens in general which are the targets for autoimmune responses [3]. Clearly it is essential to account for genetic susceptibility [ 112]. Since virus infections are common and autoimmune diseases are relatively uncommon. It is also essential to identify the factors which might lead to this outcome. The second condition is the possibility of testing the proposed contribution of the infection by a set of tests equivalent to Koch's postulates. In general terms there is a wide range of theoretical possibilities for viral infections to transform normally ephemeral autoimmune responses into persistent autoimmune diseases. As shown in the schematic diagram in Fig. 1 the risk of autoimmune disease is likely to increase in line with viral persistence. This conclusion is based on two considerations. Firstly, many studies have failed to show any link between acute virus infections and the onset of autoimmune diseases. Secondly, proposed mecha-

134

Table 3. Proposed general mechanisms for virus-induced autoimmunity Mechanism

A) Non-specific hostfactors Cytokine dysregulation Complement deficiency Disturbed apoptosis Mitogenic effects ("superstimulation") Disturbed lymphocyte traffic Lymphocyte killing Disturbed antigen presentation B) Specific hostfactors Dysregulated autoantigen expression Neo-antigen production

C) Immunological recognition Mimicry Cross-presentation Epitope spreading Bystander involvement Immune deviation Polyclonal stimulation Superantigen Adjuvant effects D) Viralfactors Persistence - complete or defective? Virulent or attenuated Coding potential Susceptibility to environmental factors

Reference

[69] [64] [ 100] speculative [77] [ 101] [61] [ 102] mainly speculative [103] [ 104] [86] [ 105] [ 106] [ 107] [ 108] [ 109] [97] [ 110] (The EBV model) [55] [ 111] [52]

nisms such as auto-reactive T cell responses are unlikely to arise after acute infections. In the following sections we consider the most plausible of the proposed mechanisms.

8.2. Altered Autoantigens Theories based on a viral aetiology must take detailed account of the natural history of viral infections of postulated relevance. Persistent infection with the continuous production of viral encoded proteins is more likely to elicit conventional T cell or antibody responses to self antigens. Latent infection may not induce the production of any viral antigens but could still disrupt the regulation of immune responses to unrelated antigens. Indeed the most significant pathological consequences of

graduated evolution of autoimmunity after a viral infection

I

f

j l

.....

f

J

f

I

j

a cute

!!

....

persistent

I I , autoimmune

Figure 1. In this schema autoimmune disease after viral infection is envisaged as a late complication of immunopathology generated by persistent infection. persistent herpesvirus infections concern gamma herpesviruses which, alone of this group, are able to persist in latent form. Another central issue is the starting point of proposed virus-induced autoimmune disease. One set of theories propose that the autoimmune response is driven by a normal immune response to autoantigens whose expression is qualitatively or quantitatively abnormal. According to the alternative, classical view the primary abnormality lies in the immune system. Indeed the most frequently advocated theories of virus induced autoimmune disease concentrate in altered patterns of antigen recognition by T cells secondary to continued immune stimulation by unaltered auto-antigens. Many mechanisms have been suggested by which auto-antigen expression might be affected by viral infection but there is tittle experimental information bearing on this issue. Viral infection might increase auto-antigen expression through its effects on the genetic regulation of this process. No structural changes in these antigens need be postulated and viral infection would operate in a manner analogous with autoimmune disease resulting from transgenic manipulation of antigen expression or experimentally induced exposure to an abnormal cytokine environment. This remains speculative. Altemafively, auto-antigens might be physically altered by direct association with viral antigens or by other

effects of contiguous viral replication. This hypothesis would be more persuasive if viruses implicated in the pathogenesis of autoimmune diseases used common target auto-antigens in their replication cycle. There is some evidence that this may occur. Antinuclear autoantibody characteristic of many systemic autoimmune diseases react with aminoacyl-tRNA synthetases which are also used in virus replication [ 113] thereby stimulating the notion that the autoreactivity might be primarily anti-viral. In contrast there is limited information concerning the strategies evolved by different viruses which allow the viral genome to obtain access to intra-nuclear replication sites. When it is forthcoming, the information is very surprising and encourages the idea that associations through this mechanism may stimulate autoreactivity. Intact adenovirus 2 is too large to penetrate pores in the nuclear membrane. The virus docks with the nuclear pore complex receptor CAN/Nup214, slowly disassembles, and transfers its genetic information through the nuclear pore while still attached to the nuclear pore receptor. The passage of viral genetic information to the cell nucleus is further assisted by the linker histone HI protein which normally travels backwards and forwards between the nucleoplasm and cytoplasm and acts as a transport vehicle in infected cells [ 102, 111]. Histones are the target for autoantibodies in many general autoimmune diseases and an asso-

135

ciation with viral components could explain these responses. Indeed there is experimental evidence that viral antigens can act as haptens allowing T cells to initiate autoantibody responses to histones and nucleosomes by B cells encoding these autoantibodies as part of their normal repertoire [15]. These experiments suggest that T cells reactive with histone or the polyoma T antigen provide help for autoreactive B cells. Interestingly the frequency of reactive T cells and the complementarity determining sequences in the T cell receptors were similar in normal individuals and SLE patients. However this kind of speculation raises further issues such as the manner in which these antigens would become accessible to T cell recognition and the threshold necessary to elicit such a response. There is no evidence that any of the many virus infections implicated in the immunopathogenesis of type I diabetes disrupt the genes encoding target auto-antigens for the immune destruction of pancreatic islet beta cells. However targeted disruption of the gene encoding the protein tyrosine phosphataselike molecule IA-2 induces hyperglycaemia and impaired glucose tolerance in mice [ 114]. Anti-IA-2 autoantibodies may appear years before human type I diabetes becomes clinically manifest. It is intriguing to speculate that virus infections might produce similar disruption and insidious consequences. Classical immunological theory proposed that the release of sequestered antigens was a plausible mechanism for inducing autoimmune disease. This notion was upset by the realisation that tissue damage commonly results in temporary loss of peripheral tolerance. It has been partially revived by the discovery that autoimmune disease accompanies inherited or acquired defects in immune complex clearance including complexes between auto-antigens and autoantibodies. It is reasonable to propose that the destruction of virus infected cells either secondary to immune destruction or apoptosis might overwhelm clearance mechanisms especially in individuals with defective clearance. However there is little supporting evidence in general or in viral infections. There is some evidence that granzyme B secreted by cytotoxic lymphocytes cleaves auto-antigens into novel fragments during the process of cell death. Furthermore, these fragments are not generated during apoptosis initiated by other mechanisms. The inflammatory infiltrate in

136

Table 4. Proposed "Koch's" postulates for autoimmune diseases 1) Sequencehomology (molecular mimicry) 2) Epitopepresented in immuogenic form after pathogen processing by antigen-presenting cells 3) Immunogenicconfiguration in vivo 4) T cell and/or B cell activation demonstrable in vitro 5) T cell and/or B cell activation demonstrable in vivo 6) Experimentalmodel shows that inappropriate recognition and activation produces an autoimmune disease 7) Evidencein a human autoimmune disease that cross reactivity contributes to the disease process.

salivary glands affected by Sjrgren's syndrome generates novel fragments of La, alpha-fodrin, and type 3 muscarinic actylcholine receptor. Presentation of alpha-fodrin and possibly other auto-antigens by class II molecules to lymphocytes can be blocked by inhibitors of cathepsin S leading to reduced salivary gland inflammation in a murine model of Sj/Sgren's disease [ll5]. These observations are of general interest but there is no information on the extent to which the presentation of intact or cleaved autoantigens in persistent viral infection contributes to autoimmune-mediated inflammation.

8.3. Molecular Mimicry Molecular mimicry is the process by which T cells recognise identical sequences in antigens encoded in viral infections and self antigens thereby providing help for autoreactive B cells. This mechanism is commonly proposed as mechanism for post-viral autoimmune disease but the available evidence remains fragmentary and unsatisfying. Much of it relies on sequence homologies derived from computer based searches. It is possible, indeed essential to devise "Koch's postulates" for attributing autoimmune diseases to this mechanism (Table 4). Some of these criteria can be satisfied fairly easily. However it is very difficult to validate in vitro findings either in experimental models or in human disease. Thus glutamic acid decarboxylase (GAD65) is a major autoantigen in type I diabetes and cross reacts with a peptide in human cytome-

galovirus DNA-binding protein. Furthermore the cytomegalovirus-derived peptide is processed by antigen presenting cells and stimulates T cells [116]. However the other criteria are notoriously difficult to satisfy.

8.4. Epitope Spreading Epitope spreading is defined as the process in which specific anti-viral responses in the early stages of infection becomes less sharply defined and extend to self antigens. Most of the evidence for this theory is derived from experimental models of virus-induced disease. There are many difficulties in applying the findings in experimental systems to clinical situations. Autoantibody specificity remains constant in organ-specific and generalised autoimmune diseases. This is not what one would predict from the theory of epitope spreading. Nor does it account for the restricted range of auto-antigens which are the target for immune attack, a point consistently and correctly emphasised. Credibility depends on demonstrating a peculiar vulnerability of these auto-antigens to immune attention in virus-induced immunopathology and so far there is little evidence to this effect.

8.5. Bystander Effect The bystander effect is defined as a process whereby the continued immune response to infection and attendant inflammation allows exposure of normally sequestered auto-antigens to the immune response. Theoretically this could operate by T cell recognition resulting in help for potentially auto-reactive B cells. Alternatively auto-antigens could bypass T cell help by stimulating B cells responding polyclonally to mitogens generated secondary to tissue damage or virus infection. The distinction between this mechanism and epitope spreading is a fine one and adds little to the difficulties of dissecting the pathogenesis of human autoimmune diseases.

8.6. Virus Induced Apoptosis Genetically determined defects in apoptosis clearly account for immunoproliferative B cell diseases leading to autoimmunity in mice and similar defects

may operate in human autoimmune diseases. In contrast novel treatment which augments B cell apoptosis through Fas-and TNF receptor-independent mechanisms suppresses autoimmunity [ 117]. In addition, defective removal of apoptotic cells predisposes to autoimmunity as occurs in Ciq defective mice and humans. Accumulating complexes trigger autoantibody production by B cells which form immune complexes and also convert anti-inflammatory reactions by macrophages into pro-inflammatory ones [118]. The situation in organ specific autoimmune diseases is still more complicated. One interpretation of thyroid autoimmune disease proposes that increased thyrocyte apoptosis contributes to this process. However this might be the primary event or secondary to an autoimmune attack. It has also been proposed that up-regulation of the TNF receptor superfamily is the primary event which attracts pro-apoptotic T cell ligands [119]. Furthermore thyrocytes themselves may be able to kill other thyrocytes through caspase mediated apoptosis independent of lymphocyte-mediated apoptosis [120]. Alternatively ganzyme B released by infiltrating T cells might increase apoptosis[121]. Furthermore Fas-mediated apoptosis may also lead to cell death. However the pattern of Fas and FasL expression by infiltrating lymphocytes and thyrocytes is different in Hashimoto's thyroiditis and hyperthyroidism suggesting more complex explanations [ 119, 122]. One can argue simplistically that viruses interfere with apoptosis as a strategy for survival because prolonging host cell life favours viral persistence [ 100]. Viruses modulate both the apoptotic machinery of the cell and also the extrinsic pathway mediated by TNF-alpha. However some viral infections increase apoptosis through their direct effects on infected cells. Interestingly too, they may provoke apoptosis through indirect mechanisms which may be relevant to human autoimmune disease. Scleroderma is marked by major micro-vascular damage, endothelial cell apoptosis, and fibroblastic proliferation. Sera from scleroderma patients contain an autoantibody against the surface integrin-NAG-2 complex on endothelial cells which induces their apoptosis. This antibody also reacts with the homologous cytomegalovirus late protein UL94 suggesting virus mimicry secondary to infection by this virus [123]. Viruses also abet their persistence by

137

killing lymphocytes through activating apoptosis. The lymphopenia accompanying many infections not only blunts specific immunity to the invading virus but also disrupts other cell populations with unpredictable consequences [ 101]. In contrast, killing virus-infected cells enables the host to eliminate infection at the cost of losing that cell's normal function. Currently one can only speculate about the possibility that increased apoptosis secondary to viral infection contributes to autoimmune disease in individuals with overtly or subtly defective clearance mechanisms. 8.7. Other Mechanisms

In considering currently popular theories we should not lose sight of other possibilities. One interesting requirement for autoreactive T cells to induce disease is their ability to home to the target organ. Lymphocyte homing through interactions with endothelial cells is a major area of study. Less considered is the need for virus specific CD8 T cells to home accurately to infected tissues as part of an efficient host response. For example herpesvirus type 2 specific T cells have been shown to express cutaneous lymphocyte-associated antigen (CLA) thereby suggesting a means by which they can home to cells infected by this virus [124]. The intriguing possibility that deviant homing may also occur is unexplored.

9. MODELS OF VIRUS INDUCED AUTOIMMUNE DISEASE

Relatively few animal models have been used to explore virus-induced autoimmune disease. Even in these models there are difficulties in distinguishing in ascribing a primary role to autoimmunity. Indeed the complexities perversely reproduce the difficulties in dissecting the human diseases they are intended to help clarify. 9.1. Viral Myocarditis

Coxsackie B3 virus induces myocarditis in mice which closely resembles the human disease but the contribution of autoimmunity is difficult to delineate. The model is potentially useful in some respects

138

since the pathogenesis in mouse and man have features in common. CB3 damages cardiac muscle in the majority of infected strains irrespective of their genetic background yet chronic myocarditis develops only in a minority of strains. The outcome of the infection is dependent on host anti-viral irmnunity and manipulating this response reveals some of the factors predisposing to chronic inflammation. For example transgenic mice expressing interferon gamma coding genes in their islet cells have raised circulating interferon levels and are protected from myocarditis [125]. Although it is not clear which of the proposed mechanisms for virus-induced myocarditis are operating, it seems most likely that autoimmunity is induced by antigens released from muscle fibres damaged by viral infection rather than mimicry [106]. Nevertheless these experiments do not exclude the possibility that autoimmunity is a secondary response proportional to the extent of viral; replication and the resulting inflammatory reaction. 9.2. Theiler's Virus

Theiler's murine encephalomyelitis agent, generally termed Theiler's virus, induces chronic inflammation and demyelination of the spinal cord in mice [ 126, 127]. It provides an attractive model for those investigators who believe that multiple sclerosis and other human demyelinating diseases result from virus-induced autoimmunity. Virus persistence is a crucial feature essential as the agent can readily be isolated. The innate and specific immune response to the virus are genetically controlled and determine the levels of viral replication. T cell responses to basic myelin protein antigens can readily be demonstrated. Epitope spreading has been invoked as the most likely mechanism operating in this model by analogy with experimental autoimmune encephalomyelitis induced by immunisation with the dominant epitope of myelin [105]. The T cell responses to myelin antigens seems to progress in a regular order determined by the efficiency with which these antigens are processed and presented to autoreactive T cells [128]. Furthermore the disease can be adoptively transferred to normal mice by autoreactive T cells. Unfortunately, the absolute requirement for viral persistence makes it difficult to decide whether anti-myelin autoreactivity contributes to

the lesions or is a secondary phenomenon irrespective of the mechanism or mechanisms responsible for its induction. Certainly, adoptive transfer shows that the sensitised T cells damage myelin in the new host but it is difficult to extrapolate too uncritically about their role in the original host.

9.3. Herpesvirus Keratitis Herpesvirus 1 (HV-1) induces stromal keratitis in the eyes of infected mice [8, 104, 129]. Not all mouse strains are susceptible. Genetic factors determine both the level of viral replication and the nature of the host immune response; both these factors determine disease susceptibility. In this model viral persistence is also indispensable. Although HV-1 infection initiates the disease, corneal inflammation is delayed for 1-2 weeks after the infection by which time persistent local infection is difficult to demonstrate. However this is not surprising in the face of a fully established host anti-viral immune response. Autoimmune reactions also appear at this stage of inflammation. Some T cells in the lesions react with a corneal antigen and others with an IgG2a sequence. Furthermore T cell clones stimulated in vitro with HV-1 infected cells recognise a viral peptide identical with this IgG sequence. Infected cells infected with an HV-1 variant which do not encode this viral peptide fail to induce T cell reactivity with the IgG determinant, suggesting that the viral peptide is essential for inducing the inflammatory disease. An additional argument in favour of the pathogenic importance of this viral peptide is the inability of HV-1 infected immunodeficient mice to develop herpetic stromal keratitits unless immunised with the same peptide. At first sight these observations seem to indicate that the inflammation is mediated by virus-reactive T cells which recognise an auto-antigen through molecular mimicry. However other observations make this argument less compelling. Firstly, viral infection of the host or stimulating cell lines is indispensable. Secondly, the target corneal antigen for the postulated autoimmune attack has not been identified. Finally there is other evidence that herpes stromal keratitis can be induced in mice which are unable to mount a T cell response to the allegedly crucial HV-1 or IgG antigens. Other evidence suggests that quantitative factors also determine susceptibility to

the disease. Specific HV-1 infection is essential for disease induction in mice with low numbers of autoreactive T cells. Furthermore, non-specific stimuli suffice to induce disease in mice with high numbers of autoreactive T cells [99, 130]. Cross reaction between the retinal S antigen and viral pepetides has also been proposed mainly on the basis of T cell proliferative responses [131] but the evidence remains fragmentary.

9.4. Murine Cytomegalovirus Infection and SjSgren's Syndrome Cytomegalovirus infection in mice (MCMV) provides an interesting model of the systemic autoimmune disease Sj6gren's syndrome [132]. The human disease is characterised by chronic destructive inflammation of the salivary and lacrimal glands, and B cell proliferation with autoantibody production notably to the Ro/SSA and La/SSB auto-antigens. Four different mouse strains infected with MCMV developed acute sialadenitis. However chronic sialadenitits with the characteristic autoantibodies developed only in the B6-1pr/lpr strain with a genetic background of defective Fas mediated apoptosis. Infectious MCMV could not be detected after 100 days post-infection. This group reported a similar outcome in MCMV infected B6-gld/gld mice which carry a defective ligand (FasL) and hence have impaired Fas-mediated apoptosis [ 133]. The chronic sialadenitis was partially reversed by local FasL treatment. This model is of particular interest because, although MCMV persistence may be a prerequisite for disease development, this postulate is difficult to prove. Furthermore the pattern of inflammatory disease in target tissues and the autoantibody profile closely resemble the spontaneous human disease.

10. GENERAL EVIDENCE FOR A VIRAL A E T I O L O G Y IN HUMAN AUTOIMMUNE DISEASES There have been many attempts to identify viruses contributing in whole or in part to the pathogenesis of human autoimmune disease [8]. The stimulus for a continued search for a viral aetiology of human autoimmune diseases rests largely on theoretical

lqQ

speculation and clinical observation of post-viral autoimmune phenomena. Acute viral infections commonly induce transient autoimmune responses usually directed at circulating bone marrow progeny namely platelets, neutrophils, and red cells.There is also a high incidence of transient autoantibodies and disease manifestations such as arthritis after viral infection or immunisation [134] but these rarely progress into established autoimmune disease. For example the polyarthritis provoked by parvovirus B 19 or rubella infection rarely progresses to chronic arthritis such as rheumatoid arthritis. These almost invariably transient events are conventionally attributed to polyclonal B cell activation but this is likely to prove an over-simplification. If viral infections do contribute in whole or in part to these diseases, we must postulate delayed mechanisms which only become evident a long time after the initial infection. The most plausible proposition is viral persistence. If the viral genome persists in latent form without any transcription of viral proteins, there might be an indefinite silent period without any disease manifestations. However even in these circumstances there is the theoretical possibility that the viral genome may alter the expression of host antigens. Indeed it is theoretically possible that transient infection may alter the regulation and expression of host genes encoding autoantigens without the need for persistence of any viral genes. However the lessons learnt from viral oncology make it more likely that viruses would need to persist in some form to elicit autoimmune diseases. In the current state of knowledge it seems more likely that some expression of virus-coded proteins would be needed to induce an autoimmune response. Many human viruses persist indefinitely which are transcribed continually or intermittently. Most prominent among these agents are herpes viruses, measles virus, and retroviruses. There have been many attempts to link viral infections with autoimmune diseases. Many studies have used classical, indirect techniques, notably antibody titres. To date these studies have been negative or inconclusive. Early studies compared anti-viral antibody titres in patients and controls but the results were often difficult to interpret. Thus the polyclonal hypergammaglobulinaemia of systemic lupus erythematosus (SLE) generates a non-specific increase in anti-viral antibody titres which reflect accumulated immunological memory but

140

not necessarily infections relevant to the aetiology of the disease. Antibody titres to viral antigens are often a poor guide to long term immunopathological consequences even when the link with disease is indisputable. For example there is a poor correlation between antibody titres and recurrent "cold sores" induced by herpes virus type 1 or Epstein-Barr (EBV) infection and Burkitt's lymphoma. Both viruses persist life long in most individuals yet cold sores affect only 10% of the population and the incidence of Burkitt's lymphoma is low even in susceptible populations. Yet EBV infection admirably illustrates the difficulties in relying on conventional markers of viral persistence as evidence for the role of a given virus in disease pathogenesis. The association between EBV infection and Burkitt's lymphoma was established mainly on epidemiological grounds. There is now direct evidence that selective EBV gene latency in B cells provokes proliferation of the infected B memory cells and evasion of the cytotoxic T cell response lymphoma cells thereby allowing the emergence of Burkitt's lymphoma clones [ 135]. Whether EBV acts in a similar manner to generate conventionally benign but autonomous autoantibody producing B cell clones is speculative. Almost universal infection EBV makes it very difficult to correlate classical markers of infection such as anti-viral antibody titres with autoimmune diseases which almost certainly develop many years after primary infection. Attempts to link EBV infection with multiple sclerosis illustrate this dilemma [136, 137]. These studies established a strong association between high titres of anti-VCA and EBNA2 antibodies encoded by EBV and multiple sclerosis. However it is difficult to determine whether this increase results from an increased virulence of common virus infections in patients destined to develop the disease, a direct aetiological relationship, or the role of EBV as an adjuvant in an unrelated pathological process. Similarly the detection of the EBV genome in cells involved in an autoimmune disease does not establish an aetiological role for the virus. The same difficulties arise from attempts to implicate EBV infection in the pathogenesis of systemic lupus erythematosus (SLE). These studies have been encouraged by other evidence such as sequence homology between the EBV peptide PPPGRRP and the peptide PPPGMRPP of the SM B'B antigen of the human spliceosome. Cross

reactivity to these sequences might account for the Sm reactive autoantibody encountered in SLE patients. A higher percentage of peripheral blood B cells from SLE patients than normal controls carry the EBV genome [138], but it is quite possible that this is secondary to the polyclonal B cell activation characteristic of this disease. In addition anti-EBV antibodies are detectable in a higher percentage of SLE patients than controls [139], but this could easily be a secondary event. There is evidence from other situations that viruses may act as adjuvants rather than prime causes of immune-mediated disease. An analysis of heart transplant rejection illustrates this dilemma. Although this process is not strictly analogous with autoimmune myocarditis, the initiating immune disorder at least has a known starting point. Heart transplant rejection in children is increased 6.5 fold if the transplant is infected by viruses, notably adenoviruses [ 140]. The evidence from studies using other techniques is still more fragmentary. The results of attempted viral isolation from target tissues and organs involved in autoimmune diseases are contentious and also difficult to interpret because any immunoproliferative disorder is likely to reactivate latent virus infections. Many studies have sought evidence based on molecular mimicry and other mechanisms invoked in experimental models of virus-induced autoimmune diseases. The same problems of interpretation arise in clinical studies. Attempts to incriminate specific viral infections in human autoimmune disease by invoking molecular mimicry are not persuasive if the claims are based solely on sequence homology. It is relatively easy to obtain these data from sequence banks. It is far more difficult to establish that the viral sequence induces a cross reactive T cell response in vivo. Many other considerations come into play including tertiary structure, accessibility to T cells, and evidence that the sequence is presented in immunogenic form. Nevertheless viral infections which indubitably provoke systemic diseases would be considered primary autoimmune diseases in the absence of this information. Hepatitis C virus infection induces a disease with many features resembling idiopathic SLE. Furthermore, a possible mechanism involving molecular mimicry has been identified [ 141 ]. CD8 T cells reactive with liver cytochrome P450 peptide

Table 5. Human autoimmune diseases investigated for a viral aetiology Disease

Reference

A) Organ or system specific

Type I diabetes Thyroiditis Myocarditis Multiple sclerosis Peripheral neuropathy Polymyositis

[143] [144] [106] [127] [145] [ 146]

B) Systemic

Rheumatoid arthritis Systemic lupus erythematosus Sj6gren' s syndrome

[147] [ 148] [149]

sequences also react to homologous sequences encoded by hepatitis C virus. The reactivity is HLA class II restricted. Although the features of hepatitis C virus associated SLE differ in some respects from those encountered in idiopathic SLE, there are many common features [ 142].

11. EVIDENCE FROM SPECIFIC HUMAN AUTOIMMUNE DISEASES The autoimmune diseases which have been most intensively investigated for a viral aetiology are listed in Table 5. A review of the evidence for a viral aetiology in some of these diseases illustrates the difficulties. 11.1. Type 1 Diabetes Virus-induced experimental models of type I diabetes highlight the difficulties in investigating the human disease [150]. The M variant of the picornavirus encephalomyocarditis virus (EMCM) induces a syndrome resembling human type I diabetes in genetically susceptible mice. The small nucleotide differences distinguishing the diabetogenic from other strains have been characterised and centre on the crucial position of an alanine in a highly conserved, strongly hydrophilic part of the sequence. This site governs viral attachment to pancreatic islet beta cells and subsequent infection. Heavily infected islet cells are rapidly destroyed by

141

viral replication and innate immune mechanisms, notably macrophages and TNF-alpha. Low titre systemic and islet cell infection by virus results in their more protracted destruction by macrophages activated through Src kinases. The Kilham rat (KRV) parvovirus causes diabetes in rats which are genetically resistant to the spontaneous disease. However, in contrast with the EMC-M model, KRV does not infect islet cells but initiates autoimmune destruction of these cells through macrophage activation and the preferential activation of cytotoxic T cells. These observations indicate that selected virus strains can induce diabetes in genetically susceptible hosts through mechanisms which do not invariably involve T cell destruction of infected islet cells. There is no guarantee that the model has any relevance to the human disease. If they are relevant, the challenges for investigators are formidable. Routine antibody screening is unlikely to detect infection by diabetogenic strains. Islet cells from diabetic patients are hardly ever available for analysis. Even if material was available, the KRV model indicates that even the most detailed virological analysis might be unrewarding. It is therefore not surprising that the search for a viral aetiology in the human disorder has been at best inconclusive [143]. The polygenic susceptibility of patients to autoimmune destruction of islet beta cells has been established beyond reasonable doubt. However no specific infection has been identified which might be a plausible primary cause of the disease or might exacerbate a primary autoimmune process unrelated to infection. The hope of progress depends on the development of culture systems with tissues from genetically well characterised donors which permit the study of interactions between islet cells, potentially diabetogenic virus strains, and the immune system.

11.2. Autoimmune Thyroid Diseases Autoimmune thyroid diseases are important for many reasons. They are the commonest diseases in which autoimmunity is unequivocally the most likely key to their aetiology. The incidence of autoimmune thyroiditis in middle age women, the most susceptible group, is around 2%. Intriguingly, the average incidence of thyroid autoantibodies in different populations in this group is about 15%; in

142

the majority of these individuals anti-thyroid autoantibodies have no apparent clinical significance but they may nevertheless be attributable to thyroiditis. Despite formidable technical difficulties, the antigen-autoantibody systems in autoimmune thyroid disease have been largely characterised [144]. The main target autoantigen in autoimmune hypothyroidism is the cell surface protein thyroid peroxidase (TPO). The main target autoantigen in hyperthyroidism is the thyrotrophin receptor (TSHR), a G protein with seven membrane-spanning segments. In cultured cells, the extra-cellular subunit A is shed from the cell surface during cleavage. TPO is the target of T-cell mediated autoimunity as well as autoantibodies. TPO-reactive autoantibodies have been isolated from immunoglobulin gene recombinatorial libraries constructed from thyroid infiltrating B lymphocytes. The germline H and L genes are similar to those in many other antibodies. However the H chain genes show a high degree of somatic mutation characteristic of antigen driven maturation and consistent with their high affinity. Defective receptor editing may also contribute to the generation of TPO-reactive autoantibodies [151]. Four largely overlapping epitopic domains have been identified. Interestingly, the epitope specificity of TPO autoantibodies from individual patients remains constant for many years with no evidence of epitope spreading. Because of technical difficulties, TSHR antibodies and their epitope specificities have been less fully characterised. TPO-reactive autoantibodies from patients with hyperthyrodism preferentially select different VH domains from those associated with hypothyroidism [151]. There is little information on the epitopes recognised by infiltrating T cells. No convincing link with viral infection has been established by indirect means. There is meagre information about the long term natural history of patients with acute, self-limited thyroiditis. A detailed survey showed no association with common viral infections of childhood or immunisation against these infections [152]. In thyroid disease, this failure can not be attributed to lack of material which has been readily available from surgical or biopsy material. Thyroiditis may be associated with hepatitis C virus infection even before interferonalpha treatment is started [ 153]. However, although

the incidence of thyroid autoantibodies [12.1%] was higher than in controls [4.0%], the prevalence of autoantibodies to TPO in infected individuals was the same. Nor was the increase associated with detectable thyroid dysfunction. In general, detailed knowledge of the auto-antigens and autoantibodies contributing to thyroid autoimmune disease has not resulted in any substantial clues to a viral aetiology.

11.3. Post-Infectious Polyneuropathy and Related Neurological Syndromes Campylobacter jejuni is the commonest identifiable cause of post-infective polyneuropathy (Guillain Barre syndrome) but viral infection is implicated in some 20% of patients. In theory nerve damage following Campylobacter jejuni infection may result from autoantibodies to about 20 distinct gangliosides provoked by molecular mimicry with similar carbohydrate sequences in bacterial lipopolysaccharides. The fine specificity of these autoantibodies is probably crucial. Anti-GQlb ganglioside autoantibodies demonstrably damage the motor nerve terminal through a complement dependent mechanism. A particularly striking association between these autoantibodies and neuropathy is seen in Miller-Fisher syndrome in which patients develop ataxia, areflexia, and ophthalmoplegia [154]. In contrast, anti-GM2 ganglioside antibodies are less specifically associated with neuropathy and their neurotoxic potential is more dubious [145]. These observations point to the possibility that similar mimicries might account for post-viral peripheral neuropathy but emphasise the difficulties in ascribing pathological significance to post-infectious autoantibodies even when these are detected.

11.4. Polymyositis, Myocarditis and Related Diseases The search for a viral aetiology for polymyositis, muocarditis, and related diseases has produced tantalising clues but no convincing solution. Coxsackieviruses have received particular attention because some strains induce myocarditis in mice and myalgia is a prominent symptom of myalgia in human coxsackievirus infections. However it is fair to conclude that the search for persistent or latent virus in affected striated or heart muscle, serological

studies, and assays of anti-viral T cell reactions have produced conflicting and often controversial results. Coxsackievirus infections are common and 60-80% of a given population have antibodies against the prevalent strains. In contrast polymyositis is uncommon with a prevalence ranging from 2.4-10.7 per 100,000 [146]. Clearly, if an association between coxsackievirus infection and polymyositis exists, it results from the interaction of a peculiar strain with an unusual host. This requirement is underscored by early attempts to establish a viral model of polymyositis which was eventually achieved by infecting CD1 Swiss mice less than 48 hours old with the Tucson strain of coxsackie B 1 virus [155]. Interesting features of the model were the disappearance of detectable virus despite persistent myositis and the difficulties in distinguishing myopathic from nonmyopathic virus strains by their virological properties. Subsequent studies have shown that myotropic, myopathic clones of coxsackie B 1 virus differ from myotropic but non-myopathic clones by 20 nucleotides. Only myopathic clones induced anti-muscle and anti-nuclear antibodies [ 156]. Non-viral models of polymyositis introduce the important concept that virus infections may initiate an inflammatory response which is perpetuated by autoimmunity unrelated to classical therories. For example an initial increased expression of class I HLA antigens might be the critical event [11]. Intriguingly, auto-antigenic aminoacyl-tRNA synthetases released from damaged muscles may be chemotactic for inflammatory cells thereby stimulating chronic autoreactive inflammation [157]. These observations are important because they emphasise the potential contribution of immune mechanisms to autoimmune disease which are not mediated by auto-reactive T and B cell clones. Further grist for the idea of viral initiation without persistence comes from the observation that adeno-associated virus, a plausible cause of inflammatory myositis, is less likely to be found in myositic muscle than in muscles from normal individuals or patients with non-inflammatory myopathies [158].

11.5. Anti-Phospholipid Syndrome The anti-phosphoplipid syndrome (APS) causes thrombotic episodes with a protean range of clinical problems including cerebral ischaemic

143

episodes, early foetal loss, and pulmonary infarction. Although its florid clinical presentations are well recognised, it may be involved in many other disorders whose pathology involves endothelial cell damage and small vessel occlusion. The responsible autoantibody is directed at beta2-glycoprotein-1. Many microbial infections have been implicated in this syndrome including hepatitis C virus and EBV and there is strong evidence that these infections induce antibodies to microbial peptides which cross-react with anti-beta2-glycoprotein- 1 [ 159].

11.6. Chronic Urticaria Justifiable preoccupation with severe, often life threatening autoimmune diseases is liable to distract attention from other problems which, although less devastating, are nevertheless common, chronic, and often very distressing. The rash and angioedema of chronic urticaria come into this category and can be so severe as to necessitate treatment with immunosuppressive drugs or plasma exchange. Specific causes such as drug allergy can be identified in many patients but in some 70% there is no obvious cause. However some 20% of patients give a history of a preceding infection which is usually upper respiratory and has features suggesting a viral infection. In one third of patients chronic urticaria results from histamine release mediated by autoantibodies to the high affinity IgE receptor FcepsilonR1 or IgE itself [160]. As interest in the pathogenesis of chronic urticaria shifts from a preoccupation with food intolerance and "pseudoallergy", observations of this kind rightly draw attention to common mechanisms which may underlie allergic and autoimmune diseases [ 17].

12. EVIDENCE THAT THE PRIMARY EVENT IN ORGAN SPECIFIC AUTOIMMUNITY MAY NOT BE IMMUNOLOGICALLY MEDIATED A major problem in autoimmune diseases is to determine the extent to which autoimmunity is the primary event in diseases undoubtedly accompanied by autoimmune reactions. This remains an issue even in many virus-induced models of autoimmune diseases where viral growth and per-

144

sistence are exacerbated by hosts with defective innate or specific immunity. It can be argued that autoimmunity contributes to tissue damage but only as a secondar3, event. There are also situations in which autoimmune disease is provoked by viral infections because of pre-existing or virus-induced defects which are unrelated to host immunity. For example abnormalities in the dystrophin coding gene predispose to cardiomyopathy. Coxcsackie B3 virus (CVB3) encodes a protease which cleaves dystrophin and disrupts the dystrophin-glycoprotein complex with consequences resembling those encountered in hereditary disease. CVB3 causes more severe cardiomyopathy in dystrophin deficient than in wild mice [ 161 ]. There are similar situations in human disease. Thrombotic thrombocytopenic purpura is a disease characterised by intra-vascular destruction of red cells and platelets which results in the formation of platelet micro-thrombi and small vessel obstruction. It follows viral infections in seemingly normal individuals or in those already affected by chronic inflammatory diseases such as juvenile chronic arthritis. Susceptibility to this chain of events results from an inherited defect in the proteolytic breakdown of the clotting factor von Willebrand factor so that the formation of microthrombi is encouraged [162]. Interactions between virus infection, apoptosis, and neuro-degenerative disease are a still more subtle illustration of degenerative disease which can later be interpreted as autoimmune in origin. A provirus insertion in apoptosis-inducing factor (Aif) in Harlequin mice interferes with the apoptosis of neurones damaged by oxidative stress. Damaged cells which would normally be removed by apoptosis re-enter the cell cycle and their exposure to oxidative stress results in neurodegeneration [163]. Observations of this kind enjoin caution about necessarily accepting that autoimmune aggression is the primary cause of common diseases associated with autoimmunity. The association between thyroiditis, the commonest organ specific autoimmune disease, and Down's syndrome, the commonest genetic disorder is well established but the implications of a seemingly nonimmunological genetic disorder for later autoimmunity are unexplained [ 164, 165].

13. RETROVIRUSES Retroviruses in many different species are recognised causes of degenerative and immunodeficiency diseases accompanied by some autoimmune features, especially directed at red cells and platelets. HIV infection in man exacerbates some inflammatory disorders with a possible autoimmune component such as psoriasis and Reiter's syndrome. The variable temporal relationship between autoimmune platelet destruction and other manifestations of infection indicates that autoimmune mechanisms may be subtle [166]. There have been suggestions that the characteristic depletion of CD4 T cells is mediated at least in part through autoimmunity induced by viral mimicry [167, 168], and not exclusively or even predominantly through direct viral infection of susceptible cells. However the view that HIV is irrelevant to the pathogenesis of AIDS has lost whatever credence it first enjoyed. An infectious retrovirus designated HRV-5 has been detected in synovial tissue from rheumatoid arthritis patients and blood mononuclear cells of patients with this disease or SLE [ 169]. However in general attempts to attribute human autoimmune diseases such as rheumatoid arthritis and SLE to retroviruses transmitted by conventional infection have been unsuccessful [ 148, 170]. A more difficult subject to address is the possibility that inherited or acquired endogenous retroviruses are involved in the pathogenesis of autoimmune diseases [171]. Theoretically, these sequences could disrupt the regulation of the myriad of the components of innate and acquired immunity which contribute to these diseases. Endogenous retroviral sequences have been detected in normal and rheumatoid synovial membranes. However there is evidence that these sequences may be selectively expressed in rheumatoid arthritis. This expression could be linked to the intriguing histological observation that synovial cell proliferation in early rheumatoid arthritis appears to precede the immunopathological events which later dominate the histological picture. However, as always in inflammatory lesions a major problem is to determine whether the altered expression of endogenous retroviral gene sequences is a primary event or the consequence of immunoproliferative disease. Retrotransposons have also been linked with

autoimmune diseases. These genetic elements resemble retroviruses but lack the env gene. They can be transmitted between genes. A retrotransposon inserted in thefas gene of MRL lpr/Ipr accounts for the destructive synovial hyperplasia accompanying the lymphoproliferative disease characteristic of this strain. Retrotransposons have been consistently detected in rheumatoid synovial memebranes. Their RNA sequences are similar to ORF2/L1 and THE1 retrotransposons, human endogenous retrovirus (ERV)-E, and other elements [ 147]. In theory, these sequences could up-regulate the genes controlling the cytokines and kinases responsible for synovial inflammation. However these changes could also be secondary to hyperplasia induced by totally unrelated events. Cytokine activation of retroviral superantigens secondary to conventional virus infection has also been postulated as the mechanism responsible for pancreatic beta cell destruction in diabetes [109]. As yet these issues remain unresolved and there is no conclusive evidence that retroviruses in any form cause human autoimmune diseases.

14. IMMUNOLOGICAL SURVEILLANCE Apoptosis is a proven strategy by which the host eliminates virus-infected cells. This is achieved by activating intrinsic pathways or by extrinsic inflammatory cells and their products. It is also possible that apoptosis is a mechanism for removing cells damaged more subtly by viral infection which do not display any obvious evidence of viral infection. Retroviruses are often cited as persistent agents with the capacity to damage cells directly as well as to induce immunopathological mechanisms. Indeed the concept of immunological surveillance of endogenous retroviral sequences has been made the basis of a general theory for autoimmune diseases [172]. Autoimmune diseases are thereby considered the price of controlling DNA damage. The indirect or delayed consequences of viral infection in general might not be revealed by conventional studies of primary infections and their outcome. For example, although B19 parvovirus infection does not induce chronic destructive arthritis, there is some evidence that it induces an invasive phenotype in normal human synovial fibroblasts [173].

145

Apoptosis secondary to immune mediated inflammarion and cell lysis by traditional immunological mechanisms could be perceived as part of a general defence strategy. However these ideas are currently mainly speculative.

15. E X T E N D E D KOCH'S POSTULATES

The complexities of host defence mechanisms and the many viral strategies for persistence make any attempt to adapt Koch's postulates to this field extremely difficult. Attempts to devise credible presuppose that the disease in question presuppose that its autoimmune nature has been unequivocally established. We suggest the following scheme: 1) The virus or its products can be consistently identified in patients with a given autoimmune disease. 2) The virus persists in a form in which it is able to initiate the autoimmune mechanisms responsible for the disease. 3) There are demonstrable mechanisms by which the virus induces the autoimmune disease. 4) The autoimmune reactions must be related to viral infection. 5) There are host features in patients with autoimmune disease attributed to the virus which confer susceptibility to the disease and distinguish them from other individuals infected by the same virus who do not develop the disease. 6) The virus induces comparable disease in an animal model.

16. CONCLUSIONS There are obvious difficulties in se problems in attempting to ascribe a viral aetiology to autoimmune diseases [7]. Nevertheless there are certain points relating to this issue which are generally recognised as fundamental, even if these are hard to resolve [ 19]. Any theory must account for autoantigenic specificity and the polygenic factors conferring susceptibility to disease. The classical dogmas about loss of T or B cell tolerance or a combination of these cell populations are still valid but the factors which lead to the breakdown of tolerance are

146

more complex than was formerly envisaged. The complexities of host defence against viral infections are increasingly apparent. Furthermore the classical, sharp distinction between innate, non-specific immunity and specific immunity is no longer tenable since these contribute to host defence in an integrated manner. Auto-reactivity is part of the normal T and B cell repertoire of responses and contributes to anti-viral immunity. In a sense, peripheral tolerance is regularly broken as part of the host strategy for destroying infected cells. Nevertheless the central tenet of classical tolerance remains unshaken; autoimmunity may be common but persistent autoimmune diseases are seemingly rare after viral infection. Autoimmune disease results from a wide range of very different defects. Some are entirely genetic and distort the proliferation of potentially reactive immune cells. Others are only apparent after immune stimulation including microbial infection. Although the general principle of abnormal immune regulation has been validated, early monotheistic views attributing autoimmune diseases to a failure of suppressor T cells have been abandoned now that so many pathways for excessive inflammation have been discovered. To add to the possibilities, viral strategies for persistence further disrupt the control of inflammation with varying risks for abetting sustained autoimmune response. There is little evidence that viruses are the primary cause of most organ specific and general autoimmune diseases. Experimental models of virus-induced autoimmune diseases have served mainly to uncover new complexities of observation and interpretation. A major difficulty is to discern whether autoimmunity in these models is the initiating event or secondary to inflammation. Classical techniques for implicating viruses in human disease have produced hints but no solutions. One difficulty is the likely heterogeneity of many human diseases which have been given a single label. The immune complex disease and autoimmune processes produced by hepatitis C virus, for example, may satisfy rather arbitrary criteria for the diagnosis of SLE. However clearly SLE does not have a single cause. It is more likely that a common immune defect confers susceptibility to this disease from a variety of causes including viruses. Nevertheless the slow progress in this field is insufficient reason for abandoning an investigative route which remains

plausible and suggestive evidence is still forthcoming. Put pithily, persistent inflammation is caused by infection when we know the agent and attributed to autoimmunity when we do not [ 174]. The prospects for progress depend on advances in specific areas. As discussed, we need a modem version of Koch's postulates which not only identifies the suspected agent but also indicates why ubiquitous viruses might cause autoimmune disease only in a susceptible individuals. It is unlikely than novel observations will disclose a simple cause and effect relationship between agent and disease. Indeed this has not proved to be the case in the classic example of tuberculosis. Analysis of viral and host gene expression and the products they encode is crucial to this process. Otherwise issues such as molecular mimicry and linked immunogenic expression of host and viral encoded genes will remain unsolved. Furthermore the chain of events which lead from initial infection to sustained autoimmunity is likely to depend on quantitative factors which can never be spelt out adequately in descriptive terms. New technologies giving new data and insights into this question. In particular mapping the human genome and the science of proteomics are key scientific developments which will enable us to determine the genetic coding and degree of expression of self antigens already recognised as targets for autoimmune attack. Furthermore microarray technologies may well identify new targets for autoimmune attack and should help to distinguish between auto-antigens encoded by host genes and viral genes. Moreover these auto-antigens may prove to be encoded by viral genes incorporated into the genome following conventional infection or transmitted vertically in the germ line. The enormous mass of information on viral life span, viral interactions with the many components of the host response, the cross-talk between these components, and the genetic control of each step of this process continues to generate a wealth of defensible hypotheses but few certainties. Only an integrated, computed model of the interactions between virus and host responses will suffice [175]. From a different perspective, traditional ideas about viruses as infective agents may have underestimated the vast range of outcomes of virus infection. The intellectual separation of "viruses" and "genes" which has characterised so much of our

thinking about autoimmune diseases may be largely illusory. "The virus, instead of being single-minded agents of disease and death, now begin to look more like mobile genes. Evolution is still an infinitely long and tedious biological game, with only the winners staying at the table, but the rules beginning to look more flexible. We live in a dancing matrix of viruses; they dart, rather like bees, from organism to organism, from plant to insect to mammal to me and back again, and into the sea, tugging along pieces of this genome, strings of genes from that, transplanting grafts of DNA. passing around heredity as though at a great party. They may be a mechanism for keeping new, mutant kinds of DNA in the widest circulation amongst us. If this true, the odd virus disease, on which we all focus so much of our attention in medicine, may be looked on as an accident, something dropped" [ 176].

ACKNOWLEDGEMENTS The authors would like to thank Dr. Evelyn Denman and Dr. Tatiana Dvorkin for their invaluable help in preparing the manuscript. This work was supported in part by a grant (BRZ) from the Center for the Study of Emerging Diseases and the Israel Science Foundation.

REFERENCES 1. 2. 3. 4.

5.

6. 7.

Burnet M. Self and Not-Self. 1st Ed. London: Cambridge Univ. Press, 1969. CohenIR. On Autoimmunity. Tending Adam's garden. Academic, 2000;197-239. Plotz PH. The autoantibody repertoire: searching for order. Nat Rev Immunol 2003;3:73-78. Sakaguchi S, Sakaguchi N, Shimizu J, Yamazaki S, Sakihama T, Itoh M e t al. Immunologic tolerance maintained by CD25+ CD4+ regulatory T cells: their common role in controlling autoimmunity, tumor immunity, and transplantation tolerance. Immunol Rev 2001; 182:18-32. SilversteinAM. The Clonal Selection Theory: what it really is and why modern challenges are misplaced. Nat Immuno12002;3:793-796. BoytonRJ, Altmann DM. Transgenic models of autoimmune disease. Clin Exp Immuno12002;127:4-11. WhittonJL, Fujinami RS. Viruses as triggers of auto-

147

8.

9. 10. 11.

12. 13. 14.

15.

16.

17. 18.

19. 20.

21.

22.

23.

immunity: facts and fantasies. Curr Opin Microbiol 1999;2:392-397. Olson JK, Croxford JL, Miller SD. Virus-induced autoimmunity: potential role of viruses in initiation, perpetuation, and progression of T-cell-mediated autoimmune disease. Viral Immuno12001; 14:227-250. Zimmer C. Do chronic diseases have an infectious route? Science 2001;293:1974-1977. On Bacteriological Research. Berlin: Hirschwald, 1890. Nagaraju K, Raben N, Loeffler L, Parker T, Rochon PJ, Lee E et al. Conditional up-regulation of MHC class I in skeletal muscle leads to self-sustaining autoimmune myositis and myositis-specific autoantibodies. Proc Natl Acad Sci USA 2000;97:9209-9214. Davidson A, Diamond B. Autoimmune diseases. N Engl J Med 2001;345:340-350. Vincent A. Unravelling the pathogenesis of myasthenia gravis. Nat Rev Immuno12002;2:797-804. Stefanova I, Dorfman JR, Germain RN. Serf-recognition promotes the foreign antigen sensitivity of naive T lymphocytes. Nature 2002 ;420: 429-434. Andreassen K, Bendiksen S, Kjeldsen E, Van Ghelue M, Moens U, Arnesen E et al. T cell autoimmunity to histones and nucleosomes is a latent property of the normal immune system. Arthritis Rheum 2002;46: 1270-1281. Jutel M, Watanabe T, Klunker S, Akdis M, Thomet OA, Malolepszy Jet al. Histamine regulates T-cell and antibody responses by differential expression of HI and H2 receptors. Nature 2001 ;413:420-425. Benoist C, Mathis D. Mast cells in autoimmune disease. Nature 2002;420:875-884. Lee DM, Friend DS, Gurish MF, Benoist C, Mathis D, Brenner M. Mast cells: a cellular link between autoantibodies and inflammatory arthritis. Science 2002;297: 1689-1692. Marrack P, Kappler J, Kotzin BL. Autoimmune disease: why and where it occurs. Nat Med 2001;7:899-905. Morgan DJ, Kurts C, Kreuwel HT, Hoist KL, Heath WR, Sherman LA. Ontogeny of T cell tolerance to peripherally expressed antigens. Proc Natl Acad Sci USA 1999;96:3854-3858. Anderton SM, Wraith DC. Selection and fine-tuning of the autoimmune T-cell repertoire. Nat Rev Immunol 2002;2:487-498. Di Nola J, Neuberger MS. Altering the pathway of immunoglobulin hypermutation by inhibiting uracilDNA polymerase. Nature 2002;419:43-48. Santulli-Marotto S, Retter MW, Gee R, Mamula MJ, Clarke SH. Autoreactive B cell regulation: peripheral induction of developmental arrest by lupus-associated autoantigens. Immunity 1998;8:209-219.

148

24. Ergas D, Tsimanis A, Shtalrid M, Duskin C, Berrebi A. T-gamma large granular lymphocyte leukemia associated with amegakaryocytic thrombocytopenic purpura, Sjrgren's syndrome, and polyglandular autoimmune syndrome type II, with subsequent development of pure red cell aplasia. Am J Hemato12002;69:132-134. 25. Miyamoto A, Nakayama K, Imaki H, Hirose S, Jiang Y, Abe Met al. Increased proliferation of B cells and autoimmunity in mice lacking protein kinase Cdelta. Nature 2002;416: 865-869. 26. Cornall RJ, Cyster JG, Hibbs ML, Dunn AR, Otipoby KL, Clark EA et al. Polygenic autoimmune traits: Lyn, CD22, and SHP-1 are limiting elements of a biochemical pathway regulating BCR signaling and selection. Immunity 1998;8:497-508. 27. Cohen PL, Caricchio R, Abraham V, Camenisch TD, Jennette JC, Roubey RA et al. Delayed apoptotic cell clearance and lupus-like autoimmunity in mice lacking the c-mer membrane tyrosine kinase. J Exp Med 2002;196:135-140. 28. Halonen M, Eskelin P, Myhre AG, Perheentupa J, Husebye ES, Kampe O et al. AIRE mutations and human leukocyte antigen genotypes as determinants of the autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy phenotype. J Clin Endocrinol Metab 2002;87:2568-2574. 29. Vogel A, Strassburg CP, Obermayer-Straub P, Brabant G, Manns MP. The genetic background of autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy and its autoimmune disease components. J Mol Med 2002;80:201-211. 30. Ramsey C, Winqvist O, Puhakka L, Halonen M, Moro A, Kampe O et al. Aire deficient mice develop multiple features of APECED phenotype and show altered immune response. Hum Mol Genet 2002; 11:397-409. 31. Watanabe N, Iloata K, Nisitani S, Chiba T, Honjo T. Activation and differentiation of autoreactive B-1 cells by interleukin 10 induce autoimmune hemolytic anemia in Fas-deficient antierythrocyte immunoglobulin transgenic mice. J Exp Med 2002;196:141-146. 32. Rohowsky-Kochan C, Skurnick J, Molinaro D, Louria D. HLA antigens associated with susceptibility/ resistance to HIV-1 infection. Hum Lrnmunol 1998;59: 802-815. 33. Grene E, Pinto LA, Cohen SS, Mac TC, Trivett MT, Simonis TB et al. Generation of alloantigen-stimulated anti-human immunodeficiency virus activity is associated with HLA-A*02 expression. J Infect Dis 2001; 183: 409--416. 34. Gillespie GM, Kaul R, Dong T, Yang HB, Rostron T, Bwayo JJ et al. Cross-reactive cytotoxic T lymphocytes against a HIV-1 p24 epitope in slow progressors with B'57. AIDS 2002;16:961-972.

35. Moore CB, John M, James IR, Christiansen FT, Witt CS, Mallal SA. Evidence of HIV-1 adaptation to HLArestricted immune responses at a population level. Science 2002;296:1439-1443. 36. Harcourt G, Hellier S, Bunce M, Satsangi J, Collier J, Chapman R et al. Effect of HLA class II genotype on T helper lymphocyte responses and viral control in hepatitis C virus infection. J Viral Hepat 2001;8:174-179. 37. Leon JS, Godsel LM, Wang K, Engman DM. Cardiac myosin autoimmunity in acute Chagas' heart disease. Infect Immun 2001 ;69:5643-5649. 38. Cunha-Neto E, Kalil J. Heart-infiltrating and peripheral T cells in the pathogenesis of human Chagas' disease cardiomyopathy. J Autoimmun 2001 ;34:187-192. 39. Anguita J, Samanta S, Barthold SW, Fikrig E. Ablation of interleukin-12 exacerbates Lyme arthritis in SCID mice. Infect Immun 1997;65:4334-4336. 40. Hengartner H. Transgenic mice to study viral pathogenesis and autoimmunity. Int J Expl Pathol 1996;77: 279-282. 41. Taneja V, David CS. Lessons from animal models for human autoimmune diseases. Nat Immunol 2001;2: 781-784. 42. Cervera R. The epidemiology and significance of autoimmune diseases in health care. Scand J Clin Lab Invest Supp12001 ;27-30. 43. Magnusson Y, Wallukat G, Waagstein F, Hjalmarson A, Hoebeke J. Autoimmunity in idiopathic dilated cardiomyopathy. Characterization of antibodies against the beta 1-adrenoceptor with positive chronotropic effect. Circulation 1994;89:2760-2767. 44. Zinkernagel RM. Maternal antibodies, childhood infections, and autoimmune diseases. N Engl J Med 2001 ;345:1331-1335. 45. von Herrath M, Bach JF. Juvenile autoimmune diabetes: a pathogenic role for maternal antibodies? Nat Med 2002;8:331-333. 46. Greeley SA, Katsumata M, Yu L, Eisenbarth GS, Moore DJ, Goodarzi H et al. Elimination of maternally transmitted autoantibodies prevents diabetes in nonobese diabetic mice. Nat Med 2002;8:399-402. 47. Leslie D, Lipsky P, Notkins AL. Autoantibodies as predictors of disease. Jf Clin Invest 2001;108:1417-1422. 48. Newkirk MM. Rheumatoid factors: host resistance or autoimmunity? Clin Immuno12002; 104:1-13. 49. Moalem G, Leibowitz-Amit R, Yoles E, Mor F, Cohen IR, Schwartz M. Autoimmune T cells protect neurons from secondary degeneration after central nervous system axotomy. Nat Med 1999;5:49-55. 50. Gonzalez A, Andre-Schmutz I, Carnaud C, Mathis D, Benoist C. Damage control, rather than unresponsiveness, effected by protective DX5+ T cells in autoimmune diabetes. Nat Immuno12001;2:1117-1125.

51. von Herrath MG, Oldstone M, Homann D, Christen U. Is activation of autoreactive lymphocytes always detrimental? Viral infections and regulatory circuits in autoimmunity. Curr Dir Autoimmun 2001;4:91-122. 52. Bach JF. The effect of infections on susceptibility to autoimmune and allergic diseases. N Engl J Med 2002;347:911-920. 53. Regner M, Lambert PH. Autoimmunity through infection or immunization? Nat Immuno12001 ;2:185-188. 54. Shoenfeld Y, Aron-Maor A. Vaccination and autoimmunity-'vaccinosis': a dangerous liaison? J Autoimmun 2000;14:1-10. 55. Alcami A, Koszinowski UH. Viral mechanisms of immune evasion. Immunol Today 2000;21:447-455. 56. Xu XN, Screaton GR, McMichael AJ. Virus infections: escape, resistance, and counterattack. Immunity 2001; 15:867-870. 57. Ciurea A, Klenerman P, Hunziker L, Horvath E, Odermatt B, Ochsenbein AF et al. Persistence of lymphocytic choriomeningitis virus at very low levels in immune mice. Proc Nat Acad Sci USA 1999;96: 11964-11969. 58. Janeway CA Jr. How the immune system protects the host from infection. Microbes Infect 2001 ;3:11671171. 59. Carroll M. Innate immunity in the etiopathology of autoimmunity. Nat Immuno12001 ;2:1089-1090. 60. Cerwenka A, Lanier LL. Natural killer cells, viruses and cancer. Nat Rev Immunol 2001;1:41-49. 61. Yewdell JW, Hill AB. Viral interference with antigen presentation. Nat Immuno12002;3:1019-1025. 62. Reddehase MJ. Antigens and immunoevasins: opponents in cytomegalovirus immtme surveillance. Nat Rev Immunol 2002;2:831-844. 63. Welsh RM, Lin MY, Lohman BL, Varga SM, Zarozinski CC, Selin LK. Alpha beta and gamma delta T-cell networks and their roles in natural resistance to viral infections. Immunol Rev 1997; 159:79-93. 64. Lachmann PJ, Davies A. Complement and immunity to viruses. Immunol Rev 1997; 159:69-77. 65. Shortman K, Liu YJ. Mouse and human dendritic cell subtypes. Nat Rev Immuno12002;2:151-161. 66. Mackay CR. Chemokines: immunology's high impact factors. Nat Immuno12001 ;2:95-101. 67. Murphy PM. Viral exploitation and subversion of the immune system through chemokine mimicry. Nat Immunol 2001;2:116-122. 68. Yang D, Chertov O, Oppenheim JJ. The role of mammalian antimicrobial peptides and proteins in awakening of innate host defenses and adaptive immunity. Cell Mol Life Sci 2001;58:978-989. 69. Katze MG, He Y, Gale M Jr. Viruses and interferon: a fight for supremacy. Nat Rev Immunol 2002;2:675-

149

70.

71. 72.

73.

74.

75.

76. 77. 78.

79.

80. 81.

82.

83.

84.

85.

687. Bendelac A, Bonneville M, Kearney JF. Autoreactivity by design: innate B and T lymphocytes. Nat Rev Immuno12001; 1:177-186. Mak TW, Yeh WC. Immunology: a block at the toll gate. Nature 2002;418:835-836. Kobayashi K, Hernandez LD, Galan JE, Janeway CA Jr, Medzhitov R, Flavell RA. IRAK-M is a negative regulator of Toll-like receptor signaling. Cell 2002; 110: 191-202. Akira S, Takeda K, Kaisho T. Toll-like receptors: critical proteins linking innate and acquired immunity. Nat Immuno12001 ;2:675-680. Van Noort H, Bsibsi M, Ravid R, Gveric D. Toll-like receptors in the human CNS, dynamic regulators of inflammation and repair. Immunology 2002; 107(supp. 1):43. Le Y, Li B, Gong W, Shen W, Hu J, Dunlop NM et al. Novel pathophysiological role of classical chemotactic peptide receptors and their communications with chemokine receptors. Immunol Rev 2000; 177:185-194. Arend WP. The innate immune system in rheumatoid arthritis. Arthritis Rheum 2001;44:2224-2234. Moser B, Loetscher P. Lymphocyte traffic control by chemokines. Nat Immunol 2001 ;2:123-128. Qin S, Rottman JB, Myers P, Kassam N, Weinblatt M, Loetscher M e t al. The chemokine receptors CXCR3 and CCR5 mark subsets of T cells associated with certain inflammatory reactions. J Clin Invest 1998; 101: 746-754. Frigerio S, Junt T, Lu B, Gerard C, Zumsteg U, Hollander GA et al. Beta cells are responsible for CXCR3mediated T-cell infiltration in insulitis. Nat Med 2002;8: 1414-1420. O'Shea JJ, Ma A, Lipsky P. Cytokines and autoimmunity. Nat Rev Immunol 2002;2:37-45. Flodstrom M, Maday A, Balakrishna D, Cleary MM, Yoshimura A, Sarvetnick N. Target cell defense prevents the development of diabetes after viral infection. Nat Immunol 2002;3:373-382. Seo SH, Hoffmann E, Webster RG. Lethal H5N1 influenza viruses escape host anti-viral cytokine responses. Nat Med 2002;8:950-954. Coggeshall KM. Regulation of signal transduction by the Fc gamma receptor family members and their involvement in autoimmunity. Curr Dir Autoimmun 2002;5:1-29. Raftery MJ, Schwab M, Eibert SM, Samstag Y, Walczak H, Schonrich G. Targeting the function of mature dendritic cells by human cytomegalovirus: a multilayered viral defense strategy. Immunity 2001; 15:997-1009. Grosjean I, Caux C, Bella C, Berger I, Wild F, Banchereau J et al. Measles virus infects human den-

150

dritic cells and blocks their allostimulatory properties for CD4+ T cells. J Exp Med 1997;186:801-812. 86. Heath WR, Carbone FR. Cross-presentation in viral immunity and self-tolerance. Nat Rev Immuno12001; 1: 126-134. 87. Welsh RM, Stepp SE, Szomolanyi-Tsuda E, Peacock CD. Tumor viral escape from inhibited T cells. Nat Immuno12002;3:112-114. 88. Moser JM, Gibbs J, Jensen PE, Lukacher AE. CD94NKG2A receptors regulate antiviral CD8(+) T cell responses. Nat Immunol 2002;3:189-195. 89. Orange JS, Fassett MS, Koopman LA, Boyson JE, Strominger JL. Viral evasion of natural killer cells. Nat Immuno12002;3:1006-1012. 90. Belkald Y, Piccirillo CA, Mendez S, Shevach EM, Sacks DL. CD4+ CD25+ regulatory T cells control Leishmania major persistence and immunity. Nature 2002;420:502-507. 91. Borghans JA, De Boer RJ. Memorizing innate instructions requires a sufficiently specific adaptive immune system. Int Immuno12002;14:525-532. 92. Chen HD, Fraire AE, Joris I, Brehm MA, Welsh RM, Selin LK. Memory CD8+ T cells in heterologous antiviral immunity and immunopathology in the lung. Nat Immuno12001;2:1067-1076. 93. Takaki A, Wiese M, Maertens G, Depla E, Seifert U, Liebetrau A et al. Cellular immune responses persist and humoral responses decrease two decades after recovery from a single-source outbreak of hepatitis C. Nat Med 2000;6:578-582. 94. Homann D, Teyton L, Oldstone MB. Differential regulation of antiviral T-cell immunity results in stable CD8+ but declining CD4+ T-cell memory. Nat Med 2001;7:913-919. 95. Appay V, Dunbar PR, Callan M, Klenerman P, Gillespie GM, Papagno L et al. Memory CD8+ T cells vary in 'differentiation phenotype in different persistent virus infections. Nat Med 2002;8:379-385. 96. Chang J, Braciale TJ. Respiratory syncytial virus infection suppresses lung CD8+ T-cell effector activity and peripheral CD8+ T-cell memory in the respiratory tract. Nat Med 2002;8:54-60. 97. Cash E, Charreire J, Rott O. B-cell activation by superstimulatory influenza virus hemagglutinin: a pathogenesis for autoimmunity? Immunol Rev 1996;152:67-88. 98. Aichele P, Bachmann MF, Hengartner H, Zinkernagel RM. Immunopathology or organ-specific autoimmunity as a consequence of virus infection. Immunol Rev 1996;152:21-45. 99. Panoutsakopoulou V, Cantor H. On the relationship between viral infection and autoimmunity. J Autoimmun 2001; 16:341-345. 100. Benedict CA, Norris PA, Ware CE To kill or be killed:

viral evasion of apoptosis. Nat Immunol 2002;3:10131018. 101. Jameson SC. Maintaining the norm: T-cell homeostasis. Nat Rev Immuno12002;2:547-556. 102. Trotman LC, Mosberger N, Fornerod M, Stidwill RP, Greber UF. Import of adenovirus DNA involves the nuclear pore complex receptor CAN/Nup214 and histone HI. Nat Cell Biol 2001;3:1092-1100. 103. Barnaba V. Viruses, hidden self-epitopes and autoimmunity. Immunol Rev 1996;152:47--66. 104. Benoist C, Mathis D. Autoimmunity provoked by infection: how good is the case for T cell epitope mimicry? Nat Immuno12001;2:797-801. 105. Vanderlugt CL, Miller SD. Epitope spreading in immune-mediated diseases: implications for immunotherapy. Nat Rev Immuno12002;2:85-95. 106. Rose NR. Viral damage or 'molecular mimicry'-placing the blame in myocarditis. Nat Med 2000;6:631--632. 107. Welsh RM, Selin LK. No one is naive: the significance of heterologous T-cell immunity. Nat Rev Immunoly 2002;2:417-426. 108. Hunziker L, Recher M, Macpherson AJ, Ciurea A, Freigang S, Hengartner H et al. Hypergammaglobulinemia and autoantibody induction mechanisms in viral infections. Nat Immunol 2003;4:343-349. 109. Stauffer Y, Marguerat S, Meylan F, Ucla C, Sutkowski N, Huber B et al. Interferon-alpha-induced endogenous superantigen. A model linking environment and autoimmunity. Immunity 2001;15:591--601. 110. Kelly G, Bell A, Rickinson AB. Epstein-Barr virusassociated Burkitt lymphomagenesis selects for downregulation of the nuclear antigen EBNA2. Nat Med 2002;8:1098-1104. 111. Horwitz MS. Adenovirus immunoregulatory genes and their cellular targets. Virology 2001 ;279:1-8. 112. Fujinami RS. Viruses and autoimmune disease - two sides of the same coin? Trends Microbiol 2001;9: 377-381. 113. Marguerie C, Bunn CC, Beynon HL, Bernstein RM, Hughes JM, So AK et al. Polymyositis, pulmonary fibrosis and autoantibodies to aminoacyl-tRNA synthetase enzymes. Q J Med 1990;77:1019-1038. 114. Saeki K, Zhu M, Kubosaki A, Xie J, Lan MS, Notkins AL. Targeted disruption of the protein tyrosine phosphatase-like molecule IA-2 results in alterations in glucose tolerance tests and insulin secretion. Diabetes 2002;51:1842-1850. 115. Saegusa K, Ishimaru N, Yanagi K, Arakaki R, Ogawa K, Saito I et al. Cathepsin S inhibitor prevents autoantigen presentation and autoimmunity. J Clin Invest 2002; 110:361-369. 116. Hiemstra HS, Schloot NC, van Veelen PA, Willemen SJ, Franken KL, van Rood JJ et al. Cytomegalovirus

in autoimmunity: T cell crossreactivity to viral antigen and autoantigen glutamic acid decarboxylase. Proc Natl Acad Sci USA 2001;98:3988-3991. 117. Sun Y, Chen HM, Subudhi SK, Chen J, Koka R, Chen L et al. Costimulatory molecule-targeted antibody therapy of a spontaneous autoimmune disease. Nat Med 2002;8: 1405-1413. 118. Savill J, Dransfiled I, Gregory C, Haslett C. A blast from the past: clearance of apoptotic cells regulates immune responses. Nat Rev Im_munol 2002;2:965-975. 119. Salmaso C, Bagnasco M, Pesce G, Montagna P, Brizzolara R, Altrinetti V et al. Regulation of apoptosis in endocrine autoimmunity: insights from Hashimoto's thyroiditis and Graves' disease. Ann NY Acad Sci 2002 ;966:496-501. 120. Hammond LJ, Palazzo FF, Shattock M, Goode AW, Mirakian R. Thyrocyte targets and effectors of autoimmunity: a role for death receptors? Thyroid 2001;11" 919-927. 121. Froelich CJ, Dixit VM, Yang X. Lymphocyte granule-mediated apoptosis: matters of viral mimicry and deadly proteases. Immunol Today 1998; 19:30-36. 122. Stassi G, De Maria R. Autoimmune thyroid disease: new models of cell death in autoimmunity. Nat Rev Immuno12002;2:195-204. 123. Lunardi C, Bason C, Navone R, Millo E, Damonte G, Corrocher R et al. Systemic sclerosis immunoglobulin G autoantibodies bind the human cytomegalovirus late protein UL94 and induce apoptosis in human endothelial cells. Nat Med 2000;6:1183-1186. 124. Koelle DM, Liu Z, McClurkan CM, Topp MS, Riddell SR, Pamer EG et al. Expression of cutaneous lymphocyte-associated antigen by CD8[+] T cells specific for a skin-tropic virus. J Clin Invest 2002; 110:537-548. 125. Horwitz MS, La Cava A, Fine C, Rodriguez E, Ilic A, Sarvetnick N. Pancreatic expression of interferongamma protects mice from lethal coxsackievirus B3 infection and subsequent myocarditis. Nat Med 2000;6: 693-697. 126. Monteyne P, Bureau JF, Brahic M. The infection of mouse by Theiler's virus: from genetics to immunology. Immunol Rev 1997;159:163-176. 127. Drescher KM, Pease LR, Rodriguez M. Antiviral immune responses modulate the nature of central nervous system (CNS) disease in a murine model of multiple sclerosis. Immunol Rev 1997;159:177-193. 128. Tompkins SM, Fuller KG, Miller SD. Theiler's virusmediated autoimmunity: local presentation of CNS antigens and epitope spreading. Ann NY Acad Sci 2002;958:26-38. 129. Zhao ZS, Granucci F, Yeh L, Schaffer PA, Cantor H. Molecular mimicry by herpes simplex virus-type 1" autoimmune disease after viral infection. Science

151

1998;279:1344-1347. 130. Panoutsakopoulou V, Sanchirico ME, Huster KM, Jansson M, Granucci F, Shim DJ et al. Analysis of the relationship between viral infection and autoimmune disease. Immunity 2001;15:137-147. 131. Singh VK, Kalra HK, Yamaki K, Abe T, Donoso LA, Shinohara T. Molecular mimicry between a uveitopathogenic site of S-antigen and viral peptides. Induction of experimental autoimmune uveitis in Lewis rats. J Immunol 1990;144:1282-1287. 132. Fleck M, Kern ER, Zhou T, Lang B, Mountz JD. Murine cytomegalovirus induces a Sj6gren's syndrome-like disease in C57B1/6-1pr/lpr mice. Arthritis Rheum 1998;41:2175-2184. 133. Fleck M, Zhang HG, Kern ER, Hsu HC, Muller-Ladner U, Mountz JD. Treatment of chronic sialadenitis in a murine model of Sj6gren's syndrome by local fasL gene transfer. Arthritis Rheum 2001 ;44:964-973. 134. Schnitzer TJ, Penmetcha M. Viral arthritis. Curr Opin Rheumatol 1996;8:341-345. 135. Speck SH. EBV framed in Burkitt lymphoma. Nat Med 2002;8:1086-1087. 136. Ascherio A, Munger KL, Lennette ET, Spiegelman D, Hernan MA, Olek MJ et al. Epstein-Barr virus antibodies and risk of multiple sclerosis: a prospective study (Comment). JAMA 2001 ;286:3083-3088. 137. Levin LI, Munger KL, Rubertone MV, Peck CA, Lennette ET, Spiegelman D et al. Multiple Sclerosis and Epstein-Barr Virus. JAMA 2003;289:1533-1536. 138. Gross AJ, Thorley-Lawson DA. Very high frequencies of Epstein Barr virus latently infected B lymphocytes in the blood of patients with systemic lupus erythematosus. Arthritis Rheum 2002;46:$280. 139. James JA, Neas BR, Moser KL, Hall T, Bruner GR, Sestak AL et al. Systemic lupus erythematosus in adults is associated with previous Epstein-Barr virus exposure. Arthritis Rheum 2001;44:1122-1126. 140. Shirali GS, Ni J, Chinnock RE, Johnston JK, Rosenthal GL, Bowles NE et al. Association of viral genome with graft loss in children after cardiac transplantation. N Engl J Med 2001;344:1498-1503. 141. Kammer AR, van der Burg SH, Grabscheid B, Hunziker IP, Kwappenberg KM, Reichen Jet al. Molecular mimicry of human cytochrome P450 by hepatitis C virus at the level of cytotoxic T cell recognition. J Exp Med 1999;190:169-176. 142. Ramos-Casals M, Font J, Garcia-Carrasco M, Cervera R, Jimenez S, Trejo O et al. Hepatitis C virus infection mimicking systemic lupus erythematosus: study of hepatitis C virus infection in a series of 134 Spanish patients with systemic lupus erythematosus. Arthritis Rheum 2000;43:2801-2806. 143. Notkins AL, Lernmark A. Autoimmune type 1 diabetes:

152

resolved and unresolved issues. J Clin Invest 2001;108: 1247-1252. 144. Rapoport B, McLachlan SM. Thyroid autoimmunity. J Clin Invest 2001;108:1253-1259. 145. O'Hanlon GM, Veitch J, Gallardo E, Ilia I, Chancellor AM, Willison HJ. Peripheral neuropathy associated with anti-GM2 ganglioside antibodies: clinical and immunopathological studies. J Autoirnmunity 2000;32: 133-144. 146. Cronin ME, Plotz PH. Idiopathic inflammatory myopathies. Rheum Dis Clinf North Am 1990;16:655-665. 147. Neidhart M, Rethage J, Kuchen S, Kunzler P, Crowl RM, Billingham ME et al. Retrotransposable L1 elements expressed in rheumatoid arthritis synovial tissue: association with genomic DNA hypomethylation and influence on gene expression. Arthritis Rheum 2000;43: 2634-2647. 148. Herrmann M, Hagenhofer M, Kalden JR. Retroviruses and systemic lupus erythematosus. Immunol Rev 1996;152:145-156. 149. Venables PJ, Rigby SP. Viruses in the etiopathogenesis of Sj6gren's syndrome. J Rheumatol 1997;245:3-5. 150. Jun HS, Yoon JW. The role of viruses in type I diabetes: two distinct cellular and molecular pathogenic mechanisms of virus-induced diabetes in animals. Diabetologia 2001;44:271-285. 151. Chardes T, Chapal N, Bresson D, Bes C, Giudicelli V, Lefranc MP et al. The human anti-thyroid peroxidase autoantibody repertoire in Graves' and Hashimoto's autoimmune thyroid diseases. Immunogenetics 2002;54:141-157. 152. Lindberg B, Ahlfors K, Carlsson A, Ericsson UB, Landin-Olsson M, Lernmark A et al. Previous exposure to measles, mumps, and r u b e l l a - but not vaccination during adolescence- correlates to the prevalence of pancreatic and thyroid autoantibodies. Pediatrics 1999; 104:12. 153. Peoc'h K, Dubel L, Chazouilleres O, Ocwieja T, Duron F, Poupon R et al. Polyspecificity of antimicrosomal thyroid antibodies in hepatitis C virus-related infection. Am J Gastroentero12001 ;96:2978-2983. 154. O'Hanlon GM, Plomp JJ, Chakrabarti M, Morrison I, Wagner ER, Goodyear CS et al. Anti-GQlb ganglioside antibodies mediate complement-dependent destruction of the motor nerve terminal. Brain 2001; 124:893-906. 155. Strongwater SL, Dorovini-Zis K, Ball RD, Schnitzer TJ. A murine model of polymyositis induced by coxsackievirus B1 (Tucson strain). Arthritis Rheum 1984;27:433-442. 156. Fontan D, Tam P, Messner R. Induction of autoantibodies by myopathic virus in a model of coxsackievirusinduced chronic inflammatory myopathy. Arthritis Rheum 2002;46:$397.

157. Howard OMZ, Dong HF, Yang D, Raben N, Nagaraju K, Rosen A et al. Histidyl-tRNA synthetase and asparaginyl-tRNA synthetase, autoantigens in myositis, activate chemokine receptors on T lymphocytes and immature dendritic cells. J Exp Med 2002;196: 781-791. 158. Tezak Z, Nagaraju K, Plotz P, Hoffman EE Adenoassociated virus in normal and myositis human skeletal muscle. Neurology 2000;55:1913-1917. 159. Blank M, Krause I, Fridkin M, Keller N, Kopolovic J, Goldberg I et al. Bacterial induction of autoantibodies to beta2-glycoprotein-I accounts for the infectious etiology of antiphospholipid syndrome. J Clin Invest 2002;109: 797-804. 160. Sabroe RA, Grattan CE, Francis DM, Barr RM, Kobza BA, Greaves MW. The autologous serum skin test: a screening test for autoantibodies in chronic idiopathic urticaria. Br J Dermatol 1999;140:446-452. 161. Xiong D, Lee GH, Badorff C, Dorner A, Lee S, Wolf P et al. Dystrophin deficiency markedly increases enterovirus-induced cardiomyopathy: a genetic predisposition to viral heart disease. Nat Med 2002;8:872-877. 162. Levy GG, Nichols WC, Lian EC, Foroud T, McClintick JN, McGee BM et al. Mutations in a member of the ADAMTS gene family cause thrombotic thrombocytopenic purpura. Nature 2001 ;413:488-494. 163. Klein JA, Longo-Guess CM, Rossmann ME Seburn KL, Hurd RE, Frankel WN et al. The harlequin mouse mutation downregulates apoptosis-inducing factor. Nature 2002;419:367-374. 164. Karlsson B, Gustafsson J, Hedov G, Ivarsson SA, Anneren G. Thyroid dysfunction in Down's syndrome: relation to age and thyroid autoimmunity. Arch Dis Child 1998;7913]:242-245. 165. Prasher VE Down syndrome and thyroid disorders:

a review. Down Syndrome: Research and Practice 1999;6:25-42. 166. Yospur LS, Sun NC, Figueroa P, Niihara Y. Concurrent thrombotic thrombocytopenic purpura and immune thrombocytopenic purpura in an HIV-positive patient: case report and review of the literature. Am J Hematol 1996;51:73-78. 167. Atlan H, Gersten MJ, Salk PL, Salk J. Mechanisms of autoimmunity and AIDS: prospects for therapeutic intervention. Res Immunol 1994;145:165-183. 168. Dalgleish AG, O'Byrn e KJ. Chronic immune activation and inflammation in the pathogenesis of AIDS and cancer. Adv Cancer Res 2002;84:231-276. 169. Brand A, Griffiths DJ, Herve C, Mallon E, Venables PJ. Human retrovirus-5 in rheumatic disease. J Autoimmun 1999;13:149-154. 170. Denman AM. Systemic lupus erythematosus - is a viral aetiology a credible hypothesis? J Infect 2000;40: 229-233. 171. Nakagawa K, Harrison LC. The potential roles of endogenous retroviruses in autoimmunity. Immunol Rev 1996;152:193-236. 172. Boyd GW. An evolution-based hypothesis on the origin and mechanisms of autoimmune disease. Immunol Cell Biol 1997;75:503-507. 173. Ray NB, Nieva DR, Seftor EA, Khalkhali-Ellis Z, Naides SJ. Induction of an invasive phenotype by human parvovirus B 19 in normal human synovial fibroblasts. Arthritis Rheum 2001 ;44:1582-1586. 174. Zinkernagel RM. Antiinfection immunity and autoimmunity. Ann NY Acad Sci 2002;958:3-6. 175. Perelson AS. Modelling viral and immune system dynamics. Nat Rev Immuno12002;2:28-36. 176. Thomas L. The Lives of a Cell. Bantam Books, 1974.

153


How Transgenic Mouse Models Contribute to a Better Understanding of Virus-Induced Autoimmunity /

Philippe K r e b s 1,2 and Burkhard Ludewig ~

IKantonal Hospital St. Gallen, Research Department, St. Gallen, Switzerland; 2Institute of Experimental Immunology, Department of Pathology, University Hospital Ziirich, Ziirich, Switzerland

Abbreviations: Ad-LacZ: adenovirus recombinant for [3-galactosidase, CNS: central nervous system, CTL: cytotoxic T lymphocyte, CTLP: precursor CTL, HA: influenza virus hemagglutinin protein, HBV: hepatits B virus, IDDM: insulin-dependant diabetes mellitus, LCMV: lymphocytic choriomeningitis virus, LCMV-GP: glycoprotein of LCMV, LCMV-NP: nucleoprotein of LCMV, MBP: myelin basic protein, MS: multiple sclerosis, OVA: ovalbumin, RIP: rat insulin promoter, SM: smooth muscle promoter, TCR: T cell receptor, VSV." vesicular stomatitis virus; VE." vaccinia virus.

1. INTRODUCTION The prevalence of autoimmune diseases in the Western World is high with approximately 3-5% of the general population [1, 2] and the incidence of major autoimmune disorders such as multiple sclerosis, systemic lupus erythematosus, myasthenia gravis and primary biliary cirrhosis has been steadily increasing over recent years [1]. The frequent association of autoimmune reaction with viral and bacterial infections suggests a causative link between infection and autoimmunity [3]. However, attempts to establish a direct linkage between viral infections and autoimmune diseases have been impeded by the fact that patients usually have gone through infections with several pathogens before an autoimmune disease is finally diagnosed. Furthermore, antimicrobial immune responses are detectable at the time of onset of the autoimmune disease; viral or bacterial antigens, however, are often barely

detectable. It is therefore important to delineate the infection-associated initiating events that lead to the break of tolerance and eventually provoke autoimmune diseases. The majority of self-reactive T cells is eliminated in the thymus through clonal deletion of highaffinity autoreactive T cells (Fig. 1A). However, T cells with specificity for autoantigens exclusively expressed in the periphery and low-affinity T cells may escape thymic negative selection and therefore peripheral tolerance mechanisms are required to control potential autoimmune disease-mediating T cells. The simplest scenario of peripheral tolerance is that self-reactive T cells remain quiescent because the antigen is not presented in secondary lymphoid organs in sufficient amounts, a process that has been termed immunological ignorance. A further mechanism of peripheral tolerance is the induction of T cell anergy where functional inactivation of T cells is usually induced by TCR triggering in the absence of costimulation [4, 5] (Fig. 1B). In addition, longlasting presence of antigen in the periphery and lymphoid organs may result in the physical deletion of antigen-specific T cells, most likely after induction of initial functional impairment [6, 7]. Autoimmunity is most likely initiated in the course of an infection when target tissue inflammation is provoked either by direct cytopathic effects or by immunopathological reactions against a persisting microbial agent. The consequence may be the initiation of a dominant immune response against a single self-epitope that may broaden and thereby spread to other regions of the molecule and to other target molecules of the same tissue,

155

Figure 1. Viruses break peripheral tolerance. (A) T cells that bind to self antigens with high affinity are negatively selected in the thymus. Autoreactive T cells specific for sequestered peripheral tissue antigens may escape clonal deletion and migrate towards the periphery. (B) Peripheral auto-reactive T lymphocytes either immunologicallyignore their cognate self antigens or become anergized due to the lack of efficient costimulation. (C) Following viral infection, professional antigen-presenting cells (APC) may display viral determinants that share homology with self antigens and hence activate autoreactive T cells (molecular mimicry). Alternatively, the inflammatory environment elicited by the infection may produce cytokines and chemokines that activate self-reactive bystander T cells. Due to tissue damage caused by the virus itself or by virus-specific immune cells, self antigens are released into the inflammatory milieu that trigger autoreactive T lymphocytes (epitope spreading). i.e. "epitope spreading" [8]. Infection-associated inflammation, for example, involves the release of cytokines and chemokines which attract antiviral effector cells and lymphocytes of other specificities. This "bystander effect" can be sufficient to activate lymphocytes directed against self antigens [9] leading to initiation and/or exacerbation of autoimmune disease. A third mechanism that may lead to the initiation of autoimmunity in the course of an infection is the activation of T or B cells via antigenic determinants shared between the pathogen and the host which has been termed "molecular mimicry" [10] (Fig. 1C). Cross-reactivities between pathogen-derived and self antigens have been described for human autoimmune diseases such as insulindependent diabetes, multiple sclerosis, and Guillain-Barre syndrome.

156

Transgenic technology enabled establishment of mouse models with microbial antigens present in potential autoimmune target organs with clinical manifestations that resemble the phenotype of human autoimmune diseases [11]. These models have the advantage that the onset and progression of disease can be usually controlled. Furthermore, transgenic mice with viral or bacterial antigen expression in peripheral tissues are useful to address fundamental questions on the pathogenesis of autoimmune diseases such as the contribution of the genetic background, the nature of immunogenic self-antigens, or the role of immunoregulatory molecules and cells (Table 1). In this article, we will discuss how genetically engineered transgenic mice with defined tissue expression of viral antigens have contributed to our understanding of human

Table 1. Advantages of transgenic animals models for virus-induced autoimmunity

9 Focusingof autoimmunity to a single organ through tissue-specific transgene expression 9 Use of well-characterized antigens facilitates tracking of autoreactive T and B cells 9 Elucidation of epitope spreading from the known initiating antigenic determinants 9 Analysisof the single gene effects and their combination by crossing with other transgenic or knock-out mice 9 Valuablefor the design and the assessment of new potential therapies

autoimmune disease. The different models will be described and discussed according to the human disease they mimic.

2. TRANSGENIC MODELS FOR VIRUSINDUCED AUTOIMMUNE INSULINDEPENDENT DIABETES MELLITUS

A diabetes model of virus-induced autoimmunity illustrates the phenomenon of immunological ignorance of an extrathymically displayed neo-self antigen [12, 13]. The transgenic mouse lines express the viral glycoprotein (GP) of the leukocytic choriomeningitis virus (LCMV) under the control of the rat insulin promoter (RIP) exclusively in pancreatic islet cells (see Table 2). RIP-GP mice did neither spontaneously develop insulin-dependent diabetes mellitus (IDDM) nor did they delete potentially autoreactive GP-specific T cells. The latter was demonstrated by the fact that GP-specific immune responses can be induced by LCMV infection leading to autoimmune destruction of GP-expressing pancreatic islet cells and autoimmune diabetes in only 8 to 14 days. In a second transgenic mouse model expressing the LCMV nucleoprotein (RIPNP mice) both in I~ islet cells and thymus, diabetes developed slowly within three to six months after LCMV infection thus modeling slow onset autoimmune diabetes that is mediated mainly by low avidity T cells [14]. Interestingly, a less immunogenic LCMV-GP recombinant vaccinia virus (VV) elicited only a mild insulitis in RIP-GP mice without signifi-

cant elevation of blood glucose levels [ 15]. In addition, important observations on quantitative aspects of autoimmunity have been made in RIP-GP mice. Elevation of CTLp frequencies in double transgenic mice expressing the LCMV-GP in the pancreas and a specific TCR on their CTL (RIP-GPxTCR) strongly accelerated disease and compensated for the weak CTL induction by LCMV-GP recombinant VV [ 12, 15]. VV-GP-induced autoimmune diabetes could also be generated when the B7 costimulatory molecule was locally expressed in the 13-islets of RIP-GP mice (RIP-GPxRIP-B7) [ 16, 17], or when TNF-t~ was expressed in pancreatic islets (RIPGPxRIP-TNF-t~) [15, 18]. These mouse models present not only the characteristic hallmarks of IDDM in humans, namely hyperglycemia, hypoinsulinemia, and mononuclear cell infiltration in the ~l islets, but can also be used for the assessment of novel immunotherapeutical approaches [ 19-21 ] or the effect of regulatory T cells in the pathogenesis of autoimmune diabetes [22]. Transgenic mice expressing the influenza virus hemagglutinin (HA) as a neo-self antigen in the pancreatic islet [~ cells (Ins-HA mice) revealed that peripheral antigen may efficiently anergize self-reactive T cells [23]. Since thymocyte development is not impaired in Ins-HA mice [24], the absence of autoimmune destruction in pancreatic islets after influenza virus infection suggested efficient peripheral tolerization of self-reactive T cells. Indeed, HA-specific TCR-transgenic CD8 § T cells were activated and proliferated exclusively in the draining lymph nodes of the pancreas [25] and were subsequently functionally deleted [26]. It is most likely that HA is cross-presented by bone marrow-derived antigen presenting cells in the local lymphoid tissue in Ins-HA mice. Interestingly, not only naive but also memory CD8 § T cells may be tolerized under these conditions by the peripherally expressed antigen [27].

3. MODELING VIRUS-INDUCED LIVER DISEASE

Infection with the hepatitis B virus (HBV) or hepatitis C virus (HCV) causes severe inflammatory liver disease of variable duration and severity. Persistently infected patients with ongoing liver disease

157

Table 2. Transgenic models Disease modeled

Transgene construct (promoter and antigen)

Infectious agent

Findings

References

IDDM

(RIP)-LCMV-GP

LCMV

Peripherally expressed self-antigen may be immunologically ignored

[12, 13]

IDDM

(RIP)-LCMV-GP

LCMV-GP recombinant VV

Amount of self-reactive T cells and inflammatory milieu in the target organs determine the extent of autoimmune disease

[15-18]

IDDM

(RIP)-LCMV-NP

LCMV

Low affinity CTL mediate slow-onset IDDM

[14]

IDDM

(RIP)-Influenza HA

Influenza and HA-VV

Peripheral antigen may anergize self-reactive CTL in the local lymph node; memory CTL may be tolerized

[23, 25-27]

Hepatitis

(Albumin)-HBV envelope proteins

HBV recombinant VV

Induction of autoantibodies, but absence of HBV-specific CTL and liver disease

[29, 30]

Hepatitis

(Albumin)-LCMV-GP

LCMV

Break of tolerance and self-limited hepatitis after LCMV infection only if TCR transgenic CTL had been adoptively transferred

[31]

Multiple sclerosis

(MBP)-LCMV-GP and -NP

LCMV

Induction of CNS inflammatory and demyelinating disease through repeated infection

[32]

VSV-OVA

Adoptive transfer of IL-12-treated effector CTL augmented myocarditis

[37]

Virus-induced tolerization of self-reactive CTL through functional paralysis and/or exhaustion.

Krebs & Ludewig (in preparation)

Myocarditis (Murine cardiac ormyosin heavy-chain)OVA

Myocarditis (SM-22)-~l-galactosidase Ad-LacZ

Abbreviations: Ad-LacZ, adenovirus recombinant for ~-galactosidase; CNS, central nervous system; CTL, cytotoxic T lymphocyte; HA, influenza virus hemagglutinin protein; HBV, hepatits B virus; IDDM, insulin-dependant diabetes mellitus; LCMV, lymphocytic choriomeningitis virus; LCMV-GP, glycoprotein of LCMV; LCMV-NP, nucleoprotein of LCMV; MBP, myelin basic protein; MS, multiple sclerosis; OT-I, ovalbumin-specific transgenic CTL; OVA, ovalburnin; RIP, rat insulin promoter; SM, smooth muscle promoter; TCR, T cell receptor; VSV, vesicular stomatitis virus; VV, vaccinia virus.

may develop cirrhosis, hepatocellular carcinoma, and, in the case of chronic HCV infection, may also develop autoimmune liver disease. Transgenic mice constitutively expressing the HBV envelope proteins containing hepatitis B surface antigen (HBsAg) in the liver under the transcriptional control of the mouse albumin promoter represent a well-established model for virus-induced immunopathological liver damage [28]. In these mice, HBsAg was detected in virtually all hepatocytes and was also secreted into the blood. Repetitive infection with HBV envelope-recombinant VV leads to production of low titers of T cell-dependent antiHBV IgG autoantibodies that clear HBsAg from the

158

blood, but not to activation of HBsAg-specific CTL [29]. In this model, immunization of the transgenic animals failed to induce ongoing autoimrnune liver disease, whereas adoptive transfer of effector CTL elicited fulminant and resolving hepatitis [30]. A similar model has been established by Voehringer et al [31] who expressed the GP33 peptide of the LCMV-GP under the control of the albumin promoter (ALB 1-GP33 mice). Partial thymic deletion of GP33-specific T cells resulted in reduced GP33-specific CTL responses after LCMV infection and most likely helped to avoid LCMV-induced liver damage in ALB 1-GP33 mice [31]. It is interesting to note that adoptively transferred TCR trans-

genic CTL recognizing GP33 ignored the peripheral antigen expressed in hepatocytes, whereas adoptive transfer of activated GP33-specific CTL elicited a significant fiver disease. Virus-induced autoimmune hepatitis could be induced by adoptive transfer of naive TCR-transgenic CTL followed by LCMV infection [31]. Overall, these studies indicate that the liver may serve as a target organ for virusinduced autoimmune disease.

4. VIRUS-INDUCED AUTOIMMUNITY IN THE CENTRAL NERVOUS SYSTEM Human autoimmune demyelinating diseases such as multiple sclerosis is most likely a CD4 + T-ceU mediated disease that is associated with viral infections [8]. Virus-induced autoimmunity in the central nervous system has been studied by Evans et al. [32] who expressed LCMV viral antigens (glycoprotein and nucleoprotein) as transgenes in oligodendrocytes of the central nervous system (CNS) under the control of the myelin basic protein (MBP) (MBP-GP and MBP-NP mice). Similar to RIP-GP mice, autoreactive lymphocytes that escaped thymic negative selection were present in the periphery and were activated by LCMV infection. Virally activated autoreactive CTL were able to cross the blood-brain barrier, migrated into the CNS and lysed transgeneexpressing oligodentrocytes. Following a second LCMV infection, the amount of infiltrating cells massively increased, leading to significant motor dysfunction in infected transgenic animals. The experimental disorder in MBP-GP and MBP-NP mice disorder resembles some characteristics of human demyelinating disease [33] suggesting that relapses in multiple sclerosis that often occur after viral infections could be caused by a reactivation of oligodendrocyte-specific T cells that were initially generated through molecular mimicry.

5. VIRAL INFECTION AND AUTOIMMUNE MECHANISMS OF MYOCARDITIS Dilated cardiomyopathy is one of the leading causes of heart failure and is most likely a sequel of myocarditis induced by infectious agents such as Coxsackievirus B (CVB) or cytomegalovirus [34]. Dis-

ease in mice induced by serotype 3 CVB resembles the human situation because infection of susceptible mouse strains elicits first an acute myocarditis that resolves around day 14 post infection, followed by a chronic phase with persistent low level inflammation of the cardiac muscle. The immune system not only plays an important protective role against the infection and subsequent heart disease induced by CVB3, but may also contribute to cardiac damage by attacking heart cells. For example, cardiac damage is dramatically reduced in mice lacking a functional T cell response [35], indicating that immunopathological damage contributes crucially to myocardial injury during CVB infection. It has been suggested that molecular mimicry between viral pathogens and myocardial proteins leads to induction of cross-reactive T cells directed against heart antigens [36]. However, other studies indicate that CVB-induced "bystander activation" of selfreactive T cells is the major immunopathological mechanism in insulin-dependent diabetes [9]. It is thus still an open question to which extent the different immunopathological mechanisms contribute to virus-mediated acute and chronic heart disease. Recently, a mouse line has been developed that expresses cardiac myocyte-restricted membranebound ovalbumin (CMy-mOVA) [37]. Despite no detectable transgene expression the thymus, these transgenic mice were tolerant to OVA as shown by the lack or OVA-specific immune responses following infection with OVA-expressing vesicular stomatitis virus (VSV-OVA). However, adoptive transfer of naive OVA-specific TCR transgenic CTL and subsequent infection with VSV-OVA induced myocarditis in CMy-mOVA mice. Adoptive transfer experiments revealed that OVA-specific effector CTL require IL-12 during their in vitro stimulation to acquire full pathogenic potential. A transgenic mouse model with defined T cellmediated cardiovascular immunopathology has been recently established by our group [38]. SMLacZ mice express the bacterial 13-galactosidase (I]-gal) antigen in cardiomyocytes of the fight heart and in arterial smooth muscle cells [39]. The [~-gal transgene is immunologically ignored in these mice, despite widespread expression in the vascular system. Repetitive priming of SM-LacZ mice with dendritic cells (DC) presenting [3-gal peptide caused acute vascular immunopathology with

159

strong lymphocytic infiltration in lung arteries and aorta (arteritis) and in the fight heart (myocarditis). In the chronic phase, despite ceasing immunization with 13-gal peptide-loaded DC, SM-LacZ mice show a severe loss of functional heart tissue and fibrosis that eventually leads to dilated cardiomyopathy [40]. This transgenic model is therefore well-suited for the characterization of pathological mechanisms in cardiovascular diseases [41]. Interestingly, following intravenous administration of replicationdeficient [~-gal-recombinant adenovirus (Ad-LacZ), only limited cellular infiltrations were observed in lungs and myocardium of SM-LacZ mice (Krebs and Ludewig, manuscript in preparation). Moreover, SM-LacZ mice displayed only weak [3galactosidase-specific CTL response compared to wild type C57BL/6 mice. In this particular model, peripheral tolerance is thus most likely established by anergization and/or clonal deletion of specific CTL because high amounts of Ad-LacZ-encoded [3galactosidase antigen are presented for too long in peripheral and lymphoid organs, leading to exhaustive activation of 13-galactosidase-specific CTL.

3.

4.

5. 6.

7.

8.

9.

10. 6. C O N C L U S I O N Viruses may disturb the fine-tuned balance of the immune system by acting as "adjuvant" and/or by stimulating cross-reactive T and B cells with specifity for both viral and self antigens. Transgenic mouse models have helped to uncover the basic rules how virus interfere with self tolerance. Furthermore, the different transgenic mouse models of virus-induced autoimmunity described here represent valuable tools to delineate basic pathogenic mechanisms and to evaluate therapeutical strategies to intervene with early detrimental processes that lead to manifest autoimmune disease.

11.

12.

13.

14.

15.

REFERENCES 1.

2.

160

Jacobson DL, Gange SJ, Rose NR, Graham NM. Epidemiology and estimated population burden of selected autoimmune diseases in the United States. Clin Immunol Immunopathol 1997;84(3):223-243. Marrack P, Kappler J, Kotzin BL. Autoimmune disease: why and where it occurs. Nat Med 2001;7(8):899-905.

16.

Benoist C, Mathis D. Autoimmunity provoked by infection: how good is the case for T cell epitope mimicry? Nat Immunol 2001 ;2(9):797-801. Jenkins MK, Schwartz RH. Antigen presentation by chemically modified splenocytes induces antigen-specific T cell unresponsiveness in vitro and in vivo. J Exp Med 1987;165(2):302-319. SchwartzRH. A cell culture model for T lymphocyte clonal anergy. Science 1990;248(4961): 1349-1356. Kyburz D, Aichele P, Speiser DE, Hengartner H, Zinkemagel RM, Pircher H. T cell immunity after a viral infection versus T cell tolerance induced by soluble viral peptides. Eur J Immunol 1993;23(8): 1956-1962. Wherry EJ, Blattman JN, Murali-Krishna K, van der MR, Ahmed R. Viral persistence alters CD8 T-cell immunodominance and tissue distribution and results in distinct stages of functional impairment. J Virol 2003;77(8):4911-4927. Vanderlugt CL, Miller SD. Epitope spreading in immune-mediated diseases: implications for imnmnotherapy. Nat Rev Immunol 2002;2(2):85-95. Horwitz MS, Bradley LM, Harbertson J, Krahl T, Lee J, Sarvetnick N. Diabetes induced by Coxsackie virus: initiation by bystander damage and not molecular mimicry. Nat Med 1998;4(7):781-785. Oldstone MB. Molecular mimicry and autoimmune disease. Cell 1987;50(6):819-820. Boyton RJ, Altmann DM. Transgenic models of autoimmune disease. Clin Exp Immunol 2002;127(1): 4-11. Ohashi PS, Oehen S, Buerki K, Pircher H, Ohashi CT, Odermatt Bet al. Ablation of "tolerance" and induction of diabetes by virus infection in viral antigen transgenic mice. Cell 1991;65(2):305-317. Oldstone MB, Nerenberg M, Southern P, Price J, Lewicki H. Virus infection triggers insulin-dependent diabetes mellitus in a transgenic model: role of anti-self (virus) immune response. Cell 1991;65(2):319-331. Von Herrath MG, Dockter J, Oldstone MB. How virus induces a rapid or slow onset insulin-dependent diabetes mellitus in a transgenic model. Immunity 1994;1(3): 231-242. Ohashi PS, Oehen S, Aichele P, Pircher H, Odermatt B, Herrera Pet al. Induction of diabetes is influenced by the infectious virus and local expression of MHC class I and tumor necrosis factor-or. J Immunol 1993;150(11): 5185-5194. Harlan DM, Hengartner H, Huang ML, Kang YH, Abe R, Moreadith RW et al. Mice expressing both B7-1 and viral glycoprotein on pancreatic beta cells along with glycoprotein-specific transgenic T cells develop diabetes due to a breakdown of T-lymphocyte unresponsive-

ness. Proc Natl Acad Sci USA 1994;91(8):3137-3141. 17. Von Herrath MG, Guerder S, Lewicki H, Flavell RA, Oldstone MB. Coexpression of B7-1 and viral ("self") transgenes in pancreatic beta cells can break peripheral ignorance and lead to spontaneous autoimmune diabetes. Immunity 1995;3(6):727-738. 18. Higuchi Y, Herrera P, Muniesa P, Huarte J, Belin D, Ohashi P e t al. Expression of a tumor necrosis factor alpha transgene in murine pancreatic beta cells results in severe and permanent insulitis without evolution towards diabetes. J Exp Med 1992; 176(6): 1719-1731. 19. Aichele P, Kyburz D, Ohashi PS, Odermatt B, Zinkemagel RM, Hengartner H et al. Peptide-induced T-cell tolerance to prevent autoimmune diabetes in a transgenic mouse model. Proc Natl Acad Sci USA 1994;91(2): 444--448. 20. Bot A, Smith D, Bot S, Hughes A, Wolfe T, Wang Let al. Plasmid vaccination with insulin B chain prevents autoimmune diabetes in nonobese diabetic mice. J Immuno12001;167(5):2950-2955. 21. Wolfe T, Bot A, Hughes A, Mohrle U, Rodrigo E, Jaume JC et al. Endogenous expression levels of autoantigens influence success or failure of DNA immunizations to prevent type 1 diabetes: addition of IL-4 increases safety. Eur J Immuno12002;32(1): 113-121. 22. Homann D, Jahreis A, Wolfe T, Hughes A, Coon B, van Stipdonk MJ et al. CD40L blockade prevents autoimmune diabetes by induction of bitypic NK/DC regulatory cells. Immunity 2002; 16(3):403-415. 23. Lo D, Freedman J, Hesse S, Palmiter RD, Brinster RL, Sherman LA. Peripheral tolerance to an islet cell-specific hemagglutinin transgene affects both CD4+ and CD8+ T cells. Eur J Immunol 1992;22(4):1013-1022. 24. Morgan DJ, Liblau R, Scott B, Fleck S, McDevitt HO, Sarvetnick N et al. CD8(+) T cell-mediated spontaneous diabetes in neonatal mice. J Immunol 1996;157(3): 978-983. 25. Morgan DJ, Kurts C, Kreuwel HT, Hoist KL, Heath WR, Sherman LA. Ontogeny of T cell tolerance to peripherally expressed antigens. Proc Natl Acad Sci USA 1999;96(7):3854-3858. 26. Morgan DJ, Kreuwel HT, Sherman LA. Antigen concentration and precursor frequency determine the rate of CD8+ T cell tolerance to peripherally expressed antigens. J Immunol 1999; 163(2):723-727. 27. Kreuwel HT, Aung S, Silao C, Sherman LA. Memory CD8(+) T cells undergo peripheral tolerance. Immunity 2002;17(1):73-81. 28. Chisari FV, Ferrari C. Hepatitis B virus immunopathogenesis. Annu Rev Immunol 1995; 13:29-60:29-60. 29. Wirth S, Guidotti LG, Ando K, Schlicht HJ, Chisari bag". Breaking tolerance leads to autoantibody production but not autoimmune liver disease in hepatitis B virus

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

envelope transgenic mice. J Immunol 1995;154(5): 2504-2515. Ando K, Moriyama T, Guidotti LG, Wirth S, Schreiber RD, Schlicht HJ et al. Mechanisms of class I restricted immunopathology. A transgenic mouse model of fulminant hepatitis. J Exp Med 1993;178(5):1541-1554. Voehringer D, B laser C, Grawitz AB, Chisari FV, Buerki K, Pircher H. Break of T cell ignorance to a viral antigen in the liver induces hepatitis. J Immunol 2000; 165(5):2415-2422. Evans CF, Horwitz MS, Hobbs MV, Oldstone MB. Viral infection of transgenic mice expressing a viral protein in oligodendrocytes leads to chronic central nervous system autoimmune disease. J Exp Med 1996;184(6): 2371-2384. Von Herrath MG, Evans CF, Horwitz MS, Oldstone MB. Using transgenic mouse models to dissect the pathogenesis of virus-induced autoimmune disorders of the islets of Langerhans and the central nervous system. Immunol Rev 1996;152:111-43:111-143. Fairweather D, Kaya Z, Shellam GR, Lawson CM, Rose NR. From infection to autoimmunity. J Autoimmun 2001;16(3):175-186. Opavsky MA, Penninger J, Aitken K, Wen WH, Dawood F, Mak T et al. Susceptibility to myocarditis is dependent on the response of alphabeta T lymphocytes to coxsackieviral infection [see comments]. Circ Res 1999;85(6):551-558. Schwimmbeck PL, Huber SA, Schultheiss HP. Roles of T cells in coxsackievirus B-induced disease. Curr Top Microbiol Immunol 1997;223:283-303:283-303. Grabie N, Delfs MW, Westrich JR, Love VA, Stavrakis G, Ahmad F et al. IL-12 is required for differentiation of pathogenic CD8+ T cell effectors that cause myocarditis. J Clin Invest 2003;111(5):671-680. Ludewig B, Freigang S, Jaggi M, Kurrer MO, Pei YC, Vlk L e t al. Linking immune-mediated arterial inflammation and cholesterol-induced atherosclerosis in a transgenic mouse model. Proc Natl Acad Sci USA 2000;97(23): 12752-12757. Moessler H, Mericskay M, Li Z, Nagl S, Paulin D, Small JV. The SM 22 promotor directs tissue-specific expression in arterial but not in venous or visceral smooth muscle cells in transgenic mice. Development 1996; 122:2415-2425. Ludewig B, Ochsenbein AF, Odermatt B, Paulin D, Hengartner H, Zinkernagel RM. Immunotherapy with dendritic cells directed against tumor antigens shared with normal host cells results in severe autoimmune disease. J Exp Med 2000 2000;191(5):795-804. Ludewig B, Zinkemagel RM, Hengartner H. Arterial inflammation and atherosclerosis. Trends Cardiovasc Med 2002;12(4): 154-159.

161

9 2004 Elsevier B. V. All rights reserved. Infection and Autoimmurfity Y. Shoenfeld and N.R. Rose, editors

Epstein-Barr Virus and Autoimmunity Michael P. Pender

Neuroimmunology Research Centre, School of Medicine, The University of Queensland, and Department of Neurology, Royal Brisbane and Women's Hospital Brisbane, Queensland, Australia

1. I N T R O D U C T I O N There is a large body of evidence that infection with the Epstein-Barr virus (EBV), the aetiological agent of infectious mononucleosis, has a role in the pathogenesis of many human chronic autoimmune diseases. This chapter will review the evidence for the role of EBV in each of these diseases and also focus on the features that are common to the different human chronic autoimmune diseases, with the aim of providing an explanation for what appears to be a unique role for EBV in the pathogenesis of these diseases.

2. G E N E R A L ASPECTS OF HUMAN C H R O N I C A U T O I M M U N E DISEASES Human chronic autoimmune diseases share a number of common features. The various autoimmune diseases have similarities in their patterns of genetic susceptibility. The major histocompatibility complex (MHC) class II region contributes to this genetic susceptibility, and each autoimmune disease is associated with particular MHC class II genes [1]. However, there is increasing evidence that another important genetic component is susceptibility to 'autoimmunity-in-general'. People with one particular autoimmune disease such as multiple sclerosis (MS) have an increased risk of developing other autoimmune diseases, and their first-degree relatives also have an increased risk of developing other autoimmune diseases [2]. Studies on autoimmune family pedigrees have led to the proposal that autoimmunity is an autosomal dominant trait with

penetrance (disease expression) in -92% of females and 49% of males carrying the abnormal gene [3, 4]. Furthermore, people with organ-specific autoimmune diseases, such as insulin-dependent diabetes mellitus [5], autoimmune thyroid disease [6], MS [7] and inflammatory bowel disease [8] have an increased incidence of antinuclear antibodies. I have recently proposed that the genetic susceptibility to 'autoimmunity-in-general' is mediated by susceptibility to the effects of B-cell infection by EBV [9]. Human autoimmune diseases are generally more common in females than males and tend to be exacerbated in the post-partum period. Many chronic autoimmune diseases have a relapsing-remitting course, for example rheumatoid arthritis (RA), ulcerative colitis and MS, suggesting fluctuations in the autoimmune attack. Other autoimmune diseases, such as insulin-dependent diabetes mellitus and autoimmune hypothyroidism, do not become clinically apparent until much of the target organ has been destroyed; fluctuating autoimmune attack might also be occurring in these diseases but would not be clinically evident. Some chronic autoimmune diseases are manifested clinically by a primary progressive course, such as primary progressive MS, where there is progressive clinical deterioration without clear relapses or remissions. In such diseases there still could be fluctuations in the level of autoimmune attack but these could be masked by a lack of target organ repair and a subsequent lack of any periods of clinical improvement. There is also evidence of similarities in the environmental factors that predispose to or exacerbate different chronic autoimmune diseases; for example, exacerbations can be triggered by a variety of infections.

163

3. GENERAL ASPECTS OF EBV INFECTION

EBV has the unique ability to infect, activate and latently persist in B lymphocytes. When EBV infects resting B cells in vitro, it drives them into activation and proliferation independently of T-cell help. Infection of B cells from normal individuals in vitro results in the production of monoclonal autoantibodies reacting with antigens in multiple organs [10]. This accounts for the transient appearance of autoantibodies during the course of infectious mononucleosis [11]. Usually, the proliferating infected B cells are eventually eliminated by EBV-specific cytotoxic CD8+ T cells, but latently infected non-proliferating memory B cells persist in the individual for life [12]. Antigen-driven differentiation of latently infected memory B cells into plasma cells might trigger entry into the lytic cycle with the production of infectious virus [ 12].

4. POSSIBLE MECHANISMS BY WHICH EBV INFECTION COULD PROMOTE AUTOIMMUNE DISEASE

EBV infection could promote autoimmune disease by: inducing cross-reactive immune responses against self antigens; infection of organs with resultant tissue damage and release of antigens and secondary immune sensitization; non-specific general upregulation of the immune system; infection of autoreactive B cells which could produce autoantibodies and act as professional antigen-presenting cells in the target organ. There is evidence for T-cell or antibody cross-reactivity between EBV antigens and self antigens, for example myelin basic protein in MS [13, 14], La antigen in Sjrgren's syndrome [15], SmD in systemic lupus erythematosus (SLE) [16] and self MHC-derived peptides in oligoarticular juvenile idiopathic arthritis [17]. However, cross-reactivity between self antigens and viral antigens is a phenomenon applicable to all infectious agents and is therefore unlikely to be the primary mechanism for the unique role that EBV appears to have in the pathogenesis of autoimmune diseases such as MS and SLE. Infection of organs with resultant tissue damage, release of antigens and secondary immune sensitization is also a mechanism that potentially could occur following infections

164

with many different agents. Similarly, non-specific general upregulation of the immune system, for example through upregulation of cytokines and adhesion molecules, could also occur following any infection. In contrast, the ability of EBV to infect and immortalize B cells, including autoreactive B cells, is unique and therefore a likely explanation for a unique pathogenic role of EBV in human chronic autoimmune diseases [9]. EBV-infected autoreactive B cells could produce pathogenic autoantibodies. They could also act as professional antigen-presenting cells in the target organ where they could provide a costimulatory survival signal to autoreactive T cells that have been activated in peripheral lymphoid organs by cross-reactivity with infectious agents and that would otherwise undergo activation-induced apoptosis when they enter the target organ [ 18-20]. On receiving a costimulatory survival signal from the EBV-infected B cells, the autoreactive T cells could instead proliferate and produce cytokines, which recruit other inflammatory cells, with resultant target organ damage and chronic autoimmune disease [9].

5. RELATIONSHIPS BETWEEN EBV INFECTION AND PARTICULAR AUTOIMMUNE DISEASES 5.1. Multiple Sclerosis (MS)

In 1980 Sumaya et al [21] reported a higher frequency of EBV seropositivity and a higher prevalence of high anti-EBV antibody titres in patients with MS compared to controls. Subsequent studies have shown that patients with MS are almost universally seropositive for EBV, raising the possibility that EBV infection might be a prerequisite for the development of MS. A review of eight published case-control studies comparing EBV serology in MS patients and controls revealed that 99% of MS patients were EBV-seropositive compared to 90% of controls; the summary odds ratio of MS comparing EBV-seropositive individuals with EBV-seronegative individuals was 13.5 (95% confidence interval = 6.3-31.4) [22]. This difference does not apply to other herpes viruses [23]. Furthermore, a definite clinical history of infectious mononucleosis, which indicates primary infection with EBV with a high

frequency of infected B cells [ 11] further increases the risk of MS in EBV-seropositive subjects (eightfold, if infection occurs before the age of 18 years) [24]. Levin et al [25], in a study of blood samples collected from US military personnel before the onset of MS, have shown that the presence of high titres of antibodies to EBV increases the risk 34-fold for developing MS. In some cases the first attack of MS has occurred at the time of primary EBV infection [26]. Interestingly, elevated anti-EBV antibody levels were found in a child who developed MS at the age of 10 months [27]. Anti-EBV antibodies occur more often in the cerebrospinal fluid (CSF) of MS patients than controls [28], but MS patients exhibit local central nervous system (CNS) production of antibodies to various viruses [29]. Some patients have CSF oligoclonal bands of IgG reacting with EBV nuclear antigen-1 (EBNA-1) [30]. In 1979 Fraser et al [31] reported that patients with clinically active MS had an increased tendency to spontaneous in-vitro B-lymphocyte transformation compared to healthy subjects and patients with clinically quiescent MS. This could result from an increased frequency of circulating EBV-infected B cells or from defective control of outgrowth of EBV-transformed B cells in vitro by EBV-specific cytotoxic T cells. Wandinger et al [23] found EBV DNA in the sera of patients with clinically active MS but not in those with clinically stable disease. They interpreted this as evidence of an association between disease activity and EBV replication, which was supported by the finding of increased IgM and IgA responses to EBV early antigens in the patients with clinically active disease. Analysis of the CSF from MS patients using the polymerase chain reaction has not detected EBV DNA [32]; this makes it unlikely that EBV is a major target for immune attack in the CNS but does not exclude the presence of EBV-infected B cells that could act as professional antigen-presenting cells in the CNS. I have suggested [9] that EBV-infected B cells could be the source of the monoclonally expanded B cells present in the CSF of MS patients [33] and be responsible for the development of primary B-cell lymphoma in the CNS in MS [34]. Patients with MS have defective T-cell control of EBV-infected B cells [35]. One possible mechanism for this is decreased MHC class I expression on B cells, which has been reported to occur in patients

with MS [36] and other autoimmune diseases [37], although it remains unclear whether the reported decrease is sufficient to cause decreased EBV-specific CD8+ T-cell cytotoxicity. A recent study found an increased frequency of CD8+ T cells responding to two immunodominant EBV epitopes in MS patients but it was not determined whether these T cells were cytotoxic [38]. EBV-specific CD8+ T cells are enriched in MS brain lesions compared to the peripheral blood, but such enrichment is also found for EBV-specific and cytomegalovirus-specific CD8+ T cells in other inflammatory lesions of the brain and other organs, including non-autoimmune inflammatory lesions [39]. This might simply reflect the accumulation of activated T cells in any chronic inflarmnatory lesion and does not necessarily imply that the virus-specific T cells are recognizing viral antigen or cross-reacting self antigen in the inflamed organ. There is evidence of T-cell cross-reactivity between EBV antigens and the myelin antigen, myelin basic protein [13]. A CD4+ T-cell clone from an MS patient has been found to react with both a DPO35*0101-restricted EBV peptide and a DRB 1" 1501-restricted myelin basic protein peptide [14]. Furthermore, EBV infection induces the B-cell expression of ctB-crystallin, a small heat-shock protein [40], which has been reported to be present in MS lesions and to be an immunodominant myelin antigen for T cells from healthy subjects and MS patients [41]. These findings have been interpreted as evidence that T cells generated in response to t~B-crystallin expressed and presented by EBVinfected B cells might be pathogenic for CNS myelin expressing the same stress-induced protein [40].

5.2. Systemic Lupus Erythematosus (SLE) In 1971 Evans et al [42] reported elevated levels of anti-EBV antibodies in the sera of patients with SLE. Subsequent studies have shown that 99% of SLE patients are seropositive for EBV [43, 44]. The association of EBV-seropositivity with SLE is particularly striking in young patients, 99% of whom are seropositive compared to 70% of agematched controls (odds ratio 49.9, 95% confidence interval 9.3-1025, P < 0.00000000001) [43]. Seroconversion rates for other herpes viruses do not

165

differ between SLE patients and controls [43, 44]. SLE can develop immediately after EBV-induced infectious mononucleosis [45]. T cells from patients with SLE cannot control the numbers of EBV-infected B cells from SLE patients or normal subjects but T cells from normal EBV-seropositive subjects can control infected B cells from SLE patients [46]; this indicates impaired T-cell control of EBV-infected B cells in SLE. This might be explained by the reported decrease in MHC class I expression on B cells in patients with SLE [37]. Patients with SLE have autoantibodies that bind an amino acid sequence which is shared between SmD, a small nuclear ribonucleoprotein, and EBNA-1 [16].

5.3. Rheumatoid Arthritis (RA) Patients with RA have increased anti-EBV antibody levels in their sera compared to healthy subjects [47]. They also have an increased frequency of circulating EBV-infected B cells, as determined by the frequency of spontaneously transforming B cells [48]. A recent study using real-time polymerase chain reaction has demonstrated a 10-fold increase in the EBV DNA load in the peripheral blood mononuclear cells of patients with RA compared to normal controls [49]. The high frequency of EBV-infected B cells in patients with RA is not due to increased uptake of the virus by B cells [48] but might be explained by the defective control of infected B cells by EBV-specific T cells [50, 51]. This might be explained by the reported decrease in MHC class I expression on B cells in patients with RA [37]. A study using a highly sensitive in-situ hybridization technique to detect EBV-encoded small nuclear RNAs (EBERs) in synovial membrane biopsy samples of patients with RA concluded that there was a lack of evidence for involvement of EBV [52]. Yet, the study actually found EBERs in seven (19%) of 37 patients with RA and in zero of 51 patients with other joint diseases; cells expressing EBERs were B cells and plasma cells. These results could also be interpreted as supporting a role for EBV infection of B cells in the pathogenesis of RA if the negative results in the other patients with RA were due to the limitations imposed by sampling. EBV-specific CD8+ T cells are enriched in the inflamed joints of patients with RA compared to the peripheral blood,

166:

but such enrichment is also found for cytomegalovirus-specific CD8+ T cells in the inflamed joints and for EBV-specific and cytomegalovirus-specific CD8+ T cells in autoimmune and non-autoimmune inflammatory lesions in other organs [39].

5.4. Sjiigren's Syndrome Patients with Sj6gren's syndrome have increased levels of anti-EBV antibodies in their sera [53, 54], an increased tendency to spontaneous in-vitro B-lymphocyte transformation from the peripheral blood [54] and an increased frequency of shedding of EBV from the oropharynx [54]. They also have decreased EBV-specific T-cell cytotoxicity [55] which accounts for the impaired ability to abort in-vitro outgrowth in regression assays of EBVinduced B-cell transformation [54, 55]. Decreased EBV-specific T-cell cytotoxicity might be explained by the reported decrease in MHC class I expression on B cells of patients with Sj6gren's syndrome [37]. EBV-infected B cells could be the source of the monoclonally expanded B cells in the salivary glands in Sj6gren's syndrome [56] and be responsible for the increased risk of the development of B-cell lymphoma in the salivary glands in Sj6gren's syndrome [57]. Moreover, antibodies to the La autoantigen of Sj6gren's syndrome also react with EBERs complexed with protein [15].

5.5. Autoimmune Thyroid Disease Patients with autoimmune thyroiditis have increased titres of anti-EBV antibodies in their sera compared to healthy subjects [58]. Thyrotoxicosis can develop immediately after infectious mononucleosis due to primary EBV infection, and autoimmune hypothyroidism can develop in association with acute EBV infection [59]. Intrathyroidal EBV-infected B cells could be the source of the monoclonally expanded B cells in the thyroid gland in autoimmune thyroiditis [60] and might be responsible for the increased risk of development of B-cell lymphoma in the thyroid gland in patients with autoimmune thyroiditis [61 ].

5.6. Scleroderma Patients with scleroderma have defective T-cell control of EBV-infected B cells [62]. Progressive

systemic sclerosis has developed in an infant five months after infectious mononucleosis [63].

5.7. Autoimmune Liver Disease There is evidence for a role of EBV in both primary biliary cirrhosis and autoimmune hepatitis. Patients with primary biliary cirrhosis have increased levels of EBV DNA in their peripheral blood mononuclear cells, fiver and saliva compared to controls [64]. They also have defective T-cell control of EBVinfected B cells [65]. Autoimmune hepatitis can develop soon after infectious mononucleosis due to primary EBV infection [66]. 5.8. Inflammatory Bowel Disease Latently and productively EBV-infected B cells are present at a higher frequency in the colonic mucosa of patients with ulcerative colitis than controls [67, 68]. Patients with Crohn's disease also have a higher frequency of EBV-infected B cells in the colonic mucosa than controls [67].

organs; an increased risk of developing B-cell lymphoma in the target organs of chronic autoimmune disease; and T-cell and antibody cross-reactivity between EBV antigens and self antigens. These findings can be explained by the hypothesis that chronic autoimmune diseases occur in individuals genetically susceptible to the effects of B-cell infection by EBV, resulting in an increased frequency of latently EBV-infected autoreactive B cells. EBV-infected autoreactive B cells could produce pathogenic autoantibodies; they could also act as professional antigen-presenting cells in the target organ where they could provide a costimulatory survival signal to autoreactive T cells that have been activated in peripheral lymphoid organs by cross-reactivity with infectious agents and that would otherwise undergo activation-induced apoptosis in the target organ. On receiving a costimulatory survival signal from the EBV-infected B cells, the autoreactive T cells could proliferate and produce cytokines, which recruit other inflammatory cells, with resultant target organ damage and chronic autoimmune disease.

5.9. Cryptogenic Fibrosing Alveolitis

REFERENCES

Patients with cryptogenic fibrosing alveolitis have increased serum levels of antibodies against EBV, but not against herpes simplex virus or cytomegalovirus, compared to controls [69]. Furthermore, EBV DNA is detected in lung tissue more frequently in patients with cryptogenic fibrosing alveolitis than in controls [70].

1.

6. CONCLUSION

There is a large body of evidence indicating that EBV infection has a major role in the pathogenesis of organ-specific and non-organ-specific human chronic autoimmune diseases. This evidence includes: a high frequency and high levels of circulating anti-EBV antibodies; triggering of the first attack of autoimmune disease by infectious mononucleosis due to primary EBV infection; an increased frequency of circulating EBV-infected B cells; defective T-cell control of EBV-infected B cells; an increased level of EBV DNA in target tissues; monoclonal B-cell expansion in the target

GebeJA, Swanson E, Kwok WW. HLA class II peptidebinding and autoimmunity. Tissue Antigens 2002;59: 78-87. 2. HendersonRD, Bain CJ, Pender MP. The occurrence of autoimmune diseases in patients with multiple sclerosis and their families. J Clin Neurosci 2000;7:434--437. 3. Bias WB, Reveille JD, Beaty TH, Meyers DA, Arnett FC. Evidence that autoimmunity in man is a Mendelian dominant trait. Am J Hum Genet 1986;39:584-602. 4. McCombe PA, Chalk JB, Pender MP. Familial occurrence of multiple sclerosis with thyroid disease and systemic lupus erythematosus. J Neurol Sci 1990;97:163-171. 5. Notsu K, Note S, Nabeya N, Kuno S, Sakurami T. Antinuclear antibodies in childhood diabetics. Endocrinol Jpn 1983;30:469--473. 6. Morita S, Arima T, Matsuda M. Prevalence of nonthyroid specific autoantibodies in autoimmune thyroid diseases. J Clin Endocrinol Metab 1995;80: 1203-1206. 7. Collard RC, Koehler RP, Mattson DH. Frequency and significance of antinuclear antibodies in multiple sclerosis. Neurology 1997;49:857-861. 8. Folwaczny C, Noehl N, Endres SP, Heldwein W, Loeschke K, Fricke H. Antinuclear autoantibodies

167

9.

10.

11.

12. 13.

14.

15.

16.

17.

18.

19.

20.

21.

in patients with inflammatory bowel disease. High prevalence in first-degree relatives. Dig Dis Sci 1997 ;42:1593-1597. Pender MP. Infection of autoreactive B lymphocytes with EBV, causing chronic autoimmune diseases. Trends Immunol 2003;24:584-588. Garzelli C, Taub FE, Scharff JE, Prabhakar BS, Ginsberg-Fellner E Notkins AL. Epstein-Barr virustransformed lymphocytes produce monoclonal autoantibodies that react with antigens in multiple organs. J Virol 1984;52:722-725. Rickinson AB, Kieff E. Epstein-Barr virus. In: Knipe DM, Howley PM, eds. Field's Virology. Philadelphia: Lippincott Williams & Willdns, 2001 ;2575-2627. Thorley-Lawson DA. Epstein-Barr virus: exploiting the immune system. Nat Rev Immuno12001;1:75-82. Wucherpfennig KW, Strominger JL. Molecular mimicry in T cell-mediated autoimmunity: viral peptides activate human T cell clones specific for myelin basic protein. Cell 1995;80:695-705. Lang HLE, Jacobsen H, Ikemizu Set al. A functional and structural basis for TCR cross-reactivity in multiple sclerosis. Nat Immunol 2002;3:940-943. Lerner MR, Andrews NC, Miller G, Steitz JA. Two small RNAs encoded by Epstein-Barr virus and complexed with protein are precipitated by antibodies from patients with systemic lupus erythematosus. Proc Natl Acad Sci USA 1981;78:805-809. Sabbatini A, Bombardieri S, Migliorini E Autoantibodies from patients with systemic lupus erythematosus bind a shared sequence of StuD and Epstein-Barr virus-encoded nuclear antigen EBNA 1. Eur J Immunol 1993;23:1146-1152. Massa M, Mazzoli F, Pignatti P et al. Proinflammatory responses to self HLA epitopes are triggered by molecular mimicry to Epstein-Barr virus proteins in oligoarticular juvenile idiopathic arthritis. Arthritis Rheum 2002;46:2721-2729. Tabi Z, McCombe PA, Pender ME Apoptotic elimination of V~18.2+ cells from the central nervous system during recovery from experimental autoimmune encephalomyelitis induced by the passive transfer of VI38.2+ encephalitogenic T cells. Eur J Immunol 1994;24:2609-2617. Pender MP. Genetically determined failure of activation-induced apoptosis of autoreactive T cells as a cause of multiple sclerosis. Lancet 1998;351:978-981. Pender MP. Activation-induced apoptosis of autoreactive and alloreactive T lymphocytes in the target organ as a major mechanism of tolerance. Immunol Cell Biol 1999;77:216-223. Sumaya CV, Myers LW, Ellison GW. Epstein-Barr virus antibodies in multiple sclerosis. Arch Neurol 1980;37:

168

94-96. 22. Ascherio A, Munch M. Epstein-Barr virus and multiple sclerosis. Epidemiology 2000; 11:220-224. 23. Wandinger K-P, Jabs W, Siekhaus Aet al. Association between clinical disease activity and Epstein-Barr virus reactivation in MS. Neurology 2000;55:178-184. 24. Martyn CN, Cruddas M, Compston DAS. Symptomatic Epstein-Barr virus infection and multiple sclerosis. J Neurol Neurosurg Psychiatry 1993;56:167-168. 25. Levin LI, Munger KL, Rubertone MV et al. Multiple sclerosis and Epstein-Barr virus. J Am Med Assoc 2003;289:1533-1536. 26. Bray PF, Culp KW, McFarlin DE, Panitch HS, Torkelson RD, Schlight JP. Demyelinating disease after neurologically complicated primary Epstein-Barr virus infection. Neurology 1992;42:278-282. 27. Shaw CM, Alvord EC Jr. Multiple sclerosis beginning in infancy. J Child Neurol 1987;2:252-256. 28. Bray PF, Luka J, Bray PF, Culp KW, Schlight JP. Antibodies against Epstein-Barr nuclear antigen (EBNA) in multiple sclerosis CSF, and two pentapeptide sequence identities between EBNA and myelin basic protein. Neurology 1992;42:1798-1804. 29. Norrby E, Link H, Olsson J-E, Panelius M, Salmi A, Vandvik B. Comparison of antibodies against different viruses in cerebrospinal fluid and serum samples from patients with multiple sclerosis. Infect Immun 1974; 10: 688-694. 30. Rand KH, Houck H, Denslow ND, Heilman I~I. Epstein-Barr virus nuclear antigen-1 (EBNA-1) associated oligoclonal bands in patients with multiple sclerosis. J Neurol Sci 2000;173:32-39. 31. Fraser KB, Haire M, Millar JHD, McCrea S. Increased tendency to spontaneous in-vitro lymphocyte transformation in clinically active multiple sclerosis. Lancet 1979;11:715-717. 32. Martin C, Enbom M, Soderstrom M et al. Absence of seven human herpesviruses, including HHV-6, by polymerase chain reaction in CSF and blood from patients with multiple sclerosis and optic neuritis. Acta Neurol Scand 1997;95:280-283. 33. Qin Y, Duquette P, Zhang Y, Talbot P, Poole R, Antel J. Clonal expansion and somatic hypermutation of VH genes of B cells from cerebrospinal fluid in multiple sclerosis. J Clin Invest 1998;102:1045-1050. 34. Bender GP, Schapiro RT. Primary CNS lymphoma presenting as multiple sclerosis. A case study. Minn Med 1989;72:157-160. 35. Craig JC, Haire M, Merrett JD. T-cell-mediated suppression of Epstein-Barr virus-induced B lymphocyte activation in multiple sclerosis. Clin Immunol Immunopathol 1988;48:253-260. 36. Li F, Linan MJ, Stein MC, Faustman DL. Reduced

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

expression of peptide-loaded HLA class I molecules on multiple sclerosis lymphocytes. Ann Neurol 1995;38: 147-154. Fu Y, Nathan DM, Li F, Li X, Faustman DL. Defective major histocompatibility complex class I expression on lymphoid cells in autoimmunity. J Clin Invest 1993;91: 2301-2307. H611sberg P, Hansen HJ, Haahr S. Altered CD8+ T cell responses to selected Epstein-Barr virus immunodominant epitopes in patients with multiple sclerosis. Clin Exp Immuno12003;132:137-143. Scotet E, Peyrat M-A, Saulquin X et al. Frequent enrichment for CD8 T cells reactive against common herpes viruses in chronic inflammatory lesions: towards a reassessment of the physiopathological significance of T cell clonal expansions found in autoimmune inflammatory processes. Eur J Immunol 1999;29: 973-985. Van Sechel AC, Bajramovic JJ, Van Stipdonk MJB, Persoon-Deen C, Geutskens SB, Van Noort JM. EBV-induced expression and HLA-DR-restricted presentation by human B cells of txB-crystallin, a candidate autoantigen in multiple sclerosis. J Immunol 1999;162:129-135. Van Noort JM, Van Sechel AC, Bajramovic JJ et al. The small heat-shock protein txB-crystallin as candidate autoantigen in multiple sclerosis. Nature 1995;375: 798-801. Evans AS, Rothfield NF, Niederman JC. Raised antibody titres to E.B. virus in systemic lupus erythematosus. Lancet 1971 ;I: 167-168. James JA, Kaufman KM, Farris AD, Taylor-Albert E, Lehman TJ, Harley JB. An increased prevalence of Epstein-Barr virus infection in young patients suggests a possible etiology for systemic lupus erythematosus. J Clin Invest 1997; 100:3019-3026. James JA, Neas BR, Moser KL et al. Systemic lupus erythematosus in adults is associated with previous Epstein-Barr virus exposure. Arthritis Rheum 2001;44: 1122-1126. Verdolini R, Bugatti L, Giangiacomi M, Nicolini M, Filosa G, Cerio R. Systemic lupus erythematosus induced by Epstein-Barr virus infection. Br J Dermatol 2002; 146:877-881. Tsokos GC, Magrath IT, Balow JE. Epstein-Barr virus induces normal B cell responses but defective suppressor T cell responses in patients with systemic lupus erythematosus. J Immunol 1983; 131:1797-1801. Alspaugh MA, Henle G, Lennette ET, Henle W. Elevated levels of antibodies to Epstein-Barr virus antigens in sera and synovial fluids of patients with rheumatoid arthritis. J Clin Invest 1981;67:1134-1140. Tosato G, Steinberg AD, Yarchoan R et al. Abnormally

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

elevated frequency of Epstein-Barr virus-infected B cells in the blood of patients with rheumatoid arthritis. J Clin Invest 1984;73:1789-1795. Balandraud N, Meynard JB, Auger I e t al. EpsteinBarr virus load in the peripheral blood of patients with rheumatoid arthritis: accurate quantification using real-time polymerase chain reaction. Arthritis Rheum 2003 ;48:1223-1228. Tosato G, Steinberg AD, Blaese RM. Defective EBVspecific suppressor T-cell function in rheumatoid arthritis. N Engl J Med 1981;305:1238-1243. McChesney MB, Bankhurst AD. Cytotoxic mechanisms in vitro against Epstein-Barr virus infected lymphoblastoid cell lines in rheumatoid arthritis. Ann Rheum Dis 1986;45:546-552. Niedobitek G, Lisner R, Swoboda B e t al. Lack of evidence for an involvement of Epstein-Barr virus infection of synovial membranes in the pathogenesis of rheumatoid arthritis. Arthritis Rheum 2000;43: 151-154. Origgi L, Hu C, Bertetti E et al. Antibodies to EpsteinBarr virus and cytomegalovirus in primary Sj6gren's syndrome. Boll Ist Sieroter Milan 1988;67:265-274. Yamaoka K, Miyasaka N, Yamamoto K. Possible involvement of Epstein-Barr virus in polyclonal B cell activation in Sj6gren's syndrome. Arthritis Rheum 1988;31:1014-1021. Whittingham S, McNeilage J, Mackay IR. Primary Sj6gren's syndrome after infectious mononucleosis. Ann Intern Med 1985;102:490-493. Fishleder A, Tubbs R, Hesse B, Levine H. Uniform detection of immunoglobulin-gene rearrangement in benign lymphoepithelial lesions. N Engl J Med 1987;316:1118-1121. Biasi D, Caramaschi P, Ambrosetti A et al. Mucosaassociated lymphoid tissue lymphoma of the salivary glands occurring in patients affected by Sj6gren's syndrome: report of 6 cases. Acta Haematol 2001;105: 83-88. Vrbikova J, Janatkova I, Zamrazil V, Tomiska F, Fucikova T. Epstein-Barr virus serology in patients with autoimmune thyroiditis. Exp Clin Endocrinol 1996;104:89-92. Coyle PV, Wyatt D, Connolly JH, O'Brien C. EpsteinBarr virus infection and thyroid dysfunction. Lancet 1989;1:899. Matsubayashi S, Tamai H, Morita T et al. Hashimoto's thyroiditis manifesting monoclonal lymphocytic infiltration. Clin Exp Immunol 1990;79:170-174. Pedersen RK, Pedersen NT. Primary non-Hodgkin's lymphoma of the thyroid gland: a population based study. Histopathology 1996;28:25-32. Shore A, Klock R, Lee P, Snow KM, Keystone EC.

169

63.

64.

65.

66.

170

Impaired late suppression of Epstein-Barr virus (EBV)-induced immunoglobulin synthesis: a common feature of autoimmune disease. J Clin Immunol 1989;9: 103-110. Urano J, Kohno H, Watanabe T. Unusual case of progressive systemic sclerosis with onset in early childhood and following infectious mononucleosis. Eur J Pediatr 1981;136:285-289. Morshed SA, Nishioka M, Saito I, Komiyama K, Moro I. Increased expression of Epstein-Barr virus in primary biliary cirrhosis patients. Gastroenterol Jpn 1992;27: 751-758. James SP, Jones EA, Hoofnagle JH, Strober W. Circulating activated B cells in primary biliary cirrhosis. J Clin Immunol 1985;5:254-260. Vento S, Guella L, Mirandola F et al. Epstein-Barr virus

67.

68.

69.

70.

as a trigger for autoimmune hepatitis in susceptible individuals. Lancet 1995;346:608--609. Spieker T, Herbst H. Distribution and phenotype of Epstein-Barr virus-infected cells in inflammatory bowel disease. Am J Patho12000;157:51-57. Bertalot G, Villanacci V, Gramegna M e t al. Evidence of Epstein-Barr virus infection in ulcerative colitis. Dig Liver Dis 2001;33:551-558. Vergnon JM, Vincent M, de The G, Mornex JF, Weynants P, Brune J. Cryptogenic fibrosing alveolitis and Epstein-Barr virus: an association? Lancet 1984;1I: 768-771. Stewart JP, Egan JJ, Ross AJ et al. The detection of Epstein-Barr virus DNA in lung tissue from patients with idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 1999;159:1336-1341.


HIV and Autoimmunity Gisele Zandman-Goddard ~and Yehuda Shoenfeld ~,2

;Center for Autoimmune Diseases, Department of Medicine 'B', Sheba Medical Center, Tel-Hashomer; Sackler Faculty of Medicine, Tel-Aviv University, Israel; 2Incumbent of the Laura Schwartz-Kipp Research Chair in Autoimmune Diseases, Tel Aviv University, Israel

1. INTRODUCTION The combination of immune dysfunction in patients with HIV infection and AIDS and the development of autoimmune diseases is intriguing. Yet, the spectrum of reported autoimmune phenomena in these patients is increasing [1, 2]. This wide range of reported autoimmune diseases is due to different patient selection, and the association with the development of AIDS. An infectious trigger for immune activation is one of the postulated mechanisms in autoimmunity and derives from molecular mimicry [3, 4]. During frank loss of immunocompetence, autoimmune diseases that are predominantly T cell subtype CD8 driven may predominate. Multiple anti-retroviral drug therapy for patients with AIDS provides prolonged survival and immune restoration, a setting where autoimmune diseases develop. We propose a staging of autoimmune manifestations related to HIV/AIDS manifestations and CD4 count that may be beneficial in identifying the type of autoimmune disease and establishing the proper therapy (Table 1). During Stage I there is the acute HIV infection, and the immune system is intact. In this stage, autoimmune diseases may present. While Stage II is a quiescent period without overt manifestations of AIDS, there is a declining CD4 count indicative of some immunosuppression. Autoimmune diseases are not found. During Stage III further immunosuppression is encountered manifested by a low CD4 count. However, diseases where T cell subtype CD8 predominant such as psoriasis and diffuse immune lymphocytic syndrome (Sjtigren's-

like syndrome) may present or even be the initial manifestation of AIDS. Autoimmune diseases are not found. In Stage IV there is restoration of immune competence following highly active antiretroviral therapy (HAART). In this setting, there may be a resurgence of autoimmune diseases. This review describes the various autoimmune diseases that develop in HIV/AIDS patients through possible mechanisms related to immune activation.

2. AUTOIMMUNE DISEASES IN HIV INFECTION The frequency of rheumatological syndromes in HIV patients varies from less than 1% to 60% [2, 5-7]. The list of reported autoimmune diseases in HIV/AIDS is found in Table 2.

2.1. Systemic Lupus Erythematosus (SLE) The unrestrained state of immune activation may contribute to chronic inflarmnatory and autoimmune sequelae in HIV-infected individuals. Several rheumatic entities, such as Reiter's syndrome, psoriatic arthritis, SjSgren's-like syndrome, myopathy and HIV-related vasculitis are often correlated with the severity of the HIV infection and improve with anti-retroviral therapy. However, other entities, such as SLE and sarcoidosis [8, 9], have a decreased incidence in the HIV infected population than would be expected in the general population. This inconsistency suggests that the immunosuppressive effect of HIV may inhibit the development of autoimmune

171

Table

1. HIV and autoimmunity

Stage

Stagedescription

CD4 count

Viral load

AIDS

Autoimmunity

I II III IV

Clinical latency Cellularresponse Immunedeficiency Immunerestoration

High (> 500) Normal/Low(20(0-499) Low (< 200) High (> 500)

High High High Low

No No Yes Controlled

Autoimmune disease Immune-complex, vasculitis Spondyloarthropathy Autoimmune disease

Autoimmune disease can occur with a preserved immune system requiting B and T cell interactions (normal CD4 count). Therefore, autoimmunity is possible in Stages I, II, and IV. With profound immunodeficiency (low CD4 count) autoimmune diseases are not found. Stage IV (high CD4 count) describes HIV-infected patients with immune restoration but possibly altered immunoregulation enabling the resurgence of autoimmune diseases.

Table 2. Autoimmune diseases in HIV patients Autoimmune disease

References

SLE Antiphospholipid syndrome Autoimmune thrombocytopenia Vasculitis Polymyositis Graves' disease Primary biliary cirrhosis Raynaud' s phenomena, Behqet's disease

[8-14] [17-29] [33] [2, 34-36] [37-39] [40] [41] [42, 43]

diathesis. However, HIV infection that is controlled by protease inhibitors and other anti-retroviral agents renders the immune system no longer immunodeficient. There is immune restoration with normalization of the CD4 count and functional T cell reconstitution [ 10], so that a genetically predisposed host can develop autoimmunity. This has been postulated in the coexistence of HIV with SLE [ 11 ]. Systemic lupus erythematosus (SLE) may be influenced by human immunodeficiency virus type-1 (HIV) infection. It has been suggested that the immunosuppression resulting from HIV infection can prevent the emergence of SLE. There appear to be fewer cases of SLE in the HIV infected population than would be predicted, based on the overall incidence of SLE. One case report described a female patient with systemic lupus erythematosus (SLE) who was infected with HIV; using stored serum, the precise timing of HIV seroconversion was determined and the early effects of HIV infection on SLE examined.

172

The infection resulted in clinical improvement and the disappearance of autoantibody production [ 12]. Another case report reported a patient with HIV infection who developed SLE after the initiation of highly active antiretroviral therapy [ 13]. To date, 29 cases of association between the two diseases have been reported, but the diagnosis was simultaneous in just two of these and only 18 fulfilled the ACR criteria for the diagnosis of SLE. Most patients experienced an improvement in their SLE after development of their HIV associated immunosuppression and a reactivation of lupus manifestations also were noted after immunological recovery secondary to antiretroviral therapy [ 14]. A number of clinical and laboratory features of HIV infection are found in systemic lupus erythematosus (SLE). The presence of circulating antibodies to small nuclear ribonucleoproteins (snRNP) in both diseases was analyzed [15]. Sera from 44 HIV-infected children, from 22 patients with childhood-onset SLE, and from 50 healthy children were studied. Results included the detection of anti-snRNP antibodies by ELISA in 30 HIV-infected patients (68.1%) and 19 SLE patients (86.3%). These antibodies were directed against U1-RNP (61.3% and 77.2%, respectively), Sm (29.5% and 54.5%, respectively), 60 kDa Ro/SSA (47.7% and 50%, respectively), and La/SS-B proteins (18.1% and 9%, respectively). None of the HIV-infected children and 11 SLE patients (50%) showed anti-snRNP antibodies by counter immunoelectrophoresis (CIE). None of the HIV-infected patients showed anti-70 kDa U1-RNP or anti-DSm antibodies by immunoblotting. No differences between the two groups were noted on the presence

of nonprecipitating anti-snRNP antibodies. No such reactivities were observed among the normal sera tested. The authors concluded that non-precipitating anti-snRNP antibodies in HIV-infected children are as frequent as in childhood-onset SLE. The significance of these antibodies is not clear at present. Although polyreactive and low-affinity antibodies and a mechanism of molecular mimicry may explain these results, a specific stimulation of B cells by nuclear antigens could not be excluded. SLE patients produce high titer antibodies to various retroviral proteins, including Gag, Env, and Nef of HIV and HTLV, in the absence of overt retroviral infection. In particular, the role of HTLV1-related endogenous sequence (HRES-1) should be considered in SLE. Molecular mimicry may be a mechanism between HRES-1 and the small ribonucleoprotein complex that initiate the production of autoantibodies, leading to immune complex formation, complement fixation, and pathological tissue deposition [16].

2.2. Antiphospholipid Syndrome (APS)/ Anti-Cardiolipin Antibodies/ Anti-[i2 GPI Antibodies In 1992, the association of aCL antibodies with HIV infection in male homosexuals was reported [17]. Since then, many studies have alluded to this specific combination [18-21]. We described an unusual presentation of APS associated with acute HIV infection. The APS in this patient was characterized by elevated titers of aCL antibodies and anti-132GPI, necrotic lesions in the lower extremities and testicular necrosis requiting orchiectomy. The patient had no history of AIDS, no previous opportunistic infections, and was not on any retroviral medications. The CD4 count was only minimally decreased (CD4-322) indicating that the patient had an acute infection and was not immunosuppressed [ 18]. In another case, a 42 year old woman with a 12 year history of HIV infection developed gangrene of both forefeet. A skin biopsy revealed intracapillary thrombi and severe necrosis of the hypodermis with no evidence of vasculitis. Elevated titers of IgA antibodies were detected [22]. Anti-cardiolipin antibodies and stroke was reported in 2 HIV infected patients [23, 24]. A 33-year-old female with AIDS, a prior small cer-

ebrovascular accident, thrombocytopenia, and a coagulopathy suddenly developed left upper quadrant pain and tendemess due to splenic infarction associated with a high titer of anticardiolipin antibodies. Possible clinical manifestations of anticardiolipin antibodies in this patient include recurrent thromboembolism, coagulopathy, and thrombocytopenia. This case report suggests that anticardiolipin antibodies are associated with splenic infarction and that anticardiolipin antibodies associated with AIDS may sometimes be clinically significant [25]. Another study reported four cases with acute livedo reticularis, avascular necrosis of the femoral head, thrombosis of the inferior vena cava and pulmonary embolus, and a major pulmonary embolus [26]. Avascular necrosis (AVN) in HIV infection associated with aCL was reported in 3 cases. While no other risk factor for thromboembolic event was known, hyperlipidemia (associated with antiretroviral therapy) may have been an additional risk factor for AVN [27, 28]. Other manifestations of APS found in HIV infected patients that are yet to be elucidated are thrombotic microangiopathy (TMA) and pulmonary hypertension [29]. The aCL described in HIV patients are of both the pathogenic one (~I2GPI cofactor dependent) and the infectious type (non-132GPI dependent). It seems that following infections, one may see both types of aCL as well as all isotypes and diversity of aPL including anti-PS [30]. Antiphospholipid antibodies have previously been detected in HIV patients. The presence of lupus anticoagulant (LAC), aCL antibodies, anti-prothrombin antibodies, and anti~12 glycoprotein I antibodies were investigated in 61 HIV patients and 45 patients with APS. LAC was present in 72% of HIV patients and 81% of APS patients. Anticardiolipin antibodies were detected in 67% of the HIV patients and 84% of APS patients. The detection of anti-prothrombin and anti-132 GPI antibodies was significantly less in HIV patients [31]. A recent study demonstrated a high frequency of anti-prothrombin antibodies in a group of 100 HIV infected black patients in South Africa. These variations may be due to the study population due to a different strain of HIV encountered and predominating in infected South African patients [31]. In another study, the phospholipid specificity, avidity, and reactivity with ~2 GPI in 44 patients

173

with HIV infection and compared to the results in 6 SLE patients with secondary APS, 30 SLE patients without APS, and 11 patients with primary APS was investigated [20]. Interestingly, the prevalence of aCL, anti-phosphotidyl serine, anti-phosphotidyl inositol, and anti-phosphotidyl choline (36%, 56%, 34%, and 43% respectively) was similar to that found in the SLE/APS and primary APS patients. The prevalence of these antibodies was significantly higher than that observed in SLE/non-APS patients. Anti-132 GPI antibodies occurred in only 5% of HIV-1 infected patients. A significant decrease of aPL binding after treatment with urea and NaC1 was observed in the sera of HIV- 1 infected patients when compared to APS patients, indicating that aPL antibodies from HIV patients have low resistance to dissociating agents. Anti-132 GPI antibody isotype and IgG subclass in APS patients and a variety of other thrombotic and non-thrombotic disorders including infections was studied [21 ]. Elevated levels of IgM anti-~2 GPI antibodies were observed in 65% of patients with APS and 27% of patients with HIV infection. In another study, the distribution of aCL isotypes and requirement of protein cofactor in viral infections including HIV was investigated. The isotype distribution of anti-cardiolipin antibodies in the sera from 40 patients, with infection caused by HIV-1, was studied by ELISA in the presence and absence of protein cofactor (mainly [32-GPI). The prevalence of one or more aCL antibody isotypes in serum of patients with HIV-1 infection was 47%. Most of these antibodies were mainly cofactor independent [32].

2.3. Autoimmune Thrombocytopenia Immune thrombocytopenic purpura (ITP)occurs in as many as 40% of patients infected with the human immunodeficiency virus (HIV). The evaluation of the effect of highly active antiretroviral therapy (HAART) on platelet counts in 11 homosexual men with HIV-associated ITP patients was sought. At initial evaluation, 7 patients were antiretroviral naive, 2 were taking zidovudine alone, and 2 were receiving combination antiretroviral therapy for known HIV infection. For 6 patients with < 30 • 109 platelets, prednisone was initially co-administered with HAART. The primary outcome measure was

174

the platelet count response to HAART, which was measured weekly until counts had normalized on 3 consecutive occasions, then every 3 months while on HAART. Secondary outcome measures were HIVviral RNA levels and CD4+ cell counts. The results were that one month after the initiation of HAART, 10 patients had an increase in mean platelet count. This statistically significant improvement was sustained at 6 and 12 months' follow-up for 9 of 10 patients. There were no thrombocytopenic relapses at a median follow-up of 30 months. The 9 longterm platelet responders maintained on HAART, at 12 months, had a mean reduction of > 1.5 log 10 in HIV viral RNA serum levels and a marked improvement in CD4+ T-lymphocyte cell count. HAART was effective in improving platelet counts in the setting of HIV-associated ITP, enhanced CD4+ cell counts, and reduced HIV viral loads [33].

2.4. Vasculitis Different types of vasculitis are associated with HIV infection. Co-infections inducing vasculitis have been reported, including hepatitis B and C. Systemic necrotizing vasculitis, leukocytoclastic vasculitis, cryoglobulinemia, and CNS vasculitis have been reported [2, 34, 35]. Panarteritis nodosum more frequently affects the neuromuscular system and skin. Antineutrophilic cytoplasmic antibodies are found less commonly. Vasculitis of the peripheral nerve may cause mononeuritis multiplex or polyneuropathy, sometimes the presenting symptom of HIV infection or after the development of AIDS [35]. HIV antigens and HIV-particles can be identified by electron microscope and positive in situ hybridization studies for HIV have been reported in perivascular cells [2]. Evidence of HIV pathogenicity was described in a 32 year old HIV patient without coronary heart disease risk factors who developed acute coronary vasculitis resulting in a fatal myocardial infarction. Histological analysis of two coronary arteries on autopsy showed a dense infiltration of lymphocytes with necrosis of the intima. In sire hybridization showed sparse intense staining indicating the presence of HIV-1 sequences within the arterial wall [36].

2.5. Polymyositis and Dermatomyositis

in HIV-1 patients without hyperthyroidism.

HIV-associated polymyositis was first described in 1983, and many reports in the past several years confirm this association [2, 37]. Dermatomyositis is also seen in HIV infection [38]. The clinical course, laboratory and electromyography findings are similar to the idiopathic form [2]. Polymyositis in 64 HIV/AIDS patients referred for the presence of elevated creatine kinase (CK) levels or muscle weakness was evaluated. Patients underwent neurologic and rheumatologic evaluation, electromyography, and muscle biopsy after exclusion for recreational drug or alcohol use, metabolic/endocrine disorders, zidovudine therapy, and other infections. Thirteen patients (20%) had biopsy-proven myositis. The median duration of HIV infection prior to diagnosis of myositis was 4.3 years. Six patients had concomitant diffuse infiltrative lymphocytosis syndrome. There was no correlation of severity of weakness, stage of HIV infection, or retroviral treatment with the CK level at diagnosis. Eight patients received prednisone (60 mg/day) with 5 attaining complete resolution of myositis. The remaining 3 patients received immunosuppressive therapy (azathioprine or methotrexate and intravenous immunoglobulin) and had normalization of strength and CK. Four patients had spontaneous resolution of their myositis without treatment. In this study, HIV-associated myositis occurred at any stage of HIV infection, had a relatively good prognosis, and responded well to immunosuppressive therapy [39].

2.7. Primary Biliary Cirrhosis

2.6. Thyroid Disease/Graves' Disease/ Anti-Thyroglobulin Antibodies/ Anti-Thyroid Peroxidase Antibodies The kinetics of CD4 cells, HIV viral load, and autoantibodies in AIDS patients with Graves' disease after immune restoration on (HAART) was investigated [40]. Five patients were diagnosed with Graves' disease after 20 months on HAART, several months after the plasma HIV viral load was undetectable, and when the CD4 count had risen from 14 to 340x 106 cells/L. Anti-thyroid peroxidase (anti-TPO) and anti-TSHR antibodies appeared 14 months after starting HAART and 12 months after the rise in the CD4 count. No other autoantibodies were detected. The autoantibodies were not detected

The role of retroviruses in the development of primary biliary cirrhosis was sought by utilization of immunoblots [41]. Western blot tests were performed for HIV-1 and the human intracisternal A-type particle (HIAP), on serum samples from 77 patients with primary biliary cirrhosis, 126 patients with chronic liver disease, 48 patients with systemic lupus erythematosus, and 25 healthy volunteers. HIV-1 p24 gag seroreactivity was found in 35% patients with primary biliary cirrhosis, 29% patients with systemic lupus erythematosus, 50% of patients with chronic viral hepatitis, and 39% patients with either primary sclerosing cholangitis or biliary atresia, compared with only 4% of 24 patients with alcohol-related liver disease or alphal-antitrypsindeficiency liver disease, and 4% healthy volunteers (p = 0.003). Western blot reactivity to more than two HIAP proteins was found in 51% of patients with primary biliary cirrhosis, in 58% of patients with SLE, in 20% of patients with chronic viral hepatitis, and in 17% of those with other biliary diseases. None of the 23 patients with either alcohol-related liver disease or alphal-antitrypsin deficiency, and only one of the healthy controls showed the same reactivity to HIAP proteins (p 6 months, > 2 symptoms of the Meltzer's triad (purpura, arthralgias, weakness), detection of high rheumatoid factor (RF) activity and/or low C4 levels, in the absence of coexistent diseases (autoimmune, infectious, lymphoproliferative) that may account for the cryoglobulinemia [4]. Recently, Ferri and his colleagues have also proposed a set of classification criteria for MC [5]. None of these criteria though have been validated in a standardized fashion, thus the need for a consensus in formulating classification criteria becomes

189

evident. Despite the absence of accepted criteria for HCVassociated CV, most authors agree that for an accurate diagnosis a number of serologic, pathologic and clinical findings should be present [5-8]: 9 Active chronic HCV infection should be estabfished by the presence of anti HCV antibodies and HCV RNA in the serum by established methods [8]. 9 Circulating mixed cryoglobulins (type II or III, according to the classification proposed by Brouet) [9] measured by an appropriate method must be present in the serum of infected patients (cryoglobulinemia). Since cryoglobulinemia is a common laboratory finding of patients with chronic HCV infection (35-55%), clinical and pathological findings strongly suggestive of vasculitic involvement should be also present. 9 Clinical findings suggestive of vasculitis including purpura, neuropathy (symmetric distal polyneuropathy or mononeuritis multiplex), membranoproliferative glomerulonephritis (MPGN) and skin ulcerations or digital necrosis. Rarely, clinical findings suggestive of gastrointestinal, cardiac or CNS vasculitic involvement may be present. 9 Pathological findings of small-vessel vasculitis affecting the skin (leukocytoclastic purpura), nerves, muscles or other involved organs are extremely helpful in making the correct diagnosis. The documentation of the presence of immune complexes in affected vessels either by immunofluorescence or electron microscopy (kidneys) is an additional important diagnostic tool.

1.2. Epidemiology The frequency of HCV-associated CV has not been investigated in large epidemiological studies. The worldwide prevalence of HCV infection has been estimated to be approximately 3% [10]. Although, as mentioned earlier, the prevalence of cryoglobulinemia is reported consistently in the range between 35 to 55% in HCV infected individuals, the frequency of HCV-associated CV is significantly lower [8]. Although earlier studies, have indicated a high frequency of HCV-associated CV in HCV

190

patients with cryoglobulinemia [11], recent studies including large number of patients have reported a much lower frequency of vasculitis [ 12]. Cacoub et al in a prospective study of 1614 patients with HCV infection found an overall frequency of vasculitis of 1% while in patients with cryoglobulinemia the frequency was 2-3% [12]. Similar low rates of HCV-associated CV (0-11%) have been reported in recent studies [13-15]. Extrapolating from these data, one can assume that the prevalence of HCV-associated CV in the general population should range between 0.010.3% (based on the prevalence of HCV infection in the studied population). This estimate though is probably an overestimation, since most studies examining the frequency of cryoglobulinemia and HCV-associated CV in HCV patients are biased (referral and patient selection bias) [16]. Moreover, there is clearly a shortage of well designed population-based studies on the frequency of this type of vasculitis. In a recent retrospective study of a well-defined population of Northwestern Spain (--250,000 people), Gonzalez-Gay et al found only one case of HCV-associated CV over a 10-year period [ 17]. Geographical variation exists with the disease being more common in Southern Europe compared to Northern Europe and America [6]. The mean age of patients with HCV-associated CV is approximately 50 years (range 40-60 years) while there appears to be a female predominance (female" male ratio = 3:1) [5, 18, 19].

1.3. Clinical Characteristics 1.3.1. Skin manifestations Deposition of cryoglobulins (with or without associated HCV) in skin vessels leads to a localized inflammatory reaction that is manifested clinically by a number of skin findings including purpura, leg ulcers and more rarely digital necrosis [5, 18-21]. The hallmark of HCV-associated CV is the appearance of purpuric lesions in the lower extremities. Purpura has been reported in 65-90% of patients and in most cases represents the presenting manifestation of the disease [5, 18-21]. Typically its appearance follows an intermittent pattern while the lesions tend to be non-pruritic with a lower extrem-

Table 2. Demographic and clinical characteristics of patients with HCV-associated mixed cryoglobulinemia Characteristic Mean age (range, years) Female : male ratio Purpura Weakness (asthenia) Arthralgia Leg ulcers Peripheral neuropathy Renal involvement " Sicca syndrome Raynaud's phenomenon

Table3. Laboratory findings of patients with HCVassociated mixed cryoglobulinemia Characteristic

50 (40-60) 3:1

Rheumatoid factor (RF)

68-75 %

Low C4

50-85%

65-90% 45-90% 40-80% 30-40% 8-55 % 20-35% 6-36% 3-40%

ANA

12-32%

SMA

-23%

AMA

-10%

Anti-thyroid antibodies

- 10

Data from Refs. [5, 18-21].

ity predilection [22]. Purpuric lesions are found more commonly in limbs with venous insufficiency and their disappearance is followed by residual skin hyperpigmentation which can last for prolonged period of time [23]. Biopsies of the purpuric lesions show the typical findings of leukocytoclastic vasculitis with a predominance of mononuclear cells and neutrophils [19, 24]. Immunofluorescent studies reveal the deposition of IgM and C3 in the vessel walls in approximately 80% of the cases [ 19, 25]. Attempts to detect HCV RNA in skin biopsies from patients with HCV-associated CV has given conflicting results, with some studies showing the presence of HCV virions in endothelial cells [24, 26] or vessel walls [26] whereas other studies failed to reproduce these findings [27]. In the study by Agnello et al, HCV RNA was detected in most cases in complexes with IgM and/or IgG antibodies [26]. The appearance of leg ulcers is another common skin manifestation of HCV-associated CV (30-40%, Table 1) [ 19, 20, 22]. The ulcers typically are localized in the lower extremities (above the malleoli) in association with purpuric lesions. Other less common skin manifestations of HCVassociated CV include digital necrosis, nodules and urticarial lesions [24] with variable histopathologic findings [24].

Data from Refs. [5, 7, 19, 21].

1.3.2. Nerve involvement

The predominant form of nerve involvement in HCV-associated CV is peripheral neuropathy that is detected in 8-55% of HCV patients with cryoglobulinemia (Table 2) [5, 18-21]. Peripheral nerve involvement occurs either as a symmetric distal polyneuropathy (--80%) or as mononeuritis multiplex (--10%) [28]. The role of mixed cryoglobulins in the pathogenesis of peripheral neuropathy appears crucial, since the development of neuropathy in non-cryoglobulinemic HCV infected patients is an uncommon event [28, 29]. Patients with distal polyneuropathy present with a painful symmetric neuropathy with predominant sensory findings (paresthesias) [28]. Electromyographic studies reveal an axonal sensory neuropathic process while nerve or muscle biopsies from the affected areas show inflammatory vascular lesions in the majority of cases (-83%) [28]. These inflammatory lesions take the form of vasculitis of the small and/or medium-size vessels or infiltration of vessel wall by mononuclear cells without necrosis [29]. A direct pathogenetic role for HCV has been postulated based on the detection of HCV RNA by sensitive assays in biopsied material [28, 30, 31]. HCV RNA has been found in endothelial cells, infiltrating mononuclear cells or in immune complexes deposited in the arterial wall [28]. Despite its presence though, localized viral replication as evidenced by the detection of its replicative (negative) strand has not been documented so far [28]. Mononeuritis multiplex is another less common

191

neurological manifestation with prominent inflammatory vascular lesions in pathological specimens [28] and IgM deposition by immunofluorescent studies [32]. In a recent study by Authier et al, HCV RNA was not present in muscle or nerve biopsies of three patients with mononeuritis multiplex [28]. CNS involvement has been rarely reported in patients with HCV-associated cryoglobulinemia [19, 33-35]. A number of clinical manifestations have been observed including cerebrovascular accidents, seizures, encephalopathy, dizziness and dementia [19, 33-35]. Interpretation of these limited data is problematic since detailed analysis including angiography and/or brain biopsy has not been performed in each case. 1.3.3. Renal involvement

Kidney involvement is present in 20-35% of HCV patients with mixed cryoglobulinemia (Table 2) [5, 18-21]. The most common form of renal involvement is that of MPGN (55-80%) [36--38]. Less common forms include messangial proliferative glomerulopathy, membranous nephropathy and focal segmental glomerulosclerosis [37]. Patients with MPGN typically present with hypertension (~80%), proteinuria (--55%, usually in the nephrotic range), hypoalbuminemia and mild to moderate renal insufficiency [37, 38]. Typically, renal involvement develops during the evolution of HCV-associated systemic MC [36, 38] with only 15% of the cases displaying a concomitant renal and extrarenal involvement at presentation [39]. HCV-associated MPGN usually follows a fluctuating clinical course with frequent episodes of exacerbation [39]. The true incidence of end-stage renal disease in these patients is unknown. In the largest study in the literature, Tarantino et al reported that approximately 15% of 105 patients developed end-stage renal disease requiting dialysis during a 10 year follow-up [40]. This group of patients though had a high mortality rate (40%) during the same follow-up period, indicating a possible patient selection bias (inclusion of referred patients with more severe renal disease). Although data on the clinical course and prognosis of non-MPGN forms of HCV-associated renal disease are limited [36, 40], no significant differences with MPGN have been observed.

192

Renal biopsies in patients with MPGN reveal the typical histological findings of an immune-complex mediated glomerulonephritis characterized by hypercellular glomeruli (mainly by infiltrating monocytes/macrophages), subendothelial and endocapillary deposits, and IgM/IgG and C3 glomerular deposition [36, 39]. In some cases, characteristic intraluminal thrombi composed of deposited immune complexes are noted [36, 39]. Vasculitis of small and medium size vessels is present in one third of cases [39]. As is the case with peripheral neuropathy in patients with HCV associated CV, HCV RNA has been detected in kidney tissues in a number of studies but its direct pathogenetic role has not been proven [37, 41]. Search for HCV-encoded proteins in kidney biopsies has given inconsistent results so far [37, 42]. 1.3.4. Other clinical findings

A number of other clinical manifestations have been described in patients with HCV-associated MC including arthralgias (16-83%), arthritis (10%), sicca syndrome (6-36%) and Raynaud's phenomenon (3--40%) [5, 18-21]. Since some of these manifestations occur also in cryoglobulin-negative patients with chronic HCV infection, its true association with HCV-associated CV is unknown.

1.4. Laboratory Findings The hallmark of HCV-associated CV is the presence of mixed cryoglobulins in the serum. Cryoglobulins are immunoglobulins with distinct physicochemical characteristics illustrated by their tendency to precipitate at temperatures below 37 ~ (see recent reviews, [5, 20]). Cryoglobulins in chronic hepatitis C are immune complexes composed of IgM with RF activity either monoclonal (type II) or polyclonal (type HI) directed against polyclonal IgG immunoglobulins. Intermediate forms of mixed cryoglobulins composed of oligoclonal IgM-RF have been also observed [5, 20]. Special attention to blood draw and sample handling for the accurate measurement of circulating cryoglobulins has been emphasized [5, 20]. Standardized assays for the qualitative measurements of cryoglobulins would assist in better characterizing HCV patients with

cryoglobulinemia. A number of studies have examined the frequency of type II or HI cryoglobulins in patients with chronic HCV infection [43]. The majority of patients with predominant liver disease without associated extrahepatic diseases, demonstrate more commonly type III cryoglobulins [43]. In contrast, patients with symptomatic HCV-associated CV are more comlnonly positive for type II cryoglobulins [5, 15, 19]. Similarly, in a study by Donada et al patients with chronic HCV infection and type II circulating cryoglobulins, were older and more likely to develop manifestations of HCV-associated CV such as purpura, neuropathy and nephropathy compared to patients with type III cryoglobulins [ 15]. Patients with HCV-associated CV display a number of autoimmune laboratory findings (see Table 2). Among them the detection of elevated titers of RF (68-75%) is the most prevalent [5, 7, 19, 21]. Other laboratory findings include low levels of C4 indicating immune-complex formation and tissue deposition, presence of autoantibodies such as ANA, SMA and more rarely AMA or anti-thyroid antibodies (Table 2). It should be mentioned that patients with chronic hepatitis C demonstrate a similar array of autoantibodies, indicating a chronic polyclonal activation of B lymphocytes in these patients. Patients with HCV-associated CV and cryoglobulinemia in general display much higher titers or percentage of positive RF tests, compared to HCV patients without circulating cryoglobulins [43, 44] (Vassilopoulos D/Calabrese LH, unpublished data). The activity and chronicity of the underlying liver disease in patients with circulating cryoglobulins or HCV-associated CV is a debatable issue. In a recent meta-analysis, Kayali et al reported an overall incidence of cirrhosis of 40% in patients with cryoglobulinemia compared to 17% in patients without cryoglobulins [43]. The difference remained significant even after adjustment for age, gender and disease duration [43, 44].

1.5. Pathogenesis The precise pathogenetic mechanisms that lead to the production of cryoglobulins and furthermore, HCV-associated CV during chronic HCV infection are unknown. A number of epidemiological, clinical

and laboratory observations combined with recent demonstrations of specific gene rearrangements in patients with HCV-associated CV have provided more insight in this complex process. The prevailing theory is that the development of HCV-associated CV is a sequential process (Fig. 1) [7, 45]. Chronic B cell stimulation by HCV or its antigens, leads initially to polyclonal B cell proliferation and production of type HI cryoglobulins characterized by the presence of IgM RF with polyclonal activity. In a certain subset of patients, years after the initial exposure to the virus, a monoclonal or oligoclonal B cell subpopulation arises. These B cells are located preferentially in the liver or bone marrow and produce monoclonal or oligoclonal RF (type II or intermediate typelI/I/I). A number of recent studies have shown that in such patients there is an enrichment in B cells bearing the t(14;18) translocation associated with an overexpression of the Bcl-2 anti-apoptotic protein [46-48]. These long-lived B cells may play a significant role in the production of mono- or oligo-clonal IgM RF that constitute the predominant autoantibody in type II or type IIBII cryoglobulins (Fig. 1). Deposition of monoclonal IgM RF (type II) with or without complexed IgG molecules and HCV RNA in different vascular beds leads to local complement activation and chemoattraction of neutrophils and/or monocytes/macrophages. The firing of this inflammatory cascade is responsible for the various clinical manifestations of HCV-associated CV. Although this theory is based on solid epidemiological, clinical and experimental data, a number of unanswered questions remain. It is unclear why the syndrome of HCV-associated CV is so uncommon despite the frequent presence of circulating cryoglobulins (~50%) in the large HCV infected population worldwide. Additional genetic, immunologic and viral factors have been implicated as additional necessary co-factors but none of them seems to play an exceptional role. Second, there are no well performed prospective studies with long term follow-up that validate these laboratory and experimental findings. In a study by Donada et al, among 102 patients with type III cryoglobulins only 4 developed type II cryoglobulins over a 2 year follow-up period [ 15]. During the same period, none of the cryoglobulin negative patients devel-

193

Figure 1. Pathogenesis of HCV-associated cryoglobulinemic vasculitis (CV). The prevailing theory of cryoglobulin formation and vasculitis during the course of chronic HCV infection is illustrated. Chronic stimulation of B cells by HCV (directly or indirectly) leads to their polyclonal proliferation and production of IgM rheumatoid factor (RF) with polyclonal activity (type III cryoglobulins). During the chronic evolution of the disease, a number of monoclonal or oligoclonal RF arise (mRF) and expand (type II or II/lII cryoglobulins). The role of Bcl-2 translocation may be critical in that direction. Immune complexes (IC) containing mRF, polyclonal IgG _+HCV are deposited in vessel walls (mainly skin, nerves) followed by a localized immune response leading to the development of vasculitis. Similarly, deposition of mRF in certain tissues (mainly glomeruli) leads to in situ formation of IC, chemoattraction of mononuclear cells (MNC) and polymoprhonuclear neutrophils (PMNs) that cause tissue damage (glomerulonephritis). oped cryoglobulins. Furthermore, in a recent study by Persico et al, none of a small number of patients with circulating cryoglobulins developed HCVassociated CV over a 7 year follow-up period [ 14]. Third, a direct role for HCV in this process, beyond its well-accepted participation in the generation of cryoglobulins, has not been proven. Detailed comparison of the clinical and laboratory findings of HCV positive vs. HCV negative patients with symptomatic cryoglobulinemia, shows only minor differences in a recent study by Rieu et al [19]. These findings indicate that deposition of circulating cryoglobulins is the main pathogenetic factor in the development of symptomatic disease, regardless of the presence of HCV. The absence of replicating HCV RNA in the majority of tissue specimens from

194

patients with HCV-associated CV further support these clinical and laboratory findings.

1.6. Therapy The therapy of patients with HCV-associated CV is a challenging task for the involved physicians [8]. The goals of therapy are clear: eradication of the responsible causative agent (HCV) and suppression of the vasculitic inflammatory process [49]. The recent advances in the antiviral treatment of chronic hepatitis C offer new therapeutic options for patients with HCV-associated CV.

1.6.1. Antiviral therapy Interferon-a. Interferon-a (IFN-a) remains the most important agent in the treatment of chronic HCV infection [8]. Bonomo et al used IFN-a for the treatment of HCV-associated CV, even before the discovery of HCV [50]. Following the discovery of HCV and its clear association with MC [51, 52], a number of small randomized and open trials have examined the role of standard IFNa therapy in patients with HCV-associated CV [8]. About 75% of the patients demonstrated partial or complete clinical response at the end of therapy (6-12 months), but approximately 70% of these patients relapsed after treatment discontinuation [8]. Furthermore, the clinical improvement was noted predominantly in skin lesions (purpura) and less so in renal and neurological manifestations [8, 53]. The inability of standard IFN-a to provide a sustained clinical response is directly related to the low sustained virological response rate achieved in these chronically infected HCV patients (-15%) [8].

Combination therapy (interferon a and ribavirin). Multicenter randomized clinical trials at the end of last decade confirmed the superiority of a combination scheme consisting of standard IFN-a and ribavirin over standard IFN-a monotherapy in patients with chronic hepatitis C [54]. As expected though, in these large randomized studies patients with HCV-associated vasculitis were excluded. Data on the efficacy of combination antiviral therapy in patients with HCV-associated CV are derived from single case reports or small uncontrolled studies [37, 55-61]. In the largest study of 27 patients with HCV-associated vasculitis by Cacoub et al, a complete or partial clinical response was seen in 85% of patients treated with the combination of IFN-a (mean duration = 20 months) and ribavirin (mean duration = 14 months) [60]. The major determinant of sustained clinical response to combination therapy was the rate of sustained virological response (only 1 patient with persistent viremia showed clinical remission) [60]. It should be noted that approximately 40% of patients received corticosteroids and plasmapheresis as part of their initial therapeutic scheme [60]. The introduction of pegylated interferons instead of standard IFN-a has increased the efficacy of the

combination scheme in patients with chronic hepatitis C. Currently, the combination of a pegylated IFN-a and ribavirin for 6-12 months is the optimal therapeutic approach for naive patients with chronic hepatitis C [62]. With this regimen 50 to 80% of patients (determined mainly by the HCV genotype) clear the virus [62]. There are currently no short or long-term data on the safety and efficacy of pegylated IFN-a treatment in patients with HCVassociated CV. Antiviral treatment (mono- or combination therapy) in patients with an underlying vasculitis should be administered with great caution and knowledge of its potential for severe side effects. Apart from the known contraindications and side-effects of IFN-a and ribavirin treatment [8], IFN-a has also the potential to exacerbate underlying skin [63], nerve [64] or renal [65] lesions in patients with HCVassociated CV. In a recent study, 22% of patients with HCV-associated vasculitis, discontinued IFN-a treatment due to side effects [60]. Ribavirin is contraindicated in patients with moderate to severe renal dysfunction (creatinine clearance < 50 ml/min) while reduction in its dose is frequently needed due to hemolytic anemia [49]. About 1/3 of patients with HCV-associated vasculitis had to reduce their ribavirin dose due to hemolytic anemia in a recent study by Cacoub et al [60].

1.6.2. Immunosuppressive therapy The goals of immunosuppressive therapy are to suppress the production of the pathogenic cryoglobulins by B cells and to downregulate the host immune response that is responsible for the localized vascular inflammatory process [8]. Prior to the discovery of HCV, patients with severe "essential" MC were frequently treated with a combination of corticosteroids, cyclophosphamide and plasmapheresis in an uncontrolled fashion with mixed results [23]. The administration of immunosuppressive therapy in patients with a chronic viral infection raises reasonable concerns about their short and long-term side effects [8]. Short term corticosteroid use in patients with chronic hepatitis C is associated with transient increase in HCV RNA levels but acute deterioration of liver function during or after therapy is rare [8, 60]. Similarly, short courses of cyclophosphamide therapy have not

195

been linked to acute liver failure in a large study of Italian patients with HCV-associated MPGN [39]. On the other hand, there is accumulating evidence that long-term immunosuppressive therapy leads to accelerated rates of cirrhosis in HCV infected patients [8]. Collectively, these limited data suggest that short term immunosuppressive therapy is not associated with acute liver toxicity in patients with HCV-associated vasculitis. Long-term continuous immunosuppressive therapy though may enhance chronic liver damage leading to cirrhosis.

1.6.3. Apheresis Plasmapheresis is used temporally in patients with severe/life-threatening manifestations of HCV-associated CV including rapidly deteriorating glomerulonephritis, skin necrosis, CNS or motor neuropathy and hyperviscosity syndrome [6, 39]. It is usually administered in combination with corticosteroids and/or cyclophosphamide. Controlled data on its efficacy are not available but anecdotal evidence supports its use in patients with life-threatening disease.

1.6.4. New immunosuppressive agents Biologic agents that specifically target elements of the immune system that participate in the pathogenesis of HCV-associated CV (see Fig. 1) are currently under investigation for the treatment of this disorder. So far, published data are available only for rituximab [66, 67]. Rituximab is a chimeric monoclonal antibody (anti-CD20) that specifically targets and depletes B cells from the circulation. This agent is already being used in clinical practice for the treatment of B-cell Non-Hodgkin's lymphomas and autoimmune cytopenias (autoimmune hemolytic anemia and idiopathic thrombocytopenic purpura) [68, 69]. Recently, two studies reporting on the efficacy of rituximab in Italian patients with MC were published [66, 67]. In both studies, patients with treatment resistant HCV-associated MC (12 and 20 patients respectively) were treated with 4 weekly intravenous infusions of standard dose Rituximab (375 mg/m 2) [66, 67]. The majority of patients had mild to moderate disease activity at baseline

196

(purpura = 75-80%, neuropathy = 33--60%, renal disease = 5-17%). This initial course of therapy was associated with significant clinical response mainly in the skin (-75%) and nerve (50-100%) manifestations [66, 67]. The medication was well tolerated without significant side effects. Relapses were noted in 25% of patients in the study by Sansonno et al [66] while 1/3 of patients in the study by Zaja et al [67] had to be retreated. There were no significant changes in ALT levels during therapy in both studies but increases in HCV RNA levels were noted in one study (mainly in clinically responding patients) [66]. More studies with particular emphasis on the long term safety of this promising agent are needed.

1.6.5. Treatment guidelines for HCV-associated CV Management of HCVoassociated CV should be designed on an individual basis. Careful initial assessment of the severity of vasculitic involvement, the status of the underlying liver disease and the virological characteristics of each patient are necessary prior to the initial decision making. In nah've patients with mild to moderate disease activity (purpura, arthralgias/arthritis, mild sensory neuropathy, mild proteinuria/hematuria with normal creatinine values), combination therapy with IFN-a and ribavirin should be offered in addition to symptomatic therapy [5, 7, 8, 53]. Given the latest data on the enhanced efficacy of pegylated interferons, these agents should be tried first in combination with ribavirin. Regular follow-up of the patients with sensitive measurements of HCV RNA levels is required as well as increased vigilance for therapy related side effects. The optimal duration of treatment should be based on the viral genotype and the clinical/virological response to therapy. In patients with severe or life-threatening disease (rapidly progressive glomerulonephritis, motor neuropathy, CNS, gastrointestinal or myocardial involvement, digital necrosis), combination therapy with immunosuppressive and antiviral therapy should be tried. Immunosuppressive therapy should be given for a short period of time (2-4 weeks) followed by antiviral treatment. Corticosteroids and/or cyclophosphamide (IV or per os) therapy are the agents most commonly used in this setting. In

patients with life-threatening disease, plasmapheresis can be also offered in combination with the immunosuppressive therapies. In resistant cases, inclusion of patients in carefully designed study protocols in referral centers is strongly recommended. In patients with relapsing disease that show clinical response to a second course of antiviral treatment, long term antiviral treatment (> 1 year) may be necessary (preferably with pegylated interferons).

3. H C V - A S S O C I A T E D L A R G E V E S S E L VASCULITIS Symptoms suggestive of giant cell arteritis in HCV infected patients have been noted in two settings. Rarely, in patients with HCV-associated CV involvement of the small vessels around the temporal artery occurs [77, 78]. This small vessel vasculitis manifests as typical giant cell arteritis. Direct involvement of the temporal artery in a patient with classical symptoms of giant cell arteritis and chronic HCV infection has been documented by Ferracioli et al [79].

2. HCV-ASSOCIATED MEDIUM S I Z E VASCULITIS ACKNOWLEDGEMENTS Medium-size vasculitis is uncommon in patients with chronic hepatitis C [12]. In most cases, cryoglobulins are also present in the circulation, so a clear distinction from HCV-associated CV with medium size vessel involvement can not be made [70]. Cacoub et al reported that patients with such characteristics present with more severe systemic disease manifestations that resemble polyarteritis nodosa (PAN) [70]. Biopsies of involved tissues demonstrate a necrotizing medium size vasculitis with mononuclear and polymorphonuclear cell infiltration. In a subset of patients, characteristic intrarenal microaneurysms were noted by angiographic studies [70, 71 ]. In unselected populations of patients with features of classic PAN, the frequency of HCV infection is < 10% [72-74]. Thus, a clear causative association between chronic HCV infection and PAN can not be made. A small number of patients with predominant skin manifestations (nodules, livedo reticularis) in the absence of systemic disease and histological findings suggestive of cutaneous PAN have been also described [75, 76]. The therapy of patients with systemic PAN-like disease does not differ from the suggested therapy for severe HCV-associated CV.

I would like to thank Dr Calabrese for his continuous support and critical review of the manuscript.

REFERENCES 1.

2.

3.

.

5. 6.

7. 8.

Vassilopoulos D, Calabrese LH. Virus-associated vasculitides: Clinical aspects. In: Hoffman GS, Weyand C, eds. Inflammatory Diseases of the Blood Vessels. New York: Marcell Dekker, 2001 ;565-597. Jennette JC, Falk RJ, Andrassy K, Bacon PA, Churg J, Gross WL et al. Nomenclature of systemic vasculitides. Proposal of an international consensus conference. Arthritis Rheum 1994;37:187-192. Pioltelli P, Maldifassi P, Vacca A, Mazzaro C, Mussini C, Migliaresi S et al. GISC protocol experience in the treatment of essential mixed cryoglobulinaemia. Clin Exp Rheumatol 1995;13 Suppl 13:S187-S190. Lamprecht P, Moosig F, Gause A, Herlyn K, Csemok E, Hansen H et al. Immunological and clinical follow up of hepatitis C virus associated cryoglobulinaemic vasculitis. Ann Rheum Dis 2001 ;60:385-390. Ferri C, Zignego AL, Piled SA. Cryoglobulins. J Clin Patho12002;55:4-13. Della RA, Tavoni A, Bombardieri S. Cryoglobulinemia. In: Hochberg MC, Silman AJ, Smolen JS, Weinblatt ME, Weisman MH, eds. Rheumatology. London: Mosby, 2003;1697-1703. Lamprecht P, Gause A, Gross WL. Cryoglobulinemic vasculitis. Arthritis Rheum 1999;42:2507-2516. VassilopoulosD, Calabrese LH. Hepatitis C virus infection and vasculitis. Implications of antiviral and immunosuppressive therapies. Arthritis Rheum 2002;46: 585-597.

197

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

198

Brouet JC, Clauvel JP, Danon F, Klein M, Seligmann M. Biologic and clinical significance of cryoglobulins. A report of 86 cases. Am J Med 1974;57:775-788. Alter MJ, Hutin YJ, Armstrong GL. Epidemiology of hepatitis C. In: Liang TJ, Hoofnagle JH, eds. Hepatitis C. San Diego: Academic, 2000; 169-183. Lunel F, Musset L, Cacoub P, Frangeul L, Cresta P, Perrin M et al. Cryoglobulinemia in chronic liver diseases: role of hepatitis C virus and liver damage [published erratum appears in Gastroenterology 1995 Feb;108(2):620]. Gastroenterology 1994;106:12911300. Cacoub P, Poynard T, Ghillani P, Charlotte F, Olivi M, Piette JC et al. Extrahepatic manifestations of chronic hepatitis C. MULTIVIRC Group. Multidepartment Virus C. Arthritis Rheum 1999;42:2204-2212. Cicardi M, Cesana B, Del Ninno E, Pappalardo E, Silini E, Agostoni A et al. Prevalence and risk factors for the presence of serum cryoglobulins in patients with chronic hepatitis C. J Viral Hepat 2000;7:138-143. Persico M, De Marino FA, Di Giacomo RG, Persico E, Morante A, Palmentieri B e t al. Prevalence and incidence of cryoglobulins in hepatitis C virus-related chronic hepatitis patients: a prospective study. Am J Gastroentero12003;98:884-888. Donada C, Crucitti A, Donadon V, Tommasi L, Zanette G, Crovatto M et al. Systemic manifestations and liver disease in patients with chronic hepatitis C and type II or III mixed cryoglobulinaemia. J Viral Hepat 1998;5: 179-185. Agnello V. Mixed cryoglobulinaemia after hepatitis C virus: more and less ambiguity. Ann Rheum Dis 1998;57:701-702. Gonzalez-Gay MA, Garcia-Porrua C. Systemic vasculitis in adults in northwestern Spain, 1988-1997. Clinical and epidemiologic aspects. Medicine (Baltimore) 1999;78:292-308. Agnello V. The etiology and pathophysiology of mixed cryoglobulinemia secondary to hepatitis C virus infection. Springer Semin Immunopathol 1997;19:111-129. Rieu V, Cohen P, Andre MH, Mouthon L, Godmer P, Jarrousse B et al. Characteristics and outcome of 49 patients with symptomatic cryoglobulinaemia. Rheumatology (Oxf) 2002;41:290-300. Dammacco F, Sansonno D, Piccoli C, Tucci FA, Racanelli V. The cryoglobulins: an overview. Eur J Clin Invest 2001 ;31:628--638. Trejo O, Ramos-Casals M, Garcia-Carrasco M, Yague J, Jimenez S, de la RG et al. Cryoglobulinemia: study of etiologic factors and clinical and immunologic features in 443 patients from a single center. Medicine (Baltimore) 2001;80:252-262. Gorevic PD, Kassab HJ, Levo Y, Kohn R, Meltzer M,

23. 24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

Prose P et al. Mixed cryoglobulinemia: clinical aspects and long-term follow-up of 40 patients. Am J Med 1980;69:287-308. Dispenzieri A, Gorevic PD. Cryoglobulinemia. Hematol Oncol Clin North Am 1999;13:1315-1349. Crowson AN, Nuovo G, Ferri C, Magro CM. The dermatopathologic manifestations of hepatitis C infection: a clinical, histological, and molecular assessment of 35 cases. Hum Patho12003;34:573-579. Daoud MS, el Azhary RA, Gibson LE, Lutz ME, Daoud S. Chronic hepatitis C, cryoglobulinemia, and cutaneous necrotizing vasculitis. Clinical, pathologic, and immunopathologic study of twelve patients. J Am Acad Dermatol 1996;34:219-223. Agnello V, Abel G. Localization of hepatitis C virus in cutaneous vasculitic lesions in patients with type II cryoglobulinemia. Arthritis Rheum 1997;40:20072015. Mangia A, Andriulli A, Zenarola P, Lomuto M, Cascavilla I, Quadri R et al. Lack of hepatitis C virus replication intermediate RNA in diseased skin tissue of chronic hepatitis C patients. J Med Virol 1999;59: 277-280. Authier FJ, Bassez G, Payan C, Guillevin L, Pawlotsky JM, Degos JD et al. Detection of genomic viral RNA in nerve and muscle of patients with HCV neuropathy. Neurology 2003;60:808-812. Lidove O, Cacoub P, Maisonobe T, Servan J, Thibault V, Piette JC et al. Hepatitis C virus infection with peripheral neuropathy is not always associated with cryoglobulinaemia. Ann Rheum Dis 2001 ;60:290-292. De Martino L, Sampaolo S, Tucci C, Ambrosone L, Budillon A, Migliaresi Set al. Viral RNA in nerve tissues of patients with hepatitis C infection and peripheral neuropathy. Muscle Nerve 2003;27:102-104. Bonetti B, Scardoni M, Monaco S, Rizzuto N, Scarpa A. Hepatitis C virus infection of peripheral nerves in type II cryoglobulinaemia. Virchows Arch 1999;434: 533-535. David WS, Peine C, Schlesinger P, Smith SA. Nonsystemic vasculitic mononeuropathy multiplex, cryoglobulinemia, and hepatitis C. Muscle Nerve 1996;19: 1596-1602. Origgi L, Vanoli M, Carbone A, Grasso M, Scorza R. Central nervous system involvement in patients with HCV-related cryoglobulinemia. Am J Med Sci 1998 ;315:208-210. Petty GW, Duffy J, Houston J, HI. Cerebral ischemia in patients with hepatitis C virus infection and mixed cryoglobulinemia. Mayo Clin Proc 1996;71:671--678. Tembl JI, Ferrer JM, Sevilla MT, Lago A, Mayordomo F, Vilchez JJ. Neurologic complications associated with hepatitis C virus infection. Neurology 1999;53:

861-864. 36. Beddhu S, Bastacky S, Johnson JE The clinical and morphologic spectrum of renal cryoglobulinemia. Medicine (Baltimore) 2002;81:398-409. 37. Sabry AA, Sobh MA, Irving WL, Grabowska A, Wagner BE, Fox S et al. A comprehensive study of the association between hepatitis C virus and glomerulopathy. Nephrol Dial Transplant 2002;17:239-245. 38. Tarantino A, Moroni G, Banff G, Manzoni C, Segagni S, Ponticelli C. Renal replacement therapy in cryoglobulinaemic nephritis. Nephrol Dial Transplant 1994;9: 1426-1430. 39. D'Amico G. Renal involvement in hepatitis C infection: cryoglobulinemic glomerulonephritis. Kidney Int 1998;54:650-671. 40. Tarantino A, Campise M, Banff G, Confalonieri R, Bucci, Montoli A et al. Long-term predictors of survival in essential mixed cryoglobulinemic glomerulonephritis. Kidney Int 1995;47:618-623. 41. Rodriguez-Inigo E, Casqueiro M, Bartolome J, Barat A, Caramelo C, Ortiz A et al. Hepatitis C virus RNA in kidney biopsies from infected patients with renal diseases. J Viral Hepat 2000;7:23-29. 42. Sansonno D, Gesualdo L, Manno C, Schena FP, Dammacco E Hepatitis C virus-related proteins in kidney tissue from hepatitis C virus-infected patients with cryoglobulinemic membranoproliferative glomerulonephritis. Hepatology 1997;25:1237-1244. 43. Kayali Z, Buckwold VE, Zimmerman B, Schmidt WN. Hepatitis C, cryoglobulinemia, and cirrhosis: a metaanalysis. Hepatology 2002;36:978-985. 44. Pawlotsky JM, Ben Yahia M, Andre C, Voisin MC, Intrator L, Roudot-Thoraval F et al. Immunological disorders in C virus chronic active hepatitis: a prospective case-control study. Hepatology 1994; 19:841-848. 45. Agnello V. Hepatitis C virus infection and type II cryoglobulinemia: an immunological perspective. Hepatology 1997;26:1375-1379. 46. Zignego AL, Ferri C, Giannelli E Giannini C, Caini P, Monti M e t al. Prevalence of bcl-2 rearrangement in patients with hepatitis C virus-related mixed cryoglobulinemia with or without B-cell lymphomas. Ann Intern Med 2002;137:571-580. 47. Zuckerman E, Zuckerman T, Sahar D, Streichman S, Attias D, Sabo E et al. bcl-2 and immunoglobulin gene rearrangement in patients with hepatitis C virus infection. Br J Haemato12001 ;112:364-369. 48. Kitay-Cohen Y, Amiel A, Hilzenrat N, Buskila D, Ashur Y, Fejgin Met al. Bcl-2 rearrangement in patients with chronic hepatitis C associated with essential mixed cryoglobulinemia type II. Blood 2000;96:2910-2912. 49. Vassilopoulos D, Calabrese LH. Rheumatic manifestations of hepatitis C infection. Curr Rheumatol Rep

2003;5:200-204. 50. Bonomo L, Casato M, Afeltra A, Caccavo D. Treatment of idiopathic mixed cryoglobulinemia with alpha interferon. Am J Med 1987;83:726-730. 51. Pascual M, Perrin L, Giostra E, Schifferli JA. Hepatitis C virus in patients with cryoglobulinemia type II [letter]. J Infect Dis 1990;162:569-570. 52. Ferri C, Greco F, Longombardo G, Palla P, Moretti A, Marzo E et al. Association between hepatitis C virus and mixed cryoglobulinemia. Clin Exp Rheumatol 1991;9:621-624. 53. Cacoub P, Costedoat-Chalumeau N, Lidove O, Alric L. Cryoglobulinemia vasculitis. Curr Opin Rheumatol 2002; 14:29-35. 54. McHutchison JG, Hoofnagle JH. Therapy of chronic hepatitis C. In: Liang TJ, Hoofnagle JH, eds. Hepatitis C. San Diego: Academic, 2000;203-239. 55. Donada C, Crucitti A, Donadon V, Chemello L, Alberfi A. Interferon and ribavirin combination therapy in patients with chronic hepatitis C and mixed cryoglobulinemia [letter]. Blood 1998;92:2983-2984. 56. Calleja JL, Albillos A, Moreno-Otero R, Rossi I, Cacho G, Domper F et al. Sustained response to interferonalpha or to interferon-alpha plus ribavirin in hepatitis C virus-associated symptomatic mixed cryoglobulinaemia. Aliment Pharmacol Ther 1999;13:1179-1186. 57. Misiani R, Bellavita P, Baio P, Caldara R, Ferruzzi S, Rossi P et al. Successful treatment of HCV-associated cryoglobulinaemic glomerulonephritis with a combination of interferon-alpha and ribavirin. Nephrol Dial Transplant 1999;14:1558-1560. 58. Zuckerman E, Keren D, Slobodin G, Rosner I, Rozenbaum M, Toubi E et al. Treatment of refractory, symptomatic, hepatitis C virus related mixed cryoglobulinemia with ribavirin and interferon-alpha. J Rheumato12000;27:2172-2178. 59. Garini G, Allegri L, Carnevali L, CateUani W, Manganelli P, Buzio C. Interferon-alpha in combination with ribavirin as initial treatment for hepatitis C virusassociated cryoglobulinemic membranoproliferative glomerulonephritis. Am J Kidney Dis 2001 ;38:E35. 60. Cacoub P, Lidove O, Maisonobe T, Duhaut P, Thibault V, Ghillani Pet al. Interferon-alpha and ribavirin treatment in patients with hepatitis C virus-related systemic vasculitis. Arthritis Rheum 2002;46:3317-3326. 61. Rossi P, Bertani T, Baio P, Caldara R, Luliri P, Tengattini F et al. Hepatitis C virus-related cryoglobulinemic glomerulonephritis: long-term remission after antiviral therapy. Kidney Int 2003;63:2236-2241. 62. National Institutes of Health Consensus Development Conference Statement: Management of hepatitis C: 2002; June 10-12, 2002. Hepatology 2002;36:$3-20. 63. Cid MC, Hernandez-Rodriguez J, Robert J, del Rio A,

199

64.

65.

66.

67.

68.

69.

70.

71.

200

Casademont J, Coll-Vinent B et al. Interferon-alpha may exacerbate cryoblobulinemia-related ischemic manifestations: an adverse effect potentially related to its anti-angiogenic activity. Arthritis Rheum 1999;42: 1051-1055. Boonyapisit K, Katirji B. Severe exacerbation of hepatitis C-associated vasculitic neuropathy following treatment with interferon alpha: a case report and literature review. Muscle Nerve 2002;25:909-913. Ohta S, Yokoyama H, Wada T, Sakai N, Shimizu M, Kato T et al. Exacerbation of glomerulonephritis in subjects with chronic hepatitis C virus infection after interferon therapy. Am J Kidney Dis 1999;33:1040-1048. Sansonno D, De R, V, Lauletta G, Tucci FA, Boiocchi M, Dammacco E Monoclonal antibody treatment of mixed cryoglobulinemia resistant to interferon alpha with an anti-CD20. Blood 2003;101:3818-3826. Zaja F, De Vita S, Mazzaro C, Sacco S, Damiani D, De Marchi G et al. Efficacy and safety of rituximab in type II mixed cryoglobulinemi. Blood 2003;101: 3827-3834. Silverman GJ, Weisman S. Rituximab therapy and autoimmune disorders: prospects for anti-B cell therapy. Arthritis Rheum 2003;48:1484-1492. Boye J, Elter T, Engert A. An overview of the current clinical use of the anti-CD20 monoclonal antibody rituximab. Ann Onco12003;14:520-535. Cacoub P, Maisonobe T, Thibault V, Gatel A, Servan J, Musset L et al. Systemic vasculitis in patients with hepatitis C. J Rheumato12001;28:109-118. Costedoat-Chalumeau N, Cacoub P, Maisonobe T, Thibault V, Cluzel P, Gatfosse M e t al. Renal microaneurysms in three cases of hepatitis C virus-related

vasculitis. Rheumatology (Oxf) 2002;41:708-710. 72. Cacoub P, Lunel-Fabiani F, Du LT. Polyarteritis nodosa and hepatitis C virus infection [letter]. Ann Intern Med 1992; 116:605-606. 73. Carson CW, Conn DL, Czaja AJ, Wright TL, Brecher ME. Frequency and significance of antibodies to hepatitis C virus in polyarteritis nodosa. J Rheumatol 1993;20:304-309. 74. Quint L, Deny P, Guillevin L, Granger B, Jarrousse B, Lhote F et al. Hepatitis C virus in patients with polyarteritis nodosa. Prevalence in 38 patients. Clin Exp Rheumatol 1991 ;9:253-257. 75. Vitali C, Galluzzo E, Ciancia EM, Moretti A, Marchi S. Giant cell arteritis of the leg in a patient with hepatitis C virus infection. Ann Rheum Dis 1997;56:697-698. 76. Soufir N, Descamps V, Crickx B, Thibault V, Cosnes A, Becherel PA et al. Hepatitis C virus infection in cutaneous polyarteritis nodosa: a retrospective study of 16 cases. Arch Dermatol 1999; 135:1001-1002. 77. Disdier P, Pellissier JF, Harle JR, Figarella-Branger D, Bolla G, Weiller PJ. Significance of isolated vasculitis of the vasa vasorum on temporal artery biopsy. J Rheumatol 1994;21:258-260. 78. Genereau T, Martin A, Lortholary O, Noel V, Guillevin L. Temporal arteritis symptoms in a patient with hepatitis C virus associated type II cryoglobulinemia and small vessel vasculitis. J Rheumatol 1998;25:183-185. 79. Ferraccioli GF, Mariuzzi L, Damato R, Rocco M, Pirisi M, Beltrami CA. Jaw and leg claudication in a patient with temporal arteritis, chronic sialoadenitis and previous hepatitis C virus infection. Clin Exp Rheumatol 1998;16:463-468.


HCV and Cryoglobulinemia Clodoveo Ferri I and Stefano Bombardieri 2

1Rheumatology Unit, Department of lnternal Medicine, University of Modena, Medical School Modena, Italy; 2Rheumatology Unit, Department of Internal Medicine, University of Pisa, Medical School Pisa, Italy

1. C R Y O G L O B U L I N E M I A

Cryoglobulinemia is defined as the presence of circulating immunoglobulins (Ig) that precipitate at temperatures < 37 ~ C and redissolve on re-worming [1]. Such an in vitro phenomenon is detectable in a wide number of chronic infectious and immunological disorders, as well as in some hematological malignancies [ 1-3]. The real mechanism(s) of cryoprecipitation that remains largely unknown; it could be secondary to intrinsic characteristics of both mono- and polyclonal immunoglobulin components and/or to the interaction among single components of the cryoprecipitate [ 1-5]. Cryoglobulinemia is usually classified into three subgroups according to Brouet et al [4]: type I, composed by single monoclonal Ig; type II and HI, which contained a mixture of polyclonal IgG and mono- (type II) or polyclonal (type III) IgM rheumatoid factor (RF). Cryoglobulinemia type I or monoclonal cryoimmunoglobulinemia is frequently associated to a well known hematological disorders; in particular, lymphoid tumors such as Waldenstrom's macroglobulinaemia, multiple myeloma, and immunocytoma. Monoclonal cryoimmunoglobulinemia is generally asymptomatic, in only few cases it can be complicated by hyperviscosity syndrome. Type II and III mixed cryoglobulinemia (MC) are often responsible for a clinical syndrome characterized by leucocytoclastic vasculitis of small and medium sized vessels, and multiple organ involvement [1-5]. MC is classified as essential or secondary in the absence/presence of other well defined- infectious, immunological or neoplastic - d i s e a s e s [2-5]. Table 1 shows the main biologi-

cal and clinico-pathological characteristics of cryoglobulinemia subgroups, including a newly proposed serological variant the type II-III MC [4, 6]. The analysis of cryoprecipitates is generally carded out by means of immunoelectrophoresis or immunofixation. With more sensitive methodologies, i.e. immunoblotting or 2-dimensional polyacrylamide gel electrophoresis, type II MC frequently shows a microheterogeneous composition; in particular, oligoclonal IgM or a mixture of polyclonal and monoclonal IgM can be detected [6]. Type II-III MC could fit together the most recent molecular studies showing the presence of oligoclonal B-lymphocyte proliferation's in liver and bone marrow biopsies from MC patients [7, 8]. In two third of type II MC a cross-idiotype WA monoclonal RF has been demonstrated [3, 9]. This WA (after the patient in whom it was first detected) autoantibody almost invariably express a Vk light chain derived from a single germinal gene, the human KV 325. The same WA monoclonal IgMk RF has also been detected in type II MC secondary to lymphoid malignancies, probably expression of an antigen-independent clonal B-cell lymphoproliferation.

2. M I X E D C R Y O G L O B U L I N E M I A

The so-called 'essential' MC was first described in 1966; this term refers to a distinct clinical syndrome in the absence of other well known systemic or neoplastic disorders [5]. Clinically, MC is characterized by a typical t r i a d - purpura, weakness, arthralgias - and by multisystem organ involvement including chronic hepatitis, membranoproliferative glomeru-

201

Table 1. Classification and clinico-pathological characteristics of different cryoglobulinemias Composition

Pathological findings

Clinical associations

Type I cryoglobulinaemia

monoclonal Ig, mainly IgG, or IgM, or IgA self-aggregation through Fc fragment of Ig

tissue histological alterations of underlying disorder

lymphoproliferative dis.: MM, WM, CLL, B-cell NHL

Type II mixed cryogl.

monoclonal IgM (or IgG, or IgA) with RF activity (often crossidiotype WA-mRF) and polyclonal Ig (mainly IgG)

leukocytoclastic vasculitis B-lymphocyte expansion with tissue infiltrates

infections (mainly HCV) autoimmune/lymphoproliferative dis. rarely 'essential'

Type H-Ill mixed cryogl.

oligoclonal IgM RF or mixture of poly/monoclonal IgM (often crossidiotype WA-mRF)


infections (mainly HCV) autoimmune/lymphoproliferative dis. rarely 'essential'

Type III mixed cryogl.

polyclonal mixed Ig (all isotypes) with RF activity of one polyclonal component (usually IgM)


infections (mainly HCV) more often autoimmune disorders rarely 'essential'

Lymphoproliferative disorders: MM (multiple myeloma), WM (Waldenstrom's macroglobulinemia), CLL (chronic lymphocytic leukemia), B-cell non-Hodgkin's lymphoma; Ig: immunoglobulin; RF: rheumatoid factor; HCV: hepatitis C virus.

lonephritis (MPGN), peripheral neuropathy, skin ulcers, widespread vasculitis, and less frequently lymphatic and hepatic malignancies [2-5, 10] (Table 2). A variety of circulating immune-complexes, mainly mixed cryoglobulins with RF activity, a low hemolytic complement activity, and markedly low C4 are the typical serological findings of MC [2-5, 10]. MC is considered to be a relatively rare disorder; however, its prevalence among different countries shows a great geographical etherogeneity, being the disease more frequent in Southern Europe than in Northern Europe or Northern America. Because its clinical polymorphism, MC patients are often referred to different specialties according to the main symptom(s), i.e. skin vasculitis, hepatitis, nephritis, peripheral neuropathy, etc. Consequently, the actual prevalence of the MC is probably underestimated; moreover, in some patients a correct diagnosis can be delayed or overlooked entirely. There are not available classification/diagnostic criteria for MC. In the clinical practice, the main diagnostic parameters are serum mixed cryoglobulins with RF activity, frequently associated to low C4, orthostatic skin purpura due to leukocytoclastic vasculitis of small/medium-sized blood vessels [25, 10]. This latter is an immune-complex-mediated vasculitis, with the possible contribution of hemorheological and/or local factors [2, 10]. A polyclo-

202

Table 2. Clinico-epidemiological and laboratory features of 200 MC patients .

.

.

.

.

.

.

.

Age, mean +_SD years (range)* Female/male ratio Disease duration, mean _+SD years (range)

52+_12 (29-75) 2.6 12_+6(1-35)

Purpura Weakness Arthralgias Arthritis (non-erosive) Raynaud's phenomenon Sicca syndrome Peripheral neuropathy Renal involvement Liver involvement B-cell non-Hodgkin' s lymphoma Hepatocellular carcinoma

92% 90% 81% 10% 35% 36% 36% 29% 68% 7.5% 2.5%

Cryocrit, mean __SD % 4_+8 Type IFtype Ill mixed cryoglobulins 2/1 C3, mean _+SD mg/dl (normal 60-130) 72_+30 C4, mean _+SD mg/dl (normal 20-55) 9_+15 Autoantibodies 34% Anti-HCV antibodies HCV RNA Anti-HBV antibodies HBsAg

92% 85% 38% 3%

nal or mono-oligoclonal B-lymphocyte expansion represents the underlying pathological alteration detectable in the large majority of MC patients in the serum and/or in different tissues as lymphoid aggregates, with diffuse or nodular pattern [2, 3, 7, 8, 10]. Besides the above clinico-serological and pathological findings, the presence of one or more organ involvement can be useful for a correct classification of MC syndrome [ 10]. Both type II and type III MC are characterized by large amounts of circulating immune-complexes; the fraction of cryoprecipitable immune-complexes varies greatly among MC patients and in the same patient during the course of the disease [ 10, 11 ]. In some subjects with overt MC syndrome but without detectable serum cryoglobulins it is necessary to repeat at intervals the cryocrit determination. Cryoglobulin detection and characterization are necessary for a correct classification and diagnosis; however, the amount of serum cryoglobulins generally does not correlate with the severity and prognosis of the disease [ 10, 11].

30 and 54% of MC patients, respectively [17, 18]. One year later, HCV viremia was first reported in a large series of MC patients along with a striking correlation between HCV seropositivity and viremia (91% vs 86%) [19]. A subsequent study demonstrated that HCV RNA was markedly more concentrated (1000-fold) in the cryoprecipitate than in the supernatants [20]. The prevalent role of HCV in MC has been definitely established on the basis of epidemiological, pathological, and laboratory studies [ 10, 21]. In particular, immunohistochemical and molecular biology studies, including HCV RNA detection by in situ hybridisation, have reinforced the hypothesis of a direct involvement of HCV antigens in the immune-complex-mediated cryoglobulinemic vasculitis [21-24]. Being HCV the main triggering factor of MC the term 'essential' is no longer appropriate for the majority of cases [10,21-24].

3. HCV INFECTION AND MIXED CRYOGLOBULINEMIA

The association between HCV and MC [21] together with other immune-system alterations observed in chronically HCV-infected individuals [25-34] suggested that the same virus could be the triggering factor of other extrahepatic immunological disorders. Table 3 summarizes the main organ or systemic diseases that may be related to HCV infection according to the strength of association. Besides the well-established association with the MC syndrome, HCV can be detected in a significantly high percentage of patients with autoimmune or neoplastic diseases when compared to HCV prevalence in the general population. In particular, porphyria cutanea tarda (PCT), glomerulonephritis, diabetes, thyroid disorders, and B-cell neoplasias have been widely investigated for this purpose [27-35, 10]. In patients with the sporadic variant of PCT, a metabolic disorder characterized by reduced hepatic activity of uroporphyrinogen decarboxylase, a clear-cut association with HCV infection has been demonstrated [27, 28]. The majority of patients with PCT present chronic liver involvement along with some HCV-driven clinico-serological autoimmune phenomena [28]. Renal involvement complicating HCV infection is generally represented by type I

Being liver involvement one of the most frequent clinical features of the MC (Table 2), a causative role of hepatotropic viruses in MC has long been hypothesized during the seventies [2, 10, 12, 13]. Following the demonstration of a significant association between hepatitis B virus (HBV) and another systemic vasculitis - the polyarteritis nodosa- [ 14], a possible role of HBV was suggested also for MC [12, 13]. However, the presence of HBV antigenemia is seldom recorded, while anti-HBV antibodies largely varied among different MC patients' series [12, 15]. Thereafter, it can be estimated that HBV can represent an etiological factor in a minority of individuals, generally less than 5% of MC [10] (Table 2). In 1989 hepatitis C virus (HCV) has been identified as the major etiologic agent of post-transfusion and sporadic parentally-transmitted non-A-non-B hepatitis [16]. A role of HCV in MC was initially suggested in 1990 by two distinct studies reporting the presence of antibodies against HCV (anti-HCV; first generation ELISA, Chiron, Emeryville CA) in

4. HCV-ASSOCIATED AUTOIMMUNE AND LYMPHOPROLIFERATIVE DISORDERS

203

Table 3. Hepatitis C virus infection and extrahepatic disorders Established associationa mixedcryoglobulinemia Significant associationb B-cell NHL monoclonal gammopathies porphyria cutanea tarda diabetes mellitus thyroid disorders glomerulonephriitis Possible associationc

chronic polyarthritis sicca syndrome/Sj6gren's s. lung fibrosis polyarteritis nodosa poly/dermatomyositis gonadal (erectile) dysfunction lichen planus, other skin dis. Mooren comeal ulcers

aHCV infection in the large majority of patients. bHCV in a significant percentage of patients compared to general population. cSuggested but unproven association. MPGN, more often as visceral complication of MC syndrome. Type I MPGN alone, and less frequently milder glomerulonephritis patterns, can be also observed in HCV-positive individuals [36]. With regard to some endocrinological manifestations a number of epidemiological studies have reported a significantly higher risk to develop diabetes mellitus and thyroid disorders in patients with HCV infection compared to general population [10, 37-39]. Moreover, gonadal involvement responsible for erectile dysfunction in HCV-positive males has been recently reported [40]. Patients with type II MC can develop a B-cell lymphoma, usually after a long-term follow up [3, 10, 22, 41-43]. It was shownthat these patients frequently carry lymphoid infiltrates in the liver and bone marrow characterized by peculiar clinicopathologic characteristics [10, 22]. These infiltrates have been regarded as "early lymphomas", since they are sustained by lymphoid components indistinguishable from those of B-cell chronic lymphocytic leukaemia/small lymphocytic lymphoma (B-CLL) and immunocytoma (Ic) [10, 22]. However, conversely to overt lymphomas, they tend to remain unmodified for years or even decades and are followed by malignant lymphoid tumor in about 10% of cases [41, 43]. These characteristics justify

204

the recently proposed term of "monotypic lymphoproliferative disorder of undetermined significance (MLDUS)" [10, 22]. This condition includes two main pathological patterns; namely, the B-CLLlike and the Ic-like. Since MC may be regarded as potential pre-lymphomatous disorder and HCV represents its triggering factor, a possible role of the same virus also in 'idiopathic' B-cell NHL had been suggested [44]. In 1994, HCV infection was first demonstrated in a significant percentage of Italian patients with unselected B-cell non-Hodgkin's lymphomas (B-NHLs), regardless of the histotype [45]. This association was then confirmed by several studies on various B-cell NHL patient populations from Italy and other Countries [ 10, 22, 46, 47]. An increasing number of clinico-epiderniological and immunopathological studies seems to support a causative role of HCV in the above immune system disorders [10, 48]. The actual strength of these associations is controversial; however, HCV might play a pathogenetic role for at least certain patients' subsets and in some geographical areas. Of interest, different HCV-related diseases show an intriguing clinico-serological overlap [10]. In this scenario, MC can be represent a crossing road between some classical autoimmune disorders (autoimmune hepatitis, sicca syndrome, glomerulonephritis, thyroiditis, etc.) and malignancies (B-cell lymphomas, hepatocellular carcinoma) [ 10, 49] (Fig. 1). Because of possible methodological bias it is difficult to verify whether other suggested but unproven associations between HCV and some immunological disorders (Table 3c) is coincidental or a pathogenetic link actually exists. HCV infection is rarely complicated by classic rheumatoid arthritis; whereas an intermittent, non-erosive olygoarthritis of large-medium sized joints, is often observed in a significant number of HCV-infected individuals, more often without overt MC syndrome [10, 34, 50, 51]. Generally, polyarthritis in HCV-positive patients shows a more benign clinical course and good response to low steroid dosage and hydroxychlorochine treatment [50]. In some instances, some symptoms such as polyarthritis, glomerulonephritis, or neuropathy may present as apparently isolated manifestation of HCV infection. In these patients a concomitant, subclinical hepatitis, cryoglobulinemic syndrome, or B-cell lymphoma should be carefully investigated.

Figure 1. Possible etiopathogenesis of mixed cryoglobulinaemia (MC) and other HCV-related disorders. HCV infection may exert a chronic stimulus on the immune-system; in particular, various pathogenetic mechanisms can be taken in account: a) the interaction between HCV envelope protein E2 and CD81 on both hepatocytes and lymphocytes; b) a molecular mimicry mechanism involving HCV antigens and possible autoantigens; and c) T(14;18) translocation commonly found in HCV-infected individuals, particularly in MC patients; the consequent activation of Bcl2 proto-oncogene may lead to prolonged B-cell survival. B-lymphocyte expansion may be responsible for various autoantibodies production, including rheumatoid factor and cryo- and non-cryoprecipitable immune-complexes (CIC). Consequently, various auto'immune disorders and cryoglobulinemic vasculitis may develop. The indolent B-cell proliferation underlying MC may be complicated by framkmalignant lymphoma in about 10% of patients. Moreover, HCV is the major causative factor of hepatocellular carcinoma; finally, a possible link between HCV and thyroid cancer has been also suggested. There is a clinico-serologic and pathologic overlap among different HCV-related diseases; mixed cryoglobulinaemia syndrome represents a crossroads between these autoimmune and neoplastic disorders.

As observed for MC, the prevalence of different HCV-related autoimmune-lymphoproliferative diseases shows a geographically heterogeneous distribution [10, 27-31, 36], suggesting a role of important co-factors also in these HCV-related disorders.

5. P A T H O G E N E S I S O F H C V - A S S O C I A T E D MIXED CRYOGLOBULINEMIA Following its identification, HCV has been recognized to be both hepato- and lymphotropic virus, as firstly demonstrated by the presence of active or latent viral replication in the peripheral lymphocytes of patients with type C hepatitis [52]. Of great interest, the infection of lymphoid tissue may explain the appearance of a constellation of autoimmune

205

Table 4. Treatmentof HCV-associated mixed cryoglobulinaemia proposed treatments

none

asymptomatic mild-moderate manifestations

purpura, weakness arthralgias, arthritis, peripheral sensory neuropathy

~

low dosage of steroids and/or LAC-diet other symptomatics

severe manifestations

nephropathy, skin ulcers sensory-motor neuropathy widespread vasculitis active hepatitis

attempt at HCV eradication"

~ steroids and/or plasma exchange and/or cyclophosphamide rituximab interferon + ribavirin

cancer

B-cell NHL, HCC

chemotherapy, surgery

NHL: non-Hodgkin's lymphoma; HCC: hepatocellular carcinoma; *alpha-interferon+ ribavirin; LAC-diet: low antigen content diet

and lymphoproliferative disorders in HCV-infected individuals [10, 44, 52]. HCV-related types HI and II MC are comparable with regard their organ involvement and clinical course, with the exception of their potential evolution to malignancy. Although not definitely demonstrated, they might represent two different steps of the same disorder: MC type III may evolve to benign linfoproliferative disorder, the mono-oligoclonal B-cell proliferation of MC type II, which in some individuals can be complicated by frank B-cell non-Hodgkin's lymphoma (NHL), usually after a long-term follow-up period [ 10, 22]. Circulating mixed cryoglobulins are frequently detectable in HCV-infected individuals (50%); whereas, overt cryoglobulinemic syndrome develops in only a minority of cases (5%) [10]. HCV infection presents a homogeneous diffusion worldwide, which contrasts with the geographical etherogeneity in the prevalence of HCV-related MC as well as other immune-system disorders. The involvement of particular HCV genotypes, environmental and/or host genetic factors should contribute to the pathogenesis of MC; however, the actual role of the above co-factors remains still to be demonstrated [ 10, 22,

53, 54]. The above considerations suggest that HCV p e r se might be insufficient to drive the autoim-

mune-lymphoproliferative phenomena observed in

206

a limited but significant proportion of infected individuals. HCV is a positive, single-stranded RNA virus without a DNA intermediate in its replicative cycle, so that viral genomic sequences cannot be integrated into the host genome. Therefore, it has been proposed that HCV infection exerts a chronic stimulus to the immune system, which facilitates the polyclonal B-lymphocyte expansion and the selection in some subjects of malignant B-cell clones [10, 22]. In spite of morphology and monotypic Ig light chain expression, the lymphoproliferation occurring in HCV-positive patients with type II MC should not be regarded as a real lymphomatous situation, since: 1) it is usually characterized by oligoclonality, and 2) the overt malignant lymphoma which eventually develops during follow-up, more often stems from a B-cell clone other than the ones sustaining MLDUS as suggested by the molecular analysis of liver and bone marrow lymphoid infiltrates [7, 8, 22]. More interestingly, the presence of t(14;18) translocation leading to Bcl-2 activation has been demonstrated in a significant percentage of peripheral blood lymphocytes in HCV-infected individuals, particularly in those with MC [55-57]. Besides, the recent identification of HCV envelop protein E2 able to bind CD81 molecule expressed on both hepatocytes and B-lymphocytes [58] could help to clarify the pathogenesis of HCV-related autoim-

mune and neoplastic diseases (Fig. 1). In fact, CD81 is a cell-surface protein that, on B-cell, is part of a complex with CD21, CD 19, and Leu 13. This complex reduces the threshold for B-cell activation by bridging antigen specific recognition and CD21-mediated complement recognition. It can be hypothesized that the interaction between HCV-E2 and CD81 may increase the frequency of VDJ rearrangement in antigen-reactive B-cell. One possible consequence could be the above mentioned bcl-2 activation observed in HCV-related diseases, mainly MC [55-57]. This proto-oncogene is able to inhibit the apoptosis leading to extended cell survival [59]. The aberration of bcl-2 can explain, at least in part, the B-lymphocyte expansion and the wide autoantibody production observed in HCV-infected individuals [ 10, 22, 28, 29, 31 ]. Other mechanisms such as molecular mimicry can be involved in B lymphocyte activation responsible for different hepatic and extrahepatic autoimmune disorders. On the other hand, the prolonged B-cell survival can expose these cells to other genetic aberrations leading to overt malignant lymphoma (Fig. 2). HCV exerts a well-known oncogenic potential as definitely demonstrated for hepatocellular carcinoma; the same virus seems to be also involved in the lymphomagenesis and, possibly, in other malignancies such as thyroid cancer [33, 37].

6. MANAGEMENT OF HCV-ASSOCIATED MIXED CRYOGLOBULINEMIA The clinical symptoms of MC largely vary among patients and in the same patient during the followup. Usually, MC shows a relatively benign clinical course; the disease is often oligosymptomatic for long time intervals characterized by mild weakness, arthralgias, and sporadic flares of purpura on the legs. In other cases MC syndrome may start with or may be studded by one or more severe symptoms such as renal, neurological, and/or liver involvement, widespread vasculitis, and/or neoplastic complications [ 10]. The cumulative survival of MC shows a significantly poor prognosis if compared to general population [60]. Overall, the treatment of MC syndrome is particularly challenging because of its complex etiopathogenesis [10]. For a correct therapeutic approach to HCV-related MC

HCVeradication immunosuppressors LAC-dietPlasma'exchange steroids

HCV infection I

ill

Benign B-cell expansion autoantibodies and cryoglobulin production

.........

l

~1Cryoglobulinemic vasculitis /

chemotherapy

_

" ]B-cell

,~,

_ _

_

lymphomaI

Figure 2. Mixed cryoglobulinaemia is a combination of three main clinico-pathological alterations: chronic HCV infection, B-cell lymphoproliferation, and immune-complex vasculitis. We can treat the disease at different levels by means of combined - etiologic, pathogenetic, and symptomatic - therapies. LAC-diet: low antigen content diet.

we must deal with the concomitance of conflicting conditions: HCV infection, autoimmune, and lymphoproliferative alterations. According to the cascade of events leading from HCV infection to cryoglobulinemic vasculitis (Figs. 1, 2) we can treat the disease at three different levels by means of etiologic, pathogenetic, and/or symptomatic therapies. Since HCV represents the triggering factor of the disease and probably exerts a chronic stimulus on the immune system (Fig. 1), an attempt at HCV eradication should be done in all cases of HCV-associated MC. In this respect encouraging data came from some preliminary observations: in MC patients with MLDUS repeated bone-marrow biopsies, before/after interferon therapy, showed a regression of lymphoid infiltrates along with HCV clearance [61 ]; moreover, the antiviral therapy may induce the regression of T(14; 18) beating B-cell clones in HCV-positive patients [62]. On these basis, we can hypothesize that antiviral therapy (interferon + ribavirin) may improve or treat the lymphoproliferative disorder underlying the MC. Unfortunately, HCV eradication is obtained in a small percentage of cases, while the beneficial effect observed with interferon treatment is often transient and not rarely associated with important immune-mediated complications, in particular, the peripheral sensory-motor neuropathy [63-68]. There are no parameters available for predicting

207

this hamafial complication; thus, alpha-interferon therapy should be avoided at least in those patients with clinically evident peripheral neuropathy. Similarly, in patients treated by alpha-interferon for type C hepatitis without MC syndrome, it is not rare to encounter complications such as peripheral neuropathy, thyroiditis, and rheumatoid-like polyarthritis. Probably, in predisposed subjects, alpha-interferon, both an antiviral and immunomodulating agent, can trigger or exacerbate some pre-existing, often subclinical, symptoms [66-69]. On the whole, the usefulness of alpha-interferon treatment in MC patients is limited by the low rate of responders and frequent side effects. The association of pegylated interferon and ribavirin might achieve the eradication of HCV infection in a rather significant number of treated subjects, as recently demonstrated in patients with type C chronic hepatitis [70-72]. Controlled clinical trials are necessary to definitely evaluate the usefulness of such combined antiviral therapy in HCV-related MC patients. With the rapid growth of molecular biology a vaccine against HCV might be available in the near future. The identification of the interaction between HCV envelope protein E2 and CD81 on both hepatocytes and lymphocytes [58] suggests the possibility of interfering with HCV binding to target cells. In HCV-infected individuals a vaccine-based therapy with recombinant HCV proteins [58, 73] could be able to prevent the evolution from HCV infection to both severe hepatic and extra-hepatic complications, and could possibly interrupt the selfperpetuating autoimmune mechanism underlying HCV-related disorders. Immunosuppressive treatment is still the firstline intervention in rare cases of 'essential' MC. In the setting of HCV-related MC the immunosuppressive treatment should be considered mainly in those patients who have failed to respond to alpha-interferon. An immunosuppressive treatment with cyclophosphamide often in association with steroids, and/or plasma exchange may be able to treat some severe MC complications such as nephropathy, sensory-motor neuropathy, or widespread vasculitis [2, 10, 74, 75]. Both traditional and double-filtration plasma exchange are able to achieve a dramatic reduction of circulating immune-complex levels, including the cryoglobu-

208

lins, as well as the viral loading [ 10, 76]. The beneficial effect of such 'symptomatic' treatment can be reinforced by means of oral cyclophosphamide during the tapering of apheretic sessions (50--100 mg/day for 4-8 weeks). In particular, it can prevent the rebound phenomena that may be observed after the discontinuation of apheresis. Plasma exchange is useful in severe MC complications, and particularly in active cryoglobulinemic nephropathy. Low-antigen-content diet (LAC-diet) has been employed in some immune-complex-mediated disorders, namely MC and IgA-nephropathy [77, 78]. In MC patients, this particular dietetic treatment can improve the serum clearance of immune-complexes by restoring the activity of the reticulo-endothelial system, overloaded by large amounts of circulating cryoglobulins [77]. LAC-diet and/or low dosage of steroids may be sufficient to improve mild-moderate manifestations of MC, i.e. purpura, arthralgias, peripheral sensory neuropathy, etc. [ 10]. More recently, a pathogenetic treatment with rituximab, a monoclonal chimeric antibody that binds to the B-cell surface antigen CD20, has been proposed in HCV-positive patients with type II MC [79, 80]. The selective B-cell blockade leads to the improvement of MC manifestations, including skin vasculitis, peripheral neuropathy, and lowgrade B-cell lymphoma along with a significant reduction of serum RF and cryoglobulin levels. Of interest, it has been noticed that serum HCV RNA increased approximately twice the baseline levels in the responders [80]. The impact of rituximab on HCV viremia suggests the possible use of combined therapy with this monoclonal antibody and other (antiviral?) agents. Finally, the long-term efficacy and safety of rituximab need to be investigated by controlled clinical trials. On the whole, MC treatment should be tailored for the single patient according to the severity of clinical symptoms. While asymptomatic patients usually do not need any treatment, even in the presence of high levels of cryocrit, patients with mildmoderate symptoms, such as palpable purpura, are particularly sensitive to the smallest variations of daily steroid dosage (1-2 mg). On the contrary, severe, life-threatening vasculitic manifestations must be promptly treated with a combined therapy based on plasma exchange, high doses of steroids, and/or immunosuppressors. A careful clinical moni-

toting of the disease is mandatory in all cases, with particular attention to neoplastic complications.

6.

ACKNOWLEDGEMENTS 7. We thank all the following people who actively contributed to our studies: L. La Civita, MD, G Longombardo, BS, G Porciello, MD, P Fadda, MD, M Sebastiani, D Giuggioli, M Cazzato, R. Cecchetti, G Pasero, MD: Rheumatology Unit,

Department of Internal Medicine, University of Pisa, Pisa, Italy; A.L. Zignego: Istituto Medicina Interna, University of Florence, Italy; S. Pileri: Pathologic Anatomy and Haematopathology Unit, University of Bologna, Italy; E Caracciolo, MD, M. Petrini: Cattedra di Ematologia; University of Pisa; Pisa, Italy; E Greco, MD, A. Mazzoni, MD: Blood Center, Ospedale S. Chiara, Pisa, Italy; L. Moriconi, R Puccini: Nephrology Unit, Ospedale S. Chiara, Pisa, Italy; S. Marchi, MD, and F. Costa, MD: Clinica Medica I, University of Pisa; P. Highfield, Ph, and T. Corbishley, Ph: Wellcome Diagnostic, Beckenham, UK; M.P. Manns, MD: Department of Gastroenterology and Hepatology, Zentrum Innere Medizin, Medizinische Hochschule, Hannover, Germany.

8.

9.

10. 11.

12.

13.

REFERENCES 14. 1.

2.

3.

4.

5.

Lospalluto J, Dorward B, Miller W Jr, Ziff M. Cryoglobulinemia based on interaction between a gamma macroglobulin and 7S gamma globulin. Am J Med 1962;32:142-145. Gorevic PD, Kassab HJ, Levo Y, Kohn R, Meltzer M, Prose P, Franklin E. Mixed Cryoglobulinemia: clinical aspects and long-term follow-up of 40 patients. Am J Med 1980;69:287-308. Gorevic PD, Frangione B. Mixed cryoglobulinemia cross-reactive idiotypes: implication for relationship of MC to rheumatic and lymphoproliferative diseases. Seminars in Hematology 1991;28:79-94. Brouet JC, Clouvel JP, Danon F, Klein M, Seligmann M. Biologic and clinical significance of cryoglobulins. Am J Med 1974;57:775-88. Meltzer M, Franklin EC, Elias K, McCluskey RT, Cooper N. Cryoglobulinemia. A clinical and laboratory study. II. Cryoglobulins with rheumatoid factor activity. Am J Med 1966;40:837-856.

15.

16.

17.

18.

19.

Tissot JD, Schifferli JA, Hochstrasser DE Pasquali C, Spertini F, Clement F, Frutiger S, Paquet N, Hughes GJ, Schneider P. Two-dimensional polyacrylamide gel electrophoresis analysis of cryoglobulins and identification of an IgM-associated peptide. J Immunol Methods 1994;173:63-75. Magalini AR, Facchetti F, Salvi L, Fontana M, Puoti M, Scarpa A. Clonality of B-cells in portal lymphoid infiltrates of HCV-infected livers. J Pathol 1998;185: 86-90. De Vita S, De Re V, Gasparotto D, Ballar~ M, Pivetta B, Ferraccioli G, Pileri S, Boiocchi M, Monteverde A. Type II mixed cryoglobulinemia is a non-malignant B-cell disorder. Clinical and molecular findings do not support bone marrow pathologic diagnosis of low grade B-cell lymphoma. Arthritis & Rheumatism 2000;43: 94-102. Abel G, Zhang QX, Agnello V Hepatitis C virus infection in type II mixed cryoglobulinemia (review). Arthritis Rheum 1993;36:1341-9. Ferri C, Zignego AL, Pileri SA. Cryoglobulins. J Clin Pathol. 2002 Jan;55(1):4-13. Review. Bombardieri S, Ferri C, Migliorini P, Pontrandolfo A, Puccetti A, Vitali C, Pasero G. Cryoglobulins and immune complexes in essential mixed cryoglobulinemia. La Ricerca Clin Lab 1986;16:281-8. Bombardieri S, Ferri C, Di Munno O, Pasero G. Liver involvement in essential mixed cryoglobulinemia. La Ricerca Clin Lab 1979;9:361-368. Levo Y, Gorevic PD, Kassab HJ, Zucker-Franklin D, Franklin EC. Association between hepatitis B virus and essential mixed cryoglobulinemia. N Engl J Med 1977;296:1501-1504. Gocke DJ, Hsu K, Morgan C, Bombardieri S, Lockshin M, Christian CL Association between polyarteritis and Australia antigen. Lancet 1970;2:1149-53. Popp JW, Dienstag JL, Wands JR, Bloch KJ. Essential mixed cryoglobulinemia without evidence for hepatitis B virus infection. Ann Intern Med 1980;92:379-383. Choo GL, Kuo G, Weiner AJ, Overby LR, Bradley DW, Houghton M. Isolation of a cDNA clone derived from a blood-borne non-A non-B viral hepatitis genome. Science 1989;244:359-361. Pascual M, Perrin L, Giostra E, Schifer JA. Hepatitis C virus in patients with cryoglobulinemia type II. J Infect Dis 1990;162:569-570. Ferri C, Marzo E, Longombardo G, Lombardini F, Greco F, Bombardieri S. Alpha-interferon in the treatment of mixed cryoglobulinemia patients. International Cancer Update. Focus on Interferon Alfa 2b. Cannes, France. November 1-4, 1990. Proceedings; Eur J Cancer 1991;27(4):81-82. Ferri C, Greco F, Longombardo G, Palla P, Moretti A,

209

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

210

Marzo E, Mazzoni A, Pasero G, Bombardieri S, Highfield P, Corbishley T. Association between hepatitis C virus and mixed cryoglobulinaemia. Clin Exp Rheumatol 1991;9:621--624. Agnello V, Chung RT, Kaplan L.M. A role for hepatitis C virus infection in type II cryoglubulinemia. N Engl J Med 1992;327:1490-1495. Ferri C, La Civita L, Longombardo G, Greco F, Bombardieri S. Hepatitis C virus and mixed cryoglobulinemia (review). Eur J Clin Invest 1993;23:399-405. Ferri C, Pileri S, Zignego AL. Hepatitis C virus, B-cell disorders, and non-Hodgkin's lymphoma. Chapter 19. In: Goedert JJ, ed. Infectious Causes of Cancer. Targets for Intervention. National Cancer Institute (NIH). Totowa, NJ: Humana Press, 2000;349-368. Sansonno D, De Vita S, Cornacchiulo V, Carbone A, Baiocchi M, Dammacco E Detection and distribution of hepatitis C-related proteins in lymph nodes of patients with type II cryoglobulinemia and neoplastic or non-neoplastic lymphoproliferation. Blood 1996;88: 4638--4645. Agnello V, Abel G. Localization of hepatitis C virus in cutaneous vasculitic lesions in patients with type II cryoglobulinemia. Arthritis Rheum 1997;40:2007-15. Lunel F, Musset L, Cacoub P, Frangeul L, Cresta P, Perlin M, Grippon P, Hoang C, Valla D, Piette JC. Cryoglobulinemia in chronic liver diseases: role of hepatitis C virus and liver damage. Gastroenterology 1994; 106:1291-300. Pawlotsky J, Mustapha B, Andre C, Voisin M-C, Intrator L, Roudot-Thoraval F, Deforges L, Duvoux C, Zairani E-S, Duval J. Immunological disorders in C virus chronic active hepatitis: a prospective case-control study. Hepatology 1994; 19:841-848. Fargion S, Piperno A, Cappellini MD, Sampietro M, Fracanzani AL, Romano R, Caldarelli R, Marcelli R, Vecchi L, Fiorelli G. Hepatitis C virus and porphyria cutanea tarda: evidence of a strong association. Hepatology 1992;16:1322-1326. Ferri C, Baicchi U, La Civita L, Greco F, Longombardo G, Mazzoni A, Careccia G, Bombardieri S, Pasero G, Zignego AL, Manns MP. Hepatitis C virus-related autoimmunity in patients with porphyria cutanea tarda. Eur J Clin Invest 1993;23:851-855. Lenzi M, Johnson PJ, McFarlane IG, Ballardini G, Smith HM, McFarlane BM, Bridger C, Vergani D, Bianchi FB, Williams R. Antibodies to hepatitis C virus in autoimmune liver disease: evidence for geographical heterogeneity. Lancet 1991 ;338ii:277-280. Stehman-Breen C, Willson R, Alpers CE, Gretch D, Johnson RJ. Hepatitis C virus-associated glomerulonephritis. Curr Opin Nephrol Hypertens 1995;4:287-94. Ferri C, Longombardo G, La Civita L, Greco F, Lom-

32.

33.

34.

35. 36.

37.

38.

39.

40.

41.

42.

43.

44.

bardini E Cecchetti R, Cagianelli MA, Marchi S, Monti M, Zignego AL, Manns ME Hepatitis C virus as common cause of mixed cryoglobulinemia and chronic liver disease. J Intern Med 1994;236:31-36. Ferri C, La Civita L, Zignego AL, Pasero G. Viruses and cancers: possible role of hepatitis C virus (review). Eur J Clin Invest 1997;27:711-718. Antonelli A, Ferri C, Fallahi E Thyroid cancer in patients with hepatitis C infection. JAMA 1999;281 (17): 1588. Lovy MR, Starkebaum G, Uberoi. Hepatitis C infection presenting with rheumatic manifestations: a mimic of rheumatoid arthritis. J Rheumatol 1996;23:979-83. Lamprecht P, Gause A, Gross WL. Cryoglobulinemic vasculitis (review). Arthritis Rheum 1999;42:2507-16. Fabrizi E Colucci P, Ponticelli C, Locatelli E Kidney and liver involvement in cryoglobulinemia. Semin Nephrol 2002;22:309-18. Review. Antonelli A, Ferri C, Fallahi P, Nesti C, Zignego AL, Maccheroni M. Thyroid cancer in HCV-related mixed cryoglobulinemia patients. Clin Exp Rheumatol 2002;20:693-6. Antonelli A, Ferri C, Fallahi P, Sebastiani, M, Nesti C, Barali L, Ferrannini. Type 2 diabetes in hepatitis C-related mixed cryoglobulinemia patients. Rheumatology 2004;43:238-40. Mehta SH, Brancati FL, Sulkowski MS, Strathdee SA, Szklo M, Thomas DL. Prevalence of type 2 diabetes mellitus among persons with hepatitis C virus infection in the United States. Ann Intern Med 2000;133:592-9. Ferri C, Bertozzi MA, Zignego AL. Erectile dysfunction and hepatitis C virus infection. JAMA 2002;288: 698-9. Ferri C, Monti M, La Civita L, Careccia G, Mazzaro C, Longombardo G, Lombardini F, Greco F, Pasero G, Bombardieri S, Zignego AL. Hepatitis C virus infection in non-Hodgkin's B-cell lymphoma complicating mixed cryoglobulinaemia. Eur J Clin Invest 1994;24: 781-784. Pozzato G, Mazzaro C, Crovatto M, Modolo ML, Ceselli S, Mazzi G, Sulfaro S, Franzin F, Tulissi P, Moretti M, Santini GE Low-grade malignant lymphoma, hepatitis C virus infection, and mixed cryoglobulinemia. Blood 1994;84:3047-53. Monteverde A, Ballar~ M, Pileri S. Hepatic lymphoid aggregates in chronic hepatitis C and mixed cryoglobulinemia. Springer Semin. Imunopathol 1997;19: 99-110. Ferri C, Monti M, La Civita L, Longombardo G, Greco F, Pasero G, Gentilini P, Bombardieri S, Zignego AL. Infection of peripheral blood mononuclear cells by hepatitis C virus in mixed cryoglobulinemia. Blood 1993 ;82:3701-3704.

45. Ferri C, Caracciolo F, Zignego AL, La Civita L, Monti M, Longobardo G et al. Hepatitis C virus infection in patients with non-Hodgkin's lymphoma. Br J Haematol 1994;88:392-394. 46. Luppi M, Grazia FM, Bonaccorsi G, Longo G, Nami F, Barozzi Pet al. Hepatitis C virus infection in a subset of neoplastic lymphoproliferations not associated with cryoglobulinemia. Leukemia 1996; 10:351-355. 47. Zuckerman E, Zuckerman T, Levine AM, Douer D, Gutekunst K, Mizokami M, Qian DG, Velankar M, Nathwani BN, Fong TL. Hepatitis C virus infection in patients with B-cell non-Hodgkin's lymphoma. Ann Intern Med 1997;127:423-428. 48. Dammacco F, Sansonno D, Piccoli C, Racanelli V, D'Amore FP, Lauletta G. The lymphoid system in hepatitis C virus infection: autoimmunity, mixed cryoglobulinemia, and overt B-cell malignancy. Semin Liver Dis 2000;20:143-57. Review. 49. Ferri C, La Civita L, Longombardo G, Zignego AL, Pasero G. Mixed cryoglobulinemia: a cross-road between autoimmune and lymphoproliferative disorders. Lupus 1998;7:275-279. 50. Fadda P, La Civita L, Zignego AL, Ferri C. Hepatitis C virus infection and arthritis. A clinico-serological investigation of arthritis in patients with or without cryoglobulinemic syndrome. Reumatismo 2002;54: 316-23. 51. Buskila D. Hepatitis C-associated arthritis. Curr Opin Rheumato12000;12:295-9. Review. 52. Zignego AL, Macchia D, Monti M, Thiers V, Mazzetti M, Foschi Met al. Infection of peripheral mononuclear blood cells by hepatitis C virus. J Hepatol 1992;15: 382-386. 53. Zignego AL, Ferri C, Giannini C, Monti M, La Civita L, Careccia G, Longombardo G, Lombardini F, Bombardier S, Gentilini P. Hepatitis C virus genotype analysis in patients with type II mixed cryoglobulinemia. Ann Intern Med 1996;124:31-34. 54. Lenzi M, Frisoni M, Mantovani V, Ricci P, Muratori L, Francesconi R, Cuccia M, Ferri S, Bianchi FB. Haplotype HLA-B8-DR3 confers susceptibility to hepatitis C virus-related mixed cryoglobulinemia. Blood 1998;91: 2062-6. 55. Zignego AL, Giannelli F, Marrocchi ME, Mazzocca A, Ferri C, Giannini C, Monti M, Caini P, Villa GL, Laffi G, Gentilini P. T(14;18) translocation in chronic hepatitis C virus infection. Hepatology 2000;31:474-9. 56. Zignego AL, Ferri C, Giannelli F, Giannini C, Caini P, Monti M, Marrocchi ME, Di Pietro E, La Villa G, Laffi G, Gentilini P. Prevalence of bcl-2 rearrangement in patients with hepatitis C virus-related mixed cryoglobulinemia with or without B-cell lymphomas. Ann Intern Med 2002; 137:571-80.

57. Kitay-Cohen Y, Amiel A, Hilzenrat N, Buskila D, Fejgin M, Gaber E, Safdi R, Tur-kaspa R, Lishner M. Bcl-2 rearrangement in patients with chronic hepatitis C associated with essential mixed cryoglobulinemia type II. Blood 2000;96:2910-2. 58. Pileri P, Uematsu Y, Campagnoli S, Galli G, Falugi F, Petracca R et al. Binding of hepatitis C virus to CD81. Science 1998;282:938-941. 59. Korsmeyer SJ. Bcl-2: a repressor of lymphocyte death. Immunol Today 1992; 13:285-287. 60. Ferri C, Sebastiani M, Giuggioli D, Cazzato M, Longombardo G, Alessandro Antonelli A, Rodolfo Puccini R, Claudio Michelassi C, Zignego AL. Mixed cryoglobulinemia: demographic, clinical, and serological features, and survival in 231 patients. Sem Arthritis Rheum 2004;33: in press. 61. Mazzaro C, Franzin F, Tulissi P, Pussini E, Crovatto M, Carniello GS et al. Regression of monoclonal B-cell expansion in patients affected by mixed cryoglobulinemia responsive to a-interferon therapy. Cancer 1996;77: 2604-2613. 62. Giannelli F, Moscarella S, Giannini C, Caini P, Monti M, Gragnani L, Romanelli RG, Solazzo V, Laffi G, La Villa G, Gentilini P, Zignego AL. Effect of antiviral treatment in patients with chronic HCV infection and t(14; 18) translocation. Blood 2003. 63. Ferri C, Marzo E, Longombardo G, Lombardini E La Civita L, Vanacore R, Liberati AM, Gedi R, Greco F, Moretti A, Monti M, Gentilini P, Bombardieri S, Zignego AL. Alpha-Interferon in Mixed Cryoglobulinemia patients: a randomized crossover controlled trial. Blood 81:1132-1136;1993. 64. Misiani R, Bellavita P, Fenili D, Vicari O, Marchesi D, Sironi PL, Zilio P, Vernocchi A, Massazza M, Vendramin G. Interferon alfa-2a therapy in cryoglobulinemia associated with hepatitis C virus. N Engl J Med 1994;330:751-6. 65. Casato M, Lagana B, Antonelli G, Dianzani F, Bonomo L Long-term results of therapy with interferon-alpha for type II essential mixed cryoglobulinemia. Blood 1991 ;78(12):3142-7. 66. Di Lullo L, De Rosa FG, Coviello R, Sorgi ML, Coen G, Zorzin LR, Casato M. Interferon toxicity in hepatitis C virus-associated type II cryoglobulinemia. Clin Exp Rheumatol 1998;16:506. 67. La Civita L, Zignego AL, Lombardini F, Monti M, Longombardo G, Pasero G, Ferri C. Exacerbation of peripheral neuropathy during alpha-interferon therapy in a patient with mixed cryoglobulinemia and hepatitis B virus infection. J Rheumatol 1996;23:1641-1643. 68. Lidove O, Cacoub P, Hausfater P, Wechsler B, Frances C, Leger JM, Piette JC. Cryoglobulinemia and hepatitis C. Worsening of peripheral neuropathy after interferon

211

69.

70.

71.

72.

73. 74.

212

alpha treatment. Gastroenterol Clin Biol 1999;23: 403-6. Cid MC, Hernandez-Rodriguez J, Robert J, del Rio A, Casademont J, Coll-Vinent B, Grau JM, Kleinman HK, Urbano-Marquez A, Cardellach F. Interferon-alpha may exacerbate cryoblobulinemia-related ischemic manifestations: an adverse effect potentially related to its anti-angiogenic activity. Arthritis Rheum 1999;42: 1051-5. McHutchison JG, Gordon SC, Schiff ER, Shiffman ML, Lee WM, Rustgi VK et al. Interferon alpha-2b alone or in combination with ribavirin as initial treatment for chronic hepatitis C. Hepatitis Interventional Therapy Group. N Engl J Med 1998;339:1485-92. Misiani R, Bellavita P, Baio P, Caldara R, Ferruzzi S, Rossi P, Tengattini E Successful treatment of HCVassociated cryoglobulinaemic glomerulonephritis with a combination of interferon-alpha and ribavirin. Nephrol Dial Transplant 1999; 14:1558-60. Zuckerman E, Keren D, Slobodin G, Rosner I, Rozenbaum M, Toubi E, Sabo E, Tsykounov I, Naschitz JE, Yeshurun D. Treatment of refractory, symptomatic, hepatitis C virus related mixed cryoglobulinemia with ribavirin and interferon-alpha. J Rheumato12000;27(9): 2172-8. Abrignani S, Rosa D. Perspectives for a hepatitis C virus vaccine. Clin Diagn Virol 1998; 10:181-185. Ferri C, Moriconi L, Gremignai G, Migliorini P, Pale-

75.

76. 77.

78.

79.

80.

ologo G, Fosella PV, Bombardieri S. Treatment of renal involvement in essential mixed cryoglobulinemia with prolonged Plasma Exchance. Nephron 1986;43: 246-251. Moriconi L, Ferri C, Puccini R, Casto G, Baronti R, Cecchetti R, Gremignai G, Cioni L, Bombardieri S. Double filtration plasmapheresis in the treatment of cryoglobulinemic glomerulonephritis. Int J Art Org 1989;12/$4:83-86. Kaplan AA. Apheresis for renal disease. Ther Apher 2001;5:134--41. Review. Ferri C, Pietrogrande M, Cecchetti C, Tavoni A, Cefalo A, Buzzetti G, Vitali C, Bombardieri S. Low-antigencontent diet in the treatment of mixed cryoglobulinemia patients. Am J Med 1989;87:519-24. Ferri C, Puccini R, Longombardo G, Paleologo G, Migliorini P, Moriconi L, Pasero G, Cioni L. Low-antigen-content diet in the treatment of patients with IgA nephropathy. Nephrol Dial Transpl 1993;8:1193-8. Zaja F, De Vita S, Mazzaro C, Sacco S, Damiani D, De Marchi G, Michelutti A, Baccarani M, Fanin R, Ferraccioli G. Efficacy and safety of rituximab in type II mixed cryoglobulinemia. Blood 2003;101:3827-34. Sansonno D, De Re V, Lauletta G, Tucci FA, Boiocchi M, Dammacco F. Monoclonal antibody treatment of mixed cryoglobulinemia resistant to interferon alpha with an anti-CD20. Blood 2003;101:3818-26.


Virus-Induced Systemic Vasculitides Lo'ic Guillevin, Pascal Cohen and Christian Pagnoux

Department of Internal Medicine, HOpital Cochin, University of Paris, Paris, France

1. I N T R O D U C T I O N Viruses have been demonstrated to be the etiological agent responsible for several vasculitides, which can affect vessels of various calibers and which are usually not associated with antineutrophil cytoplasmic antibodies (ANCA). Two major vasculitides can occur as a consequence of viral infection: classic polyarteritis nodosa (PAN), as a result of hepatitis B virus (HBV) infection [1], and mixed cryoglobulinemia, in patients infected with hepatitis C virus (HCV) [2]. Other viruses can also be associated, albeit less frequently, to the occurrence of vasculitis, for example human immunodeficiency virus (HIV) [3] and parvovirus B19 [4], among others. The demonstration of a close relationship between viral infection and vasculitis justified the original therapeutic approach, avoiding prolonged administration of steroids and cytotoxic agents and based on the combination of antiviral agents and plasma exchanges (PE). Furthermore, the prospective trials organized by the French Vasculitis Study Group (FVSG) validated this therapeutic strategy [5]. A specific antiviral strategy is also recommended for HCV-related cryoglobulinemia vasculitis, despite the less favorable results.

2. POLYARTERITIS NODOSA Since the first reports on HBV-related PAN (HBVPAN) [1, 6], this causal relationship has been largely confirmed based on clinical, epidemiological and therapeutic data [5]. During the 1970s, the rate of HBV infection in patients with classic PAN reached 50%. However, over the past few years,

the frequency of HBV-PAN has declined to less than 5% [7]. In France, the incidence of HBV-PAN has decreased dramatically since blood testing and donor selection have been reinforced, and large vaccination campaigns have been organized in teenagers and people at risk. Intravenous drug use is now becoming the major cause of HBV-PAN. Since 2002, we have seen very few new cases of HBVPAN in the French population but also fewer new PAN cases, observations that indirectly support the hypothesis of a viral cause of PAN. Some other viruses have been associated with PAN but could only explain the occurrence of a few cases per year. HCV is not a major etiological factor for PAN and its responsibility as such was advanced in only a few publications [8, 9]. Less than 5% of our patients are infected with HCV, which confirms our previous findings [ 10]. GB virus-C, when sought in patients with PAN, has not been found to be an agent responsible for the disease [ 11 ]. When present, HCV was often observed in association with other viruses, HBV or HIV, and also with mixed cryoglobulinemia. Several concomitant parvovirus B 19 infections have been described too [4, 12] but a systematic survey of PAN patients did not show them to have a higher frequency of parvovirus B 19 than the control population [12]. Other viruses have been incriminated in the development of PAN, including anecdotal cases of HIV infection [3, 13, 141. The immunological process responsible for HBV-PAN mainly in patients under 40 years of age usually becomes manifest less than 12 months after infection. Hepatitis is rarely diagnosed, as it remains silent before the occurrence of PAN. Clinical manifestations are of acute onset and are roughly

213

the same as those commonly observed in PAN [ 15]. HBV-PAN is certainly the purest form of PAN and no overlap with other vasculitides, especially microscopic polyangiitis, has been observed in our experience. HBe antigen (Ag) to anti-HBe antibody (Ab) seroconversion usually leads to recovery. The major sequelae are the consequence of vascular nephropathy and peripheral neuropathy but, even in patients who initially develop renal insufficiency, it is possible to cure PAN with little residual impairment of renal function.

2.1. Relapses HBV-PAN tends not to recur once remission is induced. In our series, only 6% of the patients relapsed [15]. At present, it is still not possible to identify the subgroup of patients who will relapse. The clinical pattern of relapse does not necessarily mimic the original presentation, in that previously unaffected organs can be involved at relapse. Due to the low frequency of relapses, maintenance treatment is not necessary and short-term treatment can be envisaged.

2.2. Deaths The causes of death can be divided into three categories: related to vasculitis manifestations, attributed to treatment side effects and miscellaneous causes, usually independent of the vasculitis. 2.2.1. Deaths related to vasculitis

In all vasculitides, involvement of major organs can have lethal issue. A few patients die early from multivisceral involvement, often gastrointestinal [ 16], that cannot be controlled by treatment. In such cases, the course of the disease is generally characterized by fever, rapid weight loss, diffuse pain and involvement of one or several major organs. 2.2.2. Deaths attributed to treatment side effects

Conventional treatment with steroids and cyclophosphamide jeopardizes the patient's outcome by allowing the virus to persist, stimulating its replication and thereby facilitating evolution towards chronic hepatitis and liver cirrhosis. Thus, cyclo-

214

phosphamide, like prolonged steroid treatment, is contraindicated. In addition to these long-term side effects, infections are more frequent when immunosuppressants are prescribed. Steroids are also responsible for side effects, which are not detailed here [ 16]. PE can also favor the occurrence of infections when a central venous access is necessary.

3. HCV-RELATED CRYOGLOBULINEMIA Mixed cryoglobulinemias of type ii and, more rarely, type III are the consequence of HCV infection in more than 80% of the patients [2]. Cryoglobulinemia is asymptomatic in most patients but persists for decades and the disease duration might be a factor associated with the occurrence of clinical symptoms of vasculitis. When symptoms are present, the most frequent are purpura, peripheral neuropathy, glomerulonephritis, leg ulcers, arthritis and sicca syndrome. Cryoglobulinemia vasculitis is a small vessel vasculitis, as defined by the Chapel Hill nomenclature [17]. The clinical symptoms [ 18, 19] may develop progressively and are often of moderate intensity at their onset. Neuropathy can be symmetric and limited to sensory signs, including hypoesthesia and pain. This distal neuropathy is more frequently present in the lower than upper limbs. Neuropathy can also be mononeuritis multiplex, as described in PAN. The outcome of neuropathy is chronic and, although the motor symptoms can regress, the sensory symptoms can remain definitively. Kidney involvement, when present, is glomerulonephritis but, unlike ANCA-associated vasculitides, pauci-immune glomerulonephritis is not found. Few patients progress to end-stage renal failure. Sicca syndrome is present in 20% of the patients but without the immunological features of Sj6gren's syndrome. The presence of the cryoglobulins, usually type II, IgM kappa, is characteristic of the disease. Complement, especially the C4 component is low and a rheumatoid factor may be found, and, because it is sometimes difficult to detect a cryoprecipitate, this association is highly suggestive of the diagnosis. Autoantibodies are absent, especially ANCA. The outcome of mixed cryoglobulinemia vasculitis is characterized by chronicity and relapses, even under treatment.

4. HIV-ASSOCIATED VASCULITIS

5. T R E A T M E N T

Vasculitides occurring during the course of HIV infection have been reported [3, 14]. Most of them involved skin, peripheral neuropathy or the central nervous system. The clinical spectrum and histological findings of HIV-associated vasculitis vary widely. Large-, medium- and small-sized arteries can be affected. Necrotizing arteritis, non-necrotizing arteritis, giant-cell arteritis and eosinophil arteritis have been observed [14]. According to Calabrese [ 14], the frequency is low (1%), and most of the reported cases were identified at autopsy. In our experience [3], HIV-associated vasculitis is an extremely rare entity and we have seen only a few cases in our vasculitis reference center, which is networked to several large centers specializing in the management of HIV infection. Vasculitis can develop in adults and children at any stage of HIV infection as defined by the Centers for Disease Control classification. Some cases seem to be directly caused by opportunistic infections, such as Pneumocystis carinii, cytomegalovirus or Toxoplasma gondii, non-opportunistic infectious agents or drug-induced hypersensitivity. HIV was thought to be the etiological agent in a few patients because of the in situ localization of the virus and the absence of evidence suggesting other mechanisms. However, the etiology remains unknown in most cases. The pathogenesis of HIV-associated vasculitis is heterogeneous, but at least two general mechanisms have been hypothesized: first, virus replication might induce direct injury of the vessel wall or vascular damage might be the result of an immune mechanism. These mechanisms may be cellular and/or humoral and include deposition of immune complexes (IC) and/or their in situ formation [20]. IC are frequently detected in AIDS patients and their frequency increases with advancing stages of infection [3]. Some authors have analyzed their composition and found them to contain both antibodies specific to HIV and HIV antigens. However, their role in the vasculitic process remains to be demonstrated.

5.1. Treatment of HBV-Related PAN

For many years, HBV-PAN was treated in the same way as non-vir~s-related PAN and patients received steroids, sometimes combined with cytotoxic agents, mainly cyclophosphamide. This treatment was often effective in the short-term but careful analysis of long-term results showed that relapses and complications (chronic hepatitis or liver cirrhosis) occurred because of virus persistence. According to McMahon et al [21 ], who followed Eskimos with PAN, 4 (31%) patients died during the course of PAN. In our first randomized study [22] in which patients were not selected according to their virus status, 14/71 were HBV-positive; 84% of them recovered from PAN but 2 subsequently died of liver cirrhosis. The rationale for combining PE and antiviral treatment was to obtain the following effects" initial corticosteroids to rapidly control the most severe life-threatening manifestations of PAN which are common during the first weeks of the disease, and abrupt stoppage of corticosteroids to enhance immunological clearance of HBV-infected hepatocytes and favor HBe seroconversion. PE can almost always control the course of these PAN without the addition of steroids or cyclophosphamide. An alternative therapy was also needed to lower PAN mortality and improve prognosis. In a retrospective study, we showed that, when steroids and immunosuppressants were prescribed to treat HBVPAN, the outcome was poorer than for non-viralPAN [15]. Therefore, based on the efficacies of antiviral agents against chronic hepatitis and of PE in PAN, together with Trrpo [ 1], who described the responsibility of HBV in the development of PAN, we combined the two therapies to treat HBV-PAN [23, 24].

Vidarabine. When this therapeutic strategy was first applied, the only available antiviral agent was vidarabine. After a 3-week course of vidarabine, administered after 1 week of steroids (1 mg/kg/d) and combined with PE, a full clinical recovery was obtained in three-quarters of the patients and HBe seroconversion was observed in nearly half of the patients.

215

IFN~ IFNet has replaced vidarabine and gives better results. In a series of patients, HBe seroconversion was obtained in two-thirds of the patients and HBsAg-to-anti-HBsAb seroconversion in half of them. The dose of 3 millions units, injected subcutaneously 3 times a week is recommended. Pegylated IFNct can also be prescribed. In HBV-PAN, the combination of antiviral agents (vidarabine or interferon-alpha-2a or 2-b (IFNot)) gave excellent overall therapeutic results [5] and should be preferred to conventional regimens that jeopardize the outcome, as described above. The efficacy of this strategy was confirmed in a series of 41 patients [5]. Twenty-three (56.1%) no longer exhibit serological evidence of replication and 80.5% recovered. Lamivudine. Lamivudine is an antiviral agent specifically designed for the treatment of HBV and HIV infections. In a small series of patients (personal data), we prescribed lamivudine (100 mg/day) in combination with PE, after a few days of steroids. Because lamivudine is eliminated by the kidney, its dose should be adapted to renal function and lower for patients with renal insufficiency. In that study [25], 9/10 patients recovered and 6/9 HBe seroconverted. One patient died. Plasma exchanges. In a few cases [26], the antiviral agent was prescribed alone. In our opinion, even if it is possible to obtain good clinical results in some patients, the severity of the disease in most patients requires therapy able to control immediately the severe or life-threatening manifestations of PAN. PE are able to rapidly clear the IC responsible for the disease. This rapid intervention is the most appropriate to control the disease. In our protocol, steroids are also prescribed for a few days to control as quickly as possible the clinical manifestations while waiting for IFNot or other antiviral agent efficacy to kick in. In our opinion, PE are not indicated because of their superiority to other medications but because they are, in combination with antiviral drugs, able to replace the deleterious therapies commonly used in virus-associated vasculitides with equivalent efficacy. The optimal schedule is as follows: 4 sessions/ week for 3 weeks, then 3 sessions/week for 2 to 3 weeks, followed by progressive lengthening of the

216

intervals between sessions. One plasma volume (60 ml/kg) is usually exchanged using 4% albumin as the replacement fluid. The circuit can be primed with starch. During the first weeks of treatment, the high number of PE can decrease the level of clotting factors and thereby lead to bleeding. Should bleeding occur, fresh frozen plasma can be used instead of albumin. Usually, the tolerance of PE is excellent.

Outcome and follow-up. The previously described short- and long-term outcomes of the patients showed the progressive improvement of seroconversion rates for patients receiving IFNtx. One of the major advances obtained under our antiviral strategy was the very rapid cure of HBV-PAN, even in its most severe forms. The majority of patients received the antiviral drug for a few weeks or months but PE, which were specifically given to control the acute manifestations of the disease, were stopped after 2 months. All signs of vasculitis were sometimes eliminated more quickly, with some of our patients recovering within 3 weeks. In the days following treatment onset, transaminase levels decrease. They usually return to normal within a few days or weeks. For patients who received vidarabine, a second increase of transaminase levels was observed prior to seroconversion. This usually mild immunological response was considered normal, as it attested the patient's ability to reject the virus via the hepatocytes. Nevertheless, transaminase levels can rise sharply and fulminant hepatitis can coincide with HBe seroconversion, as for one of our patients [27] who died of fulminant hepatitis several days after seroconversion. The response observed under IFNt~ is markedly different: transaminases normalize progressively and their levels do not rise after stopping the treatment, even when seroconversion has not been obtained. When HbeAb are detected, PE should be stopped to avoid the clearance of the newly synthesized antibodies. In a few cases, the antibody levels fluctuated, sometimes being present or absent. This Ag-Ab equilibrium can be very unstable and treatment should be continued. In such cases, it is more reliable to monitor virus activity by quantitative measurements of viral DNA. After recovery from the symptoms of the vasculitis, the clinician potentially faces two different

virological situations. First, replication continues, as demonstrated by the absence of HBeAb and the positivity of viral DNA, and PAN remission has been obtained but relapses may still occur. We therefore recommend prolonging IFNt~ administration for a total of 6 to 12 months, according to the viral response measured by quantitative DNA, and focusing on the treatment of viral hepatitis and not PAN, which has been cured. Second, Ab to HBe, at least, or to HBs, at best, are present, the patient can be considered cured and relapses will never occur. If, despite the presence of HBsAb, new manifestations of PAN appear, the clinician should consider the possibility of the vasculitis occurring coincidentally with virus infection but not linked to it. For patients who do not respond to one of the antiviral drugs, a combination of IFNtx and lamivudine could be tested.

5.2. Treatment of HCV-Associated Cryoglobulinemic Vasculitis Only a quarter of the patients with chronic hepatitis C achieved a sustained virological response [28]. A higher response rate was obtained with pegylated IFNtx. In the most recent study [29], 69% of the patients had responded at 48 weeks and achieved clinical recovery. Combining IFNt~ and ribavirin also increased the seroconversion rate in hepatitis C. No treatment is able to cure the majority of mixed cryoglobulinemias definitively and an optimal therapeutic strategy has not yet been clearly defined. Steroids and immunosuppressive drugs are commonly used to treat severe forms, but they have the same noxious effects as stated above. As we did for HBV-PAN, we devised a strategy associating antiviral drugs and PE for some patients [30]. For asymptomatic patients, there is no argument to treat, and monitoring could be sufficient. For patients with moderate symptoms of cryoglobulinemia vasculitis (e.g., arthralgias, purpura, sensory peripheral neuropathy), combining IFNtz and ribavirin is indicated. Ribavirin alone is not able to completely suppress viral replication but, in conjunction with IFNtx, viral replication was no longer detectable in 48% of the patients receiving the combination for 12 months [28]. We can expect that virus suppression will also be obtained in cryoglobulinemia.

Although the majority of the patients seen for symptomatic cryoglobulinemia have virus-positive polymerase chain reaction (PCR) assays, reflecting virus replication, a few of them remain serologically positive but become PCR-negative, reflecting past contamination. We also observed, in 2 of our patients with very severe vasculitis, the disappearance of the virus under antiviral treatment and PE but the persistence of clinical symptoms, which necessitated prolonged symptomatic treatment with PE.

Plasma exchanges. The indications of PE in HCVrelated cryoglobulinemia are controversial. Based on the effectiveness observed in our patients who failed to respond to other treatments and several reported failures of IFNt~, we recommend combining PE and antiviral drugs. PE should not be prescribed systematically for every newly diagnosed case of cryoglobulinemia because the majority of patients present no or very few symptoms, and we do not know, at present, whether or not treatment is indeed indicated in these pauci- or asymptomatic forms. PE are indicated for patients with symptoms requiting medical intervention. Purpura and sicca syndrome do not constitute such indications: the former regresses spontaneously and the latter is refractory to this treatment. In the case of glomerulonephritis due to cryoglobulinemia, PE combined with IFNtx can be effective but randomized controlled trials are needed to assess their contribution. PE are mainly indicated to treat rapidly progressing peripheral neuropathy and leg ulcers. The latter manifestation is often very severe and accompanied with pain that can require intensive therapy, including morphine. Under PE, arteriolar ulcers regress quickly and complete healing can be obtained in a few weeks. PE should be tapered progressively to avoid a rebound phenomenon due to the increased production of cryoglobulins as a consequence of the stimulation of the B-cell clones responsible for their production. Some of our patients remain PEdependent: clinical symptoms recur or worsen while tapering or after abrupt discontinuation of the sessions. Maintenance treatment should therefore be prescribed and the clinician has to try to determine the minimal number of sessions able to control the disease. When indicated, the number of sessions is not

217

clearly established. We propose the following schedule: 3 sessions a week for 3 weeks, then 2 sessions a week for 2 to 3 weeks, then 1 session every week or every 10 days until clinical symptoms disappear or the optimal clinical result is obtained.

term results are better and relapses are rare. This therapeutic approach should be applied and results will surely improve with the greater efficacy of new antiviral drugs.

5.3. Treatment of HIV-Associated Vasculitis

REFERENCES

Treatment of HIV-associated vasculitis has not yet been well defined, and steroids and immunosuppressants should be used cautiously, as they could favor the development of opportunistic infections and other clinical manifestations of AIDS. Again, the dual objective is to cure the vasculitis and to control HIV infection, and thus to avoid steroids and cytotoxic agents. The first objective is HIV-replication suppression, which is more easily obtained with the combination of 2, 3 or 4 antiviral agents. Nucleotide, non-nucleotide and protease inhibitors are the most frequently used families of drugs. Based on the presence of IC, we have proposed, as for other virus-associated vasculitides, to treat the patients with PE, using the same scheme as that for HBV-PAN. PE can clear IC and cytokines involved in the vasculitic process. In our clinical experience, this regimen was successful [3] and the patients we treated improved and vasculitis remissions were obtained. This strategy was also effective for patients with HCV and HIV or HBV and HIV coinfections. Some patients with cryoglobulinemia responded very quickly to this therapy. HCV cryoglobulinemia usually recurs after stopping PE and, unfortunately, anti-HIV drugs are not able to suppress cryoglobulin production. We also observed anti-cardiolipin Ab in a patient with HIV- and HCVrelated vasculitis. HIV-associated vasculitides appear to be a oneshot disease and do not recur and one to three months of therapy are usually sufficient to cure them.

1.

6. C O N C L U S I O N Virus-associated vasculitides are not uncommon and require a specific therapeutic strategy. The combination of antiviral agent(s) and PE is effective in the majority of patients and, because this strategy is adapted to the pathogenesis of the disease, long-

218

Tr6po C, Thivolet J. Hepatitis associated antigen and periarteritis nodosa (PAN). Vox Sang 1970;19:410-1. 2. Agnello V, Chung RT, Kaplan LM. A role for hepatitis C virus infection in type II cryoglobulinemia. N Engl J Med 1992;327:1490-5. 3. Gisselbrecht M, Cohen P, Lortholary O, Jarrousse B, Gayraud M, Lecompte I e t al. Human immunodeficiency virus-related vasculitis. Clinical presentation of and therapeutic approach to eight cases. Ann M6d Interne (Paris) 1998;149:398-405. 4. Corman LC, Dolson DJ. Polyarteritis nodosa and parvovirus B 19 infection. Lancet 1992;339:491. 5. Guillevin L, Lhote F, Cohen P, Sauvaget F, Jarrousse B, Lortholary O et al. Polyarteritis nodosa related to hepatitis B virus. A prospective study with long-term observation of 41 patients. Medicine (Baltimore) 1995;74: 238-53. 6. Prince AM, Tr6po C. Role of immune complexes involving SH antigen in pathogenesis of chronic active hepatitis and polyarteritis nodosa. Lancet 1971;1: 1309-12. 7. Mahr A, Guillevin L, Poissonnet M, Aym6 S. Prevalence of polyarteritis nodosa, microscopic polyangiitis, Wegener's granulomatosis and Churg-Strauss syndrome in a French urban population in 2000: a capturerecapture estimate [abstract]. Cleve Clin J Med 2002;69 (Suppl 2):170-1. 8. Cacoub P, Lunel-Fabiani F, Du LT. Polyarteritis nodosa and hepatitis C virus infection. Ann Intern Med 1992; 116:605-6. 9. Cacoub P, Maisonobe T, Thibault V, Gatel A, Servan J, Musset L et al. Systemic vasculitis in patients with hepatitis C. J Rheumatol 2001;28:109-18. 10. Quint L, D6ny P, Guillevin L, Granger B, Jarrousse B, Lhote F et al. Hepatitis C virus in patients with polyarteritis nodosa. Prevalence in 38 patients. Clin Exp Rheumatol 1991;9:253-7. 11. Servant A, Bogard M, Delaugerre C, Cohen P, D6ny P, Guillevin L. GB virus C in systemic medium- and small-vessel necrotizing vasculitides. Br J Rheumatol 1998;37:1292-4. 12. Eden A, Gaudet F, Waghmare A, Jaenisch R. Chromosomal instability and tumors promoted by DNA hypomethylation. Science 2003;300:455.

13. Calabrese L. The rheumatic manifestations of infection with the human imunodeficiency virus. Semin Arthritis Rheum 1989; 18:225-9. 14. Calabrese LH. Vasculitis and infection with the human immunodeficiency virus. Rheum Dis Clin North Am 1991;17:131-47. 15. Guillevin L, Lhote F, Jarrousse B, Bironne P, Barrier J, D6ny Pet al. Polyarteritis nodosa related to hepatitis B virus. A retrospective study of 66 patients. Ann M6d Interne (Paris) 1992;143:63-74. 16. Gayraud M, Guillevin L, le Toumelin P, Cohen P, Lhote F, Casassus Pet al. Long-term followup of polyarteritis nodosa, microscopic polyangiitis, and Churg-Strauss syndrome: analysis of four prospective trials including 278 patients. French Vasculitis Study Group. Arthritis Rheum 2001 ;44:666-75. 17. Jennette JC, Falk RJ, Andrassy K, Bacon PA, Churg J, Gross WL et al. Nomenclature of systemic vasculitides. Proposal of an international consensus conference. Arthritis Rheum 1994;37:187-192. 18. Cacoub P, Hausfater P, Musset L, Piette JC. Mixed cryoglobulinemia in hepatitis C patients. GERMIVIC. Ann M6d Interne (Paris) 2000; 151:20-9. 19. Rieu V, Cohen P, Andr6 MH, Mouthon L, Jarrousse B et al. Characteristics and outcome of 49 patients with symptomatic cryoglobulinaemia. Rheumatology (Oxf) 2002;41:290-300. 20. Gherardi R, Lebargy F, Gaulard P, Mhiri C, Bernaudin J, Gray F. Necrotizing vasculitis and HIV replication in peripheral nerves (letter). N Engl J Med 1989;321: 685-686. 21. McMahon BJ, Heyward WL, Templin DW, Clement D, Lanier AP. Hepatitis B-associated polyarteritis nodosa in Alaskan Eskimos: clinical and epidemiologic features and long-term follow-up. Hepatology 1989;9: 97-101. 22. Guillevin L, Jarrousse B, Lok C, Lhote F, Jais JP, Le THDD et al. Longterm followup after treatment of polyarteritis nodosa and Churg-Strauss angiitis with comparison of steroids, plasma exchange and cyclophosphamide to steroids and plasma exchange. A prospective randomized trial of 71 patients. The Coopera-

23.

24.

25.

26.

27.

28.

29.

30.

tive Study Group for Polyarteritis Nodosa. J Rheumatol 1991 ;18:567-74. Guillevin L, Merrouche Y, Gayraud M, Jarrousse B, Royer I, L6on A et al. P6riart6rite noueuse due au virus de l'h6patite B. D6termination d'une nouvelle strat6gie th6rapeutique chez 13 patients. Presse M6d 1988;17: 1522-6. Tr6po C, Ouzan D, Delmont J, Tremisi J. Sup6riorit6 d'un nouveau traitement 6tiopathog6nique gu6rissant la p6riart6rite noueuse due au virus de l'h6patite B par la combinaison d'une br~ve corticoth6rapie, de vidarabine et d'6changes plasmatiques. Presse M6d 1988;17: 1527-31. Guillevin L, Mahr A, Cohen P, Larroche L, Queyrel V, Loustaud-Ratti V e t al. Short-term corticosteroids then lamivudine and plasma exchanges to treat hepatitis B virus-related polyarteritis nodosa. Arthritis Rheum (in press) 2003. Avsar E, Savas B, T6ztin N, Ulusoy NB, Kalayci C. Successful treatment of polyarteritis nodosa related to hepatitis B virus with interferon alpha as first-line therapy [letter]. J Hepatol 1998;28:525-6. Guillevin L, Lhote F, L6on A, Fauvelle F, Vivitski L, Tr6po C. Treatment of polyarteritis nodosa related to hepatitis B virus with short term steroid therapy associated with antiviral agents and plasma exchanges. A prospective trial in 33 patients. J Rheumatol 1993;20: 289-98. Poynard T, Marcellin P, Lee S et al. Randomized trial of interferon alpha 2b plus ribavirin for 48 weeks or for 24 weeks versus alpha 2b plus placebo for 48 weeks for treatment of chronic hepatitis C virus. Lancet 1998;352: 1426--32. Zeuzem S, Feinman S, Rasenack Jet al. Peginterferon alpha-2a in patients with chronic hepatitis C. N Engl J Med 2000;343:1666-72. Cohen P, Nguyen QT, D6ny P, Ferri~re F, Roulot D, Lortholary O et al. Treatment of mixed cryoglobulinemia with recombinant interferon alpha and adjuvant therapies. A prospective study on 20 patients. Ann M6d Interne 1996;147:81-6.

219


Viral Infections and Autoimmune Hepatitis Sandro Vento ~and Francesca Cainelli

1Section of lnfectious Diseases, Department of Pathology, University of Verona, Verona, Italy

Abbrevations: AH: autoimmune hepatitis, ANA: antinuclear antibodies, ASMA: anti-smooth muscle antibodies, LKMI: liver-kidney microsomal type 1 antibodies, ASGPR: asialoglycoprotein receptor, HAV: hepatitis A virus, HCV: hepatitis C virus, CYP2D6: cytochrome P450IID6, HSV: herpes simplex virus, HBV: hepatitis B virus, CMV." cytomegalovirus, EBV: Epstein-Barr virus, TTV."TT virus.

1. INTRODUCTION Autoimmune hepatitis (AH) is (similarly to the other autoimmune diseases, with the exceptions of rheumatoid arthritis and thyroiditis) a rare disease; the etiology is unknown, the female predominance is strong and the prevalence in the population is around 0.01-0.02%. The disease is characterised serologically by a striking increase in serum IgG [1]. The natural course is marked by recurrent necro-inflammatory episodes within the liver lobules and at the interface with the portal tracts (piecemeal necrosis) eventually leading to cirrhosis and possibly liver failure, despite a responsiveness to corticosteroids (approximately 85%) which is among highest of all autoimmune diseases. Autoimmune hepatitis is divided into two main forms, types 1 and 2; the former is characterized by high titer (> 1:80) autoantibodies to nuclei (ANA, homogeneous or speckled pattern) reactive with chromatin and occasionally dsDNA and/or antibodies to substrates of smooth muscle (ASMA) reactive with F-actin microfilaments. Type 2 is overall rare (20 times less frequent than type 1) but usually occurs in younger patients and with typical type 1 antibodies to liver/ kidney microsomes (LKM1) directed against the

cytochrome P450 isoform 2D6. Despite attempts over the latest 20 years at establishing a reliable experimental animal model by Australian, Japanese and German research groups, this is still lacking. A genetic predisposition for the type 1 form of the disease is suggested by descriptions of impah'ed suppressor T cell function in patients and first degree relatives [2, 3], enhancing reactivity to autoantigens and favoring an antibody-dependent cellular form of cytotoxicity against self antigens expressed on the hepatocyte surface and by the association with HLA haplotypes, which varies however among different ethnic groups. DRB 1*0301 (DR3) and DRB 1*0401 (DR4) are associated with type 1 AH in white European and North American populations; DRB 1"0405 (DR4) is the principal association in Japanese and adult Argentine patients, whereas DRBl*0404 (DR4) is the main susceptibility allele in Mestizo Mexicans [4, 5]. Furthermore, in patients of northern European origin an association with the extended HLA haplotype A1, B8, DR3 is notable [6].

2. PROPOSED MECHANISMS OF LIVER CELL INJURY AND UNDERLYI-NG IMMUNE DEFECTS

The detection of IgG on hepatocytes isolated from liver biopsy specimens, the linear pattern of immunofluorescence [7], suggestive of a reaction against antigens diffusely distributed on the cell membrane [8], the predominance of B cells and T lymphocytes of helper phenotype in the mononuclear cell infiltrate in the liver [8] and the finding of antigen-presenting dendritic cells in the periportal areas of the lobule [9] all point to a role for antibodies synthe-

221

sized within the fiver and directed against surface membrane antigens expressed on hepatocytes in the establishment and persistence of liver cell injury in autoimmune hepatitis. In addition, antibodydependent cell-mediated cytotoxicity is the major mechanism of cytotoxicity in vitro in AH [ 10]. Although a number of autoantibodies reacting with antigens thought to be expressed on the liver cell surface have been described, especially in patients with active disease, the strongest candidate as a substantial contributor to hepatocellular damage in adult AH type 1 is directed against the asialoglycoprotein receptor (ASGPR), the liver cell-specific receptor for desialylated glycoproteins inserted on the hepatocyte sinusoidal plasma cell membrane [ 11 ]. T cells reactive to ASGPR are also detectable in patients with AH [2, 3] and T lymphocytes from healthy, unrelated subjects, added in a low ratio in culture, can specifically suppress the in vitro response to ASGPR of T cells from patients with autoimmune hepatitis type 1 [2, 3]. CD4+ve T cell inducers (specific for the asialoglycoprotein receptor) of CD8+ve suppressor T lymphocytes are present in the circulation of normal healthy subjects and defective in the patient population [12], appear to be activated in vivo in normal subjects and may play an essential role in controlling liver-directed autoreactivity. This liverspecific suppressor/inducer T cell defect might be genetically determined, as it has also been found in a high proportion of healthy relatives of patients with AH [3]. The hypothesis has been put forward [12] that the disease (in its far most frequent form, namely type 1) occurs only in those individuals who, in addition to having an ASGPR-specific suppressor/inducer defect, develop (owing to environmental triggers) T lymphocyte reactivity to the same autoantigen.

3. VIRUSES AS TRIGGERS IN ! AUTOIMMUNE HEPATITIS TYPE 1

If indeed a trigger is required to set off a sequence of events leading to autoimmune hepatitis in a predisposed individual, viruses are among the most likely candidates.

222

3.1. Measles Virus

The first candidate virus put forward was the measles virus, IgM antibodies to which were found to be raised in one study in patients with AH [ 13] and the genome of which reportedly persisted in patients' lymphocytes, suggesting an etiologic relationship [14]; however, the results reported by Robertson and colleagues were not reproduced by others [ 15] and the subsequent finding by a different group of a similar proportion of anti-measles antibody-positive individuals among patients with AH and the general population of comparable age did not support the suggestion [16]. Although measles virus appears overall not to be a frequent trigger of AH, it may occasionally act as such, and the report of a 4-yearold girl with a strong genetic predisposition who developed AH type 1 in strict temporal relation to this viral infection [ 17] supports this possibility. 3.2. Hepatitis A Virus

In 1991, following a 4-years prospective study, two cases of AH type 1 were described in association with hepatitis A virus (HAV) infection in patients who had first-degree relatives with the disease [18]. In that study two of the three patients with subclinical acute hepatitis A from a group of 58 healthy relatives of 13 patients with AH developed AH within 5 months. These 2 patients had the above mentioned suppressor/inducer T cell defect specific for the ASGPR, and antibodies to this autoantigen appeared during acute HAV infection and increased during follow-up. Anti-ASGPR antibodies are indeed commonly found transiently in acute hepatitis A [ 19] and appear before the detection of T cell immunity to the same antigen. This humoral autoimmune response is therefore unlikely to be driven by autoreactive T cells recognizing ASGPR. As immune reactions against viral antigens expressed on cell surfaces are present in patients with hepatitis A [20], it might be that helper T cells reactive to HAV antigens exposed on the surface of infected liver cells provide help for autoreactive B cells specific for an antigen (the ASGPR) coexpressed on the hepatocyte membrane. The results of the study published in 1991 also showed that T cell immunity to the ASGPR is detectable, during HAV infection, after the peak

in aminotransferase concentrations, suggesting that massive antigen release on hepatocyte damage is necessary to activate specific T helper lymphocytes. In the presence of functional ASGPR-specific suppressor/inducer T cells, this activation is transient and both specific T helper cells and the corresponding antibodies rapidly disappear. In contrast, in subjects with defective suppressor/inducer T cell control of immune responses to the ASGPR, T helper cell activation and antibody production continue and increase, and appears to be responsible for an autoimmune liver damage. Subsequently Rahaman and coworkers [21] described a middle-aged woman in whom serologically defined acute hepatitis A also triggered the onset of AH, and Huppertz and colleagues [22] reported a 7-year-old patient who developed the disease in association with preceding HAV infection and in the absence of a family history of AH. In a study reporting the clinical profile of 59 patients presenting with acute HAV infection, two women presented with concomitant AH type 1 [23]. Other similar cases have been reported in adults [24] and children [25]. Taken together, the above observations indicate that in the rare patients in whom chronic liver disease follows acute hepatitis A, the disease is AH type 1 and demonstrate conclusively that HAV infection can act, although in few cases, as a trigger for this organ-specific autoimmune disease in predisposed individuals. Which peculiar characteristics may underline the role of HAV as a trigger for an autoimmune liver disease? HAV is a non-cytopathic virus (either in vitro or in vivo), and elicits powerful cellular immune reactions against viral antigens expressed on the cell surface of infected hepatocytes. Although these reactions are considered to be responsible for the associated hepatocyte necrosis [20], it must be outlined that the liver damage resembles histologically an autoimmune damage, as even an experienced pathologist would have difficulties in distinguishing the histological appearance of the liver during acute hepatitis A from that of the liver of a patient with AH. Indeed the portal inflammatory infiltrate is similarly rich in plasma cells, piecemeal necrosis is frequently observed during acute hepatitis A (but not in other virally-induced acute liver disease) [26] and the main lymphoid subsets present in portal

areas are helper/inducer T cells and B cells [27]. Occurrence of autoantibodies typical of AH is also quite frequent during acute hepatitis A: anti-actin antibodies were first reported in 1983 and antinuclear, anti-smooth muscle and anti-ASGPR antibodies [19] are also transiently produced after the peak in aminotransferase concentrations. Moreover, in children from Argentina and Brazil (two countries with a still high prevalence of H~V infection) AH type 1 is associated with a unique HLA allele, DRBI*1301 (a particular HLA DR13 allele); the very same allele is associated with protracted liver damage and persistent high titers of anti-smooth muscle/actin antibodies following HAV infection [28]. Anti-smooth muscle/actin antibodies are not associated with HLA-DRB 1" 1301 if the infection is not prolonged, and it seems therefore that their long-lasting presence only in the protracted forms is associated with the sustained release of a self antigen related to a particular genetic background. As these children do not however progress to AH, it may be that in areas of high endemicity some unknown factors protect most individuals from progression to overt AH.

3.3. Epstein-Barr Virus In 1995, the extended follow-up of the very same cohort of relatives of patients with autoimmune hepatitis which had led to the identification of HAV as a possible trigger allowed the discovery of a second trigger, namely Epstein-Barr virus (EBV). Two women, aged 24 and 15 years, developed AH type 1 in strict temporal relation to EBV-induced infectious mononucleosis; in both cases, a defect in suppressor/inducer T cells controlling the response to the ASGPR had been identified prior to the viral infection, and anti-ASGPR antibodies persisted and increased after the viral illness [29]. As EBV is a polyclonal B lymphocyte activator and induces proliferation of specific B-cell clones [30], it may trigger AH through continuining proliferation and antibody production of ASGPR-specific B lymphocytes unchecked by defective suppressor/inducer T cells of the same specificity. Three further cases of AH type 1 following EBV infection have been described: one in a young Italian woman [31], one in an old Japanese man [32], and the latest in a 5year-old Italian girl [33]. The above reports clearly

223

define EBV as a trigger for AH type 1, albeit rare.

4. CAN A ROLE FOR VIRUSES BE ENVISAGED IN AUTOIMMUNE HEPATITIS TYPE 2?

The target of the hallmark of this form of autoimmune liver disease, i.e. liver-kidney microsomal type 1 antibody, is cytochrome P450IID6, a member of the hepatic P450 enzyme family. Manns and coauthors suggested that reactivity to the major epitope of CYP2D6 recognized by LKM1 antibodies may arise through a crossreactive response to hepatitis C virus (HCV) or herpes simplex virus type 1 (HSV1), as the aminoacids 310-324 of the envelope region E1 HCV and aminoacids 156-170 of immediate early protein IE175 of HSV1 share sequence homology with the immunodominant region, aminoacids 254-271, of CYP2D6, recognised by 85% of patients with AH type 2 [34]. Klein and coworkers demonstrated however that reactivity to the linear B-cell epitope of cytochrome P2D6 CYP2D6196_218is found in as many as 68% of patients with autoimmune hepatitis type 2 but in only 18% of LKMI+ve HCV-infected patients [35]. The complexity of the situation is further illustrated by the observation of crossreactive antibody recognition of homologous regions of HCV (NS5B HCW2985_2990)and cytomegalovirus (CMV) (EXON CMV130_135)antigens in LKMI+ve HCV-infected patients recognizing CYP2D6204_209 [36], and by the homologies between CYP2D6239_271and proteins from Salmonella typhimurium and human T lymphotropic viruses 1 and 2 [37]. Are all these crossreactivities of any importance in clinical practice, i.e. do they lead to autoimmune liver disease? Lenzi and coauthors first reported an impressively high occurrence of HCV infection in their cases of anti-LKMl-positive individuals with chronic liver disease [38], and suggested a role for HCV in the induction of AH type 2, but failed to provide direct evidence; a case report from Germany described two identical twins (HLA A1,B8,DR3), only one of whom was affected by AH and had LKM1 autoantibodies and anti-HSV 1 antibodies [39], suggesting a role for this virus in the development of disease. Dalekos and coworkers [40], while studying antibody titers and per-

224

forming epitope mapping of LKMl-positive sera from patients with chronic hepatitis C, found one young female patient with a very high LKM1 titer and autoantibodies directed against the epitope of amino acids 257-269 of CYP2D6 (preferentially recognized by patients with AH type 2) [35], who showed exacerbation of the disease under interferon treatment and was switched successfully to immunosuppressive therapy. This latter patient suffered from AH type 2 as the dominant cause for liver damage, but no proof was provided that HCV induced the appearance of the autoimmune disease. Furthermore, a survey conducted in 25 LKMI+ve patients in United Kingdom failed to demonstrate any association between AH type 2 and either antibodies to HSV- 1 or to HCV [41 ]. The most striking case of viral-induced AH type 2 is the case of a 29-year-old nurse who developed the disease following acquisition and rapid clearance of HCV [42]. The finding of IgM anti-LKM1 antibodies during acute HCV infection, the subsequent switch to IgG anti-LKM1 and the reactivity of these antibodies against amino acids 257-269 of CYP2D6 all point to a real (but unique) case of HCV-induced AH type 2. Overall, paradoxically (taking into account the immunological crossreactivities observed between CYP2D6 and a variety of microorganisms sequences) evidence for a causal role of viruses in triggering AH type 2 in humans is quite lacking, especially if one considers that the LKM1 antibodypositive cases of chronic HCV infection are generally middle aged men, whereas the true AH type 2 cases typically occur in young girls.

5. CAN OTHER VIRUSES TRIGGER, OR BE ASSOCIATED WITH, AUTOIMMUNE HEPATITIS?

Two case reports have linked hepatitis B virus infection to AH type 1: the first case followed acute HBV infection in a young woman who had eliminated the virus [43], the second developed in a chronic HBV cartier concurrently with the emergence of mutant, HBeAg-negative virus and was directly related to viral multiplication, as the disease went in remission following inhibition of HBV replication [44]. TT virus, a DNA virus first isolated in Japan from serum

of a patient with post-transfusion non-A-G hepatitis, does not appear to be related to AH in German or Japanese [45] patients. Contrasting reports refer to GB-virus/hepatitis G virus, a flavivirus unable to act alone as a cause of liver damage; while an Austrian study found a significantly increased prevalence of ongoing infection in patients with all types of AH [46], German [47] and Japanese authors [45] failed to find the same. SEN virus, a DNA virus related to the TTV family, has also been investigated and ruled out as a possible cause of AH [48]. Finally, retroviruses have been proposed as potential triggers of autoimmune diseases, including type 1 diabetes, Sj6gren's syndrome and primary biliary cirrhosis. Their role has not been investigated in AH, but it is not likely, as it is still unclear whether endogenous retroviruses can play a pathogenic role in autoimmune diseases in humans and they are more likely to be activated by the chronic inflammation associated with these diseases, the production of retroviral particles being in this latter case a mere epiphenomenon.

6. CONCLUSIONS Despite decades of complex studies employing sophisticated techniques, the obscure origins of autoimmunity and the even more obscure factors leading to overt autoimmune diseases are far from being uncovered, and genetic background, hormones and environmental agents are persistently listed as contributing factors. Viruses have been proposed as triggers for several autoimmune diseases, the most cited examples being B3 Coxsackie virus for myocarditis and B4 Coxsackie virus, cytomegalovirus and rubella for type 1 diabetes mellitus. Autoimmune hepatitis is no exception, but a role for a few viruses has been convincingly shown only in rare cases of AH type 1, and is unconvincing in AH type 2, the very form where one of the most held views in modem immunology, i.e. molecular mimicry, has been repeatedly invoked in the latest 14 years. Viruses can not act alone: even the association with the putative defect in ASGPR-specific suppressor/ inducer T cells does not necessarily induce AH, as one of the patients (an 18-year-old man) of the Italian cohort who was affected by AH following HAV infection had previously acquired EBV infection

without developing the autoimmune liver disease. It can therefore be argued that, even in patients with a genetic predisposition, a trigger must intervene at the "fight" time for the disease to develop; perhaps this "fight" time has to do with sexual hormones fluctuations or with concurrent drug administration, but again no clues are available in this respect. Immunologists need to go back to the innovative, provocative, simple and challenging experiments which used to nourish the specialty long ago and led to unexpected results, and which have now been replaced by extremely complicated laboratory investigation aiming almost exclusively at confirming trendy theories; provocative results such as those reported by Burns and coworkers [49] are the kind of nowadays rare, unexpected data which can really advance our understanding. In our opinion, the most important lesson to be learned form the studies that we described is that careful observation of single cases at the onset of viral infections and prospective follow-up of cohort of individuals at possible risk for autoimmune diseases, although painstaking, are the most reasonable ways of proceeding if a role for viruses in human autoimmune diseases (including autoimmune hepatitis) has to be uncovered. Laboratory investigation conducted in patients with established disease or only aiming (as usual) at confirming trendy theories (however attractive these may appear) will never give chances of significant breakthroughs.

REFERENCES 1.

2.

3.

4. 5.

International Autoimmune Hepatitis Group report. Review of criteria for diagnosis of autoimmune hepatitis. J Hepatol 1999;31:929-938. Vento S, Eddleston A. Autoimmunity and Liver Diseases. In: Popper H, Schaffner F, eds. Progress in Liver Diseases, vol IX. Philadelphia: WB Saunders, 1990;335-343. O'Brien CJ, Vento S, Donaldson PT et al. Cell-mediated immunity and suppressor T cell defects to liverderived antigens in families of patients with autoimmune chronic active hepatitis. Lancet 1986;1:350-353. BobergKM. Prevalence and epidemiology of autoimmune hepatitis. Clin Liver Dis 2002;6:347-359. CzajaAJ, Strettell MD, Thomson LJ et al. Associations between alleles of the major histocompatibility complex and type 1 autoimmune hepatitis. Hepatology

225

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

226

1997;25:317-323. Donaldson PT, Doherty DG, Hayllar KM et al. Susceptibility to autoimmune chronic active hepatitis: human leukocyte antigens DR4 and A1-B8-DR3 are independent risk factors. Hepatology 1991;13:701-706. Hopf U, Meyer zum Buschenfelde KH, Arnold W. Detection of a liver-membrane autoantibody in HbsAg-negative chronic active hepatitis. N Engl J Med 1976;294:578-580. Montano L, Aranguibel F, Boffill M, Goodall AH, Janossy G, Thomas HC. An analysis of the composition of the inflammatory infiltrate in autoimmune and HBVinduced chronic liver disease. Hepatology 1983;3: 292-297. Baradin KA, Desmet VJ. Interdigitating and dendritic reticulum cells in chronic active hepatitis. Histopathology 1984;8:657-662. Mieli Vergani G, Vergani D, Jenkins PJ et al. Lymphocyte cytotoxicity to autologous hepatocytes in HBsAgnegative chronic active hepatitis. Clin Exp Immunol 1979;38:16-25. Mizuno M, Brown WR, Vierling JM. Ultrastructural immunocytochemical localization of the asialoglycoprotein receptor in rat hepatocytes. Gastroenterology 1984;87:763-769. Vento S, O'Brien CJ, McFarlane IG, Williams R, Eddleston ALWF. T-cell inducers of suppressor lymphocytes control liver-directed autoreactivity. Lancet 1987;1:886-888. Christie KE, Haukenes G. Measles virus-specific IgM antibodies in sera from patients with chronic active hepatitis. J Med Virol 1983;12:267-272. Robertson DAF, Zhang SL, Guy EC, Wright R. Persistent measles virus genome in autoimmune chronic active hepatitis. Lancet 1987;2:9-11. Kalland KH, Endresen C, Haukenes G e t al. Measlesspecific nucleotide seuqences and autoimmune chronic active hepatitis. Lancet 1989;1:1390-1391. Mieli-Vergani G, Sutherland S, Mowat AP. Measles and autoimmune chronic active hepatitis. Lancet 1989;2: 688. Vento S, Cainelli F, Ferraro T, Concia E. Autoimmune hepatitis type 1 after measles. Am J Gastroenterol 1996;91:2618-2619. Vento S, Garofano T, Di Perri G, Dolci L, Concia E, Bassetti D. Identification of hepatitis A virus as a trigger for autoimmune chronic hepatitis type 1 in susceptible individuals. Lancet 1991;337:1183-1187. Vento S, McFarlane BM, McSorley CG et al. Liver autoreactivity in acute virus A, B and non-A, non-B hepatitis. J Clin Lab Immunol 1988;25:1-7. Vallbracht A, Maier K, Stierhof YD, Wiedmann KH, Flehmig B, Fleischer B. Liver-derived cytotoxic T cells

21.

22.

23.

24.

25.

26. 27.

28.

29.

30.

31.

32.

33.

34.

35.

in hepatitis A virus infection. J Infect Dis 1989;160: 209-217. Rahaman SM, Chira P, Koff RS. Idiopathic autoimmune chronic hepatitis triggered by hepatitis A. Am J Gastroenterol 1994;89:106-108. Huppertz HI, Treichel U, Gassel AM, Jeschke R, Meyer zum Buschenfelde KH. Autoimmune hepatitis following hepatitis A virus infection. J Hepatol 1995;23: 204-208. Tong MJ, E1-Farra NS, Grew MI. Clinical manifestations of hepatitis A: recent experience in a community teaching hospital. J Infect Dis 1995;171:S15-S18. Hilzenrat N, Zilberman D, Klein T, Zur B, Sikuler E. Autoimmune hepatitis in a genetically susceptible patient. Is it triggered by acute viral hepatitis A? Dig Dis Sci 1999;44:1950-1952. Ramonet MD, Gomez S, Morise Set al. Acute hepatitis A virus (HAV). Association with fulminant hepatic failure (FHF) and autoimmune hepatitis (AIH) in pediatric population. J Pediatr Gastroenterol Nutr 2000;31 (suppl 2): s116-s117. Teixeira MR Jr, Weller IV, Murray A et al. The pathology of hepatitis A in man. Liver 1982;2:53-60. Sciot R, Van den Oord JJ, De Wolf-Peeters C, Desmet VJ. In situ characterisation of the (peri)portal inflammatory infiltrate in acute hepatitis A. Liver 1986;6: 331-336. Fainboim L, Canero Velasco MC, Marcos CY et al. Protracted, but not acute, hepatitis A virus infection is strongly associated with HLA-DRB 1"1301, a marker for pediatric autoimmune hepatitis. Hepatology 2001;33:1512-1517. Vento S, Guella L, Mirandola F et al. Epstein-Barr virus as a trigger for autoimmune hepatitis in susceptible individuals. Lancet 1995;346:608-609. Sutton RNP, Emond RT, Thomas DB, Doniach D. The occurrence of autoantibodies in infectious mononucleosis. Clin Exp Immunol 1974;17:427-436. Aceti A, Mura MS, Babudieri S, Bacciu SA. A young woman with hepatitis after a sore throat. Lancet 1995 ;346:1603. Kojima K, Nagayama R, Hirama S e t al. Epstein-Barr virus infection resembling autoimmune hepatitis with lactate dehydrogenase and alkaline phosphatase anomaly. J Gastroenterol 1999;34:706-712. Nobili V, Comparcola D, Sartorelli MR, Devito R, Marcellini M. Autoimmune hepatitis type 1 after EpsteinBarr virus infection. Pediatr Infect Dis J 2003;22:387. Manns MP, Griffin KJ, Sullivan KF, Johnson EE LKM1 autoantibodies recognize a short linear sequence in P450IID6, a cytochrome P-450 monooxygenase. J Clin Invest 1991;88:1370-1378. Klein R, Zanger UM, Berg T, Hopf U, Berg PA. Over-

36.

37.

38.

39.

40.

41.

42.

lapping but distinct specificities of anti-liver-kidney microsome antibodies in autoimmune hepatitis type II and hepatitis C revealed by recombinant native CYP2D6 and novel peptide epitopes. Clin Exp Immunol 1999;118:290-297. Kerkar N, Mahmoud A, Muratori L et al. New viral triggers of autoreactivity to cytochrome P4502D6, the target of liver kidney microsomal autoantibody. J Hepatol 1999;30 (suppl 1):58. Gueguen M, Boniface O, Bernard O, Clerc F, Cartwright T, Alvarez E Identification of the main epitope on human cytochrome P450 IID6 recognized by anti-liver kidney microsome antibody. J Autoimmun 1991 ;4:607-615. Lenzi M, Ballardini G, Fusconi Met al. Type 2 autoimmune hepatitis and hepatitis C virus infection. Lancet 1990;335:258-259. Manns M, Koletzko S, Lohr H et al. Discordant manifestation of LKM-1 antibody positive autoimmune hepatitis in identical twins. Hepatology 1990;12:840. Dalekos GN, Wedemeyer H, Obermayer-Straub P et al. Epitope mapping of cytochrome P4502D6 autoantigen in patients with chronic hepatitis C during c~ interferon treatment. J Hepatol 1999;30:366-375. Gregorio GV, Bracken P, Mieli-Vergani G, Vergani D. Prevalence of antibodies to hepatitis C and herpes simplex virus type 1 is not increased in children with liver kidney microsomal type 1 autoimmune hepatitis. J Pediatr Gastroenterol Nutr 1996;23:534-537. Vento S, Cainelli F, Renzini C, Concia E. Autoimmune

43.

44.

45.

46.

47.

48.

49.

hepatitis type 2 induced by HCV and persisting after viral clearance. Lancet 1997;350:1298-1299. Laskus T, Slusarczyk J. Autoimrnune chronic active hepatitis developing after acute type B hepatitis. Dig Dis Sci 1989;34:1294-1297. Becheur H, Valla D, Loriot MA, Attar A, Bloch F, Petite JP. Concurrent emergence of hepatitis B e antigen-negative hepatitis B virus variant and autoirmnune hepatitis cured by adenine arabinoside monophosphate. Dig Dis Sci 1998;43:2479-2482. Nishiguchi S, Enomoto M, Shiomi Set al. GB virus C and TT virus infections in Japanese patients with autoimmune hepatitis. J Med Viro12002;66:258-262. Tribl B, Schoniger-Hekele M, Petermann D, Bakos S, Penner E, Muller C. Prevalence of GBV-C/HGV-RNA, virus genotypes, and anti-E2 antibodies in autoimmune hepatitis. Am J Gastroenterol 1999;94:3336-3340. Heringlake S, Tillmann HL, Cordes-Temme P, Trautwein C, Hunsmann G, Manns MP. GBV-C/HGV is not the major cause of autoimmune hepatitis. J Hepatol 1996;25:980-984. Shibata M, Wang RYH, Yoshiba M, Shih JWK, Alter HJ, Mitamura K. The presence of a newly identified infectious agent (SEN virus) in patients with liver diseases and in blood donors in Japan. J Infect Dis 2001; 184:400--404. Burns JB, Bartholomew BD, Lobo ST. In vivo activation of myelin oligodendrocyte glycoprotein-specific T cells in healthy control subjects. Clin Immunol 2002;105:185-191.

227


Viral Infections and Type 1 Diabetes Hee-Sook Jun and Ji-Won Yoon

Center for Immunologic Research and Department of Microbiology and Immunology, The Chicago Medical School North Chicago, IL, USA

1. INTRODUCTION Type 1 (insulin-dependent) diabetes mellitus results from the destruction of insulin-producing [~ cells in the pancreatic islets [1-8]. Although strong genetic predisposition is associated with the development of type 1 diabetes [7, 9-12], there is considerable evidence that environmental factors play an important role in the etiology of type 1 diabetes [13-16]. The concordance rate for type 1 diabetes in identical twins is only 25-60% [17-19]. Epidemiological studies also indicate a role for environmental factors in the development of type 1 diabetes. Variations in the incidence of type 1 diabetes have been observed among populations with similar genetic backgrounds but from different geographical areas, and between migrant populations and their indigenous population [20]. Environmental factors that are suspected to be involved in the initiation and/or progression of [3 cell destruction leading to type 1 diabetes include dietary composition, [3 cell toxins, and viruses. Viruses have long been suspected to contribute to the development of human type 1 diabetes, largely by temporal and geographic association between the disease and viral infection, serological evidence of infection in patients recently diagnosed with type 1 diabetes, and isolation of viruses from the pancreas of diabetic patients in a few cases [21]. As well, some viruses have been reported to be associated with the development of type 1 diabetes in animals. Viruses can induce type 1 diabetes either by direct infection and cytolytic killing of [~ cells or by triggering [3 cell autoimmunity with or without direct infection of the [~ cells.

In this review, we will discuss viruses that are considered to be associated with the pathogenesis of type 1 diabetes in humans and animals (Table 1) and the possible mechanisms by which they induce this disease.

2. VIRUSES AND TYPE 1 DIABETES IN HUMANS 2.1. Coxsackie B Virus and Enterovirus

Considerable evidence indicates that Coxsackie B virus, especially the B4 serotype, is associated with the development of type 1 diabetes. Several epidemiological studies have shown high frequencies of anti-Coxsackie B viral antibodies in children newly diagnosed with type 1 diabetes as compared to non-diabetic subjects [22-31]. In addition, T cell responses against a Coxsackie B viral protein were observed in new-onset type 1 diabetic patients [32-34]. While these studies support the involvement of Coxsackie B virus in the development of human type 1 diabetes, other epidemiological studies have come to the opposite conclusion. Several studies found no evidence for a correlation between the onset of type 1 diabetes and Coxsackie B viral infections [35-38], and other studies found higher levels of anti-Coxsackie B virus-specific antibodies in non-diabetic control subjects than in recent-onset type 1 diabetic patients [39, 40]. The controversy may arise from the nature of the virus and genetically determined host factors. There are different variants of virus within each serotype. For example, Prabhakar et al [41] isolated thirteen variants of

229

Table 1. Viruses associated with the development of type 1 diabetes Virus RNA VIRUSES: Picornaviridae Coxsackie B virus Encephalomyocarditis virus Mengovirus Foot-and-mouth disease virus Ljungan virus Retroviridae Retrovirus Togaviridae Rubella virus

Bovine viral diarrhoea-mucosal disease virus ParamyxovilTdae Mumps virus Reoviridae Rotavirus Reovirus DNA VIRUSES: Parvoviridae Kilham rat virus Herpesviridae Cytomegalovirus Epstein-Barr virus

Coxsackie B4 virus. In another study, four variants of Coxsackie B4 virus were tested and only one was found to be diabetogenic, while the remaining three were not [42]. This is an indication of the possible rarity of diabetogenic variants of Coxsackie B4 virus. Also, it is difficult to distinguish between diabetogenic and non-diabetogenic variants using routine neutralizing antibody or ELISA assays, since antibodies against one variant cross-react with other variants. Therefore, if a person is exposed to a more common variant of Coxsackie B4 virus before exposure to a more rare diabetogenic variant of the same serotype, that person will have already developed antibodies against the non-diabetogenic variant, which will neutralize the diabetogenic variant during a subsequent infection; thus, the person will

230

Host

Involvement of genetic factors

Humans Mice Non-human primates Mice Hamster Mice Pigs, cattle Bank voles

Not determined Yes Yes Yes Yes Yes Not determined Not determined

Humans

Mice

Not determined Yes

Humans Hamsters Rabbits Cattle

Not determined Not determined Not determined Not determined

Humans

Yes

Humans Mice

Not determined Yes

Rats

Yes

Humans Degu Humans

Not determined Not determined Not determined

not become diabetic, even if he or she is genetically predisposed to the disease. If this person is a subject in an epidemiological study, then the results will be misleading, as the lack of diabetes will not be a result of lack of exposure to a diabetogenic Coxsackie B virus, and no correlation between Coxsackie B viral infection and incidence of diabetes will be found. In contrast, outbreaks of diabetogenic virus in certain areas before outbreaks of non-diabetogenic virus will result in a high correlation between Coxsackie B viral infection and the development of diabetes. In addition, there are genetically determined differences in susceptibility to virus-induced diabetes, as has been shown in experiments using different strains of mice infected with Coxsackie B4 virus [43]. It is thought that humans, as well, will not

become diabetic when infected with diabetogenic Coxsackie B4 virus unless they are genetically predisposed to developing the disease. Thus, the correlation between Coxsackie B viral infection and the development of diabetes seen in some epidemiological studies but not in others may depend on the genetic makeup of the virus and the genetic background of the patients. In addition to these epidemiological studies, there are many anecdotal reports describing the development of type 1 diabetes in patients with recent or concurrent Coxsackie B viral infections [42, 44-52]. Direct evidence supporting the involvement of Coxsackie B viral infection in the development of type 1 diabetes comes from studies in which Coxsackie B4 and B5 viruses have been isolated from the pancreata of patients with acute-onset type 1 diabetes and the isolates have been shown to induce diabetes in susceptible strains of mice [53, 54]. A patient who died of diabetes after Coxsackie viral infection showed lymphocytic infiltration of the islets and ~ cell necrosis at autopsy (Fig. 1) and the isolated virus was able to induce diabetes in SJL mice but not in C57BL6, Balb/c or CBA mice [53]. Additional evidence includes a study in which Coxsackie B virus-specific antigens were detected in islets showing marked 13cell damage in children who developed diabetes following severe infections by these viruses and died [55]. As well, in vitro studies have shown that Coxsackie B viral infection can impair human islet cell metabolism. Infection of human 13cells with Coxsackie B3 and B4 viruses decreased insulin content beginning at 24 h after infection and the decrease in insulin roughly paralleled the increase in viral titer [53, 56, 57]. Although the exact mechanisms for Coxsackie virus-induced type 1 diabetes are not known, direct destruction of 13cells by cytolytic infection, molecular mimicry, and bystander activation of pre-existing autoreactive T cells have been suggested. Coxsackie B virus may directly infect pancreatic ~ cells and destroy them by cytolysis, as has been found in mice [43, 58]. Other studies have suggested that molecular mimicry could underlie autoimmune responses that result in [3 cell damage after Coxsackie viral infection. P-2C, a non-capsid protein of Coxsackie B4 virus, has sequence homology with glutamic acid decarboxylase (GAD), a putative autoantigen expressed by [~ cells [59]. Moreover, infection with

the virus increases expression of GAD by [~ cells [60]. Antibodies that react with both P2-C and GAD have been detected in type 1 diabetic patients [61]. However, this hypothesis is not supported by studies of antibodies produced by lymphocytes isolated from a newly diagnosed type 1 diabetic patient. Four of six antibodies studied recognized and bound to the region of GAD65 that is homologous to P2C, but none cross-reacted with P2-C itself or with any other Coxsackie B4 viral proteins. The lack of cross-reactivity between these two proteins may be due to differences in secondary or tertiary structure [62]. On the other hand, the capacity of murine T lymphocytes to cross-react with P2-C and GAD is associated with a diabetes susceptibility allele; cross-reactive T-cell recognition of GAD65 could therefore contribute to the initiation or amplification of autoimmune responses against the ~l cell, and perhaps contribute to the association of type 1 diabetes with certain human leukocyte antigen (HLA) alleles [63]. Another hypothesis is that Coxsackie B virus induces diabetes via "bystander" activation of autoreactive T cells against islet antigens. In mice with diabetes-susceptible major histocompatibility complex (MHC) alleles, these viruses did not accelerate the development of diabetes, whereas transgenic mice carrying a T cell receptor specific for an islet autoantigen rapidly became diabetic. This suggests that Coxsackie B virus induces diabetes by direct local infection, leading to inflammation, tissue damage, and the release of sequestered islet antigens that results in the re-stimulation of resting autoreactive T cells [64]. A further possibility is that a defective Coxsackie B virus, lacking the usual high lytic activity, could cause persistent infection of [3 cells, resulting in bystander activation of autoreactive T cells [65]. This hypothesis would be consistent with evidence of continuing Coxsackie B viral infection in other diseases. Interferon (IFN)-t~ production associated with hyperexpression of HLA type I antigens, was identified in the 13cells of three out of four children who died from Coxsackie B viral pancreatitis [66]. In addition, it was reported that plasma IFN-ct levels were elevated in some type 1 diabetic patients, and that this was associated with Coxsackie B viral infection [67]. Persistent Coxsackie B viral infection of the 13 cells might stimulate them to synthesis and release IFN-ct,

231

Figure 1. Pancreatic sections from a non-diabetic subject and a diabetic, Coxsackie B4 virus-infected patient. (A) Section of a pancreas from a non-diabetic subject, showing a single intact islet of Langerhans surrounded by acinar cells (x 160). (B-F) Sections from different locations of a pancreas from a Coxsackie B4 virus-infected 10 year-old boy who died after acute onset of type 1 diabetes. (B) Islet with moderate insulitis (x 230). (C) Atrophied islet with severe insulitis (x 160). (D) Lymphocytic infiltration in the periphery of the islet. (E) Islet with extensive inflammatory infiltrate, loss of islet architecture and severe islet destruction. (F) Islet with severe necrosis of the 13cells and only a few lymphocytes remaining in the islet. which in turn could induce hyperexpression of HLA class I antigens and the production of chemokines that recruit and activate macrophages and autoreactive T cells. These activated immune cells could then kill the 13 cells, resulting in type 1 diabetes (Fig. 2). Finally, Coxsackie viral infections may be

232

involved in the pathogenesis of type 1 diabetes by acting as the terminal insult in individuals who have already lost substantial [3-cell mass through ongoing autoimmune damage. Destruction of a critical number of residual cells would result in the abrupt onset of type 1 diabetes.

Figure 2. Hypothetical scheme of a possible mechanism of Coxsackie B4 virus-induced diabetes by persistent infection. Coxsackie B4 virus may persistently infect pancreatic 13cells and induce the expression of IFN-~. In turn, IFN-~ may induce the expression of chemokines and MHC-I on the 13cells. These chemokines may recruit macrophages and T cells to the pancreatic islets, which then are activated and may kill 13cells in conjunction with the hyperexpressed MHC class I molecules, resulting in the development of type 1 diabetes.

233

There is considerable information that type 1 diabetes may be associated with infection with other enteroviruses in addition to Coxsackie B virus. Epidemiological studies show a strong association between the development of type 1 diabetes and enteroviral infection [25, 68-70], and antibodies against enterovirus [26-31, 71 ] or T cell responses to enterovirus [32-34, 72] have been detected in new-onset diabetic patients. In addition, enteroviral RNA was detected with high frequency in the serum or lymphocytes of type 1 diabetes patients [44, 45, 73-75]. Some case studies have reported that type 1 diabetes developed after enteroviral infection [76-78]; however the role of enteroviruses in type 1 diabetes is still poorly understood [79]. There is also some indirect evidence that Coxsackie A virus may also be associated with type 1 diabetes [80, 81 ].

2.2. Mumps Virus Mumps virus was one of the first viruses reported to be associated with human type 1 diabetes. Several cases were reported in which mumps infection appeared to precede the onset of type 1 diabetes [82, 83]. Mumps-related type 1 diabetes may have an autoimmune basis. Some children with mumps parotitis develop islet-cell antibodies [84], and there is evidence that the virus may induce an autoimmune response against ~ cells, or might intensify a pre-existing autoimmune attack. Mumps infection of a human l-cell line (insulinoma) in vitro induced the release of interleukin (IL)-I and IL-6 and also up-regulated the expression of HLA class I and II antigens [85, 86]. The production of interleukins and increased or aberrant expression of HLA antigens by 13 cells may be crucial steps in provoking autoimmune ~ cell damage. However, studies on the impact of mumps vaccination on the development of type 1 diabetes reported that there is no association with childhood mumps vaccinations and the development of islet autoimmunity [87] or type 1 diabetes [88], suggesting that mumps virus may not be a good candidate virus for the induction of diabetes. In contrast, one study reported that the elimination of mumps infections by vaccination may have been responsible for the decreased risk of developing type 1 diabetes over the time period studied [89], suggesting that infection with mumps virus may be associated with the development of type 1 diabetes.

234

Further studies are required to determine whether mumps virus is definitely involved in the development of type 1 diabetes.

2.3. Rubella Virus Rubella virus has been implicated in type 1 diabetes, because approximately 12-20% of patients with congenital rubella syndrome (CRS) develop diabetes by 5-20 years of age [90-98]. Islet cell and anti-insulin antibodies were found in 50-80% of diabetic patients with CRS, whereas these antibodies were present in only 20% of nondiabetic CRS patients, suggesting an underlying autoimmune disorder. There was also an increased frequency of HLA-DR3 in patients with CRS, suggesting some genetic susceptibility might be involved in the development of type 1 diabetes in CRS patients [99]. While rubella virus appears to be involved in the development of type 1 diabetes in patients with CRS, more research is required to discover if infection by rubella virus after birth plays any role in the induction of type 1 diabetes. In vitro studies have shown that human islets are susceptible to rubella infection. Human fetal islets exposed to rubella virus showed rubella viral antigens in both 13and non-[~ cells and had lowered levels of insulin production [ 100], although without any observable cytopathology [ 101 ]. Several mechanisms have been suggested by which rubella viral infection may induce type 1 diabetes. Rubella viral infection may alter antigens in the plasma membrane of infected 13 cells that may be perceived as foreign by the host's immune system, resulting in the induction of ~ cell-specific autoimmunity. Altematively, rubella viral infection may induce autoimmune type 1 diabetes by molecular mimicry. It was found that a monoclonal antibody directed against a rubella capsid protein crossreacted with extracts from rat and human islets and rat insulinoma cells, and the shared epitope was shown to be part of a unidentified 52 kD [3 cell protein [ 102]. It has also been shown that T cells from type 1 diabetic patients cross-react with epitopes in rubella viral proteins and the ~ cell isoform of GAD [ 103]. These results suggest that rubella viral infection may lead to the generation of viral antigenspecific cytotoxic T cells that also recognize 13 cell antigens in susceptible individuals.

2.4. Cytomegalovirus (CMV) Case reports have described type 1 diabetes developing in CMV-infected individuals, including a child with congenital CMV infection [104] and an adult with severe CMV infection that caused acute pancreatitis and rhabdomyolysis [ 105]. In addition, CMV infection can cause [~ cell damage under certain circumstances and characteristic inclusion bodies have been found in 13 cells in children who died with disseminated CMV [ 106]. A study showed that 20% of type 1 diabetic patients had CMV genomic DNA in their lymphocytes, compared to only 2% of normal controls. Furthermore, 80% of patients who had both antiCMV antibodies and the CMV genome in their lymphocytes also had islet cell autoantibodies [107]. Another study found that non-diabetic siblings of type 1 diabetes patients had a significant association between high titres of anti-CMV antibodies and islet cell autoantibodies, but no correlation between anti-CMV antibodies and HLA-DR antigens [ 108]. Although these results are largely circumstantial, they suggest that chronic CMV infection may be associated with islet cell autoantibody production, but that other factors may be needed for the development of clinical type 1 diabetes. Evidence for molecular mimicry is the finding that human CMV can induce an islet cell antibody that reacts with a 38 kD autoantigen expressed in human pancreatic islets [ 109]. Also, a study showed that a CD4 § T cell clone reactive to GAD65 isolated from a prediabetic Stiffman syndrome patient cross-reacted with a peptide of human CMV major DNA binding protein, suggesting that human CMV may be involved in the induction of autoimmunity by molecular mimicry of the 13 cell autoantigen, GAD65 [110]. However, further investigation is required to determine whether CMV is truly involved in the development of type 1 diabetes.

2.5. Retrovirus Many studies have suggested that retroviruses may be implicated in autoimmune disease, including type 1 diabetes [ 111, 112]. Retroviral-like particles were found in patients with multiple sclerosis and Sj6gren's syndrome [ 113, 114]. Nucleotide sequence homology was found between human retrovirus and

self-antigens, in particular between ribonucleoproteins and the p30 C-type retroviral gag gene product [115-117]. Anti-insulin autoantibodies from type 1 diabetic patients and their non-diabetic, first-degree relatives have been found to cross-react with retroviral p73 antigen, suggesting that endogenous retroviruses may be involved in the pathogenesis of type 1 diabetes [ 118]. A novel human endogenous retroviral gene, designated IDDMK1.222, was reported to be expressed in the plasma of recent-onset type 1 diabetes patients but not in non-diabetic control subjects [ 119]. This virus was thought to belong to the mouse mammary tumor virus-related family of human endogenous retroviruses (HERV)-K. However, careful studies have shown that the IDDMK1.222 sequence was not present in either the plasma or peripheral lymphocytes from either diabetic or control subjects [120-122]. Instead, a related human endogenous retrovirus with 90-93% sequence homology with IDDMK1.222 was present with equal frequency in both diabetic and nondiabetic subjects [120]. These human endogenous retroviruses are therefore unlikely to play a role in the development of autoimmune type 1 diabetes in humans [120-122]. Even though it appears that the endogenous retroviral gene homologous with IDDMK1.222 is not associated with type 1 diabetes, it does not necessarily exclude the involvement of other human retroviruses or endogenous retrovims genes in the pathogenesis of autoimmune diabetes. An interesting report showed that the expression of the defective retroviral gene, the HERV-K18 provirus encoding superantigen, is induced by IFN-~ and subsequently stimulates V137 T cells [123], which was correlated with the onset of type 1 diabetes. Whether the HERV-K18 provirus is truly involved in the development of autoimmune diabetes remains to be determined.

2.6. Epstein-Barr Virus (EBV) EBV has been implicated in the etiology of several autoimmune diseases [124], and a few cases have been reported to be linked with the onset of type 1 diabetes [125]. It is suggested that EBV may be potentially capable of triggering autoimmune type 1 diabetes by molecular mimicry, since an eleven amino acid sequence of the EBV protein, BOLF1, was found to be homologous with residues in the

235

Asp-57 region of the HLA-DQW8 13 chain peptide [ 126]. It was also found that a pentapeptide sequence in the Asp-57 region of the HLA-DQI] chain is successively repeated six times in the EBV-BERF4 epitope [127, 128]. Two patients who produced antibodies against this epitope during acute EBV infection developed type 1 diabetes soon thereafter, while five individuals also acutely infected but not producing antibodies against this epitope did not develop type 1 diabetes [ 128]. Further investigation is needed to find the relationship between EBV and type 1 diabetes. 2.7. Other Viruses There is circumstantial evidence that hepatitis A virus [129], varicella zoster virus [130], measles virus [ 130], polio virus [ 130], influenza virus [ 131 ], and rotavirus [132] may be associated with the development of type 1 diabetes in humans. However, whether these viruses are truly associated with the development of type 1 diabetes remains to be determined.

3. VIRUSES AND TYPE 1 DIABETES IN ANIMALS 3.1. Encephalomyocarditis (EMC) Virus EMC virus selectively infects pancreatic 13 cells [133] and induces diabetes in genetically predisposed mice by the destruction of pancreatic [3 cells [134, 135]. This virus has been the most thoroughly studied of the diabetogenic viruses in animals [ 136]. The EMC virus has two antigenically indistinguishable variants: EMC-D virus causes diabetes by direct cytolytic destruction of 13 cells in over 90% of the animals it infects, whereas EMC-B virus is completely non-diabetogenic [137]. However, diabetes only develops after EMC-D viral infection in some strains of mice such as SJL/J, SWR/J, DBA/1J, and DBAJ2J, whereas C57BL/6J, CBA/J, and AKR/J strains are resistant. Susceptibility to EMC-D virus-induced diabetes is determined by a single autosomal recessive gene [138], which may modulate the expression of viral receptors on [3 cells [135, 139-141]. Examination of the complete nucleotide

236

sequence of the genomes of the EMC-D and EMCB variants showed that they were different in only 14 nucleotide positions [142, 143]. Further studies using several mutant viruses showed that only one amino acid, alanine at the 152nd amino acid residue of the major capsid protein VP1, is critical for diabetogenicity of the EMC virus [ 144, 145]. Threedimensional molecular modeling of the VP 1 protein showed that the van der Waals interactions are greater and the residues surrounding position 152 are more closely packed in recombinant chimeric viruses containing Thr, Ser, Pro, Asp, or Val in this position than in recombinant chimeric viruses containing Ala in the same position. The surface area surrounding Ala at position 152 of VP1 is more accessible, thus increasing the availability of the binding sites for attachment to 13 cell receptors and resulting in viral infection and the development of diabetes [ 146]. Two different animals models have been established with respect to pathogenic mechanisms for EMC virus-induced diabetes. The first model involves animals infected with a high dose (105 plaque-forming units [PFU]/mouse) of EMC virus, in which replication of the virus within the 13 cells plays a major role and recruitment of macrophages plays a minor role in 1~cell destruction. In contrast, the second model involves animals infected with a low dose (< 102 PFU/mouse) of EMC virus, in which activated macrophages that are recruited to the [3 cells play a major role and replication of virus within the [3 cells plays a minor role in [3 cell destruction [136]. In low-dose EMC virus-infected mice, it was found that macrophage-derived soluble mediators play a critical role in the destruction of pancreatic I] cells. Further study revealed that EMCD virus infects and activates macrophages, but does not replicate within them [147]. The expression of macrophage-derived soluble mediators such as IL-113, tumor necrosis factor (TNF)-c~, and inducible nitric oxide synthase (iNOS) was selectively detected in the pancreatic islets of mice infected with a low dose of EMC-D virus. In addition, treatment of EMC-D virus-infected mice with antibody against IL-1 [~ or TNF-c~ or with the iNOS inhibitor, aminoguanidine, exhibited a significant decrease in the incidence of diabetes [148]. These results suggest that macrophage-derived soluble mediators play a critical role in the destruction of pancreatic

[3 cells resulting in the development of diabetes in mice infected with a low dose of EMC-D virus. Studies have shown that tyrosine kinase signaling pathways are involved in the activation of macrophages by EMC-D virus infection. Extracellular signal-regulated kinases (ERK)I/2, p38 mitogenactivated protein kinase (MAPK), and c-Jun-terminal activation kinase (JNK) were activated in macrophages after EMC-D viral infection. Treatment of mice infected with a low dose EMC-D virus with a tyrosine kinase inhibitor, AG126, decreased the incidence of diabetes and suppressed the production of IL- 1[3, TNF-ot, and iNOS in the pancreata of these mice as compared with vehicle-treated control mice [147]. In addition, it was found that hematopoietic cell kinase (hck), a Src family kinase, is involved in the activation of macrophages by EMC-D viral infection and that treatment of EMC-D virusinfected mice with a Src kinase inhibitor prevented the development of diabetes [ 149].

3.2. Mengovirus Mengovirus induces fatal encephalitis in mice similar to EMC virus and is antigenically similar to EMC virus. Plaque purification of Mengovirus resulted in the isolated clone, Mengovirus-2T, that could cause diabetes in strains of mice resistant to EMC-D viral infection [150]. Marked ~ cell necrosis, severe inflammatory infiltration of the islets and decreased insulin content without evidence of autoimmune responses were observed in Mengovirus-2T-infected mice [ 150]. It appears that Mengovirus-2T causes diabetes by directly infecting [~ cells.

3.3. Coxsackie B4 Virus Coxsackie B virus causes diabetes in susceptible strains of mice, such as SJL/J and SWR/J mice. Coxsackie B4 virus isolated from a patient with acute-onset diabetes or patients with Coxsackie viral infection induced abnormal glucose tolerance and transient hyperglycemia in infected mice [151]. A diabetogenic Coxsackie B4 virus generated by repeated passaging of the virus in 1~ cell cultures, which enhanced its 13 cell tropism [43], resulted in lymphocytic infiltration of the islets and [3 cell destruction in infected mice. During the acute

phase of Coxsackie B4 viral infection, antigens from this virus were observed in the pancreatic islets [43]. Some clinical isolates of Coxsackie B4 virus produced 13cell damage in vivo in mice [152] or in vitro [153]. The E2 variant of Coxsackie B4 virus isolated from a child who died from systemic infection induced a diabetes-like syndrome in infected mice, such as abnormal glucose tolerance and transient hyperglycemia [ 154]. Hyperglycemia developed between 6-8 weeks after the viral infection and CD4 § T cells predominated in the pancreatic infiltrates of the infected mice [155]. However, the role of these CD4 + T cells in the development of diabetes in E2 variant-infected mice remains to be determined. Interestingly, it was reported that the expression of GAD65, which has a sequence homology with a non-structural protein (P2-C) of Coxsackie B4 virus, is increased in the islets at 72 h after E2 variant-infection [60], suggesting that molecular mimicry may play a role in the development of Coxsackie B4 virus-induced diabetes. Studies on the Coxsackie B4 viral genome have identified the amino acid residues responsible for the virulence of the virus [156]; however only preliminary work has been done on identifying the residues responsible for diabetogenicity. Sequence comparison data between the diabetogenic E2 variant of Coxsackie B4 virus and the prototype nondiabetogenic JVB strain of virus [157] revealed 111 amino acid differences [158]. Another group sequenced the genome of a ~ cell-tropic variant of Coxsackie B4 virus JVB and compared it with the sequence of the prototype strain and found only seven amino acids that were different [159]. The identification of the specific amino acids responsible for the diabetogenicity of Coxsackie B4 virus remain to be determined. In addition to causing diabetes in susceptible mouse strains, Coxsackie B4 virus can also cause a diabetes-like syndrome in certain species of monkeys. Monkey [3 cell passaged Coxsackie B4 virus was shown to impair insulin secretion and glucose tolerance in the Patas monkey, but had no apparent diabetogenic effects in other primates such as Cebus, Cynomotgus, and Rhesus monkeys [160], suggesting that genetic factors are involved in susceptibility to Coxsackie B4 virus-induced diabetes in monkeys. From the results of the above research on humans

237

and animals, it is speculated that Coxsackie B virus, especially the B4 serotype, may play a role in the development of type 1 diabetes, either by initiating the development of the disease or by operating as the final insult to [3 cells in individuals where ongoing autoimmune [3 cell destruction has already been taking place. Whatever the mechanism, evidence from studies on mice, non-human primates, and humans indicates that Coxsackie B virus affects glucose homeostasis. Research on Coxsackie B4 virus has demonstrated that antigenic changes at the epitope level occur at a frequency greater than 1/100 [41, 161 ]. This suggests that even within the same virus pool, there may be many antigenic variants with different tissue tropisms and different physiological properties, which would account for the wide spectrum of clinical disease produced by the Coxsackie B virus. Only rare variants may be diabetogenic, explaining why type 1 diabetes appears to be associated with Coxsackie B viral infection in infrequent isolated cases [68].

3.4. Kilham Rat Virus (KRV) KRV induces diabetes by autoimmune responses against [3 cells rather than by direct [3 cell cytolysis in diabetes-resistant BioBreeding (DR-BB) rats [162, 163], which are derived from diabetes-prone BB (DP-BB) rats, but do not normally develop the disease. When DR-BB rats were infected with KRV at 3 weeks of age, about 30% developed autoimmune diabetes within 2-4 weeks after infection and a further 30% showed insulitis without diabetes [ 162]. The incidence of KRV-induced diabetes could be increased to 80--100% by injection of polyI:C for three consecutive days after KRV infection [ 162]. KRV infects lymphoid organs such as the spleen, thymus, and lymph nodes, but not [3 cells. The precise mechanisms by which KRV induces autoimmune type 1 diabetes without infection of ~ cells are not clearly understood; however it is known that macrophages play an important role. Inactivation of macrophages with liposomal dichoromethylene diphosphonate results in the near complete prevention of insulitis and diabetes in KRV-infected DR-BB rats [164]. In addition, it was found that splenocytes from macrophage-depleted, KRVinfected DR-BB rats treated with polyI:C did not transfer diabetes to young DP-BB recipient rats,

238

whereas splenocytes from macrophage-containing, KRV-infected DR-BB rats treated with polyI:C transferred diabetes to 80% of the recipients [ 164]. The expression of macrophage-derived cytokines such as IL-12, TNF-o~, and IL-I~ in pancreatic islets and splenic lymphocytes was increased after KRV infection, and the increased expression of Thl cytokines such as IL-2 and IFN-y was closely correlated with an elevation in IL-12, suggesting that macrophage-derived cytokines may play a critical role in the cascade of events leading to the destruction of pancreatic 13 cells in KRV-infected DR-BB rats [ 164]. Molecular mimicry has been proposed for KRVinduced diabetes. KRV infection could generate viral peptide-specific T lymphocytes, which might cross-react with epitopes on pancreatic ~ cells and attack them, resulting in the development of diabetes. However, experimental data showed that infection of DR-BB rats with recombinant vaccinia viruses expressing various KRV peptides (VP1, VP2, NS 1, or NS2) did not cause insulitis or diabetes, even though viral peptide-specific T cells and antibodies were generated [165]. This result suggests that molecular mimicry between KRV peptides and [~ cell-specific autoantigens is unlikely to be a mechanism by which KRV induces [3 cell-specific autoimmune type 1 diabetes in DR-BB rats. An alternative hypothesis is that KRV infection of DR-BB rats might disturb the immune balance and activate 1~ cell-specific autoreactive effector T cells [ 166], which are normally held silent by immunoregulatory control involving the RT6.1 subset of T cells [ 167, 168], thus resulting in the destruction of 13 cells. Several studies have focused on identifying the population that contains the autoreactive effector T cells involved in I] cell destruction. It was found that KRV infection resulted in an increase in the percentage of CD8 + T cells, whereas the percentage of CD4 + T cells decreased. In addition, CD8 § T cells preferentially proliferated as compared with CD4 + T cells, and treatment of with OX8 monoclonal antibody, which inactivates CD8 § T cells, significantly decreased the incidence of KRVinduced diabetes. These results indicate that CD8 + T cells may play an important role in KRV-induced autoimmune diabetes. Further studies showed that the number of Th2-1ike CD45RC-CD4 + T cells was significantly reduced and the number of Thl-like

Figure 3. Schematic model of KRV-induced diabetes in DR-BB rats. KRV infection of DR-BB rats activates macrophages, resulting in the production of proinflammatory cytokines such as IL-12, IL-113, TNF-tz, and IFN-~,. These cytokines activate 13cell-specific CD8§ T cells and differentiate CD4§ T cells into Thl-like CD4§ T cells. KRV can also replicate within amplified CD4§ and CD8§ T cell populations. Ultimately, the CD8§ T cell-rnediated cytotoxic response (CTL response) and Thl-type CD4§ T cell-mediated Thl response contribute to pancreatic l~cell death, leading to diabetes in KRV-infected DR-BB rats. CD45RC§ § T cells was significantly increased in the splenocytes of KRV-infected DR-BB rats as compared with controls [165]. Adoptive transfer of Thl-like CD45RC§ § T cells and CD8 § T cells from KRV-infected DR-BB rats resulted in the development of diabetes in 88% of the recipients, suggesting that these two cell populations are major effector T cells that can induce diabetes. As well, it was recently reported that KRV infection regulated the CD25§ § regulatory T cell population [169]. Taken together, it is suggested that KRV infection activates macrophages, which then produce inflammatory cytokines. These cytokines disturb the finely tuned immune balance, resulting in the upregulation of pre-existing [3 cell-specific autoreactive CD8 § T

cells and Thl-type CD4 § T cells. The toxicity of macrophage-derived cytokines and Thl-type CD4 § T cell-derived cytokines to 13 cells, together with the damage incurred by KRV [~ cell-specific CD8 § cytotoxic T cells, may lead to 13cell destruction and autoimmune type 1 diabetes in KRV-infected DRBB rats (Fig. 3).

3.5. Retrovirus Endogenous retroviruses have been implicated in the pathogenesis of type 1 diabetes in NOD mice [170-172], but the evidence is largely circumstantial. It was found that islet cells of NOD mice express various retroviral messenger RNAs

239

(mRNAs) encoded by the gag, pol and env genes, and 1~ cells in particular express the group-specific antigen p73 of the A-type retrovirus [ 173]. In addition, the presence of both A-type and C-type retroviral particles was found in pancreatic 13 cells of NOD mice [ 171, 172, 174] and was correlated with the development of autoimmune type 1 diabetes in these animals. It is not certain how retroviruses might be involved in the pathogenesis of autoimmune type 1 diabetes in NOD mice. It is possible that a retroviral antigen expressed on the 13 cells might be recognized as foreign and presented by antigen-presenting cells, such as macrophages and dendritic cells, resulting in the development of effector T cells that can destroy the 13 cells. Another related mechanism might be the alteration of the expression of cellular genes by the retroviral genomes in the [~ cells, resulting in a ~ cell-specific altered antigen(s). An altered antigen might be recognized as foreign by immunocytes, leading to I] cell-specific autoimmunity. In addition, it is possible that cellular proteins from 13 cells taken up in the retroviral envelope may elicit an autoimmune response or that IFN-y induced as a result of retroviral infection may subsequently induce the expression of HLA-II and trigger autoimmunity through CD4 § T lymphocytes.

3.6. Reovirus Reovirus has been associated with type 1 diabetes in animals, but its mode of action is not known. Direct infection of the ~ cells has been suggested by studies in which mice infected with [~ cell-passaged reovirus type 3 showed specific viral antigens and viral particles in some ~l cells. These mice had abnormal glucose tolerance tests within 10 days after infection, but glucose tolerance returned to normal after 3 weeks [175]. Alterations in the immune system, such as induction of autoimmune mechanisms or a shift to a Th2 response, have been suggested as a mechanisms by which reoviruses might induce type 1 diabetes. It was found that mice infected with [I cell-passaged reovirus type 1 developed transient diabetes, and their sera contained autoantibodies that reacted with cytoplasmic antigens from the islets of Langerhans, the anterior pituitary, and the gastric mucosa of uninfected mice [176]. As well, administration of immunosuppressive drugs to reovirus-

240

infected SJL and NFS mice reduced or prevented the development of reovirus-induced diabetes and mortality [177], suggesting the involvement of an autoimmune response. Recent studies suggest that a Thl response induced by the increased expression of IL-12 may be responsible for the development of diabetes in newborn DBA/1 mice infected with reovirus [ 178].

3.7. Ljungan Virus It was recently found that 33% of wild bank voles (Clethrionomys glareolus), which were trapped and kept in the laboratory for one month, developed diabetes due to [3 cell lysis [179]. These diabetic animals had increased levels of GAD65, IA-2, and insulin autoantibodies. The islets of these mice stained positively for Ljungan virus, a novel picornavirus found in bank voles. When non-diabetic wild bank voles were infected with Ljungan virus in the laboratory, 15 cell lysis was induced. These results show that the development of type 1 diabetes in bank voles is associated with Ljungan virus infection.

3.8. Rubella Virus Rubella virus has been shown to induce type 1 diabetes in hamsters, apparently by direct infection of the [3 cells. Neonatal golden Syrian hamsters infected with [3 cell-passaged rubella virus developed hyperglycemia and hypoinsulinemia between 7 and 10 days of age, and their 13cells were positive for rubella virus antigen. An autoimmune process may be involved, as 40% of infected animals had cytoplasmic islet cell antibodies and 34.5% had insulitis [ 180].

3.9. Bovine Viral Diarrhoea-Mucosal Disease (BVD-MD) Virus B VD-MD virus has been reported to be associated with type 1 diabetes in cattle, however not all animals with B VD-MD viral infection develop diabetes [181]. This may be attributable to the existence of different variants of the virus or to genetic differences among the hosts. It appears that the diabetogenic effect of B VD-MD virus is not a direct effect of the virus on [I cells. Infected cattle with

Table 2. Viruses associated with the prevention of type 1 diabetes Virus

Animal model

Possible mechanism

Lymphocytic choriomeningitis virus

NOD mice BB rats

Depletion of CD4§T cell subpopulation

Mouse hepatitis virus

NOD mice

Induction of the Th2 immune response

Coxsackie B virus

NOD mice

Induction of immunoregulatory response

Encephalomyocarditis virus

NOD mice

Induction of immunoregulatory response

type 1 diabetes showed the presence of B VD-MD viral genes in the pancreas, but not in the islet cells; many of these cattle also had islet cell autoantibodies [182]. More research is needed to determine if B VD-MD virus truly induces autoimmune responses that result in type 1 diabetes in genetically susceptible animals.

4. PREVENTION OF TYPE 1 DIABETES BY VIRUSES In addition to the diabetogenic viruses described above, some viruses can prevent the development of type 1 diabetes under some circumstances (Table 2). Lymphocytic choriomeningitis virus (LCMV) [ 183, 184] and mouse hepatitis virus (MHV) [185] protected against the development of autoimmune type 1 diabetes in spontaneously diabetic DP-BB rats and nonobese diabetic (NOD) mice. Recently, it was reported that infection of NOD mice with Coxsackie B virus significantly reduced the incidence of diabetes as compared with that in mock-infected control mice [186]. Interestingly, EMC-D virus, which is diabetogenic in some strains of mice, could prevent autoimmune diabetes in NOD mice [ 187]. Possible mechanisms have been proposed for the preventive effects of these viruses: 1) viral infection may affect the immune system, resulting in the induction of immunoregulatory cells or the Th2 immune response, or 2) viral infection may deplete a CD4 § T cell subpopulation, which could interfere with the general immune response [ 188].

5. CONCLUSIONS Viruses have been considered to be an important environmental factor in the etiology of type 1 diabetes. Coxsackie virus/enterovirus, mumps virus, CMV, rubella virus, retrovirus, EBV, hepatitis A virus, varicella zoster virus, measles virus, polio virus, influenza virus, and rotavirus have been implicated as potential causal agents for human type 1 diabetes. In addition, EMC virus, KRV, Coxsackie virus, retrovirus, rubella virus, BVD-MD virus, Mengovirus, foot and mouth disease virus, and CMV are known to be associated with the development of type 1 diabetes in animals. However, the precise etiology for the involvement of viruses and their pathogenic mechanisms, particularly in human type 1 diabetes, are poorly understood. Viruses may cause type 1 diabetes by directly infecting and destroying pancreatic [3 cells. For example, susceptible strains of mice infected with a high dose of EMC-D virus develop diabetes within 3 days of infection as a result of pancreatic l] cell destruction mainly caused by replication of the virus within the [3 cells. In contrast to direct cytolytic infection of [3 cells, many viruses appear to cause type 1 diabetes by contributing to [3 cell-specific autoimmunity by a variety of mechanisms, with or without [3 cell infection. Infection of susceptible strains of mice with a low dose EMC-D virus results in the development of diabetes mainly as a result of the activation of macrophages and subsequent induction of [3 celltoxic, macrophage-derived soluble mediators. Other viruses such as retroviruses may infect [3 cells and change existing antigens into immunogenic forms or may induce new antigens, which are recognized as foreign by the immune system and lead to ~ cell-

241

specific autoimmunity. Viruses such as Coxsackie virus may cause diabetes by bystander activation of pre-existing 1~cell-specific immunocytes. In this case, it is thought that Coxsackie viral infection of 1~ cells may induce the expression of I F N - a and subsequently chemokines and cytokines, which recruits cytotoxic immunocytes to the ~ cells and results in their destruction. Viruses such as KRV may also activate pre-existing 13 cell-specific autoreactive immunocytes in genetically susceptible animals, but unlike Coxsackie virus, direct infection of the [3 cells does not occur. Instead, it is thought that KRV activates Thl-like immunocytes that disrupt the immune balance and result in the activation of I] cell-specific immunocytes that are usually held silent. Viruses such as rubella virus and CMV may exert their diabetogenic effects by molecular mimicry; effector T cells generated against viral epitopes may cross-react with homologous epitopes on pancreatic 13 cells. Lastly, several viruses such as retroviruses and mumps virus may act by inducing the expression of IFN-y in infected [3 cells, which may upregulate MHC class I and II molecules, leading to the initiation of [~ cell-specific autoimmunity. Viruses not only cause diabetes, but may also prevent the disease in autoimmune diabetes-prone animals. Infection of young NOD mice or DP-BB rats with LCMV and NOD mice with MHV, Coxsackie virus, or EMC virus prevents diabetes, probably by affecting the immune system, such as by the induction of Th2 immune responses or deletion of effector cells (i.e. Thl CD4 § T cells, CD8 § T cells, or macrophages). The identification of causative viruses in human type 1 diabetes is extremely difficult. The acute phase of viral infection may be already passed by the time that diabetes symptoms are shown. In addition, it is difficult to distinguish between diabetogenic and non-diabetogenic variants of the same virus by serological tests. A large prospective cohort study in prediabetic or genetically susceptible individuals as well as newly diabetic patients may help to understand the viral etiology of type 1 diabetes in humans. The identification of causative viruses will help to develop a preventive strategy for human type 1 diabetes.

242

REFERENCES 1.

2.

3.

4. 5. 6.

7. 8.

9.

10.

11. 12. 13.

14. 15. 16.

17.

Yoon JW, Jun HS. Insulin-dependent diabetes mellitus. In: Roitt IM, Delves PJ, eds. Encyclopedia of Immunology. 2nd Ed. London, UK: Academic, 1998: 1390-1398. Lernmark A, Falorni A. Immune phenomena, events in the islets in insulin-dependent diabetes mellitus. In: Pickup JC, Williams G, eds. Textbook of Diabetes. 2nd Ed. Oxford: Blackwell Science, 1997:15.1-15.23. BachJE Insulin-dependent diabetes mellitus as a 13cell targeted disease of immunoregulation. J Autoimmunity 1995;8:439-463. SchranzDB, Lernmark A. Immunology in diabetes: an update. Diab Metab Rev 1998;14:3-29. TischR, McDevitt H. Insulin-dependent diabetes mellitus. Cell 1996;85:291-297. VergeCE Eisenbarth GS. Prediction of type 1 diabetes: the natural history of the prediabetic period. In: Eisenbarth GS, Lafferty KJ, eds. Type 1 Diabetes: Molecular, Cellular and Clinical Immunology. New York, NY: Oxford Univ Press, 1996;230-258. Notldns AL. Immunologic and genetic factors in type 1 diabetes. J Biol Chem 2002;277:43545-43548. Kelly MA, Rayner ML, Mijovic CH, Barnett AH. Molecular aspects of type 1 diabetes. Mol Pathol 2003 ;56:1-10. Redondo MJ, Eisenbarth GS. Genetic control of autoimmunity in Type I diabetes and associated disorders. Diabetologia 2002;45:605-622. Todd JA, Bain SC. A practical approach to identification of susceptibility genes for IDDM. Diabetes 1992;41:1029-1034. She JX. Susceptibility to type I diabetes: HLA-DQ and DR revisited. Immunol Today 1996;17:323-329. Segall M. HLA and genetics of IDDM. Holism vs. reductionism? Diabetes 1988;37:1005-1008. Yoon JW, Jun HS. Role of viruses in the pathogenesis of type 1 diabetes mellitus. In: LeRoith D, Olefsky M, eds. Diabetes Mellitus: A Fundamental and Clinical Text. Philadelphia: Lippincott-Ravin, 2000;419-430. Jun HS, Yoon JW. A new look at viruses in type 1 diabetes. Diabetes Metab Res Rev 2003;19:8-31. Hyoty H, Taylor KW. The role of viruses in human diabetes. Diabetologia 2002;45:1353-1361. Akerblom HK, Vaarala O, Hyoty H, Ilonen J, Knip M. Environmental factors in the etiology of type 1 diabetes. Am J Med Genet 2002;115:18-29. Kaprio J, Tuomilehto J, Koskenvuo Met al. Concordance for type 1 (insulin-dependent) and type 2 (noninsulin-dependent) diabetes mellitus in a populationbased cohort of twins in Finland. Diabetologia 1992;35: 1060-1067.

18. Kyvik KO, Green A, Beck-Nielsen H. Concordance rates of insulin dependent diabetes mellitus: a population based study of young Danish twins. BMJ 1995;311: 913-917. 19. Barnett AH, Eft C, Leslie RD, Pyke DA. Diabetes in identical twins. A study of 200 pairs. Diabetologia 1981;20:87-93. 20. Green A. Sjolie AK, Eshoj O. The epidemiology of diabetes mellitus. In: Pickup JC, Williams G, eds. Textbook of Diabetes. 2nd Ed. Oxford: B lackwell Science, 1997 ;3.1-3.16. 21. Jun HS, Yoon JW. A new look at viruses in type 1 diabetes. Diabetes Metab Res Rev 2003; 19:8-31. 22. King ML, Shaikh A, Bidwell D, Voller A, Banatvala JE. Coxsackie B-virus specific IgM responses in children with insulin-dependent diabetes mellitus. Lancet 1983;1:1397-1399. 23. Banatvala JE, Bryant J, Schernthaner Get al. Coxsackie B, mumps, rubella and cytomegalovirus specific IgM responses in patients with juvenile-onset insulin-dependent diabetes mellitus in Britain, Austria and Australia. Lancet 1985;1:1409-1412. 24. Friman G, Fohlman J, Frisk G, Diderholm H, Ewald U, Kobbah M, Tuvemo T. An incidence peak of juvenile diabetes. Relation to Coxsackie B virus immune response. Acta Pediatr Scand 1985;320 (Suppl):14-19. 25. Gamble DR, Kinsley ML, Fitzgerald MG, Bolton R, Taylor KW. Viral antibodies in diabetes mellitus. Br Med J 1969;3:627--630. 26. Mertens T, Gruneklee D, Eggers HJ. Neutralizing antibodies against Coxsackie B viruses in patients with recent onset of type 1 diabetes. Eur J Pediatr 1983; 140: 293-294. 27. Frisk G, Fohlman J, Kobbah M et al. High frequency of Coxsackie-B-virus-specific IgM in children developing during a period of high diabetes morbidity. J Med Virol 1985;17:219-227. 28. Fohlman J, Bohme J, Rask L et al. Matching of host genotype and serotypes of Coxsackie B virus in the development of juvenile diabetes. Scand J Immunol 1987;26:105-110. 29. Frisk G, Friman G, Tuvemo T, Fohlman J, Diderholm H. Coxsackie B virus IgM in children at onset of type 1 (insulin-dependent) diabetes mellitus: evidence for IgM induction by a recent or current infection. Diabetologia 1992;35:249-253. 30. Gamble DR, Taylor KW, Cumming H. Coxsackie viruses and diabetes mellitus. Br Med J 1973;4:260262. 31. Wagenknecht L, Roseman J, Herman W. Increased incidence of insulin-dependent diabetes mellitus following an epidemic of Coxsackie virus B5. Am J Epidemiol 1991 ;133:1024-1031.

32. Varela-Calvino R, Sgarbi G, Wedderburn LR, Dayan CM, Tremble J, Peakman M. T cell activation by coxsackievirus B4 antigens in type 1 diabetes mellitus: evidence for selective TCR VI] usage without superantigenic activity. J Immuno12001 ;167:3513-3520. 33. Varela-Calvino R, Sgarbi G, Arif S, Peakman M. T-Cell reactivity to the P2C nonstructural protein of a diabetogenic strain of coxsackievirus B4. Virology 2000;274: 56-64. 34. Varela-Calvino R, Ellis R, Sgarbi G, Dayan CM, Peakman M. Characterization of the T-cell response to coxsackievirus B4: evidence that effector memory cells predominate in patients with type 1 diabetes. Diabetes 2002;51:1745-1753. 35. Palmer JP, Cooney MK, Crossley JR, Hollander PH, Asplin CM. Antibodies to viruses and to pancreatic islets in nondiabetic and insulin-dependent diabetic patients. Diabetes Care 1981;4:525-528. 36. Pagano G, Cavallo-Perin P, Cavallo TF et al. Genetic, immunologic, and environmental heterogeneity of IDDM. Incidence and 12 month follow-up of an Italian population. Diabetes 1987;36:859-863. 37. Marttila J, Juhela S, Vaarala O et al. Responses of coxsackievirus B4-specific T-cell lines to 2C proteincharacterization of epitopes with special reference to the GAD65 homology region. Virology 2001;284: 131-141. 38. Pato E, Cour MI, Gonzalez-Cuadrado S, GonzalezGomez C, Munoz JJ, Figueredo A. Coxsackie B4 and cytomegalovirus in patients with insulin-dependent diabetes. Anales de Medicina Interna 1992;9:30-32. 39. Palmer JP, Cooney MK, Ward RH et al. Reduced Coxsackie antibody titres in type 1 (insulin-dependent) diabetic patients presenting during an outbreak of Coxsackie B3 and B4 infection. Diabetologia 1982;22: 426-429. 40. Tuvemo T, Dahlquist G, Frisk G et al. The Swedish childhood diabetes study HI: IgM against Coxsackie B viruses in newly diagnosed type 1 (insulin-dependent) diabetic children - no evidence of increased antibody frequency. Diabetologia 1989;32:745-747. 41. Prabhakar BS, Haspel MV, McClintock PR, Notkins AL. High frequency of antigenic variants among naturally occurring human Coxsackie B4 virus isolates identified by monoclonal antibodies. Nature 1982;300: 374-376. 42. Yoon JW, Bachurski CJ, McArthur RG. Concept of virus as an etiological agent in the development of IDDM. Diabetes Res Clin Pract 1986:2:365-366. 43. Yoon JW, Onodera T, Notkins AL. Virus-induced diabetes mellitus. XV. Beta cell damage and insulindependent hyperglycemia in mice infected with Coxsackie virus B4. J Exp Med 1978;148:1068-1080.

243

44. Clements GB, Galbraith DN, Taylor KW. Coxsackie B virus infection and onset of childhood diabetes. Lancet 1995;346:221-223. 45. Andreoletti L, Hober D, Hober-Vandenberghe C et al. Detection of Coxsackie B virus RNA sequences in whole blood samples from adult patients at the onset of type 1 diabetes mellitus. J Med Virol 1997;52: 121-127. 46. Notkins AL, Yoon JW. Virus-induced diabetes in mice prevented by a live attenuated vaccine. New Engl J Med 1982;306:486. 47. Wilson C, Connolly JM, Thomson D. Coxsackie B2 virus infection and acute onset diabetes in a child. Br Med J 1977; 1:1008. 48. Ahmad N, Abraham AA. Pancreatic isletitis with Coxsackie virus B5 infection. Hum Pathol 1982;13: 661--662. 49. Asplin CM, Cooney MK, Crossley JR, Dornan TL, Raghu P, Palmer JP. Coxsackie B4 infection and islet cell antibodies three years before overt diabetes. J Pediatr 1982;101:398-400. 50. Orchard TJ, Becker AJ, Atchison RW et al. The development of type 1 (insulin-dependent) diabetes mellitus: two contrasting presentations. Diabetologia 1982;25: 89-92. 51. Niklasson BS, Dobersen MJ, Peters CJ, Ennis WH, Moiler E. An outbreak of Coxsackie-virus B infection followed by one case of diabetes mellitus. Scand J Infect Dis 1985;17:15-18. 52. Nigro G, Pacella ME, Patane E, Midulla M. Multisystem Coxsackievirus B-6 infection with findings suggestive of diabetes mellitus. Eur J Paediatr 1986;145: 557-559. 53. Yoon JW, Austin M, Onodera T, Notkins AL. Virusinduced diabetes mellitus: isolation of a virus from the pancreas of a child with diabetic ketoacidosis. N Engl J Med 1979;300:1173-1179. 54. Champsaur H, Bottazzo G, Bertrams J, Assan R, Bach C. Virologic, immunologic, and genetic factors in insulin-dependent diabetes mellitus. J Pediatrics 1982; 100: 15-20. 55. Gladisch R, Hoffmann W, Waldherr R. Myocarditis and insulitis following Coxsackie virus infection. Z Kardiol 1976;65:837-849. 56. Yoon JW, Onodera T, Notkins AL. Virus-induced diabetes mellitus. XI. Replication of Coxsackie virus B3 in human pancreatic beta cell cultures. Diabetes 1978;27: 778-782. 57. Szopa TM, Ward T, Taylor KW. Impaired metabolic functions in human pancreatic islets following infection with Coxsackie B4 virus in vitro. Diabetologia 1986;30:587A. 58. Toniolo A, Onodera T, Jordan G, Yoon JW, Notkins AL.

244

59.

60.

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

Virus induced diabetes mellitus: glucose abnormalities produced in mice by all six members of the Coxsackie B virus group. Diabetes 1982;31:496--499. Kaufman DL, Erlander MG, Clare-Salzler MJ, Atkinson MA, Maclaren NK, Tobin AJ. Autoimmunity to two forms of glutamate decarboxylase in insulin-dependent diabetes mellitus. J Clin Invest 1992;89:283-292. Hou J, Sheikh S, Martin DL, Chatterjee NK. Coxsackievirus B4 alters pancreatic glutamate decarboxylase expression in mice soon after infection. J Autoimmun 1993;6:529-542. Hou J, Said C, Franchi D, Dockstader P, Chatterjee NK. Antibodies to glutamic acid decarboxylase and P2-C peptides in sera from Coxsackie virus B4-infected mice and IDDM patients. Diabetes 1994;43:1260-1266. Richter W, Mertens T, Schoel B et al. Sequence homology of the diabetes-associated autoantigen glutamate decarboxylase with Coxsackie B4-2C protein and heat shock protein 60 mediates no molecular mimicry of autoantibodies. J Exp Med 1994;180:721-726. Tian J, Lehmann PV, Kaufman DL. T cell cross-reactivity between Coxsackie virus and glutamate decarboxylase is associated with a murine diabetes susceptibility allele. J Exp Med 1994;180:1979-1984. Horwitz MS, Bradley LM, Harbertson J, Krahl T, Lee J, Sarvetnick N. Diabetes induced by coxsackie virus: initiation by bystander damage and not molecular mimicry. Nat Med 1998;4:781-785. Foulis AK, Farquharson MA, Cameron SO, McGill M, Schonke H, Kandolf R. A search for the presence of the enteroviral capsid protein VPlin pancreases of patients with type 1 (insulin-dependent) diabetes and pancreases and hearts of infants who died of coxsackieviral myocarditis. Diabetologia 1990;33:290-298. Foulis AK, Farquharson MA, Meager A. Immunoreactive cz-interferon in insulin-producing 13 cells in type I diabetes mellitus. Lancet 1987 ;2:1423-1427. Chehadeh W, Weill J, Vantyghem MC et al. Increased level of interferon-alpha in blood of patients with insulin-dependent diabetes mellitus: relationship with coxsackie B infection. J Infect Dis 2000;181:1929-1939. Yoon JW, Kominek HI. Role of Coxsackie B viruses in the pathogenesis of diabetes mellitus. In: Rose NR, Friedman H, eds. Microorganisms and Autoimmune Diseases. New York: Plenum, 1996;129-158. Hiltunen M, Hyoty H, Knip M e t al. Childhood Diabetes in Finland (DiMe) Study Group. Islet cell antibody seroconversion in children is temporally associated with enterovirus infections. J Infect Dis 1997; 175:554-560. Lonnrot M, Korpela K, Knip M e t al. Enterovirus infection as a risk factor for beta-cell autoimmunity in a prospectively observed birth cohort: the Finnish Diabetes Prediction and Prevention Study. Diabetes 2000;49:

1314-1318. 71. Helfand RF, Gary HE Jr, Freeman CY, Anderson LJ, Pallansch MA. Serologic evidence of an association between enteroviruses and the onset of type 1 diabetes mellitus. Pittsburgh Diabetes Research Group. J Infect Dis 1995;172:1206-1211. 72. Juhela S, Hy6ty H, Roivainen M e t al. T-cell responses to enterovirus antigens in children with type 1 diabetes. Diabetes 2000;49:1308-1313. 73. Lonnrot M, Salminen K, Knip M e t al. Childhood Diabetes in Finland (DiMe) Study Group. Enterovirus RNA in serum is a risk factor for beta-cell autoimmunity and clinical type 1 diabetes: a prospective study. J Med Viro12000;61:214-220. 74. Nairn C, Galbraith DN, Taylor KW, Clements GB. Enterovims variants in the serum of children at the onset of Type 1 diabetes mellitus. Diabet Med 1999; 16: 509-513. 75. Yin H, Berg AK, Tuvemo T, Frisk G. Enterovirus RNA is found in peripheral blood mononuclear cells in a majority of type 1 diabetic children at onset. Diabetes 2002;51:1964-1971. 76. Otonkoski T, Roivainen M, Vaarala O et al. Neonatal Type I diabetes associated with maternal echovirus 6 infection: a case report. Diabetologia 2000:43;12351238. 77. Vreugdenhil GR, Schloot NC, Hoorens A et al. Acute onset of type I diabetes mellitus after severe echovirus 9 infection: putative pathogenic pathways. Clin Infect Dis 2000;31:1025-1031. 78. Lonnrot M, Knip M, Roivainen M, Koskela P, Akerblom HK, Hyoty H. Onset of type 1 diabetes mellitus in infancy after enterovirus infections. Diabet Med 199;15:431--444. 79. Graves PM, Norris JM, Pallansch MA, Gerling IC, Rewers M. The role of enteroviral infections in the development of IDDM: limitations of current approaches. Diabetes 1997;46:161-168. 80. Fohlman J, Friman G. Is juvenile diabetes a viral disease? Ann Med 1993;25:569-574. 81. Frisk G, Nilsson E, Tuvemo T, Friman G, Diderholm H. The possible role of Coxsackie A and echo viruses in the pathogenesis of type 1 diabetes mellitus studied by IgM analysis. J Infect 1992;24:13-22. 82. Harris HF. A case of diabetes mellitus quickly following mumps on the pathological alterations of salivary glands, closely resembling those found in pancreas, in a case of diabetes mellitus. Boston Med Sur J 1899; 140: 465. 83. Gamble DR. Relation of antecedent illness to development of diabetes in children. Br Med J 1980;2:99-101. 84. Helmke K, Otten A, Willems W. Islet cell antibodies in children with mumps infection. Lancet 1980;2:

211-212. 85. Cavallo MG, Baroni MG, Toto A et al. Viral infection induces cytoldne release by beta islet cells. Immunology 1992;75:664-668. 86. Parkkonen P, Hytity H, Koskinen L, Leinikki P. Mumps virus infects Beta cells in human fetal islet cell cultures upregulating the expression of HLA class I molecules. Diabetologia 1992;35:63-69. 87. Hummel M, Fuchtenbusch M, Schenker M, Ziegler AG. No major association of breast-feeding, vaccinations, and childhood viral diseases with early islet autoimmunity in the German BABYDIAB Study. Diabetes Care 2000;24:969-974. 88. DeStafano F, Mullooly JP, Okoro CA et al. Vaccine Safety Datalink Team. Childhood vaccinations, vaccination timing, and risk of type 1 diabetes mellitus. Pediatrics 2001;108:E112. 89. Hy6ty H, Hiltunen M, Reunanen A et al. Decline of mumps antibodies in type 1 (insulin-dependent) diabetic children and a plateau in the rising incidence of type 1 diabetes after introduction of the mumpsmeasles-rubella vaccine in Finland. Childhood Diabetes in Finland Study Group. Diabetologia 1993;36: 1303-1308. 90. Ginsberg-Fellner F, Fedun B, Cooper LZ et al. Interrelationships of congenital rubella and type 1 insulindependent diabetes mellitus. In: Jaworski MA, Molnar GD, Rajotte RV, Singh B, eds. The Immunology of Diabetes Mellitus. Amsterdam: Elsevier, 1986;279-286. 91. Patterson K, Chandra R, Jenson A. Congenital rubella, insulitis and diabetes mellitus in an infant. Lancet 1981; 1:1048-1049. 92. Schopfer K, Matter L, Flueler U, Werder E. Diabetes mellitus, endocrine autoantibodies and prenatal rubella infection. Lancet 1982;2:159. 93. Menser MA, Forrest J, Bransby R. Rubella infection and diabetes mellitus. Lancet 1978;1:57-60. 94. Forrest J, Menser MA, Burgess J. High frequency of diabetes mellitus in young adults with congenital rubella. Lancet 1971;1:332-334. 95. McEvoy R, Cooper L, Rubinstein P, Fedun B, Ginsberg-Feliner E Type I diabetes mellitus (IDDM) and autoimmunity in patients with congenital rubella syndrome (CRS): increased incidence of insulin autoantibodies. Diabetes 1986;35:187A. 96. Rabinowe S, George K, Loughlin R, Soeldner J. Predisposition to type I diabetes and organ-specific autoimmunity in congenital rubella: T cell abnormalities in young adults. Diabetes 1986;35:187A. 97. Ginsberg-Feliner F, Witt ME, Fedun B. Diabetes mellitus and autoimmunity in patients with congenital rubella syndrome. Rev Infect Dis 1985;7(Suppl 1): S170-175.

245

98. Mclntosh EDG, Menser MA. A fifty-year follow-up of congenital rubella. Lancet 1992;340:414--415. 99. Ginsberg-Fellner F, Witt ME, Yagihaski S. Congenital rubella-syndrome as a model for type 1 (insulin-dependent) diabetes mellitus: increased prevalence of islet cell surface antibodies. Diabetologia 1984;27:87-89. 100. Numazaki K, Goldman H, Wong I, Wainberg MA. Infection of cultured human fetal pancreatic islet cells by rubella virus. Am J Clin Pathol 1989;91:446--451. 101. Numazaki K, Goldman H, Seemayer TA, Wong I, Wainberg MA. Infection by human cytomegalovirus and rubella virus of cultured human fetal islets of Langerhans. In Vivo 1990;4:49-54. 102. Karounos DG, Wolinsky JS, Thomas JW. Monoclonal antibody to rubella virus capsid protein recognizes a beta-cell antigen. J Immunol 1993; 150:3080--3085. 103. Ou D, Mitchell LA, Metzger DL, Gillam S, Tingle AJ. Cross-reactive rubella virus and glutamic acid decarboxylase (65 and 67) protein determinants recognized by T cells of patients with type 1 diabetes mellitus. Diabetologia 2000;43: 750--762. 104. Ward KP, Galloway WH, Auchterlonie IA. Congenital cytomegalovirus infection and diabetes. Lancet 1979; 1: 497. 105. Yasumoto N, Hara M, Kitamoto YU, Nakayama M, Sato T. Cytomegalovirus infection associated with acute pancreatitis, rhabdomyolysis and renal failure. Intern Med 1992;31:426--430. 106. Jenson A, Rosenberg H, Notkins AL. Virus-induced diabetes mellitus XVII: Pancreatic islet cell damage in children with fatal viral infections. Lancet 1980;2: 354-358. 107. Pak CY, Eun HM, McArthur RG, Yoon JW. Association of cytomegalovirus infection with autoimmune type 1 diabetes. Lancet 1988;2:1-4. 108. Nicoletti F, Scalia G, Lunetta M e t al. Correlation between islet cell antibodies and anti-cytomegalovirus IgM and IgG antibodies in healthy first-degree relatives of type 1 (insulin-dependent) diabetic patient. Clin Immunol Immunopathol 1990;55:139-147. 109. Pak CY, Cha CY, Rajotte RV, McArthur RG, Yoon JW. Human pancreatic islet cell-specific 38 kDa autoantigen identified by cytomegalovirus-induced rnonoclonal islet cell autoantibody. Diabetologia 1990;33:569-572. 110. Hiemstra HS, Schloot NC, van Veelen PA et al. Cytomegalovirus in autoimmunity: T cell crossreactivity to viral antigen and autoantigen glutamic acid decarboxylase. Proc Natl Acad Sci USA 2001 ;98:3988-3991. 111. Krieg AM, Steinberg AD. Retroviruses and autoimmunity. J Autoimmun 1990;3:137-166. 112. Krieg AM, Gourley MF, Perl A. Endogenous retroviruses: potential etiologic agents in autoimmunity. FASEB J 1992;6:2537-2544.

246

113. Perron H, Garson JA, Bedin F et al. Molecular identification of a novel retrovirus repeatedly isolated from patients with multiple sclerosis. The Collaborative Research Group on Multiple Sclerosis. Proc Natl Acad Sci USA 1997;94:7583-7588. 114. Garry RF, Fermin CD, Hart DJ, Alexander SS, Donehower LA, Luo-Zhang H. Detection of a human intracisternal A-type retroviral particle antigenically related to HIV. Science 1990;250:1127-1129. 115. Query CC, Keene JD. A human autoimmune protein associated with U 1RNA contains a region of homology that is crossreactive with retroviral p30 gag antigen. Cell 1987;51:211-220. 116. Nyman U, Lundberg I, Hedfors E, Pettersson I. Recombinant 70-kD protein used for determination of autoantigenic epitopes recognized by anti-RNP sera. Clin Exp Immunol 1990;81:52-58. 117. Brookes SM, Pandolfino YA, Mitchell TJ et al. The immune response to and expression of cross-reactive retroviral gag sequences in autoimmune disease. Br J Rheumatol 1992;31:735-742. 118. Hao W, Serreze DV, McCulloch DK, Neifing JL, Palmer JE Insulin (auto) antibodies from human IDDM cross-react with retroviral antigen p73. J Autoimmun 1993;6:787-798. 119. Conrad B, Weissmahr RN, Boni J, Arcari R, Schupbach J, Mach B. A human endogenous retroviral superantigen as candidate autoimmune gene in type I diabetes. Cell 1997;90:303-313. 120. Kim A, Jun HS, Wong L et al. Human endogenous retrovirus with a high genomic sequence homology with IDDMK(1,2)22 is not specific for Type I (insulindependent) diabetic patients but ubiquitous. Diabetologia 1999;42:413-418. 121. Lower R, Tonjes RR, Boiler K et al. Development of insulin-dependent diabetes mellitus does not depend on specific expression of the human endogenous retrovirus HERV-K. Cell 1998;95:11-14. 122. Jaeckel E, Heringlake S, Berger D, Brabant G, Hunsmann G, Manns MP. No evidence for association between IDDMK(1,2)22, a novel isolated retrovirus, and IDDM. Diabetes 1999;48:209-214. 123. Stauffer Y, Marguerat S, Meylan F et al. Interferonalpha-induced endogenous superantigen: a model linking environment and autoimmunity. Immunity 2001; 15: 591-601. 124. Marchini B, Dolcher MP, Sabbatini A, Klein G, Migliorini P. Immune responses to different sequences of the EBNA I molecule in Epstein-Barr virus-related disorders and in autoimmune diseases. J Autoimmun 1994;7: 179-191. 125. Surcel HM, Ilonen J, Kaar ML, Hy/Sty H, Leinikki P. Infection by multiple viruses and lymphocyte abnor-

malities at the diagnosis of diabetes. Acta Paediatrica Scandinavica 1988;77:471-474. 126. Sairenji T, Daibata M, Sodi CH et al. Relating homology between the Epstein-Barr virus BOLF1 molecule and HLA-DQw8 beta chain to recent onset type 1 (insulin-dependent) diabetes mellitus. Diabetologia 1991;34:33-39. 127. Horn G, Bugawan T, Long C, Erlich H. Allelic sequence of the HLA-DQ loci: relationship to serology and to insulin-dependent diabetes susceptibility. Proc Natl Acad Sci USA 1988;85:6012-6016. 128. Parkkonen E HyOty H, Ilonen J, Reijonen H, Yla-Herttuala S, Leinikki E Antibody reactivity to an EpsteinBarr virus BERF4-encoded epitope occurring also in Asp-57 region of HLA-DQ8 13chain. Clin Exp Immunol 1994;95:287-293. 129. Makeen AM. The association of infective hepatitis type A (HAV) and diabetes mellitus. Tropical and Geographical Medicine 1992;44:362-364. 130. Notkins AL. Virus-induced diabetes mellitus. Arch Virol 1977;54:1-17. 131. Fohlman J, Friman G. Is juvenile diabetes a viral disease? Ann Med 1993;25:569-574. 132. Honeyman MC, Coulson BS, Stone NL et al. Association between rotavirus infection and pancreatic islet autoimmunity in children at risk of developing type 1 diabetes. Diabetes 2000;49:1319-1324. 133. Stefan Y, Malaisse-Lagae E Yoon JW, Notkins AL, Orci L. Virus-induced diabetes in mice: a quantitative evaluation of islet cell population by immunofluorescence technique. Diabetologia 1978; 15:395-401. 134. Craighead JE, McLane MF. Diabetes mellitus: induction in mice by encephalomyocarditis virus. Science 1968;162:913-915. 135. Yoon JW, Notkins AL. Virus-induced diabetes mellitus. VI. Genetically determined host differences in the replication of encephalomyocarditis virus in pancreatic B-cells. J Exp Med 1976;143:1170-1185. 136. Jun HS, Yoon JW. The role of viruses in type I diabetes: two distinct cellular and molecular pathogenic mechanisms of virus-induced diabetes in animals. Diabetologia 2001;44:271-285. 137. Yoon JW, McClintock PR, Onodera T, Notkins AL. Virus-induced diabetes mellitus. Inhibition by a nondiabetogenic variant of encephalomvocarditis virus. J Exp Med 1980;152:878-892. 138. Onodera T, Yoon JW, Brown K, Notkins AL. Virusinduced diabetes mellitus: evidence for control by a single locus. Nature 1978;274:693-696. 139. Kang Y, Yoon JW. A genetically determined host factor controlling susceptibility to encephalomyocarditis virus-induced diabetes in mice. J Gen Virol 1993;74: 1203-1213.

140. Yoon JW, Lesniak MA, Fussganger R, Notkins AL. Genetic differences in susceptibility of pancreatic 13 cells to virus-induced diabetes mellitus. Nature 1976;264:178-180. 141. Chairez R, Yoon JW, Notkins AL. Virus-induced diabetes mellitus. X. Attachment of encephalomyocarditis virus and permissiveness of cultured ~1 cells to infection. Virology 1978;85:606--611. 142. Bae YS, Eun HM, Yoon JW. Molecular identification of viral gene. Diabetes 1989;38:316-320. 143. Bae YS, Eun HM, Yoon JW. Genomic differences between the diabetogenic and non-diabetogenic variants of encephalomyocarditis virus. Virology 1989; 170:282-287. 144. Bae YS, Yoon JW. Determination of diabetogenicity attributable to a single amino acid, Ala-776, on the polyprotein of encephalomyocarditis virus. Diabetes 1993;42:435-443. 145. Jun HS, Kang Y, Notkins AL, Yoon JW. Gain or loss of diabetogenicity resulting from a single point mutation in recombinant encephalomyocarditis virus. J Virol 1997 ;71:9782-9785. 146. Jun HS, Kang Y, Yoon HS, Kim KH, Notkins AL, Yoon JW. Determination of encephalomyocarditis viral diabetogenicity by a putative binding site of the viral capsid protein. Diabetes 1998;47:576-582. 147. Hirasawa K, Jun HS, Han HS, Zhang M-L, Hollenberg M, Yoon JW. Prevention of encephalomyocarditis virus-induced diabetes in mice by inhibition of the tyrosine signalling pathway and suppression of nitric oxide production in macrophages. J Virol 1999;73: 8541-8548. 148. Hirasawa K, Jun HS, Maeda K et al. Possible role of macrophage-derived soluble mediators in the pathogenesis of encephalomyocarditis virus-induced diabetes in mice. J Virol 1997;71:4024--4031. 149. Choi KS, Jun HS, Kim HN et al. Role of Hck in the pathogenesis of encephalomyocarditis virus-induced diabetes in mice. J Viro12001;75:1949-1957. 150. Yoon JW, Morishima T, McClintock PR, Austin M, Notkins AL. Virus-induced diabetes mellitus: mengovirus infects pancreatic beta cells in strains of mice resistant to encephalomyocarditis virus. J Virol 1984;50: 684-690. 151. Kibrick S, Benirschke K. Severe generalized disease (encephalohepatomyocarditis) occurring in the newborn period and due to infection with coxsackie virus, group B; evidence of intrauterine infection with this agent. Pediatrics 1958;22:857-875. 152. Kuno S, Itagaki A, Yamazaki I, Katsumoto T, Kurimura T. Pathogenicity of newly isolated coxsackievirus B4 for mouse pancreas. Acta Virol 1984;28:433-436. 153. Szopa TM, Ward T, Dronfield DM, Portwood ND,

247

Taylor KW. Coxsackie B4 viruses with the potential to damage beta cells of the islets are present in clinical isolates. Diabetologia 1990;33:325-328. 154. Chatterjee NK, Haley TM, Nejman C. Functional alterations in pancreatic beta cells as a factor in virusinduced hyperglycemia in mice. J Biol Chem 1985;260: 12786-12791. 155. Chatterjee NK, Hou J, Dockstader P, Charbonneau T. Coxsackievirus B4 infection alters thymic, splenic, and peripheral lymphocyte repertoire preceding onset of hyperglycemia in mice. J Meal Virol 1992;38:124-131. 156. Caggana M, Chan P, Ramsingh A. Identification of a single amino acid residue in the capsid protein VP1 of coxsackievirus B4 that determines the virulent phenotype. J Virol 1993;67:4797-4803. 157. Jenkins O, Booth JD, Minor PD, Almond JW. The complete nucleotide sequence of Coxsackievirus B4 and its comparison to other members of picomaviridae. J Gen Virol 1987;68:1835-1848. 158. Kang Y, Chatterjee NK, Nodwell MJ, Yoon JW. Complete nucleotide sequence of a strain of Coxsackie B4 virus of human origin that induces diabetes in mice and its comparison with non-diabetogenic Coxsackie B4 JBV strain. J Med Virol 1994;44:353-361. 159. Titchener PA, Jenkins O, Szopa TM, Taylor KW, Almond JW. Complete nucleotide sequence of a betacell tropic variant of Coxsackievirus B4. J Med Virol 1994;42:369-373. 160. Yoon JW, London W, Curfinan B, Brown R, Notkins AL. Coxsackie virus B4 produces transient diabetes in nonhuman primates. Diabetes 1986;53:712-716. 161. Prabhakar BS, Menegus MA, Notkins AL. Detection of conserved and nonconserved epitopes on Coxsackie virus B4: frequency of antigenic change. Virology 1985;146:302-306. 162. Guberski DL, Thomas VA, Shek WR et al. Induction of type I diabetes by Kilham's rat virus in diabetes-resistant BB/Wor rats. Science 1991 ;254:101 0-1013. 163. Brown DW, Welsh RM, Like AA. Infection of peripancreatic lymph nodes but not islets precedes Kilham's rat virus-induced diabetes in BB/Wor rats. J Virol 1993;67: 5873-5878. 164. Chung YH, Jun HS, Hirasawa K, Lee BR, van Rooijen N, Yoon JW. Role of macrophages and macrophagederived cytokines in the pathogenesis of Kilham rat virus-induced autoimmune diabetes in diabetes-resistant BB rats. J Immunol 1997;159:466--471. 165. Chung YH, Jun HS, Son Met al. Cellular and molecular mechanism for Kilham rat virus-induced autoimmune diabetes in DR-BB rats. J Immunol 2000;165:28662876. 166. Ellerman KE, Richards CA, Guberski DL, Shek WR, Like AA. Kilham rat virus triggers T-cell-dependent

248

autoimmune diabetes in multiple strains of rat. Diabetes 1996;45:557-562. 167. Greiner DL, Mordes JP, Handler ES, Angelillo M, Nakamura N, Rossini AA. Depletion of RT6.1 § T lymphocytes induces diabetes in resistant BioBreeding/ Worcester (BB/W) rats. J Exp Med 1987;166:461-475. 168. Doukas J, Mordes JP, Swymer C et al. Thymic epithelial defects and predisposition to autoimmune disease in BB rats. Am J Pathol 1994;145:1517-1525. 169. Zipris D, Hillebrands JL, Welsh RM, Rozing J, Xie JX, Mordes JP, Greiner DL, Rossini AA. Infections that induce autoimmune diabetes in BBDR rats modulate CD4(+)CD25(+) T cell populations. J Immunol 2003; 170:3592-602. 170. Suenaga K, Yoon JW. Association of beta cell-specific expression of endogenous retrovirus with the development of insulitis and diabetes in NOD mice. Diabetes 1988;37:1722-1726. 171. Gaskins H, Prochazka M, Hamaguchi K, Serreze D, Leiter E. Beta cell expression of endogenous xenotropic retrovirus distinguishes diabetes-susceptible NOD/ Lt from resistant NON/Lt mice. J Clin Invest 1992;90: 2220-2227. 172. Nakagawa C, Hanafusa T, Miyagawa Jet al. Retrovirus gag protein p30 in the islets of nonobese diabetic mice: relevance for pathogenesis of diabetes mellitus. Diabetologia 1992;35:614--618. 173. Pak CY, Jun HS, Lee M, Yoon JW. Beta cell-specific expression of retroviral mRNAs and group-specific antigen and the development of beta cell-specific autoimmunity in non-obese diabetic mice. Autoimmunity 1995 ;20:19-24. 174. Fukino-Kurihara H, Fujita H, Hakura A, Nonaka K, Tarui S. Morphological aspects on pancreatic islets of non-obese diabetic (NOD) mice. Virchows Arch 1985;49:107-120. 175. Onodera T, Jenson AB, Yoon JW, Notkins AL. Virusinduced diabetes mellitus: reovirus infection of pancreatic beta cells in mice. Science 1978;301:529-531. 176. Onodera T, Toniolo A, Ray UR, Jenson AB, Knazek RA, Notkins AL. Virus-induced diabetes mellitus. J Exp Med 1981;153:1457-1465. 177. Onodera T, Ray UR, Melez KA, Suzuki H, Toniolo A, Notkins AL. Virus-induced diabetes mellitus: autoimmunity and polyendocrine disease prevented by immunosuppression. Nature 1982;297:66-69. 178. Hayashi T, Morimoto M, Iwata H, Onodera T. Possible involvement of IL- 12 in reovirus type-2-induced diabetes in newborn DBA/1 mice. Scand J Immuno12001;53: 572-578. 179. Niklasson B, Heller KE, Schonecker B et al. Development of type 1 diabetes in wild bank voles associated with islet autoantibodies and the novel ljungan virus. Int

J Exp Diabetes Res 2003;4:35--44. 180. Rayfield E, Kelly K, Yoon JW. Rubella virus-induced diabetes in hamsters. Diabetes 1986;35:1276-1281. 181. Tajima M, Yazawa T, Hagiwara K, Kurosawa T, Takahashi K. Diabetes mellitus in cattle infected with bovine viral diarrhea mucosal disease virus. J Vet Med A 1992;39:616--620. 182. Tajima M, Yuasa M, Kawanabe M, Taniyama H, Yamato O, Maede Y. Possible causes of diabetes mellitus in cattle infected with bovine diarrhoea virus. Zentralbl Veterinarmed B 1999;46:207-215. 183. Oldstone MBA. Prevention of type 1 diabetes in nonobese diabetic mice by virus infection. Science 1988;239:50(O502. 184. Dyrberg T, Schwimmbeek PL, Oldstone MBA. Inhibition of diabetes in BB rats by viral infection. J Clin Invest 1988;81:928-931. 185. Wilberz S, Partke HJ, Dagnaes-Hansen F, Herberg

L. Persistent MHV (mouse hepatitis virus) infection reduces the incidence of diabetes mellitus in non-obese diabetic mice. Diabetologia 1991 ;34:2-5. 186. Tracy S, Drescher KM, Chapman NM et al. Toward testing the hypothesis that group B coxsackieviruses (CVB) trigger insulin-dependent diabetes: inoculating nonobese diabetic mice with CVB markedly lowers diabetes incidence. J Viro12002;76:12097-12111. 187. Hermitte L, Vialettes B, Naquet P, Atlan C, Payan MJ, Vague P. Paradoxical lessening of autoimmune processes in non-obese diabetic mice after infection with the diabetogenic variant of encephalomyocarditis virus. Eur J Immunol 1990;20:1297-1303. 188. Oldstone MBA. Viruses as therapeutic agents: I. Treatment of nonobese insulin-dependent diabetes mice with virus prevents insulin-dependent diabetes mellitus while maintaining general immune competence. J Exp Med 1990; 101:2077-2089.

249

9 2004 Elsevier B. V All rights reserved. Infection and Aumimmunity Y. Shoenfeld and N.R. Rose, editors

Theiler's Murine Encephalomyelitis Virus-Induced Demyelinating Disease (TMEu and Autoimmunity Stephen D. Miller ~and Carol L. VanderLugt-Castaneda 2

1Department of Microbiology-Immunology and Interdepartmental Immunobiology Center, Feinberg School of Medicine, Northwestern University, Chicago, IL USA; 2Department of Biology, Indiana University Northwest, Gary, IN, USA 1. TMEV-IDD AND MS

Theiler's murine encephalomyelitis virus (TMEV) belongs to the Picornaviridae family and is a natural enteric pathogen of mice [1]. Neurovirulence upon experimental intracerebral injection varies depending on the strain of TMEV ranging from a rapidly fatal encephalitis in which grey matter neurons are infected and lysed upon infection with the GDVII strain to an initial acute phase of grey matter involvement followed by a chronic phase of viral persistence, inflammation and demyelination in the white matter of the spinal cord following infection with the BeAn or DA strains [2, 3]. The extent of acute phase grey matter involvement and mechanisms of chronic demyelination differ in BeAn and DA infection. The acute phase in DA infection of SJL mice is characterized by microglial proliferation and necrosis of neuronal motor neurons which results in flaccid paralysis (i.e., poliomyelitis) [4]. Surviving mice develop TMEV-induced demyelinating disease (TMEV-IDD). In contrast, BeAn infection results in a very limited early acute phase grey matter disease, with no clinical signs. Chronic demyelinating disease in these mice appears later (d30) and is due to the immune response itself, not to direct lysis of virally infected oligodendrocytes

[5, 6]. Intracerebral injection of the BeAn strain of TMEV into SJL mice results in a chronic demyelinating disease, TMEV-IDD, which resembles multiple sclerosis (MS) in many ways. In addition to the epidemiological data suggesting an infectious etiology of MS, histopathology, consisting of inflamma-

tory infiltrate and areas of demyelination, and clinical disease bear many similarities to MS. Although no one virus has been shown to be consistently associated with MS, early infection may trigger events, through molecular mimicry [7] or epitope spreading [8], that eventually result in autoimmune disease. Like TMEV-IDD, MS is a immune-mediated demyelinating disease characterized by perivascular CD4 § T cell and mononuclear cell infiltration, with subsequent primary demyelination of axonal tracts, leading to progressive paralysis [9]. MS is generally considered to have an autoimmune component, however a direct cause-effect relationship between myelin reactivity and disease has not been established. Interestingly, although TMEV-IDD is due initially to a persistent viral infection of the central nervous system (CNS), autoimmune anti-myelin Thl responses are seen during the chronic phase of disease [8, 10, 11].

2. PERSISTENT INFECTION AND CHRONIC DISEASE Demyelination and the resulting clinical disease symptoms are not the consequence of viral lysis of oligodendrocytes which construct the myelin sheath [12], but are immune mediated, due to a CD4 § T cell inflammatory response in the CNS [5, 13]. Initial observations using immunohistochemistry demonstrated that viral antigens could be found in abundance in the spinal cord of TMEV (BeAn) infected mice in macrophage/microglial cells and in astrocytes but not in oligodendrocytes [12, 14]. In

251

contrast to the abundance of viral antigen, infections virus in the CNS of infected mice has been consistently shown to be low. Historically, macrophage/ microglia have been thought to be the major reservoir of infectious virus. However, it had also been demonstrated that viral replication is blocked in these cells [12]. Analysis of the copy number of TMEV genomes, plus- to minus-strand ratios, and full-length detects large numbers of viral genomes [ 15]. Therefore, during the chronic phase of this disease, there is an abundance of viral antigen and viral genomes combined with very low amounts of infectious virus. This can be explained to some degree by its restricted growth in macrophages, blocked after viral RNA replication. There is some evidence that astrocytes may also serve as a reservoir of infectious virus [ 16]. Supporting the major role of the immune response itself in this chronic phase of disease, nonspecific immunosuppression with cyclophosphamide, anti-thymocyte serum, or CD4 § T cell (not CD8 § depletion after the initial viremia, reduces the inflammatory mononuclear cell infiltration into the CNS and the subsequent demyelination [ 17-20]. In fact, in vivo depletion of CD4 + T cells in SJL mice infected with the BeAn strain of TMEV results in a decreased incidence and slower onset of disease [20]. Susceptibility to TMEV-IDD correlates with chronic high levels of TMEV-specific delayed-typehypersensitivity (DTH) [6, 21] along with the presence of predominately Thl-derived cytokines [20] and transfer of TMEV-specific CD4 § Thl blasts into suboptimally infected SJL mice results in an increase in disease incidence and severity. Although it is clear that the initial anti-virus immune response is responsible for the clinical symptoms of TMEVIDD, viral persistence is required for chronic disease, since mouse strains which clear the virus do not go on to develop the chronic demyelination that characterizes TMEV-IDD [21].

252

3. FROM VIRAL INFECTION TO AUTOIMMUNITY

3.1. Virus-Specific CD4 § T Cell Responses Initiate Disease, Leading to Myelin Damage The adaptive immune response is exquisitely specific in that T and B cell responses initiated against one pathogen usually do not target other pathogens or self tissue. Two points at which some "non-specificity" can be introduced are 1) T or B cell receptor (T/BCR) interaction with their antigen and 2) the use of innate mechanisms as the effector phase of the immune response. First, T/BCRs are not as rigidly specific as once thought. Both B and T cell receptors can bind and respond to epitopes on molecules other than the original stimulant (molecular mimicry) if these epitopes include the important amino acids for TCR: peptide/MHC binding or if these epitopes fold into a similar shape with certain chemical characteristics recognized by the BCR. This mechanism (degeneracy) suggests that infectious viruses may encode within their sequence, epitopes or peptides that share homology with self-antigens. T cell activation during viral infection may thus produce T cells that cross-react with self peptides. These self-reactive T cells can then lead to self-tissue destruction and perpetuate an autoimmune response. Degeneracy in the TCR allows for the recognition of peptides with varying sequences by the same T cell clone. In fact, degeneracy in MHC class II peptide binding and TCR recognition of self-myelin peptide myelin basic protein (MBP)85-99 in the context of the human MHC class II HLA-DR2 haplotype, which is associated with MS patients, has been demonstrated [22]. Secondly, while CD8 § cytotoxic lymphocytes kill in a very specific manner, CD4 § T helper 1 (Thl) cells induce the influx and activation of macrophages, which carry out effector functions in a 'non-specific' manner. Macrophage influx and activation leads to bystander tissue destruction. During infection with TMEV, initiation of myelin damage is associated with the activation of monocyte/ macrophages by pro-inflammatory cytokines [2325] from TMEV-specific Thl cells responding to viral epitopes presented by CNS-resident antigenpresenting cells (APCs), which harbor persistent

virus for many months following infection [26]. Initially, time-course studies comparing the development of T cell responses to both virus and myelin epitopes during TMEV-IDD showed that autoreactivity to myelin epitopes is not detected prior to disease onset (30-35 days post infection) [27, 28] while immune responses to TMEV epitopes are clearly demonstrable by 5-7 days post infection [29]. Induction of peripheral tolerance to mouse spinal cord homogenate (MSCH), which effectively prevents MSCH-induced experimental autoimmune encephalomyelitis (EAE), at the time of TMEV infection does not affect the clinical onset or the development of virus-specific T cell responses in TMEV-IDD [30]. These results demonstrate that virus-specific CD4 + T cell responses initiate bystander tissue destruction (demyelination) and the clinical signs of TMEV-IDD. Infected animals exhibit a chronic progressive demyelinating disease characterized by low level persistence of TMEV in CNS microglia/ macrophages and/or astrocytes throughout the fifetime of the animals. CNS mononuclear infiltrates can be detected as early as 7 days post-infection [29] and are initially composed of peripheral macrophages and virus-specific T cells [24, 31-33]. In the SJL mouse, CD4 § T cell reactivity to the dominant myelin epitope proteolipid protein (PLP139_151), can be detected in the periphery of these animals beginning 50-55 days post-infection [8] indicating the initiation of an autoimmune component secondary to viral infection. In contrast to molecular mimicry, these data suggest that myelin debris is being processed and presented to T cells specific for autoantigens (epitope spreading).

3.2. Myelin Damage Results In Endogenous Presentation Of Myelin Epitopes And Epitope Spreading Accumulating data demonstrate that chronic immune-mediated tissue damage can lead to de novo activation of autoreactivity via epitope spreading. Epitope spreading is the process whereby epitopes distinct from, and non-cross-reactive with, an inducing epitope become major targets of an ongoing immune response. Two prominent examples of epitope spreading in CD4 + T cell-mediated autoimmune models are diabetes in NOD mice

[34-36] and R-EAE [37-39]. In addition, epitope spreading has been demonstrated following viral infections with picornaviruses, such as TMEV [8] and Coxsackie virus [40]. Initiation of myelin damage in TMEV-infected SJL mice by TMEV-specific CD4 § T cells targeting virus persisting in CNS-resident APCs leads to up-regulation of pro-inflammatory cytokines in the CNS, and is associated with the activation of CD4 § myelin-specific T cells during the chronic phase of disease. These autoreactive T cells appear to be primed via epitope spreading as determined by their late appearance in disease (> 50-60 days PI) and by the fact that there are no apparent viral epitopes that are shared with the major encephalitogenic myelin epitopes on PLP, MBP or MOG, i.e., there is no evidence for molecular mimicry in this system [8]. The spreading process was demonstrated in TMEV-IDD by observing temporal changes in the specificity of delayed-type hypersensitivity (DTH), T cell proliferative, and IFN-~, responses to viral and myelin epitopes [41] in peripheral lymphoid tissues. Anti-viral DTH and in vitro anti-viral T cell responses appear within a few days post-infection, and these responses continue throughout the disease. In contrast, myelin-specific responses can be detected beginning only 50-60 days post infection, i.e. 3-4 wk after clinical disease onset. Most interestingly, T cell responses against myelin epitopes arise in an ordered progression, initially targeting the immunodominant myelin epitope in the SJL mouse, PLP139_ls1. Reactivity toward this peptide then continues throughout disease. As disease progresses, T cell responses to PLP178_191 followed by responses to MBP84_~04arise, paralleling the relative order of their appearance in PLP139_~5~-induced R-EAE in SJL mice [42]. Reactivity toward additional myelin epitopes is also observed. The similarity in the strength and order of epitope spreading in TMEV-IDD and R-EAE suggests a hierarchy in the processing and presentation of these epitopes and/or the precursor frequency of T cells specific for the various myelin epitopes in SJL mice. The T cell precursor frequency of PLPI39_151 > PLP178_191 > MBP84_104 in SJL mice [42] correlates with the order of the appearance of specific T cell responses in the periphery of TMEVinfected SJL mice. However, the dynamics involved in the processing myelin tissue and presentation of

253

myelin epitopes by the various types of antigen presenting cells in the CNS may effect this hierarchy and are currently not known. The availability of various myelin epitopes which are targeted in the epitope spreading process in the spinal cord of infected mice was examined by adding antigen presenting cells isolated from the spinal cord (CNS APCs) to myelin- or virus-specific T cell lines or hybridomas. As expected in a chronic CNS infection, endogenous presentation of viral envelope peptides (no exogenous peptide added) was demonstrated on day 35 (the earliest time point at which sufficient CNS APCs could be collected), and endogenous expression of these epitopes on CNS APCs persisted through 150 days post infection. This was expected since the CNS APC population is known to harbor persistent viral antigen, presumably allowing these cells to present this antigen in vivo or in vitro. In contrast, the endogenous presentation of all myelin epitopes assayed could be demonstrated (using 2.5-3.5x104 APC/well) only >80 days post infection. These data suggest that CNS APCs can present myelin epitopes endogenously only after sufficient demyelination by CNS microglia/macrophages has taken place. CNS APCs were also able to endogenously present multiple myelin epitopes around the same time post-infection, which suggests that T cell precursor frequency governs the hierarchy of epitope spreading rather than a peptide hierarchy of processibility in the CNS inflammatory environment or in CNS APCs. Immunohistochemical analysis of spinal cordinfiltrating mononuclear cells reveals that the number of CD4 + T cells and activated F4/80 + macrophages/microglia increase dramatically after TMEV infection. The F4/80 + cells also increase in size and up-regulate the requisite molecules required for activation of naive CD4 + T cells (i.e., MHC class II, B7-1, and B7-2), penetrate into the parenchyma, and accumulate in the CNS. This progressive accumulation correlates with the disease severity and the increasing number of CNS APCs that can be recovered from spinal cords of infected mice. Analyses of the F4/80 + population in the CNS reveals two subpopulations based on levels of expression of CD45 - resident C D 4 5 dim microglia and CD45 bright infiltrating peripheral macrophages. The ability of either of these CNS APCs to activate PLP139_~s~-specific Thl cells can be inhibited by both anti-MHC

254

class II and by blocking costimulation with CTLA4 Ig, indicating that the presentation of endogenous myelin epitopes is B7 dependent and MHC class II-restricted [ 10]. Within the normal CNS, a variety of cells are capable of antigen presentation to T cells, including astrocytes, microglia, and macrophages. IFN"t-treated primary astrocytes [43, 44] and microglia [45, 46] cultured from neonatal mouse brain upregulate MHC class II and can present antigens to T cells in vitro, but this may not reflect the in vivo state in adult animals. Microglia directly isolated from adult rats can more efficiently present MBP to T cell lines in vitro compared with neonatally derived microglia [47]. In our hands, CNS mononuclear cells isolated from na'fve mice are inefficient at endogenously activating myelin-specific T cells. They are, however, capable of processing and presenting exogenously added myelin proteins/ peptides, albeit with less efficiency than irradiated splenic APCs [41]. The roles played by each of these CNS APCs in epitope spreading has been difficult to unravel and continues to be investigated. Freshly isolated F4/80 +, I-A s+, CD45 + plasticadherent mononuclear cells from the spinal cord of TMEV-infected mice, which include macrophages, monocytes, and microglia, were able to process and present exogenous TMEV or horse myoglobin epitope to antigen-specific T cell lines [33] and have the ability to endogenously process and present virus epitopes at both acute and chronic stages of the disease [ 10, 41 ]. However, the relative contributions of the resident microglia vs. infiltrating macrophages in antigen presentation during TMEV-IDD had not yet been clearly delineated. Recently, we isolated CNS-resident microglia and CNS-infiltrating macrophages from TMEV-infected mice based on their differential expression levels of CD45 (see above) in order to test the APC capabilities of each cell type at various stages of disease [Mack, C, manuscript submitted]. Microglia from na'fve adult mice are clearly in a "resting state" and are not competent antigen presenting cells as shown by the lack of APC markers such as MHC class II and B7 costimulatory molecules, as well as their inability to stimulate proliferation of an antigen-specific Thl line. In contrast to microglia purified from neonatal brain [48], stimulation of na'fve microglia from adult SJL mice in vitro with pro-inflammatory cytokines,

under defined conditions, resulted in a low expression of B7-2, but no detectable upregulation MHC class ILl, and the persistent inability to stimulate T cell proliferation. The finding that na'fve adult microglia remain incompetent APCs even following stimulation with pro-inflammatory cytokdnes has also been shown by other investigators [47, 49]. In contrast to microglia from naive adult mice, in the inflammatory setting of TMEV-IDD microglia become activated, express the necessary molecular machinery to serve as competent APCs and effectively stimulate T cell proliferation and cytokine production. At the time of clinical disease onset (37 days post-infection), both the CD45 dim microglia and the CD45 b~gh~ infiltrating macrophages express similar levels of APC surface markers and are capable of stimulating proliferation and IFN- 7 production of a PLP139_~51-specific Thl line to a similar degree. The role of the resident microglia at this early point in TMEV-IDD may primarily involve processing and presentation of viral epitopes. In fact, microglia cultured from neonatal SJL mice can be persistently infected with TMEV in vitro and that infection significantly upregulated expression of costimulatory (B7-1, B7-2 and CD40) and MHC class II molecules. Most significantly, TMEV-infected microglia were able to efficiently process and present both endogenous virus epitopes and exogenous myelin epitopes to inflammatory CD4 § Thl cells [48]. Microglia are activated early in response to a number of different infections or injuries to the CNS [50, 51 ]. Interestingly, during the chronic-progressive stage of TMEV-IDD (90 days post-infection), the C D 4 5 b~ight infiltrating macrophages express higher levels of APC markers and spinal cord-infiltrating macrophages are more potent stimulators of T cell proliferation when compared to the CD45 dimmicrogila. Increasing numbers of infiltrating macrophages as disease progresses leads to an increased secretion of TNF-t~ which, along with IFN-), derived from the virus and myelin peptide-specific T cells, may lead to up-regulation of MHC-class II and B7 surface expression on these cells, creating more potent APCs. The amount of TNF-oc mRNA expression in the spinal cords of SJL mice with TMEV-IDD sharply increases as the disease progresses [52]. The resident microglia may not be as responsive as the infiltrating macrophage to TNF-o~-facilitated, IFN-

q-mediated upregulation of APC surface markers and presenting function. Another explanation for the differential APC capability of macrophages versus microglia later in disease may be due to differential inhibition by nitric oxide. In support of this hypothesis, it has been reported that microglia in the setting of MOG35_55 peptide-induced EAE were competent APCs during the peak of disease [53]. However, late in this disease, after mice had partially recovered, there was a reduction in microglial APC capability that was attributable to enhanced production of nitric oxide by infiltrating macrophages. It is of major interest to determine whether T cells involved in the epitope spreading process that are specific for endogenous myelin epitopes become activated in the periphery (draining lymph nodes and spleen) and/orthe CNS. It is possible that following inflammatory disruption of the blood-brain barrier, myelin debris and/or macrophages/microglia that have ingested myelin proteins gain access to the cervical lymph nodes that drain the cerebrospinal fluid [54] or to the spleen, which concentrates blood-borne material. In support of this hypothesis, it has been reported that donor cells from alloantigen-disparate solid CNS grafts placed intracerebrally can be later identified in the host spleen and lymph nodes [55]. We are currently assessing the ability of APCs purified from the spleen and deep cervical lymph nodes of mice with chronic disease to endogenously present self epitopes. In contrast, it is also possible that T cells specific for endogenous myelin epitopes are activated in the local inflammatory environment within the CNS. In TMEVIDD, the inflammatory infiltrate is composed of T and B lymphocytes, activated microglia derived from the CNS-resident pool, and macrophages infiltrating from the peripheral blood [33, 56, 57]. Macrophages/microglia within the demyelinated areas contain phagocytized myelin debris [58] and are capable of processing and presenting myelin epitopes. Therefore, any myelin-specific T cells that enter the CNS during the anti-viral inflammatory response, whether already primed in the periphery or not, could potentially be induced to proliferate and/or to secrete pro-inflammatory cytokines in response to myelin epitopes.

255

4. DIFFERENTIAL ABILITIES OF CNS RESIDENT MICROGLIA, ENDOTHELIAL CELLS, AND ASTROCYTES TO SERVE AS INDUCIBLE ANTIGEN-PRESENTING CELLS A variety of cells within the normal CNS are capable of antigen presentation to T cells. MHC class II and costimulatory molecule expressing cells can be found in MS lesions [59-62] and human microglia have been shown to express costimulatory molecules required for activation of T cells [63, 64]. As previously discussed, microglia and macrophages differ in their APC ability depending on the microenvironment. Cerebrovascular endothelial cells (CVEs) upregulate MHC class II and B7-1 costimulatory molecule in response to IFN-y in vitro. However, murine CVEs did not elicit significant MHC class II-restricted T cell responses [65]. IFN-y-treated SJL astrocytes fed intact PLP or MP4 (a fusion protein containing PLP and MBP portions) efficiently activated lines and hybridomas specific for the immunodominant PLP139_151 epitope. However, T cell lines specific for the less immunodominant self encephalitogenic epitopes (PLP56_7o, PLP~04_~7, and PLP178_191) were not activated by IFN-y-treated astrocytes fed intact PLP, but were activated by astrocytes pulsed with the relevant autologous peptide [44]. Similar results were seen with multiple independently derived PLP peptide-specific T cell lines and hybridomas specific for the less dominant epitopes and when astrocytes were activated with a combination of TNF-cx and IFN-y, which enhances MHC class II and Ii expression above the levels seen for astrocytes stimulated with IFN-y alone. Under the pro-inflammatory conditions examined it appears that astrocytes may not play a major role in the phenomenon of epitope spreading. However, it is also possible that in the local milieu of the CNS, additional cytokines may play a role in activating astrocytes to more effectively process and present the subdominant PLP epitopes. Significantly, IFNy-treated SJL/J astrocytes pulsed with either intact MP4 or PLP139_151 were also capable of activating PLP~39_~5~-specific T cells for the adoptive transfer of R-EAE, indicating that they can induce the upregulation of the appropriate integrins and cytokines necessary for CD4 § T cells to home to the CNS and initiate the demyelinating process.

256

In addition to the inflammatory environment in which these CNS APCs reside, TMEV infection itself may also play a role in upregulation of APC function. As previously detailed, microglia from SJL mice can be persistently infected in vitro with TMEV and, as a result of this infection, these cells are activated to function as competent APCs with the ability to process and present both virus and myelin epitopes to memory CD4 § Thl cells [48]. Concomitant with the acquisition of this functional antigen .presentation capacity, TMEV infection induced the upregulation of cytokines involved in innate immune responses and of cytokdnes and costimulatory molecules required for the activation and differentiation Th 1 effector cells. Most significantly, direct TMEV infection of microglia was nearly as effective as stimulation with high levels of IFN-y in conferring APC function.

5. USING TMEV TO INVESTIGATE MOLECULAR MIMICRY, AN ALTERNATE MECHANISM FOR INDUCTION OF AUTOIMMUNITY DURING INFECTION The mechanism(s) underlying the initiation and progression of multiple sclerosis and other autoimmune diseases are not well understood, but epidemiologic studies have provided strong suggestive evidence for a role of virus infection(s) in the development and/or exacerbations of MS. The possible mechanisms by which virus infection can trigger an autoimmune response include molecular mimicry, bystander activation, and epitope spreading. In the TMEV-IDD model of MS, we have demonstrated bystander activation, the non-specific activation of autoreactive T cells resulting from the virus-specific CD4 + Thl inflammatory response itself on tissue in the target organ, followed by epitope spreading, the activation of autoreactive T cells due to the tissue damage following release of self epitopes during that immune response. Molecular mimicry, on the other hand, theoretically results following infection with a virus expressing a peptide determinant(s) that shares homology with a self peptide, resulting in activation of T cells that can crossreact with the self epitope. The discovery of TCR degeneracy, the TCR's abil-

ity to recognize multiple peptides with only a few key amino acid positions in common, has led to the widespread belief that some microbial proteins probably contain peptide sequences that are able to activate self-reactive T cells. Recent studies have shown degeneracy in the TCR specific for the human myelin basic protein MBP85_99peptide, with the TCR requiting only a few critical residues for recognition [22]. T cell clones specific for MBP 85-99 established from MS patients were shown to crossreact with viral peptides expressed by a number of viruses, including HSV, adenovirus, reovirus, and human papillomavirus [22]. Likewise, a few critical residues were shown to be necessary for recognition of PLP139_151by its TCR [66]. PLPx39_lsl-specific T cell hybridomas derived from SJL mice were also shown to cross-react with peptides expressed by various mouse pathogens, demonstrating degeneracy in the PLP139_151TCR [66]. Therefore, myelin-specific T cells have been shown by in vitro studies to have the potential to cross-react with viral epitopes, supporting the molecular mimicry model described in these studies. In order to directly investigate molecular mimicry as a potential mechanism of CD4 § T cellmediated autoimmunity, we developed an infectious model of molecular mimicry by inserting a sequence encompassing the immunodominant PLP~39_~5~epitope into the coding region of a nonpathogenic TMEV variant (PLP~39-TMEV) [67, 68]. PLP139-TMEV-infected mice developed a rapid onset paralytic inflammatory, demyelinating disease paralleled by the activation of PLPa39_~5~-specific CD4+Th 1 responses within 10-14 days post-infection. These data demonstrate that the early onset demyelinating disease induced by PLPI39-TMEV is the direct result of autoreactive PLP~39_~5~-specific CD4 § T cell responses. PLP139_~5~-specific CD4 § T cells from PLP~39-TMEV-infected mice transferred demyelinating disease to naive recipients and infection with the mimic virus at sites peripheral to the CNS induced early demyelinating disease, suggesting that the PLP139_xs~-Specific CD4 § T cells could be activated in the periphery and traffic to the CNS. Importantly, PLP139_~5~ epitope-specific tolerance before infection with PLP~39-TMEV resulted in the specific reduction of PLP139_x51-specific CD4 § Thl responses that directly correlated with a significant reduction in the incidence and severity of the early

onset demyelinating disease. In addition, mimic PLP139_151 sequences were constructed in which amino acid substitutions were made at the primary (amino acid 144) or secondary (amino acid 147) TCR contact residues [67, 68]. Infection with the virus carrying a substitution in the secondary TCR contact residue induced early-onset demyelinating disease and activated cross-reactive PLPi39_lsl-specific CD4+T cells. In contrast, infection with the virus substituted at the primary TCR contact residue (position 144) failed to induce early demyelinating disease or activation of cross-reactive PLP139_151-specificCD4§ cells. An additional mimic virus was constructed by inserting a sequence from H. influenzae that shared only 6 of 13 amino acids with the core PLP139_151epitope [67]. More significant to a role for molecular mimicry in induction of autoimmune disease, infection with this mimic virus resulted in early onset demyelinating disease and activation of Thl cells cross-reactive with the native PLP139_151determinant. This model is the first to directly demonstrate that a virus encoding a mimic of an encephalitogenic self myelin epitope could induce an autoreactive CD4§ cell response leading to a CNS demyelinating disease. The ability of any microbial peptide mimicking self to be processed is required for APC presentation to T cells. Ongoing work inserting 30mers into the virus coding region will determine if the induction of the autoimmune disease requires that the mimic epitope be processed from its native flanking regions in addition to the requirement that the core epitope be presented in an appropriate fashion to activate the self-reactive Thl response.

6. SUMMARY

The epidemiology of MS strongly suggests a role for an infectious agent, most likely a virus. Presentation of viral antigens within the CNS (leading to bystander demyelination), of neuroantigens crossreactive with viral antigens (molecular mimicry), or of neuroantigens liberated by immune or virusinduced CNS damage (epitope spreading) are all possible mechanisms by which pathogenic immune reactions could be initiated by viruses within the CNS (Fig. 1). TMEV-IDD is a well-characterized CD4 § T cell-

257

Figure 1. Possible Mechanisms of Virus-Induced T Cell-Mediated Autoimmune Disease. The figure illustrates induction of CD4§ T cell-mediated autoimmune tissue destruction via induction of a self antigen-specific cross-reactive T cell response (i.e., molecular mimicry) following peripheral virus infection (Left Panel) and via epitope spreading to self antigen-reactive T cells secondary to bystander tissue destruction and release of self antigens initiated by a specific T cell response to virus persistent in the target tissue (Right Panel).

mediated model of MS. Life-long persistent viral infection of CNS resident microglia, macrophages, and astrocytes is directly related to the development of the chronic demyelinating disease. Initial myelin damage is mediated by a bystander mechanism wherein the primary effector cells are mononuclear phagocytes (microglia/macrophages) activated by inflammatory cytokines produced from TMEVspecific Thl cells responding to viral epitopes that persist in the CNS. Early myelin destruction leads to the de novo activation of myelin-specific T cells (epitope spreading). The initial myelin response is directed toward the immunodominant PLP 139-151 epitope, and epitope spreading then leads to an ordered progression of T cell responses to multiple myelin autoepitopes which appear to play a significant role in the chronic phase of the disease by escalating the demyelinating process. The continuous presence of the virus within the CNS perpetuates this chronic inflammatory process in which epitope spreading leads to the induction of autoreactive T cells. These findings enhance our understanding of the pathogenesis of human MS. MHC class II-bearing

258

macrophages, astrocytes, and endothelial cells have been observed in or near MS lesions, together with expression of B7 costimulatory molecules. Therefore, multiple cells in MS lesions have the potential to fully activate both naive and memory T cells within the CNS. Although not the mechanism of autoimmunity in mice infected with native TMEV, mimic peptide- and natural pathogen peptide-engineered TMEV infection models demonstrate that molecular mimicry can also lead from viral infection to autoimmune disease.

REFERENCES 1. 2.

3.

4.

Theiler M. Spontaneous encephalomyelitis of mice, a new virus disease. J Exp Med 1937;65:705-719. Lipton HL. Theiler's virus infection in mice: An unusual biphasic disease process leading to demyelination. Infect Immun 1975;11:1147-1155. LiptonHL, Dal Canto MC. Chronic neurologic disease in Theiler's virus infection of SJL/J mice. J Neurol Sci 1976;30:201-207. LehrichJR, Arnason BGW, Hochberg F. Demyelinative myelopathy in mice induced by the DA virus. J Neurol

5.

6.

7.

8.

9. 10.

11.

12.

13.

14.

15.

16.

17.

Sci 1976;29:149-160. Dal Canto MC, Lipton HL. Primary demyelination in Theiler's virus infection. An ultrastructural study. Lab Invest 1975;33:626-637. Clatch RJ, Melvold RW, Miller SD, Lipton HL. Theiler's murine encephalomyelitis virus (TMEV)induced demyelinating disease in mice is influenced by the H-2D region: correlation with TMEV-specific delayed-type hypersensitivity. J Immunol 1985;135: 1408-1414. Miller SD, Olson JK, Croxford JL. Multiple pathways to induction of virus-induced autoimmune demyelination: lessons from Theiler's virus infection. J Autoimmun 2001; 16:219-227. Miller SD, Yanderlugt CL, Begolka WS, Pao W, Yauch RL, Neville KL et al. Persistent infection with Theiler's virus leads to CNS autoimmunity via epitope spreading. Nature Med 1997;3:1133-1136. Wekerle H. Immunopathogenesis of multiple sclerosis. Acta Neurologica 1991; 13:197-204. Katz-Levy Y, Neville KL, Girvin AM, Vanderlugt CL, Pope JG, Tan LJ et al. Endogenous presentation of self myelin epitopes by CNS-resident APCs in Theiler's virus-infected mice. J Clin Invest 1999;104:599-610. Miller SD, Katz-Levy Y, Neville KL, Vanderlugt CL. Virus-induced autoimmunity: epitope spreading to myelin autoepitopes in Theiler's virus infection of the central nervous system. Adv Virus Res 2001;56: 199-217. Lipton HL, Twaddle G, Jelachich ML. The predominant virus antigen burden is present in macrophages in Theiler's murine encephalomyelitis virus-induced demyelinating disease. J Virol 1995;69:2525-2533. Clatch RJ, Melvold RW, Dal Canto MC, Miller SD, Lipton HL. The Theiler's murine encephalomyelitis virus (TMEV) model for multiple sclerosis shows a strong influence of the murine equivalents of HLA-A, B, and C. J Neuroimmunol 1987;15:121-135. Dal Canto MC, Lipton HL. Ultrastructural immunohistochemical localization of virus in acute and chronic demyelinating Theiler's virus infection. Am J Pathol 1982; 106:20-29. Trottier M, Kallio P, Wang W, Lipton HL. High numbers of viral RNA copies in the central nervous system of mice during persistent infection with Theiler's virus. J Viro12001;75:7420-7428. Zheng L, Calenoff MA, Dal Canto MC. Astrocytes, not microglia, are the main cells responsible for viral persistence in Theiler's murine encephalomyelitis virus infection leading to demyelination. J Neuroimmunol 2001; 118:256-267. Roos RP, Firestone S, Wollmann R, Variakojis D, Amason BG. The effect of short-term and chronic

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

immunosuppression on Theiler's virus demyelination. J Neuroimmunol 1982;2:223-234. Lipton HL, Dal Canto MC. Theiler's virus-induced demyelination: prevention by immunosuppression. Science 1976;192:62-64. Lipton HL, Dal Canto MC. Contrasting effects of immunosuppression on Theiler's virus infection in mice. Infect Immun 1977; 15:903-909. Gerety SJ, Rundell MK, Dal Canto MC, Miller SD. Class II-restricted T cell responses in Theiler's murine encephalomyelitis virus (TMEV)-induced demyelinating disease. VI. Potentiation of demyelination with and characterization of an immunopathologic CD4 § T cell line specific for an immunodominant VP2 epitope. J Immunol 1994;152:919-929. Clatch RJ, Lipton HL, Miller SD. Class II-restricted T cell responses in Theiler's murine encephalomyelitis virus (TMEV)-induced demyelinating disease. II. Survey of host immune responses and central nervous system virus titers in inbred mouse strains. Microb Pathogen 1987;3:327-337. Wucherpfennig KW, Sette A, Southwood S, Oseroff C, Matsui M, Strominger JL et al. Structural requirements for binding of an immunodominant myelin basic protein peptide to DR2 isotypes and for its recognition by human T cell clones. J Exp Med 1994;179:279-290. Zavala F, Abad S, Ezine S, Taupin V, Masson A, Bach JE G-CSF therapy of ongoing experimental allergic encephalomyelitis via chemokine- and cytokine-based immune deviation. J Immunol 2002;168:2011-2019. Whitney LW, Ludwin SK, McFarland HF, Biddison WE. Microarray analysis of gene expression in multiple sclerosis and EAE identifies 5-1ipoxygenase as a component of inflammatory lesions. J Neuroimmunol 2001;121:40--48. Ibrahim SM, Mix E, Bottcher T, Koczan D, Gold R, Rolfs A e t al. Gene expression profiling of the nervous system in murine experimental autoimmune encephalomyelitis. Brain 2001;124:1927-1938. Lipton HL, Kratochvil J, Sethi P, Dal Canto MC. Theiler's virus antigen detected in mouse spinal cord 2 1/2 years after infection. Neurology 1984;34:1117-1119. Miller SD, Clatch RJ, Pevear DC, Trotter JL, Lipton HL. Class II-restricted T cell responses in Theiler's murine encephalomyelitis virus (TMEV)-induced demyelinating disease. I. Cross-specificity among TMEV substrains and related picornaviruses, but not myelin proteins. J Immunol 1987; 138:3776-3784. Barbano RL, Dal Canto MC. Serum and cells from Theiler's virus-infected mice fail to injure myelinating cultures or to produce in vivo transfer of disease. The pathogenesis of Theiler's virus-induced demyelination appears to differ from that of EAE. J Neurol Sci

259

1984;66:283-293. 29. Clatch RJ, Lipton HL, Miller SD. Characterization of Theiler's murine encephalomyelitis virus (TMEV)specific delayed-type hypersensitivity responses in TMEV-induced demyelinating disease: correlation with clinical signs. J Immunol 1986; 136:920-927. 30. Miller SD, Gerety SJ, Kennedy MK, Peterson JD, Trotter JL, Tuohy VK et al. Class H-restricted T cell responses in Theiler's murine encephalomyelitis virus (TMEV)-induced demyelinating disease. IH. Failure of neuroantigen-specific immune tolerance to affect the clinical course of demyelination. J Neuroimmunol 1990;26:9-23. 31. Gerety SJ, Karpus WJ, Cubbon AR, Goswami RG, Rundell MK, Peterson JD et al. Class H-restricted T cell responses in Theiler's murine encephalomyelitis virus (TMEV)-induced demyelinating disease. V. Mapping of a dominant immunopathologic VP2 T cell epitope in susceptible SJL/J mice. J Immunol 1994;152:908-918. 32. Yauch RL, Kerekes K, Saujani K, Kim BS. Identification of a major T-cell epitope within VP3 amino acid residues 24 to 37 of Theiler's virus in demyelinationsusceptible SJL/J mice. J Virol 1995;69:7315-7318. 33. Pope JG, Vanderlugt CL, Lipton HL, Rahbe SM, Miller SD. Characterization of and functional antigen presentation by central nervous system mononuclear cells from mice infected with Theiler's murine encephalomyelitis virus. J Virol 1998;72:7762-7771. 34. Tisch R, Yang XD, Singer SM, Liblau RS, Fugger L, McDevitt HO. Immune response to glutamic acid decarboxylase correlates with insulitis in non-obese diabetic mice. Nature 1993;366:72-75. 35. Kaufman DL, Clare-Salzler M, Tian J, Forsthuber T, Ting GSP, Robinson P et al. Spontaneous loss of T cell tolerance to glutamic acid decarboxylase in murine insulin-dependent diabetes. Nature 1993;366:69-72. 36. Tian J, Atkinson MA, Clare-Salzler M, Herschenfeld A, Forsthuber T, Lehmann PV et al. Nasal administration of glutamate decarboxylase (GAD65) peptides induces Th2 responses and prevents murine insulin-dependent diabetes. J Exp Med 1996;183:1561-1567. 37. Lehmann PV, Forsthuber T, Miller A, Sercarz EE. Spreading of T-cell autoimmunity to cryptic determinants of an autoantigen. Nature 1992;358:155-157. 38. McRae BL, Vanderlugt CL, Dal Canto MC, Miller SD. Functional evidence for epitope spreading in the relapsing pathology of experimental autoimmune encephalomyelitis. J Exp Med 1995;182:75-85. 39. Yu M, Johnson JM, Tuohy VK. A predictable sequential determinant spreading cascade invariably accompanies progression of experimental autoimmune encephalomyelitis: A basis for peptide-specific therapy after onset of clinical disease. J Exp Med 1996;183:1777-1788.

260

40. Horwitz MS, Bradley LM, Harbetson J, Krahl T, Lee J, Sarvetnick N. Diabetes induced by Coxsackie virus: initiation by bystander damage and not molecular mimicry. Nature Med 1998;4:781-786. 41. Katz-Levy Y, Neville KL, Padilla J, Rahbe SM, Begolka WS, Girvin AM et al. Temporal development of autoreactive Thl responses and endogenous antigen presentation of self myelin epitopes by CNS-resident APCs in Theiler's virus-infected mice. J Immunol 2000; 165:5304-5314. 42. Vanderlugt CL, Eagar TN, Neville KL, Nikcevich KM, Bluestone JA, Miller SD. Pathologic role and temporal appearance of newly emerging autoepitopes in relapsing experimental autoimmune encephalomyelitis. J Immuno12000; 164:670--678. 43. Fontana A, Fierz W, Wekerle H. Astrocytes present myelin basic protein to encephalitogenic T-cell lines. Nature 1984;307:273-276. 44. Tan LJ, Gordon KB, Mueller JP, Matis LA, Miller SD. Presentation of proteolipid protein epitopes and B7-1-dependent activation of encephalitogenic T cells by IFN-gamma-activated SJL/J astrocytes. J Immunol 1998; 160:4271-4279. 45. Frei K, Siepl C, Groscurth P, Bodmer S, Schwerdel C, Fontana A. Antigen presentation and tumor cytotoxicity by interferon-gamma-treated microglial cells. Eur J Immunol 1987;17:1271-1278. 46. Cash E, Rott O. Microglial cells qualify as the stimulators of unprimed CD4 § and CD8 § T lymphocytes in the central nervous system. Clin Exp Immunol 1994;98: 313-318. 47. Ford AL, Goodsall AL, Hickey WF, Sedgwick JD. Normal adult ramified microglia separated from other central nervous system macrophages by flow cytometric sorting. Phenotypic differences defined and direct ex vivo antigen presentation to myelin basic proteinreactive CD4+ T cells compared. J Immunol 1995; 154: 4309-4321. 48. Olson JK, Girvin AM, Miller SD. Direct activation of innate and antigen presenting functions of microglia following infection with Theiler's virus. J Virol 2001;75:9780-9789. 49. Carson MJ, Reilly CR, Sutcliffe JG, Lo D. Mature microglia resemble immature antigen-presenting cells. Glia 1998;22:72-85. 50. Shrikant P, Benveniste EN. The central nervous system as an immunocompetent organ: role of glial cells in antigen presentation. J Immunol 1996;157:1819-1822. 51. Kreutzberg GW. Microglia: a sensor for pathological events in the CNS. Trends Neurosci 1996;19:312-318. 52. Begolka WS, Vanderlugt CL, Rahbe SM, Miller SD. Differential expression of inflammatory cytokines parallels progression of central nervous system pathology

53.

54.

55.

56.

57.

58.

59.

60.

in two clinically distinct models of multiple sclerosis. J Immunol 1998;161:4437-4446. Juedes AE, Ruddle NH. Resident and infiltrating central nervous system APCs regulate the emergence and resolution of experimental autoimmune encephalomyelitis. J Immuno12001;166:5168-5175. Cserr HF, Knopf PM. Cervical lymphatics, the bloodbrain barrier and the immunoreactivity of the brain: a new view. Immunol Today 1992;13:507-512. Broadwell RD, Baker BJ, Ebert PS, Hickey WF. Allografts of CNS tissue possess a blood-brain barrier: HI. Neuropathological, methodological, and immunological considerations. Microsc Res Tech 1994;27: 471-494. Miller SD, Vanderlugt CL, Lenschow DJ, Pope JG, Karandikar NJ, Dal Canto MC et al. Blockade of CD28/B7-1 interaction prevents epitope spreading and clinical relapses of murine EAE. Immunity 1995;3: 739-745. Karandikar NJ, Vanderlugt CL, Eagar TN, Tan L, Bluestone JA, Miller SD. Tissue-specific up-regulation of B7-1 expression and function during the course of murine relapsing experimental autoimmune encephalomyelitis. J Immunol 1998;161:192-199. Dal Canto MC, Melvold RW, Kim BS, Miller SD. Two models of multiple sclerosis: Experimental allergic encephalomyelitis (EAE) and Theiler's murine encephalomyelitis virus (TMEV) infection- a pathological and immunological comparison. Microsc Res Tech 1995;32:215-229. Hofman FM, von Hanwehr RI, Dinarello CA, Mizel SB, Hinton D, Merrill JE. Immunoregulatory molecules and IL 2 receptors identified in multiple sclerosis brain. J Immunol 1986; 136:3239-3245. Traugott U, Reinherz EL, Raine CS. Multiple sclerosis:

61.

62.

63.

64.

65.

66.

67.

68.

Distribution of T cell subsets within active chronic lesions. Science 1983;219:308-310. Traugott U. Multiple sclerosis: relevance of class I and class II MHC-expressing cells to lesion development. J Neuroimmunol 1987; 16:283-302. Windhagen A, Newcombe J, Dangond F, Strand C, Woodroofe MN, Cuzner ML et al. Expression of costimulatory molecules B7-1 (CD80), B7-2 (CD80), and interleukin 12 cytokine in multiple sclerosis lesions. J Exp Med 1995;182:1985-1996. Desimone R, Giampaolo A, Giometto B, Gallo P, Levi G, Peschle C et al. The costimulatory molecule B7 Is expressed on human microglia in culture and in multiple sclerosis acute lesions. J Neuropathol Exp Neurol 1995;54:175-187. Williams K, Ulvestad E, Antel JP. B7/BB-1 antigen expression on adult human microglia studied in vitro and in situ. Eur J Immunol 1994;24:3031-3037. Girvin AM, Gordon KB, Welsh CJ, Clipstone NA, Miller SD. Differential abilities of central nervous system resident endothelial cells and astrocytes to serve as inducible antigen-presenting cells. Blood 2002;99: 3692-3701. Carrizosa AM, Nicholson LB, Farzan M, Southwood S, Sette A, Sobel RA et al. Expansion by self antigen is necessary for the induction of experimental autoimmune encephalomyelitis by T cells primed with a crossreactive environmental antigen. J Immunol 1998;161: 3307-3314. Olson JK, Croxford JL, Calenoff M, Dal Canto MC, Miller SD. A virus-induced molecular mimicry model of multiple sclerosis. J Clin Invest 2001; 108:311-318. Olson JK, Eagar TN, Miller SD. Functional activation of myelin-specific T cells by virus-induced molecular mimicry. J Immuno12002; 169:2719-2726.

261


Viruses and Multiple Sclerosis A. Achiron

Multiple Sclerosis Center, Sheba Medical Center, Tel-Hashomer, Sackler Faculty of Medicine, Tel-Aviv University, Israel

1. INTRODUCTION

2. MICROBIAL-INDUCED AUTOIMMUNITY

Multiple sclerosis (MS) is a central nervous system (CNS) white-matter demyelinating disease affecting young adults. The characteristic clinical course of MS in 85% of patients is relapsing-remitting, whereas in about 15% of patients the disease presents as a primary progressive course. Within 10 years from onset, 50% of patients with relapsing-remitting disease will advance to the secondary progressive phase, with consequent increases in neurological disability [1]. The disease has an autoimmune component, with the presence of selfreactive lymphocytes targeting myelin peptides such as myelin basic protein (MBP), proteolipid protein (PLP) and myelin oligodendrocyte glycoprotein (MOG), leading to inflammation and myelin destruction within the brain and spinal cord [2, 3]. The etiology of MS is as yet unknown, and the hypothesis that an infectious agent is responsible for triggering the disease has waxed and waned over the last two centuries (since Pierre Marie first proposed that MS often starts as an infectious process). The possible role of infectious agents has been suggested by the different temporal patterns of the disease in different geographic areas, changes in prevalence due to migration and the effect of age at migration, the relapsing-remitting course of MS, and the induction of demyelination in animal models by various viruses [4-6].

The different mechanisms by which infectious agents might activate autoreactive lymphocytes and lead to an autoimmune disease fall into two major classes: antigen-nonspecific and antigen-specific. 2.1. Antigen-Nonspecific - The 'Innocent Bystander Activation' Theory

This theory is based on nonspecific antigen activation with no particular microbial determinant implicated. Once the immune system becomes primed to attack the infecting pathogen, there is a possibility that the myelin could be inadvertently attacked in the process. The mechanisms suggested to be involved include: (a) Direct inflammatory damage caused by the inflammatory response to the microbial agent, resulting in cell destruction and the subsequent release of different cell ingredients that will be presented to the immune system at the inflammation site. These newly presented self-determinants induce an immune response that will result in an autoimmune disease. In MS, the autoimmune process within the CNS involves activation of microglia by signals originating from either activated monocytes and lymphocytes in the blood stream, or from activated macrophages or astrocytes within the brain. This microglial activation subsequently results in the release of excitotoxins, cytokines and chemokines, with further myelin destruction. (b) A microbial-induced alteration in the phenotype of antigen-presenting cells (APCs). This can result in the enhanced expression of co-stimulatory molecules, increased

263

production of inflammatory cytokines such as rumor necrosis factor (TNF)-0~ and interleukin (IL)1 that may promote retroviral replication, and the modification of lymphocyte migration patterns. (c) Provoked T-cell lines/clones induced via either a mitogen or a superantigen effect by the microbial agent. 2.2. Antigen-Specific - 'Epitope Mimicry'

The cornerstone of the antigen-specific theory is epitope mimicry; an antigenic determinant on one of the proteins of the microbe is structurally similar to a determinant of a host protein, although different enough to be recognised as foreign by the host's immune system. For T-cells, the determinants involved would be linear peptide stretches of about 8-15 amino acids long. The immune response to the microbial determinant cross-reacts with host tissue and eventually results in autoimmune destruction

[7, 8]. 3. MICROBIAL-INDUCED AUTOIMMUNITY IN MS It is currently not known whether an organism is a causative agent of MS, or merely an opportunistic pathogen that takes advantage of a disease process initiated by some other means. Several studies have showed that a naturally infectious virus encoding a myelin epitope can directly initiate organ-specific T-cell-mediated autoimmunity. Lenz et al identified a 20-mer peptide from a protein specific to Chlamydia pneumoniae, which shares a 7-aminoacid motif with a critical epitope of MBP, a major CNS antigen targeted by the immune system in MS [9]. This bacterial peptide induced a Thl response accompanied by severe clinical and histological experimental autoimmune encephalomyelitis in Lewis rats, a condition closely reflective in many aspects of MS. In a similar study, Olson et al studied the potential of virus-induced molecular mimicry to initiate autoimmune demyelination, using a nonpathogenic Theiler's murine encephalomyelitis virus (TMEV) variant that was engineered to encode a 30-mer peptide encompassing the immunodominant encephalitogenic myelin PLP(139-151) epitope [ 10]. Within 10-14 days of

264

infection with the PLP(139-151)-encoding TMEV, a rapid-onset paralytic demyelinating disease, characterised by PLP(139-151)-specific CD4+ Thl responses, was observed. Mice infected with TMEV encoding a Haemophilus influenzae mimic peptide, sharing only six of the 13 amino acids of PLP(139-151), displayed rapid-onset disease and developed cross-reactive, PLP(139-151)-specific CD4+ Thl responses. These studies suggest that the epitope mimicry mechanism may be involved in triggering MS.

4. INFECTIOUS AGENTS AND MS 4.1. Herpes Viruses

One of the greatest challenges in confirming or refuting a role for herpes viruses (e.g. Epstein-Barr virus (EBV) and human herpes virus 6 (HHV-6)) in MS is their ubiquitous nature - they are neurotropic, become latent and persist even with very limited genome expression, can be reactivated with the relapsing-remitting course of MS, and have been shown to induce demyelination. In addition, both herpes reactivations and MS exacerbations can be brought on by infections with other viruses.

4.1.1. Human herpes virus 6 HHV-6 was discovered in 1988 and consists of two subtypes, HHV-6A and HHV-6B, both of which are prevalent in the normal population [ 11]. HHV-6B is the causative agent of exanthem subitum, a common childhood illness, and has been associated with meningitis, myalgic encephalomyelitis and febrile seizures [12]. HHV-6A has not yet been shown to cause human disease. The most consistent neuropathologic changes associated with HHV-6 infections of the CNS have been demyelination, ranging from diffuse and extensive loss of myelin, to sharply circumscribed foci of demyelination [13, 14], combined with the destruction of axons within areas of the most severe pathological changes [12]. As MS is also associated with prominent demyelination combined with axonal destruction, it was suggested that there might be an association between HHV-6 infection and MS. Studies describing associations between HHV-6

and MS are based on either detection of HHV-6 antibodies in serum or cerebral spinal fluid (CSF), or amplification of HHV-6 DNA from serum or CSF of patients with MS but not control subjects [15, 16]. Following this lead, HHV-6 antigen expression was detected in MS brains and shown to be associated specifically with MS plaques [17]. Moore & Wolfson systematically reviewed the published evidence for a relationship between human HHV-6 and MS [ 18]. They searched the medical literature using MEDLINE and the Cochrane database, retrieving 28 studies according to 12 different experimental techniques used. When the technologies could not distinguish between active and latent HHV-6 infections (PCR analysis of blood leukocytes, CSFcontaining cells or CNS tissue), no differences were noted between samples from patients with MS and control subjects. In contrast, when diagnostic technologies were restricted to the detection of active HHV-6 infections (PCR analysis of acellular specimens, detection of HHV-6-specific IgM antibodies or immunohistochemical staining of CNS tissues), evidence for a relationship between HHV-6 and MS was found, but no causative relationship could be demonstrated. Thus, it has been suggested that the finding of a relationship between HHV-6 and MS is merely a result of the immune system activation or blood-brain-barrier breakdown in MS making signs of prior infection with HHV-6 more easily detectable. The mechanisms by which HHV-6 was suggested to cause MS could be related to the ability of the vires to infect and destroy oligodendrocytes [ 19, 20], or to the capability of HHV-6 to induce TNF-c~ production in blood mononuclear cells [21 ], as this pro-inflammatory cytokine is also known to mediate demyelination in MS, and its production by blood mononuclear Cells correlates with disease activity [22]. Interestingly, the antiviral drug acyclovir, which provides effective prophylaxis against HHV6 infections in bone marrow transplant patients, has been shown to significantly reduce the frequency of disease exacerbations in patients with MS [23].

4.1.2. Herpes simplex virus I Herpes simplex virus 1 (HSV-1) DNA has been found in some cases of acute MS but not in stable MS or healthy controls [24]. These data suggest

that HSV-1 reactivates in patients during clinical relapses and may be a trigger of MS relapses.

4.1.3. Human herpes virus 7 Human herpes virus 7 (HHV-7) has been found to be equally prevalent in a latent form in peripheral blood mononuclear cells of both MS patients and healthy controls [25]. Soldan et al measured the lymphoproliferative response to HHV-7-infected cell lysate and found no significant difference between patients with MS and controls [26]. Taus et al also reported no relationship between MS and HHV-7 [27].

4.1.4. Epstein-Barr virus Individual epidemiologic studies assessing the relationship between EBV and MS have been inconclusive, in part because of the high prevalence of previous EBV infection among individuals without MS. The reported prevalence of antiEBV seropositivity in patients with MS is 100%, compared with 80-95% in matched controls [28]. Higher concentrations of serum antibodies against both the EBV viral capsid antigen (VCA) and nuclear antigens (EBNA-1) have been reported in patients with MS [29]. Moreover a more frequent history of infectious mononucleosis, and late age at infection, have been described in patients with MS [30, 31 ]. Ascherio & Munch conducted a systematic review of case-control studies comparing EBV serology in patients with MS and controls [32]. Eight published studies were identified, including a total of 1005 cases and 1060 controls. The summary odds ratio of MS, comparing EBV seropositive individuals with EBV seronegative individuals, was 13.5 (95% CI = 6.3,-31.4). The strength and consistency of this association and the high sensitivity and specificity of EBV serology support a role for EBV in the etiology of MS. However, assuming that 90% of both patients with MS and controls were in fact infected by EBV, the results observed could be obtained if the specificity of the diagnostic test was 10% or less among patients with MS, and close to 100% among controls. Alternatively, under the assumption that both cases and controls were all infected, the results could be obtained if the sensitivity was 90% or less among

265

controls and 100% among patients with MS. In a recent large case-control study conducted among more than 3 million US military, Levin et al evaluated whether antibodies to EBV are elevated before the onset of MS [33]. The risk of MS increased monotonically with VCA or the nuclear antigens EBNA complex antibody titers, and a relationship between EBV infection and development of MS was therefore suggested. The ability of EBV to interfere with the normal process of T-cell repertoire [34], and the cross-reactions of anti-EBNA antibodies with epitopes of a neuroglial antigen [35], also support a causal association with MS, mainly related to EBV reactivation and disease activity in patients with MS, suggesting that EBV might trigger an underlying disease process [36].

4.1.5. Cytomegalovirus Cytomegalovirus (CMV) is a prevalent viral pathogen. The majority of patients with acute CMV will experience an inapparent infection. A primary CMV infection will cause up to 7% of cases of mononucleosis syndrome and will manifest symptoms almost indistinguishable from those of EBV-induced mononucleosis. There are no reports demonstrating a convincing link between MS and CMV. Sanders et al used a PCR approach to compare active and inactive plaques from patients with MS [37]. CMV sequences were detected in 9-22% of specimens, irrespective of disease activity. In another large population-based study, no association was found between CMV antibodies and MS [33].

4.1.6. Varicella zoster virus The association between varicella zoster virus (VZV) and MS is limited largely to clinical experience and epidemiological surveys. Both MS and varicella are most prevalent in temperate zones and rare in countries close to the equator. An early serological study found the geometric mean titer of antibodies to VZV to be significantly higher among patients with MS than among patients with other diseases and normal individuals [38]. Recently, Marrie & Wolfson reviewed the epidemiological evidence for an etiological role of VZV infection

266

in the development of MS; a MEDLINE search of English language literature published between 1965 and 1999 identified 40 studies that were classified according to strict methodological criteria [39]. Five studies that utilised the best methodology failed to show an increased risk of MS associated with varicelia or zoster infections. At the present time, there is insufficient evidence to support an important etiological role of VZV infection in the development of MS. 4.2. Retroviruses

4.2.1. Human immunodeficiency virus Only a few cases of human immunodeficiency virus (HIV)-positive patients with MS or MS-like lesions have been reported. In four of the seven patients described by Berger et al, the MS preceded the HIV infection by a long period. In the other three patients, HIV seroconversion occurred concomitantly or within 3 months of the onset of the neurological symptoms [40]. Gray et al described two patients with a fulminating demyelinating leukoencephalopathy in the early course of HIV infection [41 ], and Graber et al reported a patient with a relapsing and remitting leukoencephalopathy, who was HIV-positive at the onset of the disease, but HIV-negative 9 months earlier [42].

4.2.2. Multiple sclerosis-associated retrovirus The island of Sardinia has a high and increasing incidence of MS. Serra et al and Dolei et al searched for environmental factors that may account for this anomalously high incidence [43, 44]. They detected MS-associated retrovirus (MSRV), an exogenous member of the human endogenous retrovirus family W (HERV-W) in all patients with MS, in most patients with inflammatory neurologic diseases, and rarely in healthy blood donors. MSRV was found in the plasma and CSF of patients with MS, and was produced in vitro by their cells. Detection of MSRV is not restricted to MS, as the virus has also been found in synovial fluids of rheumatoid arthritis patients [45]. Similar to HSV-1, it has been suggested that MSRV may contribute to MS reactivation in Sardinia.

4.3. Parvovirus

Human parvovirus B19 (PVB19), the etiologic agent of Etythema infectiosum, causes transient and persistent immune derangements. PVB 19 infections have been reported to be associated with chronic immune-mediated disorders, including rheumatoid arthritis and systemic lupus erythematosus. PVB 19 can invade the CNS, possibly resulting in encephalopathy and meningitis. Only one study to date has evaluated the association between PVB 19 and MS [46]. The prevalence of serum anti-PVB 19 IgG was shown to be significantly higher in patients with MS than in healthy subjects. On the other hand, none of the patients developed an E. infectiosum infection nor had serological or molecular evidence of an active PVB 19 infection, such as the presence of anti-PVB 19 IgM or PVB 19 DNA. Furthermore, serum anti-PVB 19 IgM and PVB 19 DNA in CSF were consistently negative in patients during exacerbation of MS [46]. 4.4. TT Virus

TT virus (TTV), identified in 1998, is a widespread infectious agent of humans. In infected individuals, TTV induces persistent viremia. However, the life-cycle and pathogenic potential of TTV are still poorly understood. In only one study, 21 paired samples from CSF and serum from patients with MS were tested for TTV using real-time PCR. The majority of MS serum samples (71%) were TTVpositive as expected based on TTV prevalence and viremia levels in the general population, but none of the CSF samples of patients with MS were TTVpositive, suggesting no role for this virus in MS [47].

5. VIRAL INFECTIONS AND DISEASE ACTIVITY

A possible increase in the risk to develop MS following infections of any kind (upper respiratory tract, gastrointestinal, urinary tract) has been investigated in several studies. Hernan et al investigated these associations in a case-control study that included 301 patients with MS and matched controls [48]. Except for infec-

tious mononucleosis, which was a moderate risk factor, little association was found between the history of common viral diseases or exposure to canine distemper virus and risk of developing MS. However, a relationship between mumps and measles after 15 years of age and MS was found. The question of whether infectious diseases can induce MS disease activity was assessed by Sibley et al [49]. A total of 170 patients with MS were evaluated for a mean of 5.2 years. The results showed a 2.8-fold increase in relapse rate during risk periods associated with infection. Andersen et al reported a relative risk for relapse during infection risk periods of only 1.3 in 60 patients with MS followed for a mean of 31 months [50], and Buljevac et al, in a prospective survey of 73 patients with MS followed for a mean of 1.7 years, found a relative risk of 2.1 during infection-related risk periods [51]. These findings confirm the association between infections and relapses in MS and can be explained by the antigen nonspecific theory where activation of the host immune system by viral superantigens results in activation of autoreactive T-cells, pro-inflammatory cytokine production and activation of the disease process [52]. However, no significant changes in MRI activity related to the infections could be demonstrated.

6. VACCINATION AND MS Several case reports describing the onset or exacerbation of MS shortly after vaccination have suggested that vaccines may increase the risk of the disease. The major question raised was whether vaccine-preventable infectious diseases increase the risk of MS onset or exacerbations. DeStefano et al studied 440 patients with MS or optic neuritis and 950 matched controls for the onset of first symptoms of demyelinating disease at any time after vaccination [53]. The odds ratios of the associations between ever having been vaccinated and risk of demyelinating disease were: 0.9 (0.6-1.5) for hepatitis B vaccine; 0.6 (0.4-0.8) for tetanus vaccination; 0.8 (0.6-1.2) for influenza vaccine; 0.8 (0.5-1.5) for measles-mumps-rubella vaccine; 0.9 (0.5-1.4) for measles vaccine; and 0.7 (0.4-1.0) for rubella vaccine. The study concluded that vaccination against hepatitis B, influenza, tetanus, measles

267

or rubella is not associated with an increased risk of MS or optic neuritis. In the early 1990s, several cases of demyelinating diseases were reported in France in association with the vaccination against hepatitis B. A large scientific, regulatory and public debate took place to reassure the growing concern of the population. Even on the basis of the early findings, which appeared to be compatible with a low increase in the risk associated with the vaccination, it was apparent that the risk-benefit profile was unchanged for newborns, and was essentially unchanged for adolescents and high-risk adults [54]. To further address the safety of immunisation in patients with MS in relation to increased risk of relapses after vaccination, the MS Council for Clinical Practice Guidelines commissioned a systematic review [55]. Upon conducting a meta-analysis of 130 articles, after screening 667 citations and 280 full-text articles, strong evidence was found against an increased risk of MS exacerbation after influenza immunisation. There was no evidence to suggest that hepatitis B, varicella, tetanus or Bacille Calmette-Guerin vaccines increase the risk of MS exacerbations. Insufficient evidence was found for other vaccines.

7. C O N C L U S I O N S The viral hypothesis in MS is hampered by the lack of evidence for a specific agent, in addition to the weakness of the results of analytical studies that have tested the association between MS and previous infections. None of the organisms so far investigated for a role in MS has gained acceptance as the causative agent. Although several studies have implicated the role of viruses in the etiology of MS, with the temporal relationship sometimes being impressive, many of the associations appear less than convincing, and even for those that seem to be on solid footing, there is no real understanding of the underlying mechanism(s). It is essential to rely on well-conducted systematic studies that produce valid and reliable estimates of the risk-associated profile of a causative viral agent, as many commonly circulating viruses may be indirectly activated under the autoimmune circumstances occurring in MS, and not necessary associated to these circumstances.

268

In conclusion, the strict criteria of evidencebased medicine that includes the detection of viral DNA/RNA in the brain or spinal cord of patients with MS, confirmation of epidemiological studies in a variety of geographic regions, and validation of viral antibodies by several laboratories and in comparison with the normal population have not come close to be satisfying. Additional studies will be needed to clarify any connection.

REFERENCES

1.

WeinshenkerB. The natural history of multiple sclerosis: update. Semin Neurol 1998;18:301-7. 2. Stinissen P, Raus J, Zhang J. Autoimmune pathogenesis of multiple sclerosis: role of autoreactive T lymphocytes and new immunotherapeutic strategies. Crit Rev Immunol 1997;17:33-75. 3. Olsson T, Sun J, Hillert J, Hojeberg B, Ekre H, Andersson G, Olerup O, Link H. Increased numbers of T cells recognizing multiple myelin basic protein epitopes in multiple sclerosis. Eur J Neurol 1992;22:1083-7. 4. KurtzkeJ. MS epidemiology world wide. One view of current status. Acta Neurol Scand 1995;161(Suppl.): 23-33. 5. Sadovnick A, Ebers G. Epidemiology of multiple sclerosis: a critical overview. Can J Neurol Sci 1993;20: 17-29. 6. Steiner I, Nisipianu P, Wirguin I. Infection and the etiology and pathogenesis of multiple sclerosis. Curr Neurol Neurosci Rep 2001; 1:271-6. 7. Benoist C, Mathis D. Autoimmune provoked by infection: how good is the case of epitope mimicry? Nat Immuno12001 ;2:797-801. 8. Levin M, Lee S, Kalume F, Morcos Y, Dohan F, Hasty K, Callaway J, Zunt J, Desiderio D, Stuart J. Autoimmunity due to molecular mimicry as a cause of neurological disease. Nat Med 2002;8:509-13. Lenz D, Lu L, Conant S, Wolf N, Grrard H, WhittumHudson J, Hudson A, Swanborg R. A chlamydia pneumoniae-specific peptide induces experimental autoimmune encephalomyelitis in rats. J Immunol 2001; 167: 1803-8. 10. Olson J, Croxford J, Calenoff M, Mauro C, Dal Canto M, Miller S. A virus-induced molecular mimicry model of multiple sclerosis. J Clin Invest 2001; 108:311-8. 11. Braun D, Dominguez G, Pellett P. Human herpesvirus 6. Clin Microbiol Rev 1997;10:521-67. 12. Patnaik, M, JB P. Intrathecal synthesis of antibodies to human herpesvirus 6 early antigen in patients with meningitis/encephalitis. Clin Infect Dis 1995;21: .

715-6. 13. Drobyski W, Knox K, Majewski D, Carrigan D. Brief report: fatal encephalitis due to variant B human herpesvirus-6 infection in a bone marrow-transplant recipient. N Engl J Med 1994;330:1356-60. 14. Mccullers J, Lakeman F, Whitley R. Human herpesvims 6 is associated with focal encephalitis. Clin Infect Dis 1995;21:571-6. 15. Knox K, Brewer J, Henry J et al. Human herpesvirus 6 and multiple sclerosis: systemic active infections in patients with early disease. Clin Infect Dis 2000;31: 894-903. 16. Ielsen L, Moller-Larsen A, Munk M, Vestergaard B. Human herpesvirus-6 immunoglobulin G antibodies in patients with multiple sclerosis. Acta Neurol Scand 1997; 169(Suppl.):76-8. 17. Challoner P, Smith K, Parker J et al. Plaque-associated expression of human herpesvirus 6 in multiple sclerosis. Proc Natl Acad Sci USA 1995;92i7440-4. 18. Moore F, Wolfson C. Human herpes virus 6 and multiple sclerosis. Acta Neurol Scand 2002;106:63-83. 19. Ito M, Baker J, Mock D, Goodman A, B lumberg B, Shrier D, Powers J. Human herpesvirus 6-meningoencephalitis in an HIV patient with progressive multifocal leukoencephalopathy. Acta Neuropathol (Bed) 2000;100:337-41. 20. Novoa L, Nagra R, Nakawatase T et al. Fulminant demyelinating encephalomyelitis associated with productive HHV-6 infection inan immunocompetent adult. J Med Virol 1997;52:301-8. 21. Arena A, Liberto M, Capozza A, Foca A. Productive HHV-6 infection in differentiated U937 cells: role of TNF alpha in regulation of HHV-6. New Microbiol 1997;20:13-20. 22. Selmaj K. Tumour necrosis factor and anti-tumour necrosis factor approach to inflammatory demyelinating diseases of the central nervous system. Ann Rheum Dis 2000;59(Suppl. 1):i94-i102. 23. Lycke J, Svennerholm B, Hjelmquist E, Frisen L, Badr G, Andersson M, Vahlne A, Andersen O. Acyclovir treatment of relapsing-remitting multiple sclerosis. A randomized, placebo-controlled, double-blind study. J Neurol 1996;243:214-24. 24. Ferrante P, Mancuso R, Pagani E, Guerini F, Calvo M, Saresella M e t al. Molecular evidences for a role of HSV-1 in multiple sclerosis clinical acute attack. J Neuroviro12000;6(Suppl. 2):S 109-S 14. 25. Rotola A, Caselli E, Cassai E, Tola M, Granieri E, Luca D. Novel human herpesviruses and multiple sclerosis. J Neurovirol 2000;6(Suppl. 2):$88-$91. 26. Soldan S, Leist T, Juhng K, McFarland H, Jacobson S. Increased lymphoproliferative response to human herpesvirus type 6A variant in multiple sclerosis

patients. Ann Neurol 2000;47:306-13. 27. Taus C, Pucci E, Cartechini E, Fie A, Giuliani G, Clementi M e t al. Absence of HHV-6 and HHV-7 in cerebrospinal fluid in relapsing-remitting multiple sclerosis. Acta Neurol Scand 2000;101:224-8. 28. Munch M, Riisom K, Christensen T, Moeller-Larsen A, Haahr S. The significance of Epstein-Barr virus seropositivity in multiple sclerosis patients.? Acta Neurol Scand 1998;97:171--4. 29. Larsen P, Bloomer L, Bray P. Epstein-Barr nuclear antigen and viral capsid antigen antibody titers in multiple sclerosis. Neurology 1985;35. 30. Martyn C, Cruddas M, Compston D. Symptomatic Epstein-Barr virus infection and multiple sclerosis. J Neurol Neurosurg Psychiatry 1993;56:167-8. 31. Haahr S, Koch-Henriksen N, Moeller-Larsen A, Eriksen L, Andersen H. Increased risk of multiple sclerosis after late Epstein-Barr virus infection. Acta Neurol Scand 1997; 169(Suppl.):70-5. 32. Ascherio A, Munch M. Epstein-Barr virus and multiple sclerosis. Epidemiology 2000; 11:220--4. 33. Levin L, Munger K, Rubertone M, Peck C, Lennette E, Spiegelman D, Ascherio A. Multiple sclerosis-and Epstein-Barr virus. JAMA 2003;289:1533-6. 34. Dreyfus D, Kelleher C, Jones J, Gelfand E. EpsteinBarr virus infection of T cells: implications of altered T-lymphocyte activation, repertoire:development and autoimmunity. Immunol Rev 1996;152:89-1~10. 35. Vaughan J, Riise T, Rhodes G, Nguyen M, ~Barrett~Connor E, Nyland H. An Epstein-Barr virus-related cross reactive autoimmune response in.multiple sclerosis in Norway. J Neuroimmunol 1996;69:95-102. 36. Wandinger K, JabsW, Siekhaus A, Bubel S, Trillenberg E Wagner H et al. Association between clinicaldisease activity and Epstein, Barr virus reactivation in MS. Neurology 2000;55:178-84. 37. Sanders V, Felisan S, Waddell A, Tourtellotte W. Detection.of herpesviridae in postmortem multiple sclerosis brain tissue and controls by polymerase chain reaction. J Neurovirol 1996;2:249-58. 38. Ito M, Barron A, Olszewski W, Milgrom E Antibody titers by mixed agglutination to varicella-zoster, herpes simplex and vaccinia viruses in patients with multiple sclerosis. Proc Soc Exp~Biol Med 1975;149:.835-9. 39. Marrie R, Wolfson C. Multiple sclerosis and varicella zoster virus infection: a review. Epidemiol Infect 2001"127:315-25. 40. :Berger J, Sheremata "W,Resnick L, Atherton S, Fletcher M, Norenberg M. Multiple sclerosis-like illness occurring with human immunodeficiency virus infection. Neurology 1989;39:324-9. 41. Gray F, Chimelli L, Mohr M, Clavelou P,, Scaravilli F, Poirier J. Fulminating multiple sclerosis-like leukoen-

269

42.

43.

44.

45.

46.

47.

48.

cephalopathy revealing human immunodeficiency virus infection. Neurology 1991 ;41:105-9. Graber P, Rosenmund A, Probst A, Zimmerli W. Multiple sclerosis-like illness in early HIV infection. AIDS 2000;14:2411-3. Serra C, Sotgiu S, Mameli G, Pugliatti M, Rosati G, Dolei A. Multiple sclerosis and multiple sclerosisassociated retrovirus in Sardinia. Neurol Sci 2001;22: 171-3. Dolei A, Serra C, Mameli G, Pugliatti M, Sechi G, Cirotto M, Rosati G, Sotgiu S. Multiple sclerosis-associated retrovirus (MSRV) in Sardinian MS patients. Neurology 2002;58:471-3. Gaudin P, Ijaz S, Tuke P, Marcel F, Paraz A, Seigneurin J, Mandrand B, Perron H, Garson J. Infrequency of detection of particle-associated MSRV/HERV-W RNA in the synovial fluid of patients with rheumatoid arthritis. Rheumatology (Oxf) 2000;39:950--4. Nakashima I, Fujihara K, Itoyama Y. Human parvovirus B19 infection in multiple sclerosis. Eur Neurol 1999;42:36-40. Maggi F, Fornai C, Vatteroni M, Siciliano G, Menichetti F, Tascini C, Specter S, Pistello M, Bendinelli M. Low prevalence of TT virus in the cerebrospinal fluid of viremic patients with central nervous system disorders. J Med Viro12001;65:418-22. Hernan M, Zhang S, Lipworth L, Olek M, Ascherio A. Multiple sclerosis and age at infection with common viruses. Epidemiology 2001; 12:301-6.

270

49. Sibley W, Bamford C, Clark K. Clinical viral infections and multiple sclerosis. Lancet 1985;8:1313-5. 50. Andersen O, Lygner P, Bergstrom T, Andersson M, Vahlne A. Viral infections trigger multiple sclerosis relapses: a prospective seroepidemiological study. J Neurol 1993;240:417-22. 51. Buljevac D, Flach H, Hop W, Hijdra D, Laman J, Savelkoul H, van Der Meche F, van Doom P, Hintzen R. Prospective study on the relationship between infections and multiple sclerosis exacerbations. Brain 2002; 125(Pt 5):952-60. 52. Horwitz M, Sarvetnick N. Viruses, host responses, and autoimmunity. Immunol Rev 1999;169:241-53. 53. DeStefano F, Verstraeten T, Jackson L, Okoro C, Benson P, Black S, Shinefield H, Mullooly J, Likosky W, Chen R, Vaccine Safety Datalink Research Group, National Immunization Program, Centers for Disease Control and Prevention. Vaccinations and risk of central nervous system demyelinating diseases in adults. Arch Neuro12003;60:504-9. 54. Jefferson T, Traversa G. Hepatitis B vaccination: Riskbenefit profile and the role of systematic reviews in the assessment of causality of adverse events following Immunization. J Med Virol 2002;67:451-3. 55. Rutschmann O, McCrory D, Matchar D, Immunization Panel of the Multiple Sclerosis Council for Clinical Practice Guidelines. Immunization and MS. A summary of published evidence and recommendations. Neurology 2002;59:1837--43.


Endogenous Retroviruses as Etiological Agents in Systemic Lupus Erythematosus Miranda K. Adelman, David E. Yocum and John J. Marchalonis

College of Medicine, University of Arizona, Tucson, AZ, USA

I. I N T R O D U C T I O N Great insights into the etiopathogenesis of Systemic Lupus Erythematosus (SLE) have been made in recent years and point toward a multi-factorial origin consisting of environmental, genetic and retroviral factors. Environmental factors include UV light, which has been shown to induce skin rashes [1], or certain drugs [2, 3], including hydralazine, procainamide or the use of birth control pills, that are known to induce disease and may be associated with flares. Genetic influences are indicated by sibling and familial studies where 'the concordance rate for monozygotic twins is >20% and for dizygotic twins is 2-3% [4-6]. Moreover, the use of murine markers [7], genome-wide scans of murine and human chromosomes [4], and linkage analyses [8-10] have enabled the identification of numerous genes that predispose to SLE [4, 6], as well as to other autoimmune diseases [7]. The New Zealand B lacUNew Zealand White (NZBAV) mouse model for SLE provided the first suggestion of a retroviral association for SLE by demonstrating the presence of a retroviral envelope (Env) protein related to murine leukemia virus (MuLV), termed gp70, in deposited immune complexes [11]. Hence, the search for a retroviral etiology for SLE was begun. 1.1. Retroviral Elements

Of the 3 billion DNA bases comprising our genome, an estimated 3% code for the 30,000-40,000 genes that are translated into the proteins essential for life [12]. The evolutionary significance of the non-

coding "junk DNA" is not completely understood, although it is believed to function in packaging and gene expression [ 13]. Intriguingly, an estimated 50% and likely more [12, 14, 15] of the human genome is derived from the integration of various retroviral elements, which are divided into four groups based on their genomic organization: exogenous retroviruses, human endogenous retroviruses (HERVS), retrotransposons and retroposons. In contrast to exogenous retroviruses that are passed horizontally and require a replication cycle where proviral DNA is integrated into the host DNA, HERVs are transmitted vertically as stable Mendelian elements [ 16, 17]. Accounting for approximately 8% of the genome [12], it is generally believed that most HERVs integrated into the human lineage as exogenous progenitors prior to the divergence of hominid from Old World Primates [ 17]. The third group of retroelements, the retrotransposons, differs from both exogenous and endogenous retroviruses by lacking the env gene. Additionally, some truncated HERVs are classified as retrotransposons when lacking functional env genes. Although sequence analyses have revealed that the genomes of most HERVs are disrupted by frame shift mutations, termination codons and deletions, some HERVS are transcriptionally active, are expressed in a tissue specific manner and produce functional retroviral proteins [17-19]. The retroposons include the long interspersed elements (LINEs) and the short interspersed elements (SINEs) [17], most of which are present in high copy numbers and are found in species as diverse as sharks and humans [20]. HERVs share similar genomic structures as

271

A) Exogenous retroviruses MatrixMA~ Capsid C A Nucleocapsid N C Protease a PR

I

~

Protease a PR Surface glycoproteins SU Reverse transcriptase RT T r a n s m e m b r a n e protein T M Integrase IN

B) Endogenous retroviruses (HERVs)

gagHi pol ~ l l l l k ~ C) Retrotransposons D) Retroposons

Long interspersedelements (LINES) 5 ' ~ O1~ 1[ ~ A A A A - ~ , , Short interspersedelements (SINES)

5'91DqltRNA I-AAAAFigure 1. Genetic organization of retroviral elements. (A) Exogenous retroviruses, (B) human endogenous retroviruses (HERVs), (C) retrotransposons, and (D) retroposons all share similar genetic information. Flanked by direct repeats (I~) and long terminal repeats (LTRs) are the retroviral group specific antigen (gag) gene which encodes MA, CA, NC and sometimes PRa, the polymerase (pol) gene which encodes the RT, IN and sometimes PRa and the envelope (env) gene which encodes SU and TM. Although they share the same genomic organization, HERVs differ from exogenous infectious retroviruses in their inability to bud from the cell membrane and lack of infectivity. Retrotransposons (and some truncated HERVs) do not have the env gene and they too are consequently non-infectious. The retroposons, also called non-LTR retrotransposons, include the SINEs and LINEs. LINEs carry genes for a promotor (P), an open reading frame (ORF1) and pol, and have a polyA tail (AAAA) at their 3'end. SINEs have a promotor region, tRNA and a polyA tail, but do not encode an ORF or pol gene. SINEs consequently are dependent on LINEs for their replicative enzymes.

exogenous retroviruses in that they carry sequences homologous to the retroviral group specific antigen (gag), polymerase (pol) and env genes that are flanked by long terminal repeats (LTRs) (Fig. 1). Briefly, the gag gene codes for the matrix, capsid and structural core proteins of the virus, while the pol gene encodes reverse transcriptase (RT), which copies viral RNA into DNA, and protease and integrase, which facilitate protein cleavage and integration of proviral DNA into the host genome, respectively. Finally, the env gene codes for viral membrane proteins and mediates binding of the virus to its receptor and subsequent entry into the host cell, the necessary first steps in establishing an infection (for an exogenous retrovirus). The LTRs contain inverted repeats, the TATA box, promotors,

272

enhancers, polyadenylation signals, trans-activation regions and a tRNA primer binding site [15, 17]. However, in contrast to exogenous retroviruses, HERVs stop short of viral budding and consequently are non-infectious [ 13].

1.2. Discovery of HERV Families The first HERV was identified in 1981 by Martin et al [21]. Over the next 23 years, molecular techniques including screening human genomic libraries under low stringency conditions with DNA probes from animal retroviruses [ 13, 17, 22, 23] and polymerase chain reaction (PCR) using degenerate retroviral primers capable of recognizing all known exogenous retroviruses [24] have facilitated the iden-

Table 1. Classification of human endogenous retroviruses (HERVs) by polymerase (reverse transciptase) gene homologi and tRNA primer binding site~ Class 1:

c:.t~ related Family 1- HERV-HF HERV-H (RTVL-H, RGH) HERV-F Family 2: HERV-RW HERV-W HERV-R (ERV9) HERV-P (I-IuERS-P, HuRRS-P) F_ami!y Y: HERV-ER1 HERV-E (4-1, ERVA0 NP-2) 51-t HERV-R (ERV3) Family 4~ HBRV-T HERV-T (S71, CRTKI, CRTK6) Family 5: HERV-IP HERV-! (RTVL-D HERV-IP-T47D (ERV-FTD) Family 6': ERV-FRD ERV-FRD

Class 2: B- & D-type related, HER V-Kb Family 1: HML-I HERV-K (HML- 1.1) Family 2:HML-2 HERV-K10 HERV-K-HTDV Family 3:HML-3 HERV-K (HML3.1) Family 4:HML-4 HERV-K-T47D Family 5:HML-5 HERV-K-NMWV2 Family 6:HML-6 HERV-K (HML-6p) Family 7:HML-7 HERV-K-NMWV7 Family 8:HML-8 HERV-K-NMWV3 Family 9:HML-9 HERV-K-NMWV9 Family 10: HML- 10 HERV-KC4

"Note that this scheme omits characterized HERVs including HRES-1 and HERV-16. bAll Class 2 HERVs use lysine (K) as their tRNA primer binding site. cGrouped into the ER1 superfamily based on substantial homologies in the polymerase, envelope and group specific antigen genes of murine leukemia virus and baboon endogenous virus.

tification and characterization of over 20 species of HERVs [13] (Table 1). The analysis of human chromosomes and loci, coupled with information and knowledge gained from the human genome project, has further contributed to the identification and study of HERVs. HERVs are classified based on sequence similarity to animal retroviruses [16]. Animal retroviruses, belonging to the virus family Retroviridae, containing subfamilies oncovirinae, lentivirinae and spumavirinae, are classified by morphological and biological features, in addition to the observation of retroviral particles in infected cells, as established by the International Committee for Taxonomy of Viruses [25]. Retroviruses are further divided into A-type retroviruses, which are devoid of an envelope and subsequently are seen only in infected cells, and B-, C- and D-type retroviruses, which are enveloped, produce extracellular particles and consequently are infectious. The classification of both exogenous and endogenous retroviruses, the latter

based on similarity to exogenous retroviruses, has been difficult since many exogenous retroviruses were originally named after multiple investigators/ discoverers, or by the disease they caused or host cells they infected [17], for example, the human T cell leukemia viruses (HTLV-1 and-2). A tentative plan for naming HERVs was based on the single letter amino acid (code) tRNA primer binding site used by the virus (i.e. HERV-K uses AAU, lysine, k) [13, 15, 17]. However, distantly related HERV families use the same tRNA primer binding sites [13] and hence were grouped. Today, HERVs are divided into 3 classes based on RT/pol gene homologies to exogenous animal retroviruses [16]. Class 1 HERVs are related to mammalian C:type retroviruses and are subdivided into 6 families, 3 of which have been grouped into the ER1 superfamily based on homologies to the murine leukemia and baboon endogenous retroviruses in the conserved pol gene, as well as in the gag and env genes [13, 17]. Class 2 HERVs share homologies to mamma-

273

lian A-, B- and D-type retroviruses and are divided into 10 families based on pol gene homologies [ 13, 16, 22]. All class 2 HERVs possess a lysine (K) tRNA primer binding site, hence the name HERVK, as do B- and D-type mammalian retroviruses [17]. HERV-K retroelements are termed the most biologically active HERVs [13, 26], have long open reading frames (ORFs) encoding all of the retroviral genes [ 17, 27] and are highly expressed in over 20 human teratocarcinoma cell lines that were derived from teratocarcinomas or embryonic carcinomas [17, 27]. Lastly, class 3 HERVs are related to spumaviruses (foamy virus) and consist of only 1 member, HERV-L [13]. This nomenclature/classification scheme, however, is not without its own limitation, since it omits various HERVs that have already been characterized. In particular, HTLV-1related endogenous sequence (HRES-1) has been implicated as a potential etiological agent in SLE [15, 18, 28, 29], Multiple Sclerosis (MS) [30] and Sjrgren's Syndrome (SJS) [31], and is the source of autoimmunity in MRL/Ipr mice [32-34], as it integrated into the fas gene and thereby prevents expression of Fas protein on various cell types, including activated lymphocytes. HRES-1 is a Class 1 HERV, although it shares only limited homology to the HTLV-1 LTR region [ 13], but is less related to the Class 2 HERV-K endogenous sequences.

CD4 on T cells and macrophages that likely induces the immune abnormalities seen in HIV infection. Gpl20-stimulation of the T cell receptor (TCR) specifically results in tyrosine kinase activation of p561ck, activation of CD4 T cells, internalization (down modulation) of CD4, CD4 T cell anergy and apoptosis, and finally TCR inactivation [19]. In the case of gpl20-stimulation of macrophages, the binding of gpl20 to CD4 mediated Th2-cytokineinduced (IL-6, IL-10, TNFt~) polyclonal B cell activation and increased production of various chemokines, including RANTES and MCP-1, as well as the down modulation of CD4 [19]. Interestingly, IL-16 in SLE patients is thought to act on CD4 in a manner similar to that exerted by gp120 in HIV-infected persons [19]. Again reviewed by Sekigawa et al [19], IL-16 is produced from activated CD8 T cells and serum titers are increased in SLE [35, 36] and HIV infection [37]. Furthermore, increased serum titers of IL-16 are strongly correlated to active disease in some SLE patients [36], and perhaps serve as a marker for these individuals. The IL-16 receptor is located on the CD4 molecule on CD4 T cells and consequent binding by IL-16 may promote T cell activation, anergy [38] and perhaps apoptosis of CD4 cells, resulting in a shift in the CD4/CD8 ratio.

2.2. Low Incidence of HIV Infection in Exposed Individuals with SLE 2. C O R R E L A T I O N S B E T W E E N SLE AND HIV

2.1. Immune Abnormalities There is an impressive array of immune abnormalities in common to patients with HIV and SLE. As reviewed by Sekigawa et al [19], both diseases are characterized by polyclonal B cell activation, a decrease in the CD4/CD8 T cell ratio, T cell anergy, increased expression of major histocompatibility complex (MHC) class II by CD4 and CD8 T cells, defective CD4 and CD8 function and a similar shift in cytokine profile from T helper 1 (Th 1) to T helper 2 (Th2). In both diseases, the increased expression of MHC class II on CD4 and particularly CD8 T cells is associated with disease activity and is indicative of the early activation of T cells, which may lead to anergy [19]. It is the binding of HIV gpl20 to

274

Based on prevalence data for both diseases [39], it is estimated that approximately 400 Americans should have both SLE and HIV, as opposed to the reported 20 or so individuals with both diseases [ 19, 40]. Even though there have been numerous cases of HIV transmission with organ transplantation, SLE patients who received unscreened blood from 1978-1983 failed to develop HIV infection [41]. Additionally, in the few patients who have both SLE and HIV, increases in HIV viral load were seen after control of lupus flares with immunosuppressant therapy [42]. On the other hand, flares of SLE were seen following highly active antiretroviral therapy (HAART) for HIV infection in patients with both diseases [40]. Still yet, there have been cases when SLE resolves or improves with HIV infection or progression of HIV-related immunodeficiency [39], most likely as a result of augmentation or depletion

of CD4 T cells. Although the use of immunosuppressant drugs in SLE or HAART therapy for HIV might explain some of these findings in patients with both diseases, the numbers alone raise the possibility that SLE patients produce factors that are protective against infection with HIV since the proportion of people with both diseases is much lower than predicted. Intriguingly, IL-16 in SLE may exert protective effects against infection with HIV. Again, CD8 T cells produce IL-16 and it is thought to act upon its receptor, the CD4 molecule on CD4 T cells, in a manner analogous to that exerted by HIV gpl20. IL-16 inhibits HIV infection in vitro, most likely by repressing the HIV promotor and hence transcription via signaling incurred by the interaction of CD4 and IL-16 [19, 38, 43]. Elevated levels of IL-16 seen in SLE patients [ 19, 36], therefore, may be protective against infection with HIV perhaps by competitive inhibition of binding to CD4 by IL-16, although gpl20 and IL-16 do use different epitopes on the CD4 molecule [19, 43].

2.3. Retroviral-Type Activity in SLE Patients Major lines of evidence in support of a retroviral link to the etiopathogenesis of SLE and other autoimmune diseases include the presence of antiretroviral antibodies and the isolation of retroviral-like particles from autoimmune patients. Patients with SLE [31, 44-47], RA [48], SJS [31, 49, 50] or MS [31] are known to produce IgM and IgG antibodies reactive with various retroviral proteins, including Gag, Env, Nef (negative regulation factor) and the p24 capsid protein. However, in some of these cases, PCR using exogenous retroviral primers failed to amplify sequences related to HIV- 1 or HTLV- 1 [51 ]. Since PCR did not detect the presence of HIV-1 or HTLV-1 in autoimmune patients with measurable antiretroviral antibody titers, scientists were perplexed as to why these patients produced antiretroviral antibodies in the absence of exogenous retroviral infection and questioned what affect, if any, the antiretroviral antibodies had on the SLE disease process. Intriguingly, as many as one third of SLE patients are reported to have antibodies reactive to peptides based on the HIV p24 capsid sequence [44, 45], while up to 52% of SLE patients and 48% of patients with

other autoimmune diseases produce antibodies reactive with the HRES-1 endogenous retrovirus, as compared to approximately 4% of normal individuals [18, 52]. Furthermore, comparative sequence analysis has revealed remarkable sequence similarities between HIV genes, in particular gag, and the genes encoding common nuclear antigens [53]. To be discussed below, molecular mimicry, therefore, between autoantigens and retroviral antigens might explain the presence of antiretroviral antibodies in SLE patients [ 18, 28, 54, 55]. The first reports of retroviral-like particles isolated from the lips or salivary glands of patients with SJS [56, 57], from the synovium of patients with RA [48, 58-60], from the peripheral blood of SLE patients [48, 61] or from cells cultured from patients with MS [62-64] were reported in the 1990s. Additionally, detection of an atypical interferon characteristic of lentiviruses and RT activity in the supernatant was induced when lymphocytes from SLE patients were cultured [65, 66], perhaps indicating that retroviral gene products are involved in the SLE disease process. Thus, first the observation of antiretroviral antibodies in autoimmune patients and second the isolation of retroviral-like particles from autoimmune patients, spurred investigators to search for a retroviral etiology for SLE and other autoimmune diseases.

2.4. Retroviral Influences in SLE Murine Models Two mouse models for SLE, in particular the NZB/ W and MRL/lpr models, point to a strong retroviral influence in the etiopathogenesis of disease. Of interest with regards to both of these models is that they produce pathogenic antibodies to the MuLV-related Env protein, gp70 [ 11, 67, 68], which is involved in immune complex deposition in the kidneys [67, 69]. NZB/W mice exhibit severe proliferative glomerulonephritis resulting in thickening of the basement membrane and obliteration of the capillary lumina as a consequence of immune complex deposition, in addition to other autoimmune phenomena [34, 70]. Furthermore, the isolation of a HERV Env protein related to MuLV in deposited immune complexes of NZB/W kidneys suggests that the protein is significantly involved in pathogenesis, in particular glomerulonephritis [11, 67, 68, 70, 71]. Addition-

275

ally, cDNA microarray analyses of renal cortex RNAs taken from NZB/W mice and NZW control mice, identified the most up-regulated gene (5.5fold as compared to NZW mice) as corresponding to the endogenous MuLV-related to the Duplan retrovirus (EDV, L08395) [70]. In addition to microarray analyses, histopathology demonstrated that the increased expression of the EDV transcript occurred by ,4-8 weeks of age, prior tothe onset of inflammation in the kidneys of NZB/W mice [ 11, 70], thereby indicating that increased expression of EDV is not aresult of and:actually proceeds inflammation [70]. Thus, these findings suggest that proteins translated from the MuLV-related EDV are involved in deposited immune complexes that result in glomerulonephritis in NZB/W mice. MRL/lpr mice demonstrate direct evidence of a HERV influence, specifically HRES-1, in the etiopathogenesis of autoimmune disease. The autoimmunity in MRL/lpr mice results from the integration of HRES-I into the fas gene, located on chromosome 1 at positionq42 (Clq42) [32, 33]. Integration of HRES-1 results in decreased expression of Fas protein on cell surfaces and the consequent lack of apoptosis in certain cell populations, including activated T lymphocytes [32, 33]. Activation of Fas by its ligand initiates the necessary signaling for apoptosis of cells expressing Fas, most likely via activation-induced cell death (AICD) [32, 33, 69]. The region at Clq42 in the mouse has been identified by repeat linkage analyses as a region conferring predisposition to SLE-like disease in NZB/W mice [8, 34, 72]. Interestingly, Clq42 has been identified as a susceptibility region for human SLE as well [8-10, 32, 33, 73, 74], although it is not the gene for human fas.

3. MECHANISMS OF AUTOIMMUNITY INDUCED BY HERVS There are a variety of proposed mechanisms by which HERVs may initiate and/or perpetuate autoimmune responses (Fig. 2). Although the major focus here is molecular mimicry between HRES-1 and the common autoantigen the small ribonucleoprotein complex (snRNP) in the etiopathogenesis of SLE, other potential mechanisms and a limited number of examples utilized by HERVs in the

276

initiation of various autoimmune or inflammatory processes are explored, briefly. However, a point worth making is that no single HERV is unique to any particular autoimmune disease, but rather different autoimmune diseases share various features, potentially those brought about by HERVs. The concepts of HERV-encoded superantigens and skewing of the V~ T cell repertoire in MS and Insulin-Dependent Diabetes Mellitus (IDDM) are discussed, as are cases of retroviral integrations into genes that are critical in the control and regulation of the immune system. Still yet, the concepts of HERV-encoded cis- or trans-regulatory elements and the immunosuppressive effects of HERV proteins are considered. 3.1. Superantigens Superantigens are non-processed, non-MHC restricted peptides that are produced by many bacteria, mycoplasmas and viruses. They bind to conserved regions of the MHC class II molecule, outside of the classic peptide-binding groove, and specific TCR 13-chain variable (V]3) regions, irrespective of the antigen-specificity of the TCR. The selective expansion or deletion of T cells with specific V~ regions is an inherent feature of superantigens and they are capable of rapidly activating 106 more T cells than would be activated by presentation in the classic peptide-binding groove [75]. Activation in this fashion may lead to oligo- or polyclonal activation of certain VI3 subsets, leading to cytokine and chemokine production, systemic toxicity and suppression of the adaptive immune response. Superantigen-encoded TCR V~l expansion is associated with MS [76], RA, SLE and IDDM [77-80]. An enrichment of V137 T cells was found in pancreatic islet cells from patients with acute-onset IDDM [78]. Interestingly, these authors [78] isolated a novel HERV, (1,2) termed IDDMK(1,2)22, a member of the HERV-K family, from patients with IDDM. Conrad et al [78] hypothesized that the expansion of VI]7 T cells was induced by the IDDMK(1,2)22-encoded superantigen and that these T cells were involved in islet cell destruction. Although other groups have failed to duplicate these findings [81-87], the results are still intriguing. Genetic mapping has identified another HERV, HERV-K18, with 99.5% sequence homology in

Figure 2. The multi-factorial etiopathogenesis of SLE. Environmental, genetic and retroviral factors contribute to the overall SLE phenotype. Genetic factors are indicated by familial association and MHC haplotype, as well as the extreme gender imbalance seen in SLE. Retroviral factors are indicated by the presence of antiretroviral antibodies and retroviral-like particles in SLE patients. Significant regions of protein sequence homology between retroviral proteins and autoantigens exist and explain the presence of antiretroviral antibodies in SLE patients. Antiretroviral antibodies may serve to increase protection against infection with HIV in patients with SLE. Additionally, linkage analyses have identified a SLE susceptibility locus on chromosome 1 at position q42 (Clq42) that contains the well-characterized HRES-1 HERV. Environmental factors such as UV light, physical and/or emotional stress and the use of certain drugs may act upon genetic and retroviral factors by increasing transcription of HERVs. the 3'LTR of the env region to that encoded by IDDMK(1,2)22 [88]. It is located in the first intron of the CD48 gene on chromosome 1 and has three allelic env forms, all of which demonstrate superantigen-type activity by mediating the rapid expansion of V~7 T cells [79, 88]. Interestingly, the chromosomal region containing CD48/HERV-K18 has been identified as a susceptibility region for IDDM [89] and peripheral expansion of V137 T cells occurs prior to the onset of clinical disease [90]. Additionally, since CD48 is one of two ligands for CD2, it is conceivable that HERVs mediate expression of cell surface markers and thus may be intimately involved in the etiopathogenesis of !DDM, perhaps by impaired CD4 T cell activation [88]. Furthermore, IFN-cz was shown to up-regulate transcription of the HERVK18 env gene, resulting in the rapid expansion of

V137 T cells [79]. Since IFN-cz is produced from virus-infected cells, these findings suggest that viral infection may lead to expansion of V137 T cells by increased transcription of the HERV-K18 superantigen. The HERV-KI8 superantigen was also shown to selectively expand V~13 T cells [80], which are also correlated with IDDM [90]. A HERV-encoded superantigen may be associated with T lymphocyte immunopathophysiology in MS. MS associated retrovirus (MSRV), isolated from B lymphocyte cultures derived from MS patients or directly from cerebral spinal fluid (CSF) [62, 91], is a member of the class I HERV-RW family [92]. To investigate the possibility of MSRVmediated immunopathology through a superantigen-type mechanism, Perron et al [76] analyzed in vitro whether infection of PBLs from non-MS

277

individuals with MSRV particles or a recombinant MSRV Env protein resulted in the selective expansion or deletion of T cells bearing a particular V[3 chain. It was found that a significant polyclonal expansion or deletion of V~ 16 T cells was induced following inoculation of PBLs with MSRV particles, regardless of MHC class II haplotype [62]. Furthermore, polyclonal modification of V~16 and V[~17 T cell populations was seen after inoculation of PBLs with recombinant MSRV Env protein [62]. Thus, the immune response induced by MSRV is abnormal in that it is characterized by the polyclonal activation of na'fve T cells bearing a particular V~ chain independent of MHC class II haplotype, thereby suggesting that the HERV protein is critical to the superantigen-type activity seen in MS.

3.2. lnsertional Mutagenesis There are several examples where HERVs have integrated into or nearby genes that are critical in the control of the immune system, including fas, complement and the MHC. The MRL/lpr murine model for SLE provides such an example as HRES1 integrated into the murine fas apoptosis-promoting gene, resulting in decreased expression of Fas protein and the consequent failure of apoptosis in activated autoreactive lymphocytes [32, 33, 93]. Hence, the MRL/lpr model provides direct evidence of immunopathology as a consequence of retroviral integration. 3.2.1. MHC class I and H genes

Numerous HERVs, as well as various other retroelements, are found within the classical MHC genes [22, 94-96]. Kulski et al [95] analyzed 16 HERV sequences belonging to the HERV-16 (11 copies), HERV-L (1 copy), HERV-I (2 copies), HERV-K91 (1 copy) or HARLEQUIN (1 copy) families within 656kb of genomic sequence obtained from the ~and [3-blocks of the MHC class I region. The HERV16 copies most likely arose as a result of duplication of genomic sequences containing the human MHC class I and PERBII (MIC) genes, while sequences related to the other HERV families probably arose following duplication after a single insertional event or translocation [96]. Additionally, 4 of the 11 copies of HERV-16 and the single copies of HERV-I

278

and HARLEQUIN appear to have receptors facilitating the insertion of other retrotransposons [96], a mechanism to further increase the diversity and polymorphism of the MHC. In addition to providing evolutionary clues, the identification and characterization of HERVs within the MHC class I locus is of particular interest since this region is rich in polymorphic genes that have been associated with various autoimmune or inflammatory diseases, as well as with disease susceptibility following viral infection [97]. Chimpanzees, for example, have a large deletion/transposition in the MHC class I region [98], which probably includes the HERV-L, HERV16 or PERB 11 (MIC) sequences [95]. In contrast to the immune response to HIV in humans, chimpanzees that are actively infected with simian immunodeficiency virus (SIV) are capable of mounting an effective antibody response, preventing progression to AIDS [99]. Therefore, it is possible that the deletion of genes or HERVs in the MHC class I region in chimpanzees may influence susceptibility and progression to AIDS [95, 98]. HERV integrations into the MHC class II region have also been reported and are a factor influencing the polymorphic nature of the various DR haplotypes in particular [22, 94]. Distributed throughout the class II region are numerous (H)ERV9 LTRs that may be transcriptionally active in various cell types [22, 100]. Moreover, the (H)ERV-9 LTRs in the DR region contain regulatory elements capable of mediating retroviral basal and tissue-specific transcription, perhaps by functioning as IFN-y responsive elements [94]. Interestingly, IFN-y is the most potent inducer of gene expression in the DR region and consequently the (H)ERV-9 LTRs might act as IFN-y specific/responsive enhancers for the DR genes [79]. 3.2.2. Complement genes

Inherited deficiencies in components of the classical complement system are associated with SLE, RA and scleroderma [ 101,102]. The ability to clear pathological immune complexes and infectious organisms may occur in a state of complement deficiency, resulting in tissue deposition particularly to the basement membranes of the kidneys, as is seen in lupus nephritis. The most characterized inherited complement deficiency is the partial deficiency of

complement component C4 [ 101, 103,104]. The two isotypes of C4, C4A and C4B, differing by only five nucleotides, are encoded by two very polymorphic genes within the non-classical MHC class HI region [102]. Numerous alleles for both isotypes have been identified, including the null alleles C4AQO and C4BQO. The inheritance of at least one of the C4AQO null alleles occurs in as many as 50% of SLE patients, as compared to 25% of controls [ 101, 103] and is associated with certain ethnic groups [ 105]. Furthermore, a 30kb deletion of most of the C4 gene and the adjacent 5'-21-hydroxylase-A (21OHA) pseudogene is thought to account for twothirds of C4A deficiency in Caucasian SLE patients with the MHC haplotype B8-C4AQO-C4B1-DR3 [102]. Intriguingly, the C4A and C4B genes contain the complete 6.4kb sequence of HERV-K(C4), present in intron nine and absent in the case of the C4A/21-OHA deletion [106]. HERV-K(C4)-related sequences have also been found in the second intron of the C2 complement gene [ 107].

Furthermore, the translated cORF protein accumulates in the nucleolus, indicating that cORF harbors a functional nucleolar localization signal, as do Rev proteins [108]. With respect to functioning as a cisor trans-regulatory element, the HTDVTHERV-K Rev-related cORF protein may exert a pathogenic role by activating or suppressing genes involved in cellular growth and/or immune function. In another study, Horwitz et al [ 109] cloned a family of endogenous HIV-related sequences from the DNA of normal humans, chimpanzees and rhesus monkeys by low-stringency Southern blot hybridization and plaque screening. They identified a protein, termed EHS-2, similar in size, amino acid composition and structure to the arginine-rich RNA binding domain of Rev, complete with nucleolar localization motif [109]. Thus, the findings of HERV encoded Revrelated proteins potentially capable of trans-activating viral gene expression indicates that HERVencoded transcriptional transactivators related to Tat in lentiviruses or Tax in HTLV may exist.

3.3. Cis/Trans Activation

3.4. Immunoregulatory Proteins/Peptides

Although there are no definitive examples of cisor trans-activation of cellular genes by HERVs, there is presumptive evidence in support of such an event. Lower et al [108] explained that the human teratocarcinoma-derived particles (HTDV) present in numerous human teratocarcinoma cells lines are encoded by HERV-K (HTDV/HERV-K). Two forms of HTDV/HERV-K proviral genomes exist, type I and type II, differing by the absence (type I) or presence (type II) of nucleotides encoding the amino-terminus of the env gene and a putative signal sequence that overlaps the carboxyl-terminus of the pol gene. In type I HTDV/HERV-K genomes, the pol and env genes are fused, the proviruses are defective in env splicing and full-length transcripts therefore accumulate. On the other hand, type II transcripts are spliced, resulting in subgenomic env mRNA and a short open reading frame (cORF) of 14KDa with some sequence, structural and functional similarities to the RNA binding and effector domains of the lentivirus rev gene [108]. Like Rev, cORF contains an arginine-rich basic motif at its amino-terminus and a leucine-rich motif at its carboxyl-terminus, homologous to the Rev RNA binding and effector domains, respectively [108].

In addition to directly affecting cellular gene expression by integration, HERV gene products may contribute to autoimmune processes by their actions on cellular genes and physiology. The conserved transmembrane Env protein of C-type mammalian retroviruses and class I HERVs, p l5E, exerts an immunosuppressive effect on monocytes and lymphocytes in vitro [ 110-113] and in vivo [ 114, 115]. Haraguchi et al [ 115] demonstrated that a synthetic peptide corresponding to the p l5E Env protein, CKS-15, suppressed stimulant-induced mRNA expression of the Thl cytokines IL-2, IL-12 and IFN-% but did not suppress the Th2 cytokines IL-4, IL-5, IL-6 and IL-13. Interestingly, densitometric analyses of the RT-PCR products showed that CKS17 peptide up-regulated mRNA accumulation of ILl0, a cytokine capable of inhibiting cell-mediated immunity [116-118], while inhibiting stimulantinduced mRNA expression of IL-12 [115], a critical cytokine that induces Thl responses and inhibits Th2 responses [ 119, 120]. Furthermore, Cianciolo et al [ 121] found that a region of the HIV transmembrane Env protein, gp41, is homologous in sequence to p 15E and exerts inhibitory effects on lymphoproliferative responses when stimulated with anti-CD3

279

monoclonal antibody and IL-2. Additionally, they found that human neoplastic effusions contain proteins that are potent inhibitors of monocytes and are detected with anti-pl5E antibodies [121, 122]. These results suggest that pl5E and/or similar HERV-derived peptides may be involved in the immune dysregulation associated with autoimmune disease, HIV infection or cancer.

4. M O L E C U L A R MIMICRY IN THE ETIOPATHOGENESIS OF SLE An additional mechanism by which HERVs or HERV gene products initiate autoimmunity is by mimicry of host structures, sometimes leading to pathological autoantibody production. The literature now supports molecular mimicry between common autoantigens and retroviral proteins as a significant contributor to the overall SLE, MS and SJS autoimmune/inflammatory phenotypes [ 15, 28-31, 53, 54]. For example, database searches have identified a region of the MHC class I E antigen as having extremely high homology to the MuLV-related HERV-E clone 4-1 gag region [54], suggesting that HERV clone 4-1 transposed into the MHC class I E antigen. Interestingly, 48% of Japanese SLE patients have antibodies reactive with the clone 4-1 Gag region, while 11% have antibodies to the Env region [54]. These antibodies were not detected in normal individuals [123, 124]. The class I HERV, HRES-1, is of particular interest to SLE, as well as to MS and SJS, and is proposed to participate in autoimmune processes via its remarkable cross-reactivity with a Gag-related region of the 70KDa (U1) component of snRNP ((U1)snRNP) [18, 53], a common autoantigen associated with SLE, scleroderma and polymyositis [125]. Autoantibodies to snRNP, as well as to other nuclear proteins, are associated with immune complex formation, pathological tissue deposition and the recruitment of inflammatory cells and complement [126, 127]. We [28] and others [18, 53, 128-131] believe that molecular mimicry between HRES- 1 and (U 1)snRNP initiates the production of autoantibodies cross-reactive with both proteins, and that these autoantibodies induce pathology by the formation of immune complexes that subsequently result in tissue deposition and may serve to constantly fix complement.

280

4.1. HRES-1

HRES-1 integrated into the human genome during the time of the Old World Primates, most likely as an exogenous and as of yet unidentified retrovirus [31 ]. HRES-1 is transcriptionally active, contains a tRNA primer binding site, a polyadenylation signal, a TATA box, an HIV-1 trans-activation region, inverted repeats and is expressed in a tissue-specific manner [15, 18, 29, 131]. It has been mapped to Clq42, a region identified as a murine [8, 72] and human [8, 10, 74, 132] lupus susceptibility region (discussed below), and is present as a single haploid copy [31]. HRES-1 was (one of) the first HERV(s) identified as having ORFs with the capacity to encode a functional protein [31]. The 28KDa protein encoded by HRES-1 is believed to serve as an autoantigen for Gag-reactive antibodies in patients with autoimmune disease [ 18, 29, 31]. Another relevant feature of HRES-1 that pertains to SLE-like autoimmunity is the consequence of its integration into the murine fas apoptosis-promoting gene in MRL/lpr mice [32, 33, 93]. These mice develop severe lymphoproliferative disease and systemic autoimmune disease similar in features to human SLE [32, 133, 134]. 4.2. Anti-snRNP/HRES-1 Antibodies in SLE

The early 1990s noticed the association of antiretroviral antibodies in autoimmune disease [31, 44, 48, 60, 123, 135-139]. Approximately one-third to one-half of patients with SLE or SJS had circulating antibodies reactive to the HIV-1 p24 capsid protein, as well as to the Gag, Env and Nef proteins of HIV-1 [ 18, 44, 45, 50, 136], in the absence of retroviral infection [ 18, 51 ]. With regard to anti-HRES- 1 antibodies specifically, up to 52% (50/96) of SLE patients had antibodies to HRES-1 [ 18, 31 ], as compared to 3.6% (4/111) of normal donors and none of the 92 patients with either asymptomatic HIV or clinical AIDS [ 18]. It thus appears that anti-HRES- 1 antibodies are present in a majority of patients with SLE, but are not present in patients infected with HIV. Furthermore, Perl et al [18] demonstrated a correlation between the presence of antibodies to HRES-1 and the presence of antibodies to snRNP such that SLE patients with antibodies to HRES-1 were 2.3 times as likely to have clinically active

70KDa snRNPI 5 0 - D P R D A P MoMuLV p30Gag i p30Gag/snRNP~ HRESpl9] 14 HRESp24U

P P T R A E T R ~ E E ' R M ! E R K R R : : E K I,E R R Q Q - 8 0 5 2 3 -ETPEEI~E~R I R.RiE T E E K E - 5 4 0 { - PTRAPSGPRPP24 117 - ~ ~ G

PDRS

PR- 12 7

Figure 3. Sequence alignment of snRNP, p30Gag and HRES-1 peptides. Comparative sequence analyses have revealed regions of amino acid sequence homology between a Gag-related region (p30Gag) of the 70KDa U 1 small ribonucleoprotein complex ((U1)snRNP) (residues 50-80), the Gag region of the Moloney Murine Leukemia virus (MoMuLV) (residues 523-540) [53] and peptides based on the p19 (HRESpl9) and p24 (HRESp24) Gag proteins of HTLV-1 related endogenous sequence (HRES-1) [18, 31]. Residues in agreement between snRNP, p30Gag and HRES-1 are shaded in dark gray boxes and the cross-reactive consensus epitope is listed as the third sequence from the top. The fight gray shaded boxes represent homologous residues between snRNP and the HRESp 19 Gag peptide. Italicized residues in the HRES-1 peptides are residues homologous to the p19 and p24 Gag regions of HTLV-1 [31]. disease [ 18]. The (U1)snRNP protein contains regions with amino acid sequence homology to a portion of the retroviral Gag protein (p30Gag) [53]. Using a synthetic snRNP peptide based on the 70KDa (U1)snRNP protein, Query and Keene [53] demonstrated that when snRNP peptide was pre-incubated with anti-p30Gag serum, the anti-p30Gag serum lost its ability to bind the (U 1)snRNP protein, while reactivity to p30Gag persisted [53]. Additionally, the snRNP peptide blocked binding of affinity purified anti-snRNP antibodies to both p30Gag and the snRNP protein [53]. It therefore appears that the region of homology between (U1)snRNP and p30Gag is responsible for the immunological cross-reactivity between the two proteins (Fig. 3). Comparative sequence analyses have also revealed regions of homology between the 28KDa HRES-1 protein and the Gag-related region of (U1)snRNP [18, 31, 131]. Using synthetic HRES- 1 peptides related to the p 19 and p24 Gag proteins of HTLV-1, Banki's [31] and Perl's [18] research groups have confirmed that the HRES-1 peptide sequences are indeed immunogenic and represent antigenic epitopes cross-reactive with the HTLV-1 Gag protein. In particular, the HRES-1 p24 Gag peptide contains a highly charged RRE-domain (RRE: Arg, Arg, Glu) homologous with snRNP and HTLV-1 p24 Gag [ 18]. Interestingly, a rabbit antibody raised against the HRES-1 p24 Gag peptide recognized a peptide based on the 70KDa (U1)snRNP RREdomain (residues 67-77 of (U 1)snRNP), but did not react with a (U1)snRNP peptide lacking the RREdomain [18]. These results suggest that it is the

RRE-domain of HRES-1 and snRNP that comprise the cross-reactive epitope. In this regard, it is conceivable that molecular mimicry between HRES-1 and snRNP serves as one of the priming mechanisms to initiate an ongoing autoimmune response in SLE. Still yet, molecular mimicry between HRES-1 and snRNP explains both the association of antiretroviral antibodies in SLE patients and a mechanism for their generation.

4.3. HRES-1 and a SLE Susceptibility Region A susceptibility locus for murine and human SLE has been mapped using microsatellite markers to Clq42 [8-10, 72, 74], the integration site for HRES-1. The study of human chromosomal regions syntenic to murine susceptibility loci, coupled with information gained from the genome project and genome-wide scans of both species, has facilitated the study of susceptibility regions associated with human disease [6]. For example, genes associated with glomerulonephritis and antibodies to chromotin, histones and DNA have been mapped to the telomeric end of murine chromosome 1 [140-143]. Since SLE-like susceptibility genes may have been conserved between humans and mice, Tsao et al [8] analyzed the human chromosomal region (Clq31q42) syntenic to the murine lupus susceptibility region in 52 SLE-afflicted sibpairs from 3 ethnic groups and found a 15cM region at Clq41-q42 that was linked to disease. Within the mapped 15cM region at C lq41-q42, an estimated 500 genes are encoded, some of which may have relevance to SLE (TGF-132, ADPRT and HLX1) [8]. Moreover, it is

281

highly probable that an unidentified disease-causing gene or genes within this region at C lq41-q42 are associated with SLE pathology. Since HRES-1 is located within the 15cM region at C lq41-q42 identified by Tsao [8] and others [4, 9, 10, 74] as being linked to SLE, perhaps then HRES-1 plays a vital part in contributing to the Clq42 SLE susceptibility region in both mouse and man.

4.4. Polymorphic Genotypes of HRES-1 Correlate with SLE Disease Activity A polymorphic HindlII site defines two alleles of the HRES-1 genomic locus [29, 131]. In order to determine if allelic variation in the HRES-1 locus was associated with SLE, Magistrelli et al [29] used Southern blotting and PCR to characterize the polymorphic HindlII site mapped in the LTR of HRES-1. The probe used differentiated between three HRES1 genotypes: I) 5.5kb fragment only, II) 3.7kb and 1.8kb fragments only and III) all three fragments [29]. Interestingly, the frequency of genotype I with respect to genotype III was 3.1-fold lower in SLE patients, as compared to control donors, while genotype II was the least prevalent in all groups [29]. Likewise, the relative frequency of genotype 11I with respect to genotype I was increased significantly in SLE patients. Additionally, the presence of anti-HRES-1 antibodies was increased in genotype III SLE patients and diminished in genotype I patients [131]. These findings raise the possibility that the genotype I HRES-1 allele is protective against SLE-mediated autoimmunity.

5. CONCLUSIONS From an evolutionary perspective, the finding that upwards of fifty percent of the human genome is encoded by retroelements is extraordinary [12]. More than that bestowed by any other single entity, the conglomeration of genes, polymorphisms, pseudogenes and endless additional features contributed by retroelements have worked to build and shape the human genome. Intriguingly, the human genome can be considered in part a composite of ancient retroelements, most of which inserted prior to the emergence of vertebrates [ 12, 144]. With regard to retroviral mediators of autoimmune/inflammatory

282

diseases, coupled with genetic and environmental factors, it is clear that retroelements, in particular HERVs, play a vital and significant role in the etiopathogenesis of such diseases. Due to the genetic diversity of exogenous and endogenous retroviruses, it has been difficult to delineate or identify specific retroviruses associated with specific autoimmune or inflammatory diseases. However, sequencing of the human genome and genome-wide scans of both humans and mice support the hypothesis of a HERV based etiopathogenesis for SLE. Specifically, we propose that molecular mimicry between the Gagrelated region of (U 1)snRNP and HRES-1 initiates the production of cross-reactive autoantibodies and associated immune complexes. In addition to molecular mimicry, other mechanisms of autoimmunity brought about by HERVs include but are not limited to insertional mutagenesis, superantigen-type activity, cis- or trans-regulation of cellular genes and immunomodulation by HERV gene products. Continued research in the field of autoimmune disease and HERVs as etiological factors will certainly facilitate the understanding of the very complex disease, SLE, as well as other autoimmune or inflammatory diseases.

ACKNOWLEDGEMENTS This work was supported in part by NIH grant #AI460 and Arizona Disease Control Research Commission (ADCRC) grant # 5018 to JJM. We are extremely appreciative of the advice and guidance given by Samuel E Schluter, Ph.D., on many areas pertinent to this chapter. Thanks also to Nafees Ahmad, Ph.D., for his expertise in the field of Virology.

REFERENCES 1.

2. 3.

MongeyA, Hes EV. The role of environment in systemic lupus erythematosus and associated disorders. In: Wallance DJ, Hahn BH, eds. Dubois' Lupus Erythematosus. Baltimore: Williams and Wilkins, 1997;31-47. YungRL, Richardson BC. Drug-induced lupus. Rheum Dis Clin North Am 1994;20(1):61-86. RothfieldN. Clinical features of systemic lupus erythematosus. In: Lelley, Harris, Ruddy, Sledge, eds.

4.

5.

6.

7. 8.

9.

10.

11.

12. 13.

14.

15.

16.

Textbook of Rheumatology. Philadelphia: Saunders, 1985;1070-1097. Kelly JA, Moser KL, Harley JB. The genetics of systemic lupus erythematosus: putting the pieces together. Genes Immun 2002;3 Suppl 1:$71-85. Deapen D, Escalante A, Weinrib L, Horwitz D, Bachman B, Roy-Burman P, Walker A, Mack TM. A revised estimate of twin concordance in systemic lupus erythematosus. Arthritis Rheum 1992;35(3):311-318. Vyse TJ, Kotzin BL. Genetic susceptibility to systemic lupus erythematosus. Annu Rev Immunol 1998;16: 261-292. Vyse TJ, Todd JA. Genetic analysis of autoimmune disease. Cell 1996;85(3):311-318. Tsao BP, Cantor RM, Kalunian KC, Chen CJ, Badsha H, Singh R, Wallace DJ, Kitridou RC, Chen SL, Shen N, Song YW, Isenberg DA, Yu CL, Hahn B H, Rotter JI. Evidence for linkage of a candidate chromosome 1 region to human systemic lupus erythematosus. J Clin Invest 1997;99(4):725-731. Moser KL, Neas BR, Salmon JE, Yu H, Gray-McGuire C, Asundi N, Bruner GR, Fox J, Kelly J, Henshall S, Bacino D, Dietz M, Hogue R, Koelsch G, Nightingale L, Shaver T, Abdou NI, Albert DA, Carson C, Petri M, Treadwell EL, James JA, Harley JB. Genome scan of human systemic lupus erythematosus: evidence for linkage on chromosome l q in African-American pedigrees. Proc Natl Acad Sci USA 1998;95(25): 14869-14874. Moser KL, Gray-McGuire C, Kelly J, Asundi N, Yu H, Bruner GR, Mange M, Hogue R, Neas BR, Harley JB. Confirmation of genetic linkage between human systemic lupus erythematosus and chromosome l q41. Arthritis Rheum 1999;42(9): 1902-1907. Yoshiki T, Mellors RC, Strand M, August JT. The viral envelope glycoprotein of murine leukemia virus and the pathogenesis of immune complex glomerulonephritis of New Zealand mice. J Exp Med 1974;140(4): 1011-1027. Lander ES. Initial sequencing and analysis of the human genome. Nature 2001 ;409(6822):860-921. Nelson PN, Carnegie PR, Martin J, Davari Ejtehadi H, Hooley P, Roden D, Rowland-Jones S, Warren P, Astley J, Murray PG. Demystified. Human endogenous retroviruses. Mol Patho12003;56(1): 11-18. Kazazian HH Jr. Genetics. L1 retrotransposons shape the mammalian genome. Science 2000;289(5482): 1152-1153. Perl A. Role of endogenous retroviruses in autoimmune diseases. Rheum Dis Clin North Am 2003;29(1): 123-143, vii. Wilkinson DA, Mager DL, Leong J-AC. Endogenous human retroviruses. In: Levy JA, ed. The Retroviridae.

New York: Plenum Press, 1994;465-535. 17. Lower R, Lower J, Kurth R. The viruses in all of us: characteristics and biological significance of human endogenous retrovirus sequences. Proc Natl Acad Sci USA 1996;93(11):5177-5184. 18. Perl A, Colombo E, Dai H, Agarwal R, Mark KA, Banki K, Poiesz BJ, Phillips PE, Hoch SO, Reveille JD, Arnett FC. Antibody reactivity to the HRES-1 endogenous retroviral element identifies a subset of patients with systemic lupus erythematosus and overlap syndromes. Correlation with antinuclear antibodies and HLA class II alleles. Arthritis Rheum 1995;38(11):1660--1671. 19. Sekigawa I, Okada M, Ogasawara H, Naito T, Kaneko H, Hishikawa T, Iida N, Hashimoto H. Lessons from similarities between SLE and HIV infection. J Infect 2002;44(2):67-72. 20. Schluter SF, Marchalonis JJ. Cloning of shark RAG2 and characterization of the RAG1/RAG2 gene locus. Faseb J 2003;17(3):470--472. 21. Martin MA, Bryan T, Rasheed S, Khan AS. Identification and cloning of endogenous retroviral sequences present in human DNA. Proc Natl Acad Sci USA 1981 ;78(8):48924896. 22. Andersson ML, Lindeskog M, Medstrand P, Westley B, May F, B lomberg J. Diversity of human endogenous retrovirus class II-like sequences. J Gen Virol 1999;80 (Pt 1):255-260. 23. Franklin GC, Chretien S, Hanson IM, Rochefort H, May FE, Westley BR. Expression of human sequences related to those of mouse mammary tumor virus. J Virol 1988;62(4):1203-1210. 24. Medstrand P, Blomberg J. Characterization of novel reverse transcriptase encoding human endogenous retroviral sequences similar to type A and type B retroviruses: differential transcription in normal human tissues. J Virol 1993;67(11):6778-6787. 25. Coffin JM. Genetic diversity and evolution of retroviruses. Curr Top Microbiol Immunol 1992;176: 143-164. 26. Tonjes RR, Lower R, Boiler K, Denner J, Hasenmaier B, Kirsch H, Konig H, Korbmacher C, Limbach C, Lugert R, Phelps RC, Scherer J, Thelen K, Lower J, Kurth R. HERV-K: the biologically most active human endogenous retrovirus family. J Acquir Immune Defic Syndr Hum Retrovirol 1996;13 Suppl 1:$261-267. 27. Lower R, Boiler K, Hasenmaier B, Korbmacher C, Muller-Lantzsch N, Lower J, Kurth R. Identification of human endogenous retroviruses with complex mRNA expression and particle formation. Proc Natl Acad Sci USA 1993;90( 10):4480-4484. 28. Adelman MK, Marchalonis JJ. Endogenous retroviruses in systemic lupus erythematosus: candidate lupus viruses. Clin Immuno12002; 102(2): 107-116.

283

29. Magistrelli C, Samoilova E, Agarwal RK, Banki K, Ferrante P, Vladutiu A, Phillips PE, Perl A. Polymorphic genotypes of the HRES-1 human endogenous retrovirus locus correlate with systemic lupus erythematosus and autoreactivity. Immunogenetics 1999;49(10):829-834. 30. Rasmussen HB, Kelly MA, Francis DA, Clausen J. Association between the endogenous retrovirus HRES-1 and multiple sclerosis in the United Kingdom -evidence of genetically different disease subsets? Dis Markers 2000;16(3--4):101-104. 31. Banki K, Maceda J, Hurley E, Ablonczy E, Mattson DH, Szegedy L, Hung C, Pet A. Human T-cell lymphotropic virus (HTLV)-related endogenous sequence, HRES-1, encodes a 28-kDa protein: a possible autoantigen for HTLV-I gag-reactive autoantibodies. Proc Natl Acad Sci USA 1992;89(5):1939-1943. 32. Chu JL, Drappa J, Parnassa A, Elkon KB. The defect in Fas mRNA expression in MRL/lpr mice is associated with insertion of the retrotransposon, ETn. J Exp Med 1993; 178(2): 723-730. 33. Wu J, Zhou T, He J, Mountz JD. Autoimmune disease in mice due to integration of an endogenous retrovirus in an apoptosis gene. J Exp Med 1993;178(2):461--468. 34. Vyse TJ, Kotzin BL. Genetic basis of systemic lupus erythematosus. Curr Opin Immunol 1996;8(6):843851. 35. Sekigawa I, Matsushita M, Lee S, Maeda N, Ogasawara H, Kaneko H, Iida N, Hashimoto H. A possible pathogenic role of CD8+ T cells and their derived cytokine, IL- 16, in SLE. Autoimmunity 2000;33(1):37-44. 36. Lee S, Kaneko H, Sekigawa I, Tokano Y, Takasaki Y, Hashimoto H. Circulating interleukin-16 in systemic lupus erythematosus. Br J Rheumatol 1998;37(12): 1334--1337. 37. Bisset LR, Rothen M, Joller-Jemelka HI, Dubs RW, Grob PJ, Opravil M. Change in circulating levels of the chemokines macrophage inflammatory proteins 1 alpha and 11 beta, RANTES, monocyte chemotactic protein1 and interleukin-16 following treatment of severely immunodeficient HIV-infected individuals with indinavir. Aids 1997;11(4):485-491. 38. Center DM, Kornfeld H, Cruikshank WW. Interleukin16. Int J Biochem Cell Biol 1997;29(11):1231-1234. 39. Fox RA, Isenberg DA. Human immunodeficiency virus infection in systemic lupus erythematosus. Arthritis Rheum 1997;40(6):1168-1172. 40. Diri E, Lipsky PE, Berggren RE. Emergence of systemic lupus erythematosus after initiation of highly active antiretroviral therapy for human immunodeficiency virus infection. J Rheumatol 2000;27(11): 2711-2714. 41. Barthel HR, Wallace DJ. False-positive human immunodeficiency virus testing in patients with lupus erythe-

284

matosus. Semin Arthritis Rheum 1993;23(1):1-7. 42. Alonso CM, Lozada CJ. Effects of IV cyclophosphamide on HIV viral replication in a patient with systemic lupus erythematosus. Clin Exp Rheumatol 2000;18(4): 510-512. 43. Center DM, Kornfeld H, Cruikshank WW. Interleukin 16 and its function as a CD4 ligand. Immunol Today 1996; 17( 10):476--481. 44. Deas JE, Liu LG, Thompson JJ, Sander DM, Soble SS, Garry RF, Gallaher WR. Reactivity of sera from systemic lupus erythematosus and Sj6gren's syndrome patients with peptides derived from human immunodeficiency virus p24 capsid antigen. Clin Diagn Lab Immunol 1998;5(2): 181-185. 45. Talal N, Garry RF, Schur PH, Alexander S, Dauphinee MJ, Livas IH, Ballester A, Takei M, Dang H. A conserved idiotype and antibodies to retroviral proteins in systemic lupus erythematosus. J Clin Invest 1990;85(6): 1866-1871. 46. Phillips PE, Johnston SL, Runge LA, Moore JL, Poiesz BJ. High IgM antibody to human T-lymphotropic virus type I in systemic lupus erythematosus. J Clin Immunol 1986;6(3):234-241. 47. Blomberg J, Nived O, Pipkorn R, Bengtsson A, Erlinge D, Sturfelt G. Increased antiretroviral antibody reactivity in sera from a defined population of patients with systemic lupus erythematosus. Correlation with autoantibodies and clinical manifestations. Arthritis Rheum 1994 ;37( 1):57-66. 48. Griffiths DJ, Cooke SP, Herve C, Rigby SP, Mallon E, Hajeer A, Lock M, Emery V, Taylor P, Pantelidis P, Bunker CB, du Bois R, Weiss RA, Venables PJ. Detection of human retrovirus 5 in patients with arthritis and systemic lupus erythematosus. Arthritis Rheum 1999;42(3 ):448-454. 49. Garry RF. New evidence for involvement of retroviruses in Sj6gren's syndrome and other autoimmune diseases. Arthritis Rheum 1994;37(4):465-469. 50. Talal N, Dauphinee MJ, Dang H, Alexander SS, Hart DJ, Garry RF. Detection of serum antibodies to retroviral proteins in patients with primary Sj6gren's syndrome (autoimmune exocrinopathy). Arthritis Rheum 1990;33(6):774-781. 51. Nelson PN, Lever AM, Bruckner FE, Isenberg DA, Kessaris N, Hay FC. Polymerase chah'a reaction fails to incriminate exogenous retroviruses HTLV-I and HIV-1 in rheumatological diseases although a minority of sera cross react with retroviral antigens. Ann Rheum Dis 1994;53( 11):749-754. 52. Lefranc D, Dubucquoi S, Almeras L, De Seze J, Tourvieille B, Dussart P, Aubert JP, Vermersch P, Prin L. Molecular analysis of endogenous retrovirus HRES-1: identification of frameshift mutations in region encod-

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

ing putative 28-kDa autoantigen. Biochem Biophys Res Commun 2001 ;283(2):437--444. Query CC, Keene JD. A human autoimmune protein associated with U1 RNA contains a region of homology that is cross-reactive with retroviral p30gag antigen. Cell 1987;51(2):211-220. Ogasawara H, Kaneko H, Hishikawa T, Sekigawa I, Takasaki Y, Hashimoto H, Hirose S, Kaneko Y, Maruyama N. Molecular mimicry between human endogenous retrovirus clone 4-1 and HLA class I antigen with reference to the pathogenesis of systemic lupus erythematosus. Rheumatology (Oxf) 1999;38(11): 1163-1164. Ogasawara H, Hishikawa T, Sekigawa I, Hashimoto H, Yamamoto N, Maruyama N. Sequence analysis of human endogenous retrovirus clone 4-1 in systemic lupus erythematosus. Autoimmunity 2000;33(1): 15-21. Garry RF, Fermin CD, Hart DJ, Alexander SS, Donehower LA, Luo-Zhang H. Detection of a human intracisternal A-type retroviral particle antigenically related to HIV. Science 1990;250(4984):1127-1129. Yamano S, Renard JN, Mizuno F, Narita Y, Uchida Y, Higashiyama H, Sakurai H, Saito I. Retrovirus in salivary glands from patients with Sj6gren's syndrome. J Clin Pathol 1997;50(3):223-230. Stransky G, Vernon J, Aicher WK, Moreland LW, Gay RE, Gay S. Virus-like particles in synovial fluids from patients with rheumatoid arthritis. Br J Rheumatol 1993 ;32(12): 1044-1048. Nakagawa K, Brusic V, McColl G, Harrison LC. Direct evidence for the expression of multiple endogenous retroviruses in the synovial compartment in rheumatoid arthritis. Arthritis Rheum 1997;40(4):627-638. Ziegler B, Gay RE, Huang GQ, Fassbender HG, Gay S. Immunohistochemical localization of HTLV-I p 19- and p24-related antigens in synovial joints of patients with rheumatoid arthritis. Am J Pathol 1989;135(1):1-5. Shih A, Misra R, Rush MG. Detection of multiple, novel reverse transcriptase coding sequences in human nucleic acids: relation to primate retroviruses. J Virol 1989;63(1):64-75. Perron H, Lalande B, Gratacap B, Laurent A, Genoulaz O, Geny C, Mallaret M, Schuller E, Stoebner P, Seigneurin JM. Isolation of retrovirus from patients with multiple sclerosis. Lancet 1991 ;337(8745):862-863. Lan XY, Zeng Y, Zhang D, Hong ML, Wang DX, Zhang YL, Feng ZJ, Tang MH, Feng BZ. Establishment of a human malignant T lymphoma cell line carrying retrovirus-like particles with RT activity. Biomed Environ Sci 1994;7(1):1-12. Haahr S, Sommerlund M, Moller-Larsen A, Nielsen R, Hansen HJ. Just another dubious virus in cells from a

65.

66. 67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

patient with multiple sclerosis? Lancet 1991;337(8745): 863-864. Hen'mann M, Hagenhofer M, Kalden JR. Retroviruses and systemic lupus erythematosus. Immunol Rev 1996;152:145-156. Denman A. Induction of interferon by lentiviruses. Immunol Today 1986;7:201-203. Izui S, McConahey PJ, Clark JP, Hang LM, Hara I, Dixon FJ. Retroviral gp70 immune complexes in NZB x NZW F2 mice with murine lupus nephritis. J Exp Med 1981;154(2):517-528. Tucker RM, Vyse TJ, Rozzo S, Roark CL, Izui S, Kotzin BL. Genetic control of glycoprotein 70 autoantigen production and its influence on immune complex levels and nephritis in murine lupus. J Immunol 2000; 165(3): 1665-1672. Theofilopoulos AN, Dixon FJ. Murine models of systemic lupus erythematosus. Adv Immunol 1985;37: 269-390. Alexander JJ, Saxena AK, Bao L, Jacob A, Haas M, Quigg RJ. Prominent renal expression of a murine leukemia retrovirus in experimental systemic lupus erythematosus. J Am Soc Nephro12002;13(12):2869-2877. Andrews BS, Eisenberg RA, Theofilopoulos AN, Izui S, Wilson CB, McConahey PJ, Murphy ED, Roths JB, Dixon FJ. Spontaneous murine lupus-like syndromes. Clinical and immunopathological manifestations in several strains. J Exp Med 1978;148(5):1198-1215. Kotzin BL. Susceptibility loci for lupus: a guiding light from murine models? J Clin Invest 1997;99(4): 557-558. Tsao BP, Cantor RM, Kalunian KC, "Wallace DJ, Hahn BH, Rotter JI. The genetic basis of systemic lupus erythematosus. Proc Assoc Am Physicians 1998;110(2): 113-117. Gaffney PM, Ortmann WA, Selby SA, Shark KB, Ockenden TC, Rohlf KE, Walgrave NL, Boyum WP, Malmgren ML, Miller ME, Kearns GM, Messner RP, King RA, Rich SS, Behrens TW. Genome screening in human systemic lupus erythematosus: results from a second Minnesota cohort and combined analyses of 187 sib-pair families. Am J Hum Genet 2000;66(2): 547-556. Marrack P, Winslow GM, Choi Y, Scherer M, Pullen A, White J, Kappler JW. The bacterial and mouse mammary tumor virus superantigens; two different families of proteins with the same functions. Immunol Rev 1993;131:79-92. Perron H, Jouvin-Marche E, Michel M, Ounanian-Paraz A, Camelo S, Dumon A, Jolivet-Reynaud C, Marcel F, Souillet Y, Borel E, Gebuhrer L, Santoro L, Marcel S, Seigneurin JM, Marche PN, Lafon M. Multiple sclerosis retrovirus particles and recombinant envelope trig-

285

77.

78.

79.

80.

81.

82.

83.

84.

85.

86.

87.

286

ger an abnormal immune response in vitro, by inducing polyclonal Vbeta 16 T-lymphocyte activation. Virology 2001 ;287(2):321-332. Conrad B, Weidmann E, Trucco G, Rudert WA, Behboo R, Ricordi C, Rodriquez-Rilo H, Finegold D, Trucco M. Evidence for superantigen involvement in insulin-dependent diabetes mellitus aetiology. Nature 1994;371 (6495):351-355. Conrad B, Weissmahr RN, Boni J, Arcari R, Schupbach J, Mach B. A human endogenous retroviral superantigen as candidate autoimmune gene in type I diabetes. Cell 1997;90(2):303-313. Stauffer Y, Marguerat S, Meylan F, Ucla C, Sutkowski N, Huber B, Pelet T, Conrad B. Interferon-alphainduced endogenous superantigen, a model linking environment and autoimmunity. Immunity 2001; 15(4): 591-601. Sutkowski N, Conrad B, Thorley-Lawson DA, Huber BT. Epstein-Barr virus transactivates the human endogenous retrovirus HERV-K18 that encodes a superantigen. Immunity 2001;15(4):579-589. Lapatschek M, Durr S, Lower R, Magin C, Wagner H, Miethke T. Functional analysis of the env open reading frame in human endogenous retrovirus IDDMK(1,2)22 encoding superantigen activity. J Virol 2000;74(14): 6386-6393. Jaeckel E, Heringlake S, Berger D, Brabant G, Hunsmann G, Manns MP. No evidence for association between IDDMK(1,2)22, a novel isolated retrovirus, and IDDM. Diabetes 1999;48(1):209-214. Lower R, Tonjes RR, Boiler K, Denner J, Kaiser B, Phelps RC, Lower J, Kurth R, Badenhoop K, Donner H, Usadel KH, Miethke T, Lapatschek M, Wagner H. Development of insulin-dependent diabetes mellitus does not depend on specific expression of the human endogenous retrovirus HERV-K. Cell 1998;95(1): 11-14, discussion 16. Murphy VJ, Harrison LC, Rudert WA, Luppi P, Trucco M, Fierabracci A, Biro PA, Bottazzo GE Retroviral superantigens and type 1 diabetes mellitus. Cell 1998;95(1):9-11, discussion 16. Badenhoop K, Donner H, Neumann J, Herwig J, Kurth R, Usadel KH, Tonjes RR. IDDM patients neither show humoral reactivities against endogenous retroviral envelope protein nor do they differ in retroviral mRNA expression from healthy relatives or normal individuals. Diabetes 1999;48(1):215-218. Lan MS, Mason A, Coutant R, Chen QY, Vargas A, Rao J, Gomez R, Chalew S, Garry R, Maclaren NK. HERV-K10s and immune-mediated (type 1) diabetes. Cell 1998;95(1): 14-16, discussion 16. Muir A, Ruan QG, Marron MP, She JX. The IDDMK(1,2)22 retrovirus is not detectable in either

88.

89.

90.

91.

92.

93.

94.

95.

96.

97.

98.

mRNA or genomic DNA from patients with type 1 diabetes. Diabetes 1999;48(1):219-222. Hasuike S, Miura K, Miyoshi O, Miyamoto T, Niikawa N, Jinno Y, Ishikawa M. Isolation and localization of an IDDMK1,2-22-related human endogenous retroviral gene, and identification of a CA repeat marker at its locus. J Hum Genet 1999;44(5):343-347. Mein CA, Esposito L, Dunn MG, Johnson GC, Timms AE, Goy JV, Smith AN, Sebag-Montefiore L, Merriman ME, Wilson AJ, Pritchard LE, Cucca F, Barnett AH, Bain SC, Todd JA. A search for type 1 diabetes susceptibility genes in families from the United Kingdom. Nat Genet 1998;19(3):297-300. Luppi P, Zanone MM, Hyoty H, Rudert WA, Haluszczak C, Alexander AM, Bertera S, Becker D, Trucco M. Restricted TCR V beta gene expression and enterovirus infection in type I diabetes. Diabetologia 2000;43: 1484-1497. Perron H, Garson JA, Bedin F, Beseme F, Paranhos-Baccala G, Komurian-Pradel F, Mallet F, Tuke PW, Voisset C, Blond JL, Lalande B, Seigneurin JM, Mandrand B. Molecular identification of a novel retrovirus repeatedly isolated from patients with multiple sclerosis. The Collaborative Research Group on Multiple Sclerosis. Proc Natl Acad Sci USA 1997;94(14):7583-7588. Blond JL, Beseme F, Duret L, Bouton O, Bedin F, Perron H, Mandrand B, Mallet E Molecular characterization and placental expression of HERV-W, a new human endogenous retrovirus family. J Virol 1999;73(2): 1175-1185. Blatt NB, Glick GD. Anti-DNA autoantibodies and systemic lupus erythematosus. Pharmacol Ther 1999;83(2): 125-139. Andersson G, Svensson AC, Setterblad N, Rask L. Retroelements in the human MHC class II region. Trends Genet 1998;14(3):109-114. Kulski JK, Gaudieri S, Inoko H, Dawkins RL. Comparison between two human endogenous retrovirus (HERV)-rich regions within the major histocompatibility complex. J Mol Evol 1999;48(6):675--683. Kulski JK, Dawkins RL. The P5 multicopy gene family in the MHC is related in sequence to human endogenous retroviruses HERV-L and HERV-16. Immunogenetics 1999;49(5):404-412. Cameron PU, Mallal S A, French MAH. Central MHC genes between HLA-B and complement C4 confer risk for HIV-1 disease progression. In: Tsuji K, Aizawa M, Sasazuki T, eds. HLA 1991. New York: Oxford Univ Press, 1992(2);544-547. Leelayuwat C, Zhang WJ, Abraham LJ, Townend DC, Gaudieri S, Dawkins RL. Differences in the central major histocompatibility complex between humans and chimpanzees. Implications for development of autoim-

munity and acquired immune deficiency syndrome. Hum Immunol 1993;38( 1):30-41. 99. Heeney JL, van Els C, de Vries P, ten Haaft P, Otting N, Koornstra W, Boes J, Dubbes R, Niphuis H, Dings M, Cranage M, Norley S, Jonker S, Bontrop RE, Osterhaus A. Major histocompatibility complex class 1-associated vaccine protection from simian immunodeficiency virus-infected peripheral blood cells. J Exp Med 1994; 180(2):769-774. 100. Strazzullo M, Majello B, Lania L, La Mantia G. Mutational analysis of the human endogenous ERV9 proviruses promoter region. Virology 1994;200(2):686-695. 101. Ratnoff WD. Inherited deficiencies of complement in rheumatic diseases. Rheum Dis Clin North Am 1996;22(1):75-94. 102. Grant SF, Kristjansdottir H, Steinsson K, Blondal T, Yuryev A, Stefansson K, Gulcher JR. Long PCR detection of the C4A null allele in B8-C4AQ0-C4B 1-DR3. J Immunol Methods 2000;244(1-2):41-47. 103. Fielder All, Walport MJ, Batchelor JR, Rynes RI, Black CM, Dodi IA, Hughes GR. Family study of the major histocompatibility complex in patients with systemic lupus erythematosus: importance of null alleles of C4A and C4B in determining disease susceptibility. Br Med J (Clin Res Ed) 1983;286(6363):425--428. 104. Arnett FC. Genetic aspects of human lupus. Clin Immunol Immunopathol 1992;63(1):4-6. 105. Reveille JD, Arnett FC, Wilson RW, Bias WB, McLean RH. Null alleles of the fourth component of complement and HLA haplotypes in familial systemic lupus erythematosus. Immunogenetics 1985;21(4):299-311. 106. Dangel AW, Mendoza AR, Baker BJ, Daniel CM, Carroll MC, Wu LC, Yu CY. The dichotomous size variation of human complement C4 genes is mediated by a novel family of endogenous retroviruses, which also establishes species-specific genomic patterns among Old World primates. Immunogenetics 1994;40(6): 425--436. 107. Zhu ZB, Hsieh SL, Bentley DR, Campbell RD, Volanakis JE. A variable number of tandem repeats locus within the human complement C2 gene is associated with a retroposon derived from a human endogenous retrovirus. J Exp Med 1992;175(6):1783-1787. 108. Lower R, Tonjes RR, Korbmacher C, Kurth R, Lower J. Identification of a Rev-related protein by analysis of spliced transcripts of the human endogenous retroviruses HTDV/HERV-K. J Virol 1995;69(1):141-149. 109. Horwitz MS, Boyce-Jacino MT, Faras AJ. Novel human endogenous sequences related to human immunodeficiency virus type 1. J Virol 1992;66(4):2170-2179. 110. Oostendorp RA, Schaaper WM, Post J, Von B lomberg BM, Meloen RH, Scheper RJ. Suppression of lymphocyte proliferation by a retroviral p 15E-derived hexapep-

tide. Eur J Immunol 1992;22(6):1505-1511. 111. Ogasawara M, Cianciolo GJ, Snyderman R, Mitani M, Good RA, Day NK. Human IFN-gamma production is inhibited by a synthetic peptide homologous to retroviral envelope protein. J Immunol 1988; 141 (2):614-619. 112. Ogasawara M, Haraguchi S, Cianciolo GJ, Mitani M, Good RA, Day NK. Inhibition of murine cytotoxic T lymphocyte activity by a synthetic retroviral peptide and abrogation of this activity by IL. J Immunol 1990; 145(2):456-462. 113. Haraguchi S, Liu WT, Cianciolo GJ, Good RA, Day NK. Suppression of human interferon-gamma production by a 17 amino acid peptide homologous to the transmembrane envelope protein of retroviruses: evidence for a primary role played by monocytes. Cell Immunol 1992; 141 (2):388-397. 114. Nelson M, Nelson DS, Cianciolo GJ, Snyderman R. Effects of CKS-17, a synthetic retroviral envelope peptide, on cell-mediated immunity in vivo: immunosuppression, immunogenicity, and relation to immunosuppressive tumor products. Cancer Immunol Immunother 1989;30(2): 113-118. 115. Haraguchi S, Good RA, James-Yarish M, Cianciolo GJ, Day NK. Differential modulation of Thl- and Th2-related cytokine mRNA expression by a synthetic peptide homologous to a conserved domain within retroviral envelope protein. Proc Natl Acad Sci USA 1995;92(8):3611-3615. 116. Modlin RL, Nutman TB. Type 2 cytokines and negative immune regulation in human infections. Curr Opin Immunol 1993;5(4):511-517. 117. Sher A, Gazzinelli RT, Oswald IP, Clerici M, Kullberg M, Pearce EJ, Berzofsky JA, Mosmann TR, James SL, Morse HC 3rd. Role of T-cell derived cytokines in the downregulation of immune responses in parasitic and retroviral infection. Immunol Rev 1992;127:183-204. 118. D'Andrea A, Aste-Amezaga M, Valiante NM, Ma X, Kubin M, Trinchieri G. Interleukin 10 (IL-10) inhibits human lymphocyte interferon gamma-production by suppressing natural killer cell stimulatory factor/IL-12 synthesis in accessory cells. J Exp Med 1993;178(3): 1041-1048. 119. Clerici M, Lucey DR, Berzofsky JA, Pinto LA, Wynn TA, Blatt SE Dolan MJ, Hendrix CW, Wolf SE Shearer GM. Restoration of HIV-specific cell-mediated immune responses by intedeukin-12 in vitro. Science 1993;262(5140):1721-1724. 120. Chehimi J, Trinchieri G. Interleukin-12: a bridge between innate resistance and adaptive immunity with a role in infection and acquired immunodeficiency. J Clin Immunol 1994;14(3):149-161. 121. Cianciolo GJ, Bogerd H, Snyderman R. Human retrovirus-related synthetic peptides inhibit T lymphocyte

287

proliferation. Immunol Lett 1988; 19(1):7-13. 122. Scheeren RA, Oostendorp RA, van der Baan S, Keehnen RM, Scheper RJ, Meijer CJ. Distribution of retroviral pl5E-related proteins in neoplastic and non-neoplastic human tissues, and their role in the regulation of the immune response. Clin Exp Immunol 1992;89(1): 94--99. 123. Hishikawa T, Ogasawara H, Kaneko H, Shirasawa T, Matsuura Y, Sekigawa I, Takasaki Y, Hashimoto H, Hirose S, Handa S, Nagasawa R, Maruyama N. Detection of antibodies to a recombinant gag protein derived from human endogenous retrovirus clone 4-1 in autoimmune diseases. Viral Immunol 1997;10(3): 137-147. 124. Sekigawa I, Kaneko H, Hishikawa T, Hashimoto H, Hirose S, Kaneko Y, Maruyama N. HIV infection and SLE: their pathogenic relationship. Clin Exp Rheumatol 1998;16(2):175-180. 125. Reichlin M. Systemic lupus erythematosus. Antibodies to ribonuclear proteins. Rheum Dis Clin North Am 1994;20(1):29--43. 126. Shefner R, Manheimer-Lory A, Davidson A, Paul E, Aranow C, Katz J, Diamond B. Idiotypes in systemic lupus erythematosus. Clues for understanding etiology and pathogenicity. Chem Immunol 1990;48:82-108. 127. Tan EM. Antinuclear antibodies: diagnostic markers for autoimmune diseases and probes for cell biology. Adv Immunol 1989;44:93-151. 128. Oldstone MB. Molecular mimicry and autoimmune disease. Cell 1987;50(6):819-820. 129. P e t A. Mechanisms of viral pathogenesis in rheumatic disease. Ann Rheum Dis 1999;58(8):454-461. 130. Perl A, Banki K. Molecular mimicry, altered apoptosis, and immunomodulation as mechansims of viral pathogenesis in systemic lupus erythematosus. In: Kammer GM, Tsokos GC, eds. Lupus: Molecular and Cellular Pathogenesis. Totowa, NJ: Humana Press, 1999;43-64. 131. Perl A, Rosenblatt JD, Chen IS, DiVincenzo JP, Bever R, Poiesz B J, Abraham GN. Detection and cloning of new HTLV-related endogenous sequences in man. Nucleic Acids Res 1989;17(17):6841-6854. 132. Perl A, Isaacs CM, Eddy RL, Byers MG, Sait SN, Shows TB. The human T-cell leukemia vires-related endogenous sequence (HRES 1) is located on chromosome 1 at q42. Genomics 1991;11(4):1172-1173. 133. Koopman WJ, Gay S. The MRL-lpr/lpr mouse. A model for the study of rheumatoid arthritis. Scand J Rheumatol Suppl 1988;75:284-289. 134. Tanaka A, O'Sullivan FX, Koopman WJ, Gay S. Etio-

288

pathogenesis of rheumatoid arthritis-like disease in MRL/1 mice: II. Ultrastructural basis of joint destruction. J Rheumatol 1988;15(1):10--16. 135. Brand A, Griffiths DJ, Herve C, Mallon E, Venables PJ. Human retrovirus-5 in rheumatic disease. J Autoimmun 1999; 13( 1): 149-154. 136. Brookes SM, Pandolfino YA, Mitchell TJ, Venables PJ, Shattles WG, Clark DA, Entwistle A, Maini RN. The immune response to and expression of cross-reactive retroviral gag sequences in autoimmune disease. Br J Rheumatol 1992;31(11):735-742. 137. Dang H, Dauphinee MJ, Talal N, Garry RF, Seibold JR, Medsger TA Jr, Alexander S, Feghali CA. Serum antibody to retroviral gag proteins in systemic sclerosis. Arthritis Rheum 1991;34(10):1336--1337. 138. Pani MA, Seidl C, Bieda K, Seissler J, Krause M, Seiffied E, Usadel KH, Badenhoop K. Preliminary evidence that an endogenous retroviral long-terminal repeat (LTR13) at the HLA-DQB 1 gene locus confers susceptibility to Addison's disease. Clin Endocfinol (Oxf) 2002;56(6):773-777. 139. Ranki A, Kurki P, Riepponen S, Stephansson E. Antibodies to retroviral proteins in autoimmune connective tissue disease. Relation to clinical manifestations and fibonucleoprotein autoantibodies. Arthritis Rheum 1992;35( 12): 1483-1491. 140. Kono DH, Burlingame RW, Owens DG, Kuramochi A, Balderas RS, Balomenos D, Theofilopoulos ,aN. Lupus susceptibility loci in New Zealand mice. Proc Nail Acad Sci USA 1994;91(21):10168-10172. 141. Drake CG, Rozzo SJ, Hirschfeld HF, Smamworawong NP, Palmer E, Kotzin B L. Analysis of the New Zealand Black contribution to lupus-like renal disease. Multiple genes that operate in a threshold manner. J Immunol 1995; 154(5):2441-2447. 142. Morel L, Rudofsky UH, Longmate JA, Schiffenbauer J, Wakeland EK. Polygenic control of susceptibility to mufine systemic lupus erythematosus. Immunity 1994;1(3):219-229. 143. Morel L, Yu Y, Blenman KR, Caldwell RA, Wakeland EK. Production of congenic mouse strains carrying genomic intervals containing SLE-susceptibility genes derived from the SLE-prone NZM2410 strain. Mamm Genome 1996;7(5):335-339. 144. Brosius J. Genomes were forged by massive bombardments with retroelements and retrosequences. Genetica 1999; 107( 1-3):209-238.


Sj/igren's Syndrome - Autoimmune Epithelitis: Role of Coxsacldeviruses in Pathogenesis Dimitrios A Liakos, Efstathia K. Kapsogeorgou and Haralampos M Moutsopoulos

Department of Pathophysiology, Medical School National University of Athens, Athens, Greece

1. INTRODUCTION Sjrgren's syndrome (SS) or autoimmune epithelitis is a chronic autoimmune disorder characterized by inflammation of the salivary and lacrimal glands resulting in xerostomia and keratoconjuctivitis sica [1 ]. The disease can be seen as an entity alone (primary SS) or in association with other rheumatic autoimmune diseases (secondary SS). The prevalence of the syndrome is about 1%-2% in the adult population [2], and primarily affects females (9:1 female to male ratio) in their fourth and fifth decade of fife. Symptoms of the disease may appear six to eight years prior to the full blown clinical development of the syndrome. Progression is slow and usually involves glandular tissues. About one third of the patients can develop some extraglandular manifestation including interstitial renal disease, bronchitis sicca, autoimmune liver disorder and vasculitis. Furthermore, about 5% of patients may develop [3-cell lymphoma. Malignant lymphoma development is a major complication of the disease. The risk of lymphoma development in patients with primary SS is 40 times higher compared to the normal population and is associated with increased mortality [3].

2. CLINICAL PICTURE (REVIEWED IN REF. [21) Sjrgren's syndrome is characterized by dryness of the mouth (xerostomia) caused by decreased production of saliva. Salivary flow rate is used for the evaluation of major salivary gland function as

well as scintigraphy and digital subtraction sialography. Age, gender, medication and psychological factors may influence salivary flow measurements. In patients with xerostomia, the oral mucosa can be sticky, dry and erythematosus. Patients report problems with chewing and swallowing due to the dryness of the mouth. The tongue may be dry, with deep fissures and atrophic papillae. Mouth dryness can lead to fungal infections that can result to angular chelitis and fungal overgrowth on the tongue. Recurrent carries is a problem that is often reported, as well as discomfort with dentures. Major salivary gland enlargement occurs in patients with primary SS and although it may initially appear unilaterally, in most cases it develops bilaterally. Chronic eye dryness leads to irritation and destruction of the corneal and conjuctival epithelium. Tear secretion rate is measured by the Shirmer's test while staining corneal and conjuctival epithelial tissues with Rose Bengal and other stains reveals the extend of epithelial destruction. About one third of Sjrgren's syndrome patients at some point in the course of the disease display some extraglandular systemic manifestation. Raynaud's phenomenon precedes sicca manifestations and is found in 35% of patients. Vasculitis of the skin is presented with palpable purpuric or petechial lesions. Musculoskeletal involvement includes arthralgias, myalgias, fatigue and morning stiffness. Nonerosive arthritis and symmetric polyarthritis can be observed. In patients with primary SS, respiratory tract involvement is mild but frequent. Dry cough is quite common and is caused by xerotrachia or bronchitis sicca. Dysphagia might also occur as a result of dryness of the pharynx and esophagus. Liver

289

involvement is rare (5%) in patients with primary SS and presents either as primary billiary cirrhosis or chronic active hepatitis. About 4% of Sjrgren's syndrome patients have clinically significant renal involvement in the form of interstitial nephritis or glomerulonephritis. Peripheral sensory or sensorymotor polyneurpathy and mononeuritis multiplex occurs in 1% to 2% of patients. Furthermore, anxiety, depressed mood and personality structure disorders are frequently observed [3]. In primary Sj/Sgren's syndrome, malignant non-Hodgkin lymphoma occurs in 4% to 6% of patients. Lymphoma usually develops later in the course of the disease. Extranodal localization is quite common and is found in the salivary glands in 55% of lymphoma patients. The presence of parotid gland enlargement, palpable purpura, low C4 levels and mixed monoclonal cryoglobulinemia at the first visit are adverse prognostic factors and adequately distinguish high risk patients for the development of lymphoproliferative disorders. Patients with primary SS and adverse prognostic factors display increased mortality compared to the general population (mortality ratio: 1.15) and one to five deaths of patients with primary SS are attributed to lymphoma. The presence of the adverse prognostic factors is strongly correlated with the increased mortality.

3. IMMUNOPATHOLOGY (REVIEWED IN R E E [4]) The aetiology of Sjtigren's syndrome remains unknown. Even though environmental influences and different genetic factors are related to the disorder, no single factor can be identified as responsible for the pathogenesis of the syndrome. It is considered an autoimmune disease due to the presence of autoantibodies and the focal lymphocytic infiltrations that are observed in the lesions. Furthermore, the association of SS with other autoimmune disorders supports the autoimmune nature of the disease. Sjrgren's syndrome is characterized by B-cell reactivity and destruction of exocrine glands associated with dense lymphocytic infiltrations. B-cell activation is the most prominent immunologic feature of Sjtigren's syndrome. B-cells infiltrating

290

Figure 1. Advancedlymphocyticinfiltration in labial minor salivary gland biopsy. The bellow the image of the lesion is a list of the cells that participate in the lesion and their characteristics.

the minor salivary glands are activated since they produce increased amounts of immunoglobulins with autoantibody reactivity. Rheumatoid factor (RF) and antinuclear antibodies (ANA) are found in high frequencies in Sjrgren's syndrome patients. The analysis of specificity of anti nuclear antibodies reveals the presence of antibodies against two ribonucleoproteins, Ro(SSA) and La(SSB). While the function of Ro(SSA) remains unknown, La(SSB) is known to participate in transcription termination of RNA polymerase III and also in the translation of viral RNA. Another autoantigen in Sjrgren's syndrome is a cytoskeletal protein a-fodrin. Oligoclonal B-cell expansion takes place early in the development of the disease. Monoclonal light chains are detected in higher frequencies in patients with systemic involvement compared to patients with glandular disease. Furthermore, about one third of the patients with primary SS have high levels of mixed monoclonal/polyclonal type II cryoglobulins with rheumatoid factor activity [5]. It is possible that B-cell neoplastic transformation takes place in the gland possibly due to the chronic immunologic stimulation. This transformation probably takes place along with immunoglobulin gene rearrangements and mutations in the p53 transcription factor [6]. The lesion that is observed in labial minor salivary gland biopsies of Sjrgren's syndrome patients is characterized by the round cell infiltrates that in

Table 1. Aberrantexpression of molecules implicated in epithelial cell activation in Sjtren's syndrome Minor salivary gland biopsies

Salivary gland cultured epithelial cells

Activation markers

Protooncogenes

Not studied

Immune reactive molecules

MHC I and MHC II B7 costimulatory molecules Adhesion molecules Chemokines CD40 Proinflammatory cytokines

MHC I and MHC II B7 costimulatory molecules Adhesion molecules Chemokines CD40 Proinflammatory cytokines

Apoptosis-related molecules

FAS, FAS Ligand

FAS, FAS Ligand

early lesions surround ductal epithelial cells. As the lesions progress, the infiltrate extends and replaces the functional tissue of the salivary gland (Fig. 1). The majority of the cells are T-cells while B-cells represent about one fourth of the infiltrating cells. The majority of T-cells are CD4 positive, express the memory/inducer marker and the lymphocyte function associated molecule (LFA-1); a cell surface glycoprotein that has been associated with adhesion of lymphocytes and macrophages. Monocytes, macrophages and natural killer cells represent less than 5% of cells in the pathologic lesion. Dendritic cells, the other classical antigen presenting cells, are only found in advanced lesions. These cells express the DRC cell surface marker, are located among B and T lymphocytes and represent about 2% of mononuclear cells that are present in the lesion. This reduced number of professional antigen presenting cells suggests that this role is probably played by some other cell type in the pathologic lesion. Clinical and pathophysiologic findings show that the inflamed tissue in affected organs of Sj6gren's syndrome patients is epithelium. The epithelium plays a central role in the initiation and perpetuation of immune response. This is attested by a series of immunopathologic findings that show aberrant expression of various activation and immune response associated molecules in epithelial cells of minor salivary gland biopsies (Table 1). These cells express molecules implicated in antigen presentation, such as MHC class I (HLA-ABC), MHC class II (HLA-DR) and B7 costimulatory molecules. In addition, there is upregulated expression of mol-

ecules that mediate B and T-cell recruitment as well as molecules that are involved in the expansion of the immune response. These molecules are adhesion molecules, lymphoattractant chemokines and proinflammatory cytokines. Furthermore, the activated state of epithelium is attested by the overexpression of apoptosis related molecules (FAS, FAS ligand). Increased rates of epithelial apoptosis result to the release of intracellular antigens that are recognized by the immune system. These findings were substantiated by the detection of increased constitutive expression of the same molecules in long term cultured non-neoplastic epithelial cell lines established from SjOgren's syndrome patients' salivary glands (Table 1) [2, 7]. The establishment of long-term salivary gland epithelial cell cultures revealed the capacity of these cells to interact with immune cells through the expression of functional costimulatory molecules (B7 and CD40) [7, 8]. Moreover, it has been shown that salivary gland epithelial cells can costimulate the growth of activated T-cells, indicating that epithelial cells are able to participate in the antigen mediated activation and proliferation of the immune response. It is therefore very likely that there are intrinsic activation processes active in affected epithelial cells. The above data indicate that epithelial cells are suitably equipped to act as antigen presenting cells and support their central role in the pathogenesis of Sj6gren's syndrome.

291

oo

4. VIRUSES AND SJOGREN'S SYNDROME The implication of viruses in the development of Sj6gren's syndrome has long been suspected. There is no clear evidence so far that proves viral induction of Sj6gren's syndrome. However, there are data that link viruses and the disease. It has been shown by some groups that Sj6gren's syndrome patients express EBV-associated antigens in their salivary glands and also display an increased content of EBV-DNA in their saliva [9]. This coupled to the fact Herpesviruses like cytomegalovirus and EpsteinBarr virus (EBV) can replicate in the salivary glands can be interpreted as an indication of viral involvement in the development of the disease. This hypothesis has been challenged by other groups that have demonstrated that the frequency of EBV detection in Sj6gren's syndrome patients is similar to the frequency observed in normal populations [10, 11 ]. Another link between viruses and the disease comes from the fact that chronic lymphocytic sialadenitis that is linked to viral infection is histologically very similar to Sj6gren's syndrome and can be found in 14% to 50% of hepatitis C virus infected patients [12]. HCV RNA has been detected in salivary glands of patients with chronic HCV infection by in situ hybridization [ 13]. Furthermore, sialadenitis has been reported in transgenic animals carrying the HCV envelope genes [14]. Retroviruses can also cause sialadenitis in patients infected with human immunodeficiency virus (HIV) [15] and human Tlymphotropic virus I (HTLV-I) [16]. Furthermore, serum antibodies to the p24 capsid protein of HIV have been reported in 30% of SS patients compared to 1% to 4% in healthy controls. Although there is great similarity between sialadenitis and Sj6gren's syndrome, the two entities are different. There is a difference in the severity of tissue damage in the salivary gland that separates the two, and more importantly patients with chronic lymphocytic sialadenitis are negative for disease specific autoantibodies in contrast to SS patients. Furthermore, these types of viral infections that can give rise to sialadenitis are very uncommon in patients suffering from Sj6gren's syndrome. It is possible that viruses can act as the initiating factor in the activation of the epithelial cells in the disease. Transient or persistent infection of the epithelial cells by a putative virus may be the

292

initiating event that leads to the accumulation of T and B cells. These cells could then prime a local autoimmune response using autoantigens provided by the epithelial cells as a result of the viral infection. Finally, monoclonal expansion of B cells under selective antigenic or T-cell-induced pressure can lead to tissue destruction. To further explore this activation of the epithelial cells, it is essential to identify the factors, possibly of viral origin, that trigger the immune response. The identity of the gene products that play this activating role and their origin are essential information in the understanding of the pathogenesis of the disease. To identify genes that may contribute to primary Sj6gren's syndrome pathogenesis the differential display protocol was applied to minor salivary gland RNA samples of a patient with primary Sj6gren's syndrome and a healthy control individual. After sequencing of several differentially expressed genes a 94 bp-fragment homologous to the VP1 region of coxsackievirus B4 RNA expressed exclusively in the diseased sample was identified. The identification of this viral RNA suggests that this virus could have an active role in the pathogenesis of Sj6gren's syndrome. [Triantafyllopoulou et al, unpublished data.]

5. COXSACKIEVIRUSES Coxsackieviruses belong to the large viral family of picomaviruses. This family includes two major groups of human pathogens, the enteroviruses and rhinoviruses. Coxsackieviruses, as all members of the family of enteroviruses, are small non-enveloped RNA viruses that are classified on the basis of their antigenic response. Coxsackie viruses are divided into two major serotype groups as determined by antibody neutralization tests. The larger group contains 23 viruses (A1-A24, no A23), while the smaller group contains 6 (B l-B6) [17]. Due to the nature of their infection cycle that only utilizes RNA as the genetic material, these viruses have no mechanism that maintains sequence integrity equivalent to the DNA proofreading system. As a result, these viruses mutate very rapidly giving rise to new variants. Mutations in the coding and non-coding regions of the viral sequence can result in viruses with altered virulence [ 18].

Coxsackievirus infection gives rise to a variety of human diseases. Herpangina, acute hemorrhagic conjuctivitis and hand-foot-and-mouth disease are all caused by type A coxsackieviruses. Type B coxsackieviruses are known to cause myocarditis, pericarditis and meningoencephalitis. Both types also give rise to aseptic meningitis, respiratory and undifferentiated febrile illnesses and hepatitis [ 19]. The virion of coxsackieviruses consists of a capsid shell of 60 subunits, arranged in pentamers, each being composed of four proteins (VP1-VP4). The pentamers are arranged to form an icosahedron. These pentamer subunits surround the viral genome that is made up of a single strand of positive sense RNA. The three largest proteins (VP 1, VP2 andVP3) are very similar in structure, with the peptide backbone of the protein looping back to form a barrel of eight strands. The aminoacid chains between this ~-barrel and the ends of the protein form a series of loops that contain the main antigenic sites that are found on the surface of the virion [20]. The five subunits on each pentamer form a cleft or canyon and on the floor of this canyon is the receptor binding site that is used in the attachment of the virion to the host cell. This canyon is too narrow for deep penetration by antibody molecules [21 ]. This should protect this important site from structural variation that could result from antibody selection in hosts. The genome of the virus is an RNA molecule of about 7.4 kb that codes for a single polyprotein that is cleaved to produce the various proteins that are required for virion structure and replication. The RNA is polyadenylated at the 3' end while the 5' end is covalently bound to a small viral protein, VPg. The viral RNA contains 5' and 3' untranslated regions (UTR) flanking the large coding region of the polyprotein. In the 5' UTR the VPg binding site can be found, in addition to other sequences involved in the control of viral RNA translation and replication. This 5' UTR region is highly conserved among enteroviruses indicating that this region contains sequences with common function among this family of viruses. The polyprotein-coding region can be divided into two major regions according to the function of the smaller proteins that result from the cleavage of the single large polyprotein. The polyprotein is cleaved to produce two types of proteins, structural proteins called VP proteins and functional proteins called P proteins. The sequence

for the VP proteins is located in the 5' end of the polyprotein coding region while P protein sequence is on the 3' end [19]. Coxsackie replication takes place in the cytoplasm of the infected host cell. The first step in viral infection involves the attachment of the virion on receptors that are located on the surface of the plasma membrane. Coxsackie viruses gain entrance into cells by binding to two different receptors, coxsackie-adenovirus receptor (CAR) [22] and decayaccelerating factor (DAF) [23]. CAR is a common receptor for coxsackieviruses and adenoviruses. While in coxsackieviruses the receptor is used for both attachment and internalization, in adenoviruses CAR mediates virus attachment and subsequently viral entry is achieved through integrins. CAR is a 46 kDa membrane glycoprotein with two immunoglobulin-like extracellular domains, a transmembrane domain and a long cytoplasmic domain. CAR is expressed in many tissues but its cellular function other than as a virus receptor remains unknown. DAF is a GPI-anchored glycoprotein that functions in protecting cells from lysis by autologous complement. Expression of DAF on the cell surface permits virus attachment but not infection. This suggests that unlike CAR, DAF is incapable of mediating some important post-attachment activity essential for viral entry. CAR appears to bind to the canyon that is formed on the surface of the virus. This binding results in a conformational change of the viral proteins that leads to viral instability and uncoating [21]. This appears to result from the interaction of the receptor with an unknown pocket factor that binds in the base of the canyon and competes with the receptor for the binding site. Because the pocket factor and the receptor have overlaping binding sites, only one can bind to the canyon at a time. The pocket factor stabilizes the virus for transport from cell to cell, but as soon as the receptor competes for the binding site, the pocket factor is removed and viral uncoating can begin. During this uncoating VP4 is lost from the viral structure and the resulting conformational change leads to viral internalization and release of the viral RNA into the host cell. The viral positive strand RNA is translated in the cytoplasm to produce the single polyprotein that following cleavage results to the various structural a n d essential replication proteins. This

293

cleavage is mediated by proteases of viral origin. New viral RNA synthesis begins when the viral replication proteins, including an RNA dependent RNA polymerase, are produced. The infecting plus strand RNA is copied to produce a negative strand RNA that serves as a template for the synthesis of many plus strands. These plus strands can either be translated to form more viral proteins or get packaged into new viruses. Some are also recycled as templates for the production of more negative strand RNA molecules that through the normal replication mechanism generate more plus strands. Plus strands are bound to the VPg protein before they are packaged into virions. Viral synthesis and maturation involves several cleavage reactions. The coat precursor protein p l is cleaved to generate VP0, VP3 and VP1. When an adequate concentration of these is reached, they form pentamers which package VPg attached RNA into provirions. These non-infective provirions become infective virions after VP0 is cleaved to generate VP4 and VP2. There are two types of infections that are observed in coxsackieviral infection. The most common infection leads to cell lysis and release of virus particles through increased cell permeability. By altering cell permeability the virus leaves the cell but at the same time several cellular components escape. In addition to this type of infection, persistent infection has been observed. In this case the virus stays in the cell for long periods of time and remains active in the cell without disrupting normal cell life.

6. COXSACKIEVIRUSES AND S J O G R E N ' S SYNDROME Having detected coxsackieviral RNA in a disease tissue sample, the presence of coxsackieviruses in the minor salivary gland biopsies of a large number of samples had to be evaluated. This was achieved through semi-nested RT-PCT using primers designed to amplify the 5' untranslated region of human enteroviruses [24]. By amplifying this region with this technique it was possible to detect the majority of enteroviruses and identify other viruses of the same family that could be present in the disease tissues. Amplification products of--370

294

bp were detected and the products were sequenced in order to verify their identity. PCR amplification of the ~-actin housekeeping gene was performed for all samples in order to verify the efficiency of the reverse transcription reaction and the integrity of the cDNA [Triantafyllopoulou et al, unpublished data]. Efforts to detect sequences that identify the virus with more accuracy by amplifying regions that code for parts of the protein coat that generate the antigenic response, and characterize the virus serotype were unsuccessful. This can be explained by the fact that these parts of the virus mutate rapidly [18] as a response to antibody selection processes and are difficult to amplify. A question that remained unanswered was whether the viral RNA was present in epithelial cells since the RNA was originally detected in whole minor salivary tissue samples. Furthermore, the presence of the RNA in peripheral blood monocytes could indicate a systemic infection that could present the viral RNA in the glandular tissues through infected lymphocytes. To investigate whether epithelial cells harbour the viral sequences that were detected in the biopsy tissues, cultured salivary gland epithelial cells were used. Cultured salivary gland epithelial cells are non neoplastic cell lines that were generated with the explant outgrowth technique from one lobule of minor salivary gland obtained as a part of Sj6gren's syndrome evaluation [25]. To evaluate the presence of the viral RNA outside the glandular tissues, peripheral blood lymphocytes were isolated at the same time point as the biopsy. RT-PCR utilizing the same primers as for all other cases was used to screen the samples for viral RNA. The results are summarized in Table 2. The viral RNA was detected in minor salivary gland biopsies and long term cultures of epithelial cells originating from salivary glands. The viral RNA was not detected in peripheral blood lymphocytes, indicating that there was no systemic infection at the time of the biopsy (Fig. 2). This suggests that probably the viral RNA that was detected originates from persistent viral infection in the salivary glands of Sj6gren's syndrome patients. [Triantafyllopoulou et al and Liakos et al, unpublished data.] Since the primers that were used were designed to amplify the 5'UTR of a large group of viruses, PCR products had to be sequenced in order to iden-

Figure 2. Agarose gel electrophoresis of RT-PCR products. Lanes 1 and 4 represent control samples, while lanes 2, 3, 5, 6, 7 and 8 represent SS patient samples. 13-actinwas used as a control for the reverse transcription reaction and cDNA integrity. Table 2. Detection of enteroviruses RNA sequences using reverse transcription-PCR (RT-PCR) targeting 5' untranslated region (5'UTR) n

5'UTR RT-PCR Biopsy

Cultured SGEC

Peripheral blood lymphocytes

Primary SS

20

7/8

8/8

0/4

Control

15

0/8

0/5

0/2

Triantaffylopoulou et al and Liakos et al, unpublished data.

tify the virus type. All products were sequenced and two virus types were identified. Ten PCR products were homologous to coxsackievirus type B4, while four were homologous to coxsackievirus type A13. The sequences show 97%-99% homology to the coxsackieviruses. Coxsackie B4 has been linked with autoimmunity and especially with diabetes type I. The first implication of environmental factors in the development of type I diabetes came 70 years ago by the observation that there is seasonal variation in the diagnosis of the disease suggesting that an environmental factor, probably a virus, participates in disease pathogenesis. The first indirect association between coxsackievirus and type I diabetes came from the

examination of the sera of newly diagnosed diabetic patients [26]. These sera contained antibodies against coxsackieviruses in higher frequencies compared to controls. The first direct link came from the isolation of coxsackievirus from the pancreas of a newly diagnosed patient with diabetes ketoacidosis [27]. The advent of molecular techniques enabled direct detection of the virus through polymerase chain reaction, circumventing the indirect detection through antibodies. Furthermore, it has been demonstrated that coxsackfe B virus infection can cause diabetes in two different animal models [27, 28]. The etiology of Sj6gren's syndrome remains obscure despite considerable investigation. For the first time a link between Sj6gren's syndrome

295

Autoimmune Epithelitis

/' / i

: ..... ' . . . . . . . .

",t

Ag- Pi'esentliion

.......: " .

i

/

~] : F L I ~ O N 1 : 1 / ~

! AcrwArs.oN]

..........I~,ACTiVATI~-----I

... . - . . . . :-...~;, i~T!~:i:t

,i

I.ymllhoma

Development

Figure 3. Proposed role of epithelium in the initiation and perpetuation of immune response. Viral infection leads to activation of the epithelium. Activated epithelia recruit and activate B and T-cells that in turn further enhance epithelial activation. This leads to increased epithelial apoptosis that results in the release of intracellular antigens. Classical antigen presenting cells are recruited and further enhance the immune response. B-cell activation can lead to the development of B-cell lymphoma through mutations in p53.

and viruses has been shown. While the role of the virus in the pathogenesis of the syndrome has not been identified, the presence of viral RNA in salivary glands, and more specifically in salivary gland epithelial cells, makes a strong case for the involvement of the virus in the development of the disease. There are several examples of viral-autoantigen interactions. First of all, one of the functions of La(SSB) is to bind to polioviral RNA and initiate translation of viral RNA [29]. Furthermore it has been shown that infection with Epstein-Barr virus leads to expression of La(SSB) in the cytoplasm instead of the nucleus [30]. Moreover, cleavage of La(SSB) by poliovirus 3C protease leads to the redistribution of La from the nucleus to the cytoplasm [31 ]. The mechanism by which the virus participates in disease pathogenesis could be linked to the interaction of the virus with the La(SSB) autoantigen. Since viral RNA can bind to La(SSB) and modify it by cleavage thus altering its localization pattern, it is possible that viral RNA mediates the transfer of the autoantigen to the surface of the cells and conceivably outside the cell through apoptosis and

296

apoptotic blebs [32]. This presentation of the antigen to the cell surface or the surrounding microenvironment could lead to the generation of the immune response. Although the viral sequences that have been detected appear to be of specific origin, they represent a fraction of the viral genome and could be part of some unidentified virus that may be the result of some viral recombination effect. Another question that remains unanswered is whether the virus replicates actively in the cells or the infection is persistent. Virus characterization would give information on the virus type and could provide clues about the life cycle of the virus. The receptor that mediates cell entry for this virus should be also identified. To further characterize the role of the viral sequences in the pathogenesis of SjGgren's syndrome the virus needs better characterization and also more light should be shed to the interaction of the virus with the different cellular components and especially autoantigens. In light of this material we propose the following working hypothesis for the role of coxsackieviruses in the pathogenesis of Sj6gren's syndrome (Fig. 3).

The virus after infection of the epithelium remains latent in some genetically predisposed individuals. Through some hormonal or stress related events the virus can participate in the activation of the infected epithelium by becoming active. These events that trigger the response can be connected either to the virus life cycle or the epithelial cell-virus interactions leading to the activation of the epithelium. This activated state of epithelia initiates and perpetuates the immune response that leads to tissue damage that characterizes Sjrgren's syndrome.

10.

11.

12.

REFERENCES 1.

2.

3.

4.

5.

6.

7.

8.

9.

Moutsopoulos HM, Chused TM, Mann DL, Klippel JH, Fauci AS, Frank MM et al. Sj6gren's syndrome (Sicca syndrome): current issues. Annals of Internal Medicine 1980;92(2 Pt 1):212-26. Manoussakis MN, Moutsopoulos HM. Sjrgren's syndrome: current concepts. Advances in Internal Medicine 2001;47:191-217. Skopouli FN, Dafni U, Ioannidis JP, Moutsopoulos HM. Clinical evolution, and morbidity and mortality of primary Sj6gren's syndrome. Seminars in Arthritis and Rheumatism 2000;29( 5):296-304. Moutsopoulos HM. Sj6gren's syndrome: autoimmune epithelitis. Clinical Immunology and Immunopathology 1994;72(2):162-5. Tzioufas AG, Boumba DS, Skopouli FN, Moutsopoulos HM. Mixed monoclonal cryoglobulinemia and monoclonal rheumatoid factor cross-reactive idiotypes as predictive factors for the development of lymphoma in primary Sj6gren's syndrome. Arthritis and Rheumatism 1996;39(5):767-72. Tapinos NI, Polihronis M, Moutsopoulos HM. Lymphoma development in Sj6gren's syndrome: novel p53 mutations. Arthritis and Rheumatism 1999;42(7): 1466-72. Dimitriou ID, Kapsogeorgou EK, Moutsopoulos HM, Manoussakis MN. CD40 on salivary gland epithelial cells: high constitutive expression by cultured cells from Sj6gren's syndrome patients indicating their intrinsic activation. Clinical and Experimental Immunology 2002;127(2):386-92. Kapsogeorgou EK, Moutsopoulos HM, Manoussakis MN. Functional expression of a costimulatory B7.2 (CD86) protein on human salivary gland epithelial cells that interacts with the CD28 receptor, but has reduced binding to CTLA4. Journal of Immunology 2001 ;166(5):3107-13. Fox RI, Pearson G, Vaughan JH. Detection of Epstein-

13.

14.

15.

16.

17.

18.

19.

20.

21.

Ban: virus-associated antigens and DNA in salivary gland biopsies from patients with Sjrgren's syndrome. Joumal of Immunology 1986; 137(10):3162-8. Syrjanen S, Karja V, Chang FJ, Johansson B, Syrjanen K. Epstein-Barr virus involvement in salivary gland lesions associated with Sjrgren's syndrome. Journal For Oto-Rhino-Laryngology and Its Related Specialties 1990;52(4):254-9. Meme ME, Syrjanen SM. Detection of Epstein-Barr virus in salivary gland specimens from Sjrgren's syndrome patients. The Laryngoscope 1996;106(12 Pt 1): 1534-9. Mariette X, Zerbib M, Jaccard A, Schenmetzler C, Danon F, Clauvel JP. Hepatitis C virus and Sjrgren's syndrome. Arthritis and Rheumatism 1993;36(2): 280-1. Arrieta JJ, Rodriguez-Inigo E, Ortiz-Movilla N, Bartolome J, Pardo M, Manzarbeitia F et al. In situ detection of hepatitis C virus RNA in salivary glands. American Joumal of Pathology 2001; 158(1):259-64. Koike K, Moriya K, Ishibashi K, Yotsuyanagi H, Shintani Y, Fujie H et al. Sialadenitis histologically resembling Sj6gren's syndrome in mice transgenic for hepatitis C virus envelope genes. Proceedings of the National Academy of Sciences of the United States of America 1997;94( 1):233-6. Kordossis T, Paikos S, Aroni K, Kitsanta P, Dimitrakopoulos A, Kavouklis E et al. Prevalence of Sj6gren'slike syndrome in a cohort of HIV-l-positive patients: descriptive pathology and immunopathology. British Joumal of Rheumatology 1998;37(6):691-5. Terada K, Katamine S, Eguchi K, Moriuchi R, Kita M, Shimada H et al. Prevalence of serum and salivary antibodies to HTLV-1 in Sjrgren's syndrome. Lancet 1994;344(8930): 1116-9. King A, Brown F, Christian P, Hovi T, Hyypia T, Knowles Net al. Picomaviridae. In: Regelmortel MV, Fauquet C, Bishop D, Calisher C, Carsten E, Estes Met al, eds. Virus Taxonomy. Seventh Report of the International Committee for the Taxonomy of Viruses. New York, San Diego: Academic Press, 1999;996. Holland J, Spindler K, Horodyski E Grabau E, Nichol S, VandePol S. Rapid evolution of RNA genomes. Science 1982;215(4540): 1577-85. Melnick J. Enteroviruses: polioviruses, coxsackieviruses, echoviruses and newer enteroviruses. In: Fields B, Knipe D, Howley P, Channock R, McInick J, Monath T et al, eds. Fields of Virology. 3rd Ed. Philadelphia: Lippincott-Raven, 1996;655-712. Minor PD. Antigenic structure of picomaviruses. Current Topics in Microbiology and Immunology 1990; 161:121-54. Rossmann MG, Arnold E, Erickson JW, Frankenberger

297

22.

23.

24.

25.

26.

298

EA, Griffith JP, Hecht HJ et al. Structure of a human common cold virus and functional relationship to other picornaviruses. Nature 1985;317(6033): 145-53. He Y, Chipman PR, Howitt J, Bator CM, Whitt MA, Baker TS et al. Interaction of coxsackievirus B3 with the full length coxsackievirus-adenovirus receptor. Nature Structural Biology 2001 ;8(10):874-8. Bergelson JM, Chan M, Solomon KR, StJohn NF, Lin H, Finberg RW. Decay-accelerating factor (CD55), a glycosylphosphatidylinositol-anchored complement regulatory protein, is a receptor for several echoviruses. Proceedings of the National Academy of Sciences of the United States of America 1994;91 (13):6245-9. Zoll GJ, Melchers WJ, Kopecka H, Jambroes G, van der Poel HJ, Galama JM. General primer-mediated polymerase chain reaction for detection of enteroviruses: application for diagnostic routine and persistent infections. Journal of Clinical Microbiology 1992;30(1): 160-5. Dimitriou ID, Kapsogeorgou EK, Abu-Helu RF, Moutsopoulos HM, Manoussakis MN. Establishment of a convenient system for the long-term culture and study of non-neoplastic human salivary gland epithelial cells. European Journal of Oral Sciences 2002;110(1): 21-30. Gamble DR, Kinsley ML, FitzGerald MG, Bolton R,

27.

28.

29.

30.

31.

32.

Taylor KW. Viral antibodies in diabetes mellitus. British Medical Journal 1969;3(671):627-30. Yoon JW, Austin M, Onodera T, Notkins AL. Isolation of a virus from the pancreas of a child with diabetic ketoacidosis. The New England Journal of Medicine 1979;300(21): 1173-9. Hou J, Sheikh S, Martin DL, Chatterjee NK. Coxsackievirus B4 alters pancreatic glutamate decarboxylase expression in mice soon after infection. Joumal of Autoimmunity 1993;6(5):529-42. Craig AW, Svitkin YV, Lee HS, Belsham GJ, Sonenberg N. The La autoantigen contains a dimerization domain that is essential for enhancing translation. Molecular and Cellular Biology 1997;17(1): 163-9. Clark DA, Lamey PJ, Jarrett RF, Onions DE. A model to study viral and cytokine involvement in Sj6gren's syndrome. Autoimmunity 1994; 18( 1):7-14. Shiroki K, Isoyama T, Kuge S, Ishii T, Ohmi S, Hata S et al. Intracellular redistribution of truncated La protein produced by poliovirus 3Cpro-mediated cleavage. Journal of Virology 1999;73(3):2193-200. Casciola-Rosen LA, Anhalt G, Rosen A. Autoantigens targeted in systemic lupus erythematosus are clustered in two populations of surface structures on apoptotic keratinocytes. The Journal of Experimental Medicine 1994; 179(4): 1317-30.

9 2004 Elsevier B. V. All rights reserved.

Infection and Autoimmunity Y. Shoenfeld and N.R. Rose, editors

Viral Infection and Heart Disease: Autoimmune Mechanisms Marina Afanasyeva ~and Noel R.

R o s e 1'2

1Department of Pathology and 2Department of Molecular Microbiology and Immunology, The Johns Hopkins Medical Institutions, Baltimore, MD, USA

1. HUMAN MYOCARDITIS

1.1. Viral Etiology Myocarditis, presently defined by "the presence of an inflammatory infiltrate of the myocardium with necrosis and/or degeneration of adjacent myocytes," [ 1] is a major cause of sudden death in young adults [2]. In predisposed individuals, myocarditis may evolve into a chronic inflammatory dilated cardiomyopathy (DCM), which typically progresses to heart failure and death in the absence of cardiac transplantation [3, 4]. The etiology of myocarditis remains unknown in the majority of cases but accumulating evidence supports an association with viral infection [5, 6], most often involving coxsackieviruses [7]. Other viruses reported to cause myocarditis include adenovirus, cytomegalovirus, parvovirus, human immunodeficiency virus, measles virus, mumps virus, hepatitis A and C viruses, herpes simplex virus, and encephalomyocarditis virus among others. Among coxsackieviruses, coxsackievirus B3 (CB3) has been frequently associated with myocarditis in the United States [7]. CB3 is a member of the enterovirus genus within the Picornaviridae family, and is, therefore, a non-enveloped virus with a single-stranded positive-sense RNA. Thus, it is typically a cytolytic virus and synthesis of negative-sense RNA is required for viral replication but not for viral protein synthesis. Most coxsackievirus infections are either subclinical or present with mild upper respiratory or gastrointestinal symptoms [8]. However, a small percentage of infected individuals demonstrate signs and symptoms of acute myocar-

ditis. Evidence supporting the role of coxsackieviruses in the development of myocarditis and DCM comes from epidemiologic studies demonstrating an association between prior coxsackievirus infection and subsequent cardiomyopathy [9]. Furthermore, there is serologic evidence of infection in patients with DCM [10-12] and viral PdqA has been isolated from cardiac tissue of myocarditis and DCM patients [12, 13]. These studies also demonstrated that a proportion of patients with DCM have actively replicating virus in the heart as detected by the presence of minus-strand RNA. A significant proportion of patients, however, have only positive-strand RNA, indicating latent viral persistence without replication.

1.2. Evidence for Autoimmunity Autoimmune features of human myocarditis and DCM include familial aggregation [14, 15], a weak association with human leukocyte antigen (HLA)DR4 [16], upregulated expression of HLA class II on cardiac endothelium [17], increased levels of circulating cytokines [18, 19], and presence of cardiac-specific autoantibodies of the IgG class in the blood [20].

1.2.1. Autoantibodies Cardiac autoantibodies provide the strongest evidence for autoimmunity in myocarditis and include those specific to t~ and [3 isoforms of cardiac myosin (CM) heavy chain [21-23], antibodies against some mitochondrial antigens, such as the adenine nucleotide translocator and the branched-chain or-

299

ketoacid dehydrogenase dihydrolipoyl transacylase [24, 25], and antibodies to cardiac receptors, such as [~l-adrenoreceptor and M2 muscarinic receptor [26, 27]. An important question has been whether these cardiac-specific antibodies mediate disease or represent a marker, or epiphenomenon, resulting from damage to cardiomyocytes. It is of interest that autoantibodies have been detected in family members of patients with DCM years before the development of disease [16]. Lauer et al [28] found that the presence of anti-CM IgG in serum of patients with chronic myocarditis was associated with the deterioration of both systolic and diastolic function during a six-month follow-up. It has been suggested that anti-receptor antibodies, particularly anti-J31 adrenoreceptor IgG, could either stimulate or block the receptor thereby affecting myocyte contractility [26, 29]. Muller et al [30] demonstrated beneficial effects of IgG adsorption in DCM patients with high-titer anti-J31 receptor antibody. Clinical improvement upon immunoadsorption treatment correlated with the reduction in anti-131 antibody activity in the serum as assessed by an in vitro bioassay involving cultured neonatal rat cardiomyocytes. Similar results were obtained in a different randomized study by Felix el al [31] where a significant hemodynamic improvement was observed in DCM patients treated with immunoadsorption compared to controls. A later study, however, has demonstrated that the beneficial effects of immunoadsorption could not be attributed to the reduction in anti-131 antibody levels since anti-~l antibody-positive and antibody-negative patients benefited equally from such treatment [32].

0.45 [33]. The drawback of the study was that the patients were not discriminated in terms of the presence of an active viral infection in the heart. Such discrimination might have been important since immunosuppression could improve the harmful autoimmune response but impede the protective anti-viral response. A more recent study by Frustaci et al [34] has demonstrated the importance of viral detection in the heart in selecting patients for immunosuppressive therapy. The authors treated patients with active lymphocytic myocarditis with prednisone and azathioprine in addition to a conventional therapy. Based on the retrospective analysis of the presence of a viral genome (enterovirus, adenovirus, influenza A virus, Epstein-Barr virus, parvovirus B 19, and hepatitis C virus) in biopsy specimens and serum cardiac autoantibody at the onset of treatment, the authors concluded that immunosuppression is beneficial in patients with circulating anticardiac antibody and with no viral genome (except for the presence of hepatitis C virus) in the heart. In another clinical study, Wojnicz el al [35] have demonstrated beneficial effects of immunosuppression in a group of myocarditis patients with immunohistological evidence of HLA upregulation in the myocardium. These studies provide insights into the heterogeneity among the lymphocytic myocarditis cases and underscore the importance of further subclassification of lymphocytic myocarditis in order to optimize therapy.

2. ANIMAL MODELS

2.1. CB3-lnduced Myocarditis 1.2.2. Immunosuppressive therapy Based on the hypothesis that the autoimmune component is significant in the course of myocarditis, immunosuppressive drugs have been used in some situations as part of a treatment regimen for myocarditis patients. A large-scale myocarditis treatment trial, however, did not show any improvement in ejection fraction or mortality upon treatment with prednisone plus cyclosporine or.prednisone plus azathioprine compared to conventional therapy in patients with histopathologic evidence of myocarditis and left-ventricular (LV) ejection fraction below

300

To better understand the pathologic mechanisms of infection-triggered autoimmune response, several animal models of myocarditis have been established. CB3-induced myocarditis in mice represents one of these models. Following CB3 inoculation, mice develop an acute, inflammatory focal myocarditis with a mixed cellular infiltrate and cardiomyocyte damage, peaking in about 7 days post-infection [36-38]. The severity of the acute disease varies among strains of mice and is associated with the appearance of neutralizing antibody in the serum [39]. Inflammation in the myocardium

gradually subsides by day 21, when there is typically no histologic evidence of myocarditis. Some strains of mice, however, such as A]J and BALB/c, progress to later chronic, or autoimmune, phase of myocarditis peaking around day 35 post-infection [40--42]. The course of CB3-induced myocarditis resembles human disease since majority of humans recover from the acute viral disease without autoimmune sequelae but a small fraction of predisposed individuals, similar to susceptible strains of mice, develop a late autoimmune disease. Histologically, the late phase of myocarditis in mice differs from the acute disease and is characterized by diffuse, rather than focal, leukocyte infiltration with signs of cardiomyocyte "drop out" and fibrosis. The acute viral phase is characterized by the presence of infectious virus in the heart, whereas during the chronic phase no infectious virus can be found, although viral RNA is often present. The two phases of myocarditis also differ in terms of the associated autoantibody profiles. Early after CB3 infection, there is a moderate increase in the natural IgM antibody that binds CM but cross-reacts with skeletal myosin. Those mice that develop the autoimmune phase also produce cardiac-specific and non-crossreactive IgG antibody, with primary cardiac antigen being ct-isoform of CM heavy chain [43]. This is the predominant CM heavy chain isoform in adult mouse ventricles. Adult human ventricles, on the other hand, predominantly express the ~-isoform. In BALB/c mice, these antibodies are mainly of IgG1 subclass. The presence of CM-specific IgG in the serum represents another common feature of the disease in mice and humans [21-23].

2.2. CM-Induced Experimental Autoimmune Myocarditis (EAM) The predominant antibody reactivity to CM led to a hypothesis that autoimmune response to CM might be responsible for ongoing myocarditis. To test this hypothesis, several strains of mice were injected subcutaneously with either an emulsion of purified mouse CM in complete Freund's adjuvant (CFA), or with mouse skeletal myosin in CFA, or CFA alone [44]. CM, but not skeletal myosin or adjuvant alone, administration produced inflammation in the heart resembling that of the autoimmune phase of CB3-induced myocarditis. Remarkably,

the same strains of mice that developed the autoimmune phase of myocarditis following CB3 infection developed myocarditis following injection with CM and, conversely, the strains of mice that were not susceptible to late phase CB3-induced myocarditis did not develop disease following injection with CM. This finding suggested common genetic predisposing factors in the two models of myocarditis. Hence, CB3 appears to trigger an autoimmune disease, namely ongoing myocarditis, making it the first instance in which the antigen responsible for a postinfection autoimmune disease was identified and the disease was reproduced by injecting genetically susceptible animals with that antigen. Both CB3- and CM-induced models have been extensively used to study the pathogenesis of virusinduced autoimmune myocarditis. CB3 model is irreplaceable for studies of viral virulence, viral entry, and of interaction between the virus on one hand and a cardiomyocyte and/or immune system on the other. CM (or EAM) model is valuable to study the autoimmune phenomena which are sometimes difficult to dissect in a more complex viral model, where anti-viral and anti-immune effects often interact.

2.3. Other Models of Myocarditis Similar to human myocarditis, which can be caused by viruses other than coxsackievirus, murine myocarditis can also be induced by murine cytomegalovirus (MCMV) and encephalomyocarditis virus (EMCV) [6, 41, 45]. AJJ and BALB/c mice, which are susceptible to CB3- and CM-induced autoimmune disease, develop both acute and chronic myocarditis following MCMV infection and C57BL/6 mice, resistant to CB3- and CMinduced autoimmune disease, develop only acute phase of MCMV myocarditis. The chronic phase of MCMV myocarditis is also characterized by the presence of CM-specific IgG in the serum. EMCV infection produces myocarditis in susceptible mice (e.g. BALB/c and DBA/2) with distinct acute and chronic phases [6]. Kodama el al [46] described a CM-induced model of myocarditis in Lewis rats. This model is characterized by the presence of giant cells in the myocardium, resembling human giant-cell myocarditis. Interestingly, A/J mice often develop giant

301

Figure 1. Eosinophils and giant cells in severe CM-induced autoimmune myocarditis. A, normal myocardium; B, eosinophils in inflammatory infiltrate; C, eosinophils and giant cells (arrows) in inflammatory infiltrate. Hematoxylin and eosin stain. Original magnifications: xl00 (A), x200 (B), and x400 (C). Reprinted with permission from the American Society for Microbiology [143].

cells in the heart upon CM immunization and the presence of giant cells correlates with disease severity (Fig. 1) [47]. EAM can be reproduced in AJJ mice with porcine CM, which induces disease with the same immunohistopathologic features, antibody and cytokine profiles [48]. In some strains of mice, disease-producing epitopes from CM heavy chain have been successfully used to induce EAM; one of them is a 19 amino-acid long peptide (myhc~(334352), NH2-DSAFDVLSFTAEEKAGVYK-COOH ) which binds to I-A k and produces severe myocarditis in AJJ mice [49].

susceptibility is controlled by multiple genes (our unpublished observations). On the other hand, H-2 modifies the severity of disease since A/J, A.SW and A.CA mice develop moderate to severe disease, whereas A.BY mice, which differ only in their H-2 genes, develop mild or no disease following injection with CM.

3. DISEASE PROGRESSION: FROM VIRAL ENTRY TO HEART FAILURE 3.1. Early Events

2.4. Genetic Susceptibility to Myocarditis Susceptibility to autoimmune myocarditis, whether induced by infecting mice with CB3 or injecting them with CM, appears to be under strict genetic control. A/J mice are classical good responders as are most congenics sharing the A background. BALB/c mice are moderate responders, whereas C57BL/10 and C57BL/6 are generally not susceptible to the autoimmune form of myocarditis [39, 44]. This susceptibility is due primarily to genes that are not part of the major histocompatibility (MHC) complex. For example, A.SW mice are good responders to CM immunization, whereas B10.S mice, which share the same MHC genes (or H-2), fail to respond. Hybrids between these two strains exhibit a wide range of responses, indicating that

302

Coxsackieviruses are believed to enter cells via coxsackievirus-adenovirus receptor (CAR). As can be inferred from the name, this receptor is also important for cellular entry by adenoviruses, another group of viruses associated with myocarditis and DCM. It has been shown that the expression of CAR is low in normal hearts but is upregulated in the hearts of DCM patients [50]. Similarly, upregulation of CAR has been observed in rat models of EAM and myocardial infarction [51, 52]. The function of CAR as well as the significance of its upregulation remains unknown. Decay accelerating factor (DAF), or CD55, represents a coreceptor for CB3 entry [53]. Liu et al [54] have demonstrated in a mouse model that the sarcoma family kinase Lck (or p56 ~ck) is required for the effective CB3 replication, persist-

ence, and the ability to cause myocarditis. Interestingly, the presence of p56 ~ckin T cells was sufficient to restore susceptibility to myocarditis in p56 ~ckdeficient mice. This study suggested the importance of T cells for CB3 delivery to the heart. It was also shown by others that enteroviruses can replicate in leukocytes and leukocytes may serve as carders promoting viral spread to different organs [53]. Opavsky et al [55] have shown that activation of extracellular signal-regulated kinases 1 and 2 (ERK-1/2) downstream of p56 ~ck is important for viral replication in both T cells and cardiomyocytes. The authors suggested that ERK-1/2 activation may be linked to disease susceptibility based on the observation that such activation was more pronounced in the hearts of susceptible AJJ mice compared to resistant C57BL/6. Luo et al [56] have also found that ERK-1/2 activation is important for CB3 replication and virulence.

3.2. Innate Immunity Innate immune system serves to protect the host against invading pathogens by prompt recognition of certain pathogen-associated molecular patterns (PAMPs). Sometimes, innate immune response may be sufficient for the elimination of a pathogen but, in most cases, the adaptive immune response is necessary to finish the job initiated by the innate response and clear the infection. Innate immunity sets the stage for and determines the quality of the adaptive response.

3.2.1. Toll-like receptor (TLR) 4 Lipopolysaccharide (LPS) of Gram-negative bacteria represents a classic PAMP which is recognized by CD14 and TLR4, both of which have been shown to be expressed in the myocardium [57, 58]. LPS seems to affect the susceptibility of mice to myocarditis, since co-treatment with LPS makes typically resistant B 10.A mice susceptible to CB3-induced autoimmune myocarditis with associated high titers of CM-specific IgG antibody [59]. Expression of TLR4, the receptor for LPS, in the heart has been shown to correlate with enteroviral replication in human myocarditis [60]. In a mouse model, TLR4 deficiency resulted in reduced myocarditis and reduced viral replication in the heart

on day 12 post-CB3 infection despite significant CB3 replication in the heart on day 2 post-infection, suggesting enhanced viral clearance and/or less pro-inflammatory environment in the absence of TLR4 [61]. In support of the latter view, TLR4deficient mice had significantly suppressed production of IL-113 and IL-18 in their hearts on day 12. C3H/HeJ mice, which lack functional TLR4 due to a single missense mutation within its coding gene, seem to be highly susceptible to CB3 myocarditis [62]. These discrepancies could be accounted for by the interaction of susceptibility/resistance genes which may interfere with the interpretation and comparison of data derived from different genetic backgrounds.

3.2.2. Complement Complement, another major component of the innate immune response, affects the susceptibility of mice to developing myocarditis. Anderson et al [63] have shown that complement component C3 interacts with capsid proteins of CB3 and this interaction triggers the alternative complement pathway. The authors also proposed that C3 interaction with CB3 might be important for limiting viral load by retention of the virus in the spleen in an antibodydependent fashion. Experiments using CB3 model of myocarditis demonstrated a strain difference in response to C3 depletion with cobra venom factor with decreased inflammation in DBA/2 but not BALB/c mice upon cobra venom administration as assessed on day 7 post-infection [64]. Neither strain exhibited changes in viral load in the heart in response to cobra venom treatment. The authors suggested that the treatment mainly affected the humoral (antibody) autoimmune response and therefore had an effect in DBA/2 mice, which have a mainly antibody-mediated disease, but not in BALB/c mice, which demonstrate a more pronounced cellular autoimmunity. C3 has been shown to be critical for the development of autoimmune myocarditis in the CM model. Administration of cobra venom factor to mice that were injected with CM resulted in impaired IgG antibody responses to CM and prevented myocarditis [65]. Depletion of C3 at the time of initiation rather than progression of disease was critical since multiple injections of cobra venom

303

factor between days 1 and 9 after immunization, but not between days 10 and 18, were effective in preventing myocarditis. A major product generated during activation of the complement cascade is C3d, which acts mainly through two complement receptors, CR1 (or CD35) and CR2 (or CD21). The incidence and severity of disease are significantly reduced in CR1/CR2 double knockout (KO) mice compared with wild type mice. Similarly, blockade of CR1 and CR2 during the time of CM immunization with a monoclonal antibody (mAb), which binds to the extracellular domain shared by the two receptors, abrogated disease, dramatically reduced the production of CM-specific IgG, and was associated with decreased production of IL-1 and TNF~ by splenocytes cultured with CM. CR1 and CR2 have been shown to be present on a subset of activated/memory CD44h~ghCD62L~~ T cells and their engagement triggers T cell responses, implicating complement as an important player not only in antibody-mediated but also in T cell-mediated autoimmunity [65].

20%, and in some cases up to 50%, of the acute inflammatory cell infiltrate in the heart of CB3infected mice. Upon CB3 infection of mice, depletion of 78+ T cells resulted in increased viral titers in the heart indicating the importance of these cells in controlling viral replication [71]. y8+ T cells, particularly T4+ T cells, have been shown to be important in susceptibility to myocarditis induced by CB3. These cells can recognize MHC class I-like CDld molecules and this recognition has been proposed to mediate the susceptibility to CB3-induced myocarditis. CD 1d-deficient mice developed minimal myocardial inflammation with no significant changes in the cardiac viral titers upon CB3 infection [72], but this effect could not be explained by the lack of NKT cell response since mice deficient in invariant J~281 gene, which is expressed in NKT cells, were highly susceptible to myocarditis. Therefore, it was suggested that the lack of y4+ T cell response was responsible for the reduction in myocarditis.

3.2.3. Natural killer (NK) cells

Type I IFNs, ~ and ~, are associated with early innate immune responses and represent a part of anti-viral defense system. IFN-[3 treatment of patients with inflammatory cardiomyopathy associated with LV dysfunction and presence of either enteroviral or adenoviral genomes in the myocardium resulted in improvement of LV function and clearance of viral genomes [73]. Miric et al [74] have reported beneficial effects of IFN-~ treatment in patients with idiopathic myocarditis and idiopathic DCM which were observed in a small randomized clinical trial. There have also been anecdotal reports of successful treatment of enterovirus-induced myocarditis with IFN-~ [75]. The importance of type I IFNs has been demonstrated in a murine model of CB3induced myocarditis where mice deficient for type I IFNs showed increased mortality within 2 to 4 days after infection [76]. Oral treatment with type I IFNs suppressed the inflammatory response in MCMVinduced myocarditis in mice [77].

NK cells, another component of the innate immune response, can directly kill their target cells and represent a rich source of cytokines, which in turn can influence the inflammatory milieu and affect the adaptive immune response. Godeny et al [66] have shown that NK cells limit CB3 replication. Depletion of NKI.1 § cells exacerbated acute myocarditis induced in mice with MCMV [41 ]. Interleukin (IL)18 augments NK cell activity and its administration has been shown to improve survival, reduce viral load, and decrease myocarditis in a murine model of EMCV-induced acute myocarditis [67]. The therapeutic effect of IL-18 was associated with increased NK cell activity in the spleen. The role of NK cells in the development of the autoimmune phase of myocarditis remains unclear.

3.2.4. y8 T cells T cells expressing Y and 8 chains of the T cell receptor (TCR) have been shown to accumulate in the myocardium during fulminant myocarditis in humans [68]. Huber et al [69, 70] have demonstrated that ?8 + T cells comprise between 5% and

304

3.2.5. Type I interferons (IFNs)

3.3. Adaptive Immunity 3.3.1. T cells Autoimmune myocarditis is believed to be a T cellmediated disease. Endomyocardial biopsies from patients with myocarditis and idiopathic DCM show infiltration with CD4 § and CD8 + T cells. Transfer of peripheral blood leukocytes from patients with myocarditis and impaired LV function to mice with severe combined immunodeficiency (SCID) resulted in myocardial infiltration with human leukocytes and impaired LV function [78, 79]. Omerovic et al [80] have found that transfer of peripheral blood lymphocytes from DCM patients to SCID mice induced myocardial fibrosis and deterioration of LV function, as assessed by increased LV dimensions on echocardiography, 75 days post-transfer. Upon CB3 infection, CD4-deficient mice exhibited reduced myocardial infiltration and necrosis but the same survival as the control mice [81]. In the same study, CD8 deficiency increased the severity of disease in terms of both myocardial pathology and survival. However, CD4/CD8 double-deficient mice as well as TCR~-deficient mice showed improved survival and decreased myocarditis as observed during 28 days after infection. None of these deficiencies affected cardiac viral titers on day 7 post-infection. Other studies have also demonstrated the effects of the absence of either CD4 § or CD8 § T cells on the survival, myocarditis, and viral titers after CB3 infection [82]. Overall, however, the results are rather confusing since it is difficult to dissect the effects on viral replication, viral tropism, and inflammatory response in the heart. The possibility that lymphocytes can deliver CB3 to the heart further complicates the interpretation and underscores the complexity of the viral model. Experiments by Smith and Allen [83] showed that CD4 § T cells play a central role in the pathogenesis of CM-induced EAM. Depletion of CD4 § T cells with mAb prevented the development of disease and the disease could be reproduced in immunologically deficient SCID mice by transfer of CD4 § T cells isolated from CM-immunized mice. The heart infiltrate in EAM has greater numbers of CD4 § T cells compared to CD8 + T cells [48]. The predominance of CD4 § T cells persists to day 60 post-immunization in BALB/c mice but the ratio of

CD4 § to CD8 § T cells in the myocardium decreases over time. During the chronic phase of EAM (around day 60 post-immunization), the proportion of CD4 + T cells within the total infiltrating leukocyte population correlates with systolic dysfunction and the development of large LV volumes, the hallmarks of DCM, suggesting the role of CD4 + T cells in the development of cardiac dysfunction(our unpublished observations). The role of CD8 + T cells in EAM is less well defined. It has been demonstrated that CD8 deficiency in mice leads to exacerbation of CM-induced myocarditis [84], the finding similar to that in the CB3 model [81]. However, depletion of CD8 § T cells with a mAb reduced the severity of EAM [85, 86]. The ability of CD8 + T cells to induce myocarditis was demonstrated in mice transgenically expressing an ovalbumin peptide in the heart under cardiac-specific promoter [87]. These mice developed severe myocarditis upon the transfer of CD8 § T cells from transgenic mice that expressed TCR specific for the ovalbumin peptide. In this system, IL-12 was crucial for pathogenicity of CD8 + T cells.

3.3.2. B cells and antibody The role of antibody in the development of myocarditis is less well characterized compared to the role of T cells. In both AJJ and BALB/c mice, disease severity upon CM immunization correlates with CM-specific IgG1 [47]. IgG1 is deposited in the heart and clusters of IgGl-positive cells, which are most likely plasma cells, are found in the myocardial infiltrate on day 21 post-immunization. While it is likely that antibody contributes to the pathogenesis of myocarditis, its role in disease initiation seems to vary among different strains of mice. Transfer of sera collected on day 21 post-immunization in A.SW mice failed to induce myocarditis in A.SW recipients [88]. Liao et al [89] have demonstrated that the transfer of mAb specific for CM induced myocarditis in DBA/2 but not BALB/c mice. The authors found that DBA/2 but not BALB/c mice expressed myosin or myosin-like molecules in the extracellular matrix in the myocardium and offered this finding as a potential explanation for the strain difference. Furthermore, CM-specific IgM antibody failed to induce myocarditis in DBA/2 mice and

305

only CM-specific IgG was able to transfer the disease [90]. In another study using BALB/c mice deficient in B cells due to disruption in the IgM gene, B cells have been shown dispensable for the induction of CM-induced myocarditis [91 ].

3.4. The Role of Cytokines Cytokines are the products of activation of both innate and adaptive immune systems, and, therefore, can act early, during disease initiation, and late, during disease progression. For the purpose of discussion they are classified here as proinflammatory (TNF-tx, IL-113, and IL-6), T helper (Th) 1 (IL-12 and IFN-y), and Th2 (IL-4 and IL-10). IL-10, however, does not quite fit into the Th2 group and should rather be classified as an immunoregulatory cytokine.

3.4.1. Proinflammatory cytokines Experiments using both CB3 model and CM model of myocarditis showed that IL-1 and tumor necrosis factor (TNF) are critical for the development of myocarditis [92-95]. Treatment with either of these cytokines rendered otherwise resistant B 10.A mice susceptible to CB3-induced myocarditis [92, 93]. Blocking TNF with a mAb prevented myocarditis in A/J mice immunized with CM [94]. Furthermore, mice deficient in TNF receptor (R) p55 are resistant to the induction of CM-induced EAM [95]. In EMCV-induced myocarditis, genetic deficiency of TNF-t~ resulted in increased mortality and increased viral load but decreased inflammatory response in the heart. These findings suggest that while promoting an autoimmune response, TNF-o~ inhibits viral replication and is necessary for an effective anti-viral response. It is, therefore, important for an effective intervention to differentiate between viral and virus-triggered autoimmune phases of myocarditis. Blocking IL-1 with IL-1R antagonist reduced CB3-induced autoimmune myocarditis in A/J mice [96]. Similarly, expression of IL-1R antagonist in the mouse heart by plasmid DNA decreased myocardial inflammation in CB3-induced myocarditis [97]. More recently, Eriksson et al [98] have shown an important role for IL-1R signaling in the development of EAM by demonstrating that IL-

306

1Rl-deficient BALB/c mice were protected from disease. In the same study, the authors examined the role of dendritic cells (DCs) in antigen presentation to CD4 + T cells and initiation of disease. IL-1R1 signaling was critical for the activation of DCs and the ability of CM peptide-pulsed DCs to induce myocarditis. Transfer of IL-1Rl-sufficient DCs restored disease susceptibility in IL-1Rl-deficient mice indicating the mechanism through which IL-1 triggers an autoimmune response. IL-6, another proinflammatory cytokine, has been shown important for the development of EAM induced by a CM-derived peptide [99]. IL-6 KO mice had significantly reduced prevalence of autoimmune myocarditis; this reduction in prevalence was associated with impaired upregulation of complement. The role of IL-6 in viral myocarditis is less clear. In EMCV-induced myocarditis, administration of IL-6 improved disease outcomes but the transgenic expression of IL-6 exacerbated viral myocarditis [ 100-102].

3.4.2. Thl cytokines Depending on the microenvironment, CD4 + T cell activation can lead to their polarization into either Thl or Th2 type. These two polarized and mutually exclusive states differ mainly in terms of which cytokines are produced by activated CD4 § T cells, with Th 1 cells producing IFN-y and TNF-~ and Th2 cells producing IL-4 and IL-5. The initial cytokine milieu is critical for the Thl/Th2 differentiation of na'fve CD4 + T cells. IL-12 and, more recently, IL-23 have been shown to induce Thl responses [ 103, 104]. IL-4, on the other hand, stimulates Th2 polarization. For some time, a generally believed paradigm was that autoimmune diseases are driven by Thl responses and prevented or ameliorated by Th2 responses. While supported by some studies using animal models, this paradigm does not always explain the effects of individual cytokines on the course of an autoimmune disease [105]. The induction of autoimmune myocarditis seems to be dependent on a Thl-inducing cytokine, IL-12, but its development is suppressed by a classic Thl effector cytokine, IFN-y. This conclusion was based on the studies involving the CM model that demonstrated that BALB/c trice deficient in either IL-12 or IL-12R signaling, IL-12 p40 KO, IL-12RI31 KO

Figure 2. Autoimmunemyocarditisprogresses to DCM in IFN-~/KOmice. Left, normal heart. Right, heart with DCM from an IFN-),knockout mouse on day 23 after CM immunization.Reprinted with permission from the American Society for Microbiology [143].

and signal transducer and activator of transcription (STAT) 4 KO, were resistant to EAM [106, 107]. Furthermore, treatment with exogenous recombinant IL-12 exacerbated EAM in BALB/c mice and in Lewis rats [ 107, 108]. Many actions of IL-12 are believed to be mediated by IFN-y; however, this does not seem to be the case in autoimmune myocarditis. The role of IFN-), in EAM has been investigated using four different approaches: depleting IFN-y with a mAb; using IFN-), KO mice; using WN-yR KO mice; and treating mice with exogenous recombinant IFN-y [47, 106, 107, 109]. All these experiments support the conclusion that IFN-), suppresses the development of autoimmune myocarditis. IFN-3' KO mice develop severe acute myocarditis and pronounced cardiac dysfunction. Those mice that survive the acute stage develop extensive cardiac fibrosis, and many develop DCM and die of congestive heart failure (Figs. 2 and 3). The absence of IFN-), leads to impaired apoptosis of CD4 § T cells and subsequent expansion of activated/memory C D 4 4 high T cells, which might explain the exacerbation of myocardial inflammation (our unpublished observations). Thus, IL-12 exerts its proinflammatory effects in an IFN-~,-independent fashion, but the mechanism remains unclear. Reduced disease in the absence of IL-12R signaling was associated with reduced production of proinflammatory cytokines, IL-1

and IL-6, suggesting that IL-12 promotes disease through upregulation of these cytokines [107]. In a CB3 model, IL-12R[31 deficiency also resulted in reduced inflammation as well as reduced viral replication in the heart on day 12 post-infection, indicating that IL-12R signaling is dispensable for the anti-viral protection and contributes to the initiation of the inflammatory response [61]. In EMCV model, however, recombinant IL-12 treatment reduced mortality and decreased viral replication, whereas neutralization of IL-12 with a mAb resulted in increased mortality [110]. It is plausible that IL-12 is more important in clearing EMCV than CB3 or, alternatively, the acute intervention with a cytokine or anti-cytokine antibody has different effects compared to a genetic deficiency. The effects of IL-12R~I deficiency in a CB3 model also seem to be independent of IFN-y. ILl 2RI] 1 KO mice have reduced viral replication in the heart, whereas IFN- 7 KO mice have increased viral replication on day 12 post-infection [61]. Protective effects of IFN-~, have been shown in a number of virus-induced models. Transgenic expression of WN-y in the pancreas, the initial target of CB3, protects mice from CB3-induced myocarditis [111]. Intranasal administration of IFN-y suppresses viral replication and improves prognosis of EMCVinduced myocarditis [112]. Since IFN-~, is protective during both viral and autoimmune disease, it

307

Figure 3. Morphologic and functional presentation of DCM in mice. A and B, heart cross-sections; C and D, hematoxylin and eosin stain of heart sections, original magnification • E and F, LV function assessed by pressure-volume relations obtained by means of in vivo LV catheterization. Panels A, C, and E represent a normal heart, and panels B, D, and F represent a heart with DCM. Note the increase in volume and reduction in systolic pressure characteristic of heart failure due to DCM in E Reprinted with permission from the American Society for Microbiology [ 143].

308

may represent a potential therapeutic weapon. 3.4.3. Th2 cytokines In the CM model, the analysis of histopathological (presence of eosinophils and giant cells) (Fig. 1) and immunological (correlation of disease with CM-specific IgG1 and upregulation of total IgE responses) profiles has revealed a Th2-1ike phenotype, suggesting a pathogenic role for IL-4, which is important for eosinophil recruitment and IgG 1 class switch [47]. In support of a disease-promoting role of IL-4, treatment with an anti-IL-4 mAb reduced the severity of EAM in A/J mice and induced a shift from a Th2-1ike to a Thl-like phenotype. Such as shift was demonstrated by suppressed IgE and IgG1 responses; upregulated IgG2a response; suppressed production of IL-4, IL-5, and IL-13; and increased production of IFN-~, by cultured splenocytes in response to in vitro stimulation with CM [47]. At the same time, IL-4Rct KO mice on a BALB/c background do not demonstrate reduced severity of EAM and seem to develop the disease earlier compared to the wild type controls [106]. Since IL-13, another Th2 cytokine, signals through the same receptor subunit [ 113], it will be of interest to explore its role in the development of autoimmune myocarditis. IL-10, which is often classified as a Th2 cytokine, exerts immunoregulatory effects by inhibiting activation and effector functions of T cells and antigenpresenting cells [114]. IL-10 seems to suppresses CM-induced murine myocarditis, since an anti-IL10 mAb treatment enhanced disease [115]. This effect was observed when IL-10 was blocked relatively late (starting on day 10 after immunization) but not early (between days 0 and 12), suggesting that IL-10 is more important during the resolution rather than initiation of disease. IL-10 blockade also prevented suppression of EAM induced by nasal tolerance with intra-nasal administration of CM before immunization, implicating IL-10 as a mediator of mucosal tolerance [115, 116]. Watanabe et al [117] demonstrated the disease-suppressive effect of IL-10 in a rat model of EAM by delivering IL10-expressing plasmid vector via electroporation into the tibialis anterior muscles. Treatment with recombinant IL-10 improved the outcomes in viral myocarditis induced by EMCV without any effects

on the viral load in the myocardium [118]. These results suggest that IL-10, similar to IFN-T, can be beneficial regardless of whether the viral or autoimmune component predominates.

3.5. Myocarditis and Cardiac Dysfunction 3.5.1. Changes in the mechanical properties of the heart Human autoimmune myocarditis often presents with symptoms of cardiac dysfunction and heart failure. Echocardiographic studies in patients with myocarditis have demonstrated reduction in ejection fraction indicating deterioration of systolic function [ 119]. Many patients with myocarditis also present with diastolic dysfunction, demonstrated by altered filling patterns on echocardiography with increased peak early (E) diastolic velocity, reduced late (A) diastolic velocity, consequently increased E/A ratio, and with reduced deceleration time [120]. Reduced deceleration time has been shown to represent an increase in diastolic passive stiffness of the LV [121]. Right heart cardiac catheterization demonstrates increased pulmonary artery systolic pressure indicative of increased pulmonary capillary wedge pressure [34]. In some patients, myocarditis-associated cardiac dysfunction is transient and does not lead to chronic cardiomyopathy. In others, myocarditis can progress to DCM, which is manifested by enlarged LV dimensions on echocardiography, reduced LV wall motion, and pronounced impairment of systolic function. The acute phase of EAM is also characterized by deterioration of both systolic and diastolic function as demonstrated by pressure-volume data obtained by means of in vivo LV catheterization (Fig. 3E, F), the gold standard method for assessing cardiac function first described in mice by Georgakopoulos et al [122]. Higher histologic scores of myocarditis severity correlate with reduced cardiac output, stroke work, ejection fraction, end-systolic pressure, maximal rate of pressure development (dP/dtmax) and power output. Concurrently, diastolic dysfunction is manifested by increased end-diastolic pressures, impaired diastolic relaxation, decreased peak filling rate and, most importantly, increased passive stiffness. Extensive myocardial damage eventually leads to the development of DCM, characterized by

309

further suppression of systolic function and associated myocardial remodeling resulting in increased LV volumes. DCM leads to congestive heart failure with markedly increased end-diastolic pressures, suppressed end-systolic pressures and the inability to generate adequate cardiac outputs (Fig. 3). Nishio et al [123] assessed cardiac function using pressure-volume method in DBA/2 mice during the first 14 days after infection with EMCV and observed the most pronounced suppression of both systolic and diastolic function on day 7, which somewhat improved by day 14. LV volumes, however, were largest on day 14, indicating cardiac remodeling consistent with the progression to DCM. In this regard, it would be of interest to assess cardiac function at later time points after infection. In preliminary experiments, we assessed cardiac function during the chronic phase of CB3-induced myocarditis in BALB/c mice (day 35 post-infection) and found a significant reduction in systolic function with less pronounced diastolic dysfunction.

3.5.2. Electrical remodeling Arrhythmias and other electrocardiographic abnormalities are frequent manifestations of myocarditis in humans. Despite some differences in electrophysiologic properties between rodents and humans [124], mouse and rat models of myocarditis also demonstrated pro-arrhythmic electrical remodeling. Less et al [125] studied electrical abnormalities during the acute phase (day 21) of EAM in BALB/c mice using the whole cell patch clamp technique. The authors found prolongation of the action potential duration (APD) with a decrease in repolarizing transient outward current (Ito). Increases in APD have been shown to be arrhythmogenic and are similar to prolongation of Q-T interval in humans, which is associated with life-threatening arrhythmias. Arrhythmogenic prolongation of APD associated with repolarization abnormalities has been observed in heart failure patients [126]. Prolonged action potential was also found during the acute phase of EAM in Lewis rats and was associated with decreased levels of mRNA of Kv4.2, a channel subunit important for transient outward K § currents as a part of Ito current [127]. Patients with heart failure have been shown to exhibit abnormalities in Ca 2§ transients [128]. Cardiomyocytes from

310

heart failure patients show reduced peak amplitude of Ca 2+ transients, prolongation of resequestration of cytosolic Ca 2§ into the sarcoplasmic reticulum (SR), and reduced levels of Ca 2§ in the SR [129]. The latter may be explained by reduced levels of SR Ca2+-ATPase, which pumps Ca 2+ into the SR, increased levels of phospholamban, which inhibits SR CaZ§ and increased activity of sodiumcalcium exchanger, which transports Ca 2§ out of the cell and decreases its availability for the SR uptake [128, 130]. Studies measuring cardiomyocyte Ca 2§ transients during the acute phase of EAM failed to detect reduced Ca 2§resequestration rates and even suggested a faster rate of Ca 2§ removal from the cytoplasm in cultured cardiomyocytes [125, 131]. Protein levels of SR Ca2+-ATPase were unchanged and levels of phospholamban and sodium-calcium exchanger were actually reduced on days 18 and 35 post-immunization in EAM [131]. Ca 2§ cycling needs to be further evaluated in myocarditis models including studies of the chronic phase and in mice with signs and symptoms of heart failure. Importantly, changes in ion transients in cardiomyocytes should be compared to the mechanical performance of cardiomyocytes and of the whole LV chamber in the same mouse.

3.5.3. Direct virus-mediated damage The above mentioned functional abnormalities in myocarditis can result from myocardial damage caused either directly by the virus itself or by the virus-triggered immune response. Viral infection can directly cause cardiomyocyte lysis. A more subtle change mediated by the virus was illustrated in the study by Badorff et al [132]. The authors demonstrated the ability of 2A protease of CB3 to cleave dystrophin both in cultured cardiomyocytes and in infected murine hearts. Dystrophin is a cytoskeletal protein, which provides structural support to the cardiomyocyte and links the sarcomeric contractile apparatus to the sarcolemma and extracellular matrix. Its disruption may compromise the force-generating capability of myocytes and result in DCM. It has been suggested in humans that defects in dystrophin predispose to DCM [133].

3.5.4. Immune-mediated damage Examples of the immune-mediated damage are more abundant and include direct effects of inflammarion-associated cytokines on the heart.~The classic example is TNF-ct, which has been shown to exert negative inotropic effects on cardiomyocytes, alter intracellular signaling, and promote cardiomyopathy [ 134]. Transgenic mice with cardiac-specific overexpression of TNF-tx develop cardiac dysfunction and DCM [ 135, 136]. Other cytokines, such as IL-1 and IL-6, have been shown to affect signaling in cardiomyocytes altering their metabolism and triggering hypertrophy and/or apoptosis [137]. In addition to cytokines, certain inflammatory cells can cause direct damage to cardiomyocytes. In vitro studies have demonstrated the ability of neutrophils to cause free radical-mediated injury of cardiomyocytes and impair their shortening [ 138, 139]. Neutrophils represent a significant component of myocardial infiltrate in severe myocarditis and their numbers correlate with the severity of myocarditis (our unpublished observations). Cardiac function can also be affected by the remodeling of the extracellular matrix in response to inflammation. Different inflammatory mediators can either stimulate or inhibit collagen production and, once collagen is produced, can either promote or inhibit its degradation. Some matrix metalloproteinases (MMPs) have been shown to degrade fibrillar collagen and this process is inhibited by tissue inhibitors of metalloproteinases (TIMPs). Li et al [140] have shown the upregulation of MMP3 (a stromelysin) and MMP-9 (a gelatinase) and concomitant downregulation of their inhibitors, TIMP-1 and TIMP-4, in BALB/c mice on day 10 post-infection with CB3. There is still much to learn about the signaling pathways in the heart and how they are affected by different cytokines as well as the effects of cytokines on inotropy, cardiomyocyte hypertrophy, and myocardial remodeling including fibrosis.

4. M O L E C U L A R MIMICRY VERSUS ADJUVANT EFFECT The potential mechanisms of how CB3 or other cardiotropic viruses trigger the autoimmune response

have been extensively reviewed in the literature [42]. Despite numerous attempts to study these mechanisms, they remain unclear. For some time the molecular mimicry hypothesis attracted a lot of attention and investigators have sought evidence in its support. The concept of molecular mimicry implies that the infecting microorganism shares an epitope with the tissues of the host. For example, infections by [3-hemolytic streptococci induce an antibody that cross-reacts with streptococcal M protein and CM [141]. It was shown that an antibody can cross-react with both CM and CB3 [ 142]. However, the comparison of CB3 sequence with that of CM failed to demonstrate any significant sequence identity, suggesting a low likelihood for cross-reactivity at a T-cell level. Autoimmune myocarditis could not be induced by immunizing mice with inactivated CB3 in adjuvant (our unpublished observations), again failing to support the molecular mimicry hypothesis. Horwitz et al [ 111] argued against the molecular mimicry hypothesis and for the role of direct viral damage to the heart in triggering myocarditis. In this study, NOD mice with a pancreas-specific transgenic expression of IFN-~, were infected with CB3. They developed pancreatitis but no myocarditis despite the production of heart-specific IgG. The authors concluded that myocarditis could only occur if the virus infected the heart and the presence of the viral infection elsewhere could not initiate the autoimmune process in the heart through molecular mimicry. While the importance of molecular mimicry remains controversial, recent studies favor the alternative explanation that infecting viruses cause myocardial damage and provide an appropriate inflammatory milieu, so that presentation of CM epitopes is enhanced and such presentation does not lead to tolerance but rather to immune activation. This phenomenon could be termed "an adjuvant effect" of the viral infection [143]. Similar to CFA or any other adjuvant, viral infection produces an appropriate context for the immune system activation leading to recognition of self-epitopes and subsequent autoimmune process. The initial infection with CB3 triggers activation of the innate immune system and signaling through a set of pattern recognition receptors resulting in the production of proinflammatory mediators, such as TNF-t~ and IL-1 [3. These proinflammatory mediators then participate in the

311

activation of the adaptive immune system and promote presentation of cardiac antigens by DCs. With the activation of the adaptive immune system and further development of inflammation, the balance between the disease-promoting and disease-suppressing factors determines the outcome. Factors, such IL-12, IL-6, and possibly IL-4, may perpetuate the disease and produce severe cardiomyopathy, whereas IFN- T and IL-10 may suppress the inflammation and limit disease.

REFERENCES

5. C O N C L U S I O N

4.

Despite a great deal of effort to understand the nature of virus-triggered inflammatory heart disease, the processes that underlie the progression from viral infection to an autoimmune disease and finally to cardiomyopathy and heart failure remain poorly understood. The animal models provide an opportunity to study the complex phenomena of viral entry and replication, immune response to the viral infection, autoimmune response to cardiac antigens, the role of individual inflammatory components in disease progression, the nature of cardiac remodeling in response to viral damage and inflammation, and the development of cardiac dysfunction. A better knowledge of each of these stages of disease is needed for the improvement of therapeutic interventions. Translation of the research findings into the clinically meaningful data also requires an understanding of the advantages and limitations of individual animal models, formulation of hypotheses based on the basic research findings, and a careful design of clinical trials to address these hypotheses.

1.

2.

3.

5. 6.

7. 8.

9.

10.

11.

12. ACKNOWLEDGEMENTS We are pleased to acknowledge the contributions of David Kass to the studies of cardiac function and of DeLisa Fairweather to the studies of CB3-induced myocarditis. This research was supported by NIH research grants HL67290, HL70729, and AI51835.

13.

14.

312

Aretz HT, Billingham ME, Edwards WD, Factor SM, Fallon JT, Fenoglio JJ Jr, Olsen EG, Schoen FJ. Myocarditis: a histopathological definition and classification. Am J Cardiovasc Pathol 1987;1:3-14. Drory Y, Turetz Y, Hiss Y, Lev B, Fisman EZ, Pines A, Kramer MR. Sudden unexpected death in persons less than 40 years of age. Am J Cardiol 1991;68(13): 1388-1392. TowbinJA, Bowles KR, Bowles NE. Etiologies of cardiomyopathies and heart failure. Nature Med 1999;5: 266-267. Brown CA, O'Connell JB. Myocarditis and idiopathic dilated cardiomyopathy. Am J Med 1995;99(3):309314. Feldman AM, McNamara D. Myocarditis. N Engl J Med 2000;343(19): 1388-1398. Kawai C. From myocarditis to cardiomyopathy: mechanisms of inflammation and cell death: learning from the past for the future. Circulation 1999;99(8):1091-1100. Grist NR, Bell EJ. Coxsackie viruses and the heart. Am Heart J 1969;77(3):295-300. Whitton JL. Immunopathology during coxsackievirus infection. Springer Semin Immunopathol 2002;24(2): 201-213. Riecansky I, Schreinerova Z, Egnerova A, Petrovicova A, Bzduchova O. Incidence of Coxsackie virus infection in patients with dilated cardiomyopathy. Cor Vasa 1989;31 (3):225-230. Cambridge G, MacArthur CG, Waterson AP, Goodwin JF, Oakley CM. Antibodies to Coxsackie B viruses in congestive cardiomyopathy. Br Heart J 1979;41(6): 692-696. Muir P, Nicholson F, Illavia SJ, McNeil TS, Ajetunmobi JF, Dunn H, Starkey WG, Reetoo KN, Cary NR, Parameshwar J, Banatvala JE. Serological and molecular evidence of enterovirus infection in patients with end-stage dilated cardiomyopathy. Heart 1996;76(3): 243-249. Deguchi H, Fujioka S, Terasaki F, Ukimura A, Hirasawa M, Kintaka T, Kitaura Y, Kondo K, Sasaki S, Isomura T, Suma H. Enterovirus RNA replication in cases of dilated cardiomyopathy: light microscopic in situ hybridization and virological analyses of myocardial specimens obtained at partial left ventriculectomy. J Card Surg 2001; 16(1):64-71. Fujioka S, Kitaura Y, Ukimura A, Deguchi H, Kawamura K, Isomura T, Suma H, Shimizu A. Evaluation of viral infection in the myocardium of patients with idiopathic dilated cardiomyopathy. J Am Coil Cardio12000;36(6): 1920-1926. Michels VV, Moll PP, Miller FA, Tajik AJ, Chu JS,

15.

16. 17.

18.

19.

20.

21.

22.

23.

24.

25.

Driscoll DJ, Burnett JC, Rodeheffer RJ, Chesebro JH, Tazelaar HD. The frequency of familial dilated cardiomyopathy in a series of patients with idiopathic dilated cardiomyopathy. N Engl J Med 1992;326(2):77-82. Baig MK, Goldman JH, Caforio AL, Coonar AS, Keeling PJ, McKenna WJ. Familial dilated cardiomyopathy: cardiac abnormalities are common in asymptomatic relatives and may represent early disease. J Am Coil Cardiol 1998;31(1): 195-201. Caforio AL. Role of autoimmunity in dilated cardiomyopathy. Br Heart J 1994;72(6 Suppl):S30-S34. Caforio AL, Stewart JT, Bonifacio E, Burke M, Davies MJ, McKenna WJ, Bottazzo GE Inappropriate major histocompatibility complex expression on cardiac tissue in dilated cardiomyopathy. Relevance for autoimmunity? J Autoimmun 1990;3(2):187-200. Limas CJ, Goldenberg IF, Limas C. Soluble intedeukin-2 receptor levels in patients with dilated cardiomyopathy. Correlation with disease severity and cardiac autoantibodies. Circulation 1995;91 (3):631-634. Caforio AL, Goldman JH, Baig MK, Mahon NJ, Haven AJ, Souberbielle BE, Holt DW, Dalgleish AG, McKenna WJ. Elevated serum levels of soluble interleukin-2 receptor, neopterin and beta-2-microglobulin in idiopathic dilated cardiomyopathy: relation to disease severity and autoimmune pathogenesis. Eur J Heart Fail 2001 ;3(2): 155-163. Caforio AL, Mahon NJ, Tona F, McKenna WJ. Circulating cardiac autoantibodies in dilated cardiomyopathy and myocarditis: pathogenetic and clinical significance. Eur J Heart Fail 2002;4(4):411-417. Latif N, Baker CS, Dunn MJ, Rose ML, Brady P, Yacoub MH. Frequency and specificity of antiheart antibodies in patients with dilated cardiomyopathy detected using SDS-PAGE and western blotting. J Am Coil Cardiol 1993;22(5):1378-1384. Goldman JH, Keeling PJ, Warraich RS, Baig MK, Redwood SR, Dalla LL, Sanderson JE, Caforio AL, McKenna WJ. Autoimmunity to alpha myosin in a subset of patients with idiopathic dilated cardiomyopathy. Br Heart J 1995;74(6):598-603. Warraich RS, Dunn MJ, Yacoub MH. Subclass specificity of autoantibodies against myosin in patients with idiopathic dilated cardiomyopathy: pro-inflammatory antibodies in DCM patients. Biochem Biophys Res Commun 1999;259(2):255-261. Schulze K, Schultheiss HP. The role of the ADP/ATP carder in the pathogenesis of viral heart disease. Eur Heart J 1995;16 Suppl 0:64-7:64-67. Ansari AA, Neckelmann N, Villinger F, Leung P, Danner DJ, Brar SS, Zhao S, Gravanis MB, Mayne A, Gershwin ME. Epitope mapping of the branched chain alpha-ketoacid dehydrogenase dihydrolipoyl

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

transacylase (BCKD-E2) protein that reacts with sera from patients with idiopathic dilated cardiomyopathy. J Immunol 1994;153(10):4754--4765. Magnusson Y, Wallukat G, Waagstein F, Hjalmarson A, Hoebeke J. Autoimmunity in idiopathic dilated cardiomyopathy. Characterization of antibodies against the beta 1-adrenoceptor with positive chronotropic effect. Circulation 1994;89(6):2760-2767. Fu LX, Magnusson Y, Bergh CH, Liljeqvist JA, Waagstein F, Hjalmarson A, Hoebeke J. Localization of a functional autoimmune epitope on the muscarinic acetylcholine receptor-2 in patients with idiopathic dilated cardiomyopathy. J Clin Invest 1993;91(5): 1964-1968. Lauer B, Schannwell M, Kuhl U, Strauer BE, Schultheiss HP. Antimyosin autoantibodies are associated with deterioration of systolic and diastolic left ventricular function in patients with chronic myocarditis. J Am Coil Cardio12000;35(1): 11-18. Limas CJ, Limas C. Beta-adrenoceptor antibodies and genetics in dilated cardiomyopathy- an overview and review. Eur Heart J 1991;12 Suppl D:175-7:175-177. Muller J, Wallukat G, Dandel M, Bieda H, Brandes K, Spiegelsberger S, Nissen E, Kunze R, Hetzer R. Immunoglobulin adsorption in patients with idiopathic dilated cardiomyopathy. Circulation 2000; 101(4):385-391. Felix SB, Staudt A, Dorffel WV, Stangl V, Merkel K, Pohl M, Docke WD, Morgera S, Neumayer HH, Wernecke KD, Wallukat G, Stangl K, Baumann G. Hemodynamic effects of immunoadsorption and subsequent immunoglobulin substitution in dilated cardiomyopathy: three-month results from a randomized study. J Am Coil Cardio12000;35(6): 1590-1598. Mobini R, Staudt A, Felix SB, Baumann G, Wallukat G, Deinum J, Svensson H, Hjalmarson A, Fu M. Hemodynamic improvement and removal of autoantibodies against beta(1)-adrenergic receptor by immunoadsorption therapy in dilated cardiomyopathy. J Autoimmun 2003 ;20(4 ):345-350. Mason JW, O'Connell JB, Herskowitz A, Rose NR, McManus BM, Billingham ME, Moon TE. A clinical trial of immunosuppressive therapy for myocarditis. The Myocarditis Treatment Trial Investigators. N Engl J Med 1995;333(5):269-275. Frustaci A, Chimenti C, Calabrese F, Pieroni M, Thiene G, Maseri A. Immunosuppressive therapy for active lymphocytic myocarditis: virological and immunologic profile of responders versus nonresponders. Circulation 2003; 107(6):857-863. Wojnicz R, Nowalany-Kozielska E, Wojciechowska C, Glanowska G, Wilczewski P, Niklewski T, Zembala M, Polonski L, Rozek MM, Wodniecki J. Randomized, placebo-controlled study for immunosuppressive treat-

313

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

314

ment of inflammatory dilated cardiomyopathy: twoyear follow-up results. Circulation 2001; 104(1):39-45. Herskowitz A, Beisel KW, Wolfgram LJ, Rose NR. Coxsackievirus B3 murine myocarditis: wide pathologic spectrum in genetically defined inbred strains. Hum Pathol 1985;16(7):671-673. Rose NR, Herskowitz A, Neumann DA, Neu N. Autoimmune myocarditis: a paradigm of post-infection autoimmune disease. Immunol Today 1988;9(4): 117-120. Huber SA. Autoimmunity in myocarditis: relevance of animal models. Clin Immunol Immunopathol 1997 ;83(2):93-102. Wolfgram LJ, Beisel KW, Herskowitz A, Rose NR. Variations in the susceptibility to Coxsackievirus B3induced myocarditis among different strains of mice. J Immunol 1986;136(5): 1846-1852. Rose NR, Wolfgram LJ, Herskowitz A, Beisel KW. Postinfectious autoimmunity: two distinct phases of coxsackievirus B3-induced myocarditis. Ann NY Acad Sci 1986;475:146-156. Fairweather D, Kaya Z, Shellam GR, Lawson CM, Rose NR. From infection to autoimmunity. J Autoimmun 2001;16(3):175-186. Hill SL, Afanasyeva M, Rose NR. Autoimmune myocarditis. In: Bona CA, Theophilopoulos AN, eds. The Molecular Pathology of Autoimmune Diseases. Ann Arbor, MI: Taylor & Francis, 2002;951-964. Rose NR, Beisel KW, Herskowitz A, Neu N, Wolfgram LJ, Alvarez FL, Traystman MD, Craig SW. Cardiac myosin and autoimmune myocarditis. Ciba Found Symp 1987;129:3-24. Neu N, Rose NR, Beisel KW, Herskowitz A, GurriGlass G, Craig SW. Cardiac myosin induces myocarditis in genetically predisposed mice. J Immunol 1987; 139(11):3630-3636. Lenzo JC, Fairweather D, Cull V, Shellam GR, James Lawson CM. Characterisation of murine cytomegalovirus myocarditis: cellular infiltration of the heart and virus persistence. J Mol Cell Cardiol 2002;34(6): 629-640. Kodama M, Matsumoto Y, Fujiwara M, Masani F, Izumi T, Shibata A. A novel experimental model of giant cell myocarditis induced in rats by immunization with cardiac myosin fraction. Clin Immunol Immunopathol 1990;57(2):250-262. Afanasyeva M, Wang Y, Kaya Z, Park S, Zilliox MJ, Schofield BH, Hill SL, Rose NR. Experimental autoimmune myocarditis in A]J mice is an interleukin-4dependent disease with a Th2 phenotype. Am J Pathol 2001;159(1): 193-203. Wang Y, Afanasyeva M, Hill SL, Rose NR. Characterization of murine autoimmune myocarditis induced

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

by self and foreign cardiac myosin. Autoimmunity 1999;31:151-162. Donermeyer DL, Beisel KW, Allen PM, Smith SC. Myocarditis-inducing epitope of myosin binds constitutively and stably to I-Ak on antigen-presenting cells in the heart. J Exp Med 1995;182(5):1291-1300. Noutsias M, Fechner H, de Jonge H, Wang X, Dekkers D, Houtsmuller AB, Pauschinger M, Bergelson J, Warraich R, Yacoub M, Hetzer R, Lamers J, Schultheiss HP, Poller W. Human coxsackie-adenovirus receptor is colocalized with integrins alpha(v)beta(3) and alpha(v)beta(5) on the cardiomyocyte sarcolemma and upregulated in dilated cardiomyopathy: implications for cardiotropic viral infections. Circulation 2001;104(3): 275-280. Ito M, Kodama M, Masuko M, Yamaura M, Fuse K, Uesugi Y, Hirono S, Okura Y, Kato K, Hotta Y, Honda T, Kuwano R, Aizawa Y. Expression of coxsackievirus and adenovirus receptor in hearts of rats with experimental autoimmune myocarditis. Circ Res 2000;86(3): 275-280. Fechner H, Noutsias M, Tschoepe C, Hinze K, Wang X, Escher F, Pauschinger M, Dekkers D, Vetter R, Paul M, Lamers J, Schultheiss HP, Poller W. Induction of coxsackievirus-adenovirus-receptor expression during myocardial tissue formation and remodeling: identification of a cell-to-cell contact-dependent regulatory mechanism. Circulation 2003;107(6):876-882. Vuorinen T, Vainionpaa R, Heino J, Hyypia T. Enterovirus receptors and virus replication in human leukocytes. J Gen Virol 1999;80(Pt 4):921-927. Liu P, Aitken K, Kong YY, Opavsky MA, Martino T, Dawood F, Wen WH, Kozieradzki I, Bachmaier K, Straus D, Mak TW, Penninger JM. The tyrosine 1,'6nase p561ck is essential in coxsackievirus B3-mediated heart disease. Nat Med 2000;6(4):429-434. Opavsky MA, Martino T, Rabinovitch M, Penninger J, Richardson C, Petric M, Trinidad C, Butcher L, Chan J, Liu PP. Enhanced ERK-1/2 activation in mice susceptible to coxsackievirus-induced myocarditis. J Clin Invest 2002; 109(12): 1561-1569. Luo H, Yanagawa B, Zhang J, Luo Z, Zhang M, Esfandiarei M, Carthy C, Wilson JE, Yang D, McManus BM. Coxsackievirus B3 replication is reduced by inhibition of the extracellular signal-regulated kinase (ERK) signaling pathway. J Viro12002;76(7):3365-3373. Cowan DB, Poutias DN, Del Nido PJ, McGowan FX Jr. CDl4-independent activation of cardiomyocyte signal transduction by bacterial endotoxin. Am J Physiol Heart Circ Physiol 2000;279(2):H619-H629. Frantz S, Kobzik L, Kim YD, Fukazawa R, Medzhitov R, Lee RT, Kelly RA. Toll4 (TLR4) expression in cardiac myocytes in normal and failing myocardium. J

Clin Invest 1999; 104(3):271-280. 59. Lane JR, Neumann DA, Lafond-Walker A, Herskowitz A, Rose NR. LPS promotes CB3-induced myocarditis in resistant B10.A mice. Cell Immunol 1991;136(1): 219-233. 60. Satoh M, Nakamura M, Akatsu T, Iwasaka J, Shimoda Y, Segawa I, Hiramori K. Expression of Toll-like receptor 4 is associated with enteroviral replication in human myocarditis. Clin Sci (Lond) 2003;104(6):577-584. 61. Fairweather D, Yusung S, Frisancho S, Barrett M, Gatewood S, Steele R, Rose NR. IL-12 receptor betal and toll-like receptor 4 increase IL-lbeta- and IL-18associated myocarditis and coxsackievirus replication. J Immuno12003;170(9):4731-4737. 62. Chow LH, Gauntt CJ, McManus BM. Differential effects of myocarditic variants of Coxsackievirus B3 in inbred mice. A pathologic characterization of heart tissue damage. Lab Invest 1991 ;64(1):55-64. 63. Anderson DR, Carthy CM, Wilson JE, Yang D, Devine DV, McManus BM. Complement component 3 interactions with coxsackievirus B3 capsid proteins: innate immunity and the rapid formation of splenic antiviral germinal centers. J Virol 1997;71(11):8841-8845. 64. Huber SA, Lodge PA. Coxsackievirus B-3 myocarditis. Identification of different pathogenic mechanisms in DBA/2 and Balb/c mice. Am J Pathol 1986;122(2): 284-291. 65. Kaya Z, Afanasyeva M, Wang Y, Dohmen KM, Schlichting J, Tretter T, Fairweather D, Holers VM, Rose NR. Contribution of the innate immune system to autoimmune myocarditis: a role for complement. Nat Immuno12001 ;2(8):739-745. 66. Godeny EK, Gauntt CJ. Murine natural killer cells limit coxsackievirus B3 replication. J Immunol 1987;139(3): 913-918. 67. Kanda T, Tanaka T, Sekiguchi K, Seta Y, Kurimoto M, Wilson McManus JE, Nagai R, Yang D, McManus BM, Kobayashi I. Effect of interleukin-18 on viral myocarditis: enhancement of interferon-gamma and natural killer cell activity. J Mol Cell Cardiol 2000;32(12): 2163-2171. 68. Eck M, Greiner A, Kandolf R, Schmausser B, Marx A, Muller-Hermelink HK. Active fulminant myocarditis characterized by T-lymphocytes expressing the gammadelta T-cell receptor: a new disease entity? Am J Surg Pathol 1997;21(9):1109-1112. 69. Huber SA. Coxsackievirus-induced myocarditis is dependent on distinct immunopathogenic responses in different strains of mice. Lab Invest 1997;76(5): 691-701. 70. Huber SA, Sartini D, Exley M. Vgamma4(+) T cells promote autoimmune CD8(+) cytolytic T-lymphocyte activation in coxsackievirus B3-induced myocarditis in

71.

72.

73.

74.

75.

76.

77.

78.

79.

80.

81.

82.

mice: role for CD4(+) Thl cells. J Virol 2002;76(21): 10785-10790. Huber SA, Gauntt CJ, Sakkinen P. Enteroviruses and myocarditis: viral pathogenesis through replication, cytokine induction, and immunopathogenicity. Adv Virus Res 1998;51:35-80. Huber S, Sartini D, Exley M. Role of CDld in coxsackievirus B3-induced myocarditis. J Immunol 2003;170(6):3147-3153. Kuhl U, Pauschinger M, Schwimmbeck PL, Seeberg B, Lober C, Noutsias M, Poller W, Schultheiss HP. Interferon-beta treatment eliminates cardiotropic viruses and improves left ventricular function in patients with myocardial persistence of viral genomes and left ventricular dysfunction. Circulation 2003; 107(22):2793-2798. Miric M, Vasiljevic J, Bojic M, Popovic Z, Keserovic N, Pesic M. Long-term follow up of patients with dilated heart muscle disease treated with human leucocytic interferon alpha or thymic hormones initial results. Heart 1996;75(6):596-601. Daliento L, Calabrese F, Tona F, Caforio AL, Tarsia G, Angelini A, Thiene G. Successful treatment of enterovirus-induced myocarditis with interferon-alpha. J Heart Lung Transplant 2003;22(2):214-217. Wessely R, Klingel K, Knowlton KU, Kandolf R. Cardioselective infection with coxsackievirus B3 requires intact type I interferon signaling: implications for mortality and early viral replication. Circulation 2001;103(5):756-761. Lawson CM, Beilharz MW. Low-dose oral use of interferon inhibits virally induced myocarditis. J Interferon Cytokine Res 1999; 19(8):863-867. Schwimmbeck PL, Badorff C, Rohn G, Schulze K, Schultheiss HP. Impairment of left ventricular function in combined immune deficiency mice after transfer of peripheral blood leukocytes from patients with myocarditis. Eur Heart J 1995;16 Suppl 0:59-63. Schwimmbeck PL, Rohn G, Wrusch A, Schulze K, Doerner A, Kuehl U, Tschoepe C, Pauschinger M, Schultheiss HP. Enteroviral and immune mediated myocarditis in SCID mice. Herz 2000;25(3):240-244. Omerovic E, Bollano E, Andersson B, Kujacic V, Schulze W, Hjalmarson A, Waagstein F, Fu M. Induction of cardiomyopathy in severe combined immunodeficiency mice by transfer of lymphocytes from patients with idiopathic dilated cardiomyopathy. Autoimmunity 2000;32(4):271-280. Opavsky MA, Penninger J, Aitken K, Wen WH, Dawood F, Mak T, Liu P. Susceptibility to myocarditis is dependent on the response of alphabeta T lymphocytes to coxsackieviral infection. Circ Res 1999;85(6): 551-558. Henke A, Huber S, Stelzner A, Whitton JL. The role

315

83.

84.

85.

86.

87.

88.

89.

90.

91.

92.

93.

94.

95.

316

of CD8+ T lymphocytes in coxsackievirus B3-induced myocarditis. J Virol 1995;69(11):6720-6728. Smith SC, Allen PM. Myosin-induced acute myocarditis is a T cell-mediated disease. J Immunol 1991;147: 2141-2147. Penninger JM, Neu N, Timms E, Wallace VA, Koh DR, Kishihara K, Pummerer C, Mak TW. The induction of experimental autoimmune myocarditis in mice lacking CD4 or CD8 molecules. J Exp Med 1993;178(5): 1837-1842. Pummerer C, Berger P, Fruhwirth M, Ofner C, Neu N. Cellular infiltrate, major histocompatibility antigen expression and immunopathogenic mechanisms in cardiac myosin-induced myocarditis. Lab Invest 1991;65(5):538-547. Neu N, Pummerer C, Rieker T, Berger P. T cells in cardiac myosin-induced myocarditis. Clin Immunol Immunopathol 1993;68(2): 107-110. Grabie N, Delfs MW, Westrich JR, Love VA, Stavrakis G, Ahmad F, Seidman CE, Seidman JG, Lichtman AH. IL-12 is required for differentiation of pathogenic CD8+ T cell effectors that cause myocarditis. J Clin Invest 2003;111(5):671-680. Neu N, Ploier B, Ofner C. Cardiac myosin-induced myocarditis. Heart autoantibodies are not involved in the induction of the disease. J Immunol 1990;145(12): 4094--4100. Liao L, Sindhwani R, Rojkind M, Factor S, Leinwand L, Diamond B. Antibody-mediated autoimmune myocarditis depends on genetically determined target organ sensitivity. J Exp Med 1995;181:1123-1131. Kuan AP, Zuckier L, Liao L, Factor SM, Diamond B. Immunoglobulin isotype determines pathogenicity in antibody-mediated myocarditis in naive mice. Circ Res 2000;86(3):281-285. Malkiel S, Factor S, Diamond B. Autoimmune myocarditis does not require B cells for antigen presentation. J Immunol 1999;163(10):5265-5268. Lane JR, Neumann DA, Lafond-Walker A, Herskowitz A, Rose NR. Interleukin 1 and tumor necrosis factor can promote coxsackie B3-induced myocarditis in resistant B10.A mice. J Exp Med 1992;175:1123-1129. Lane JR, Neumann DA, Lafond-Walker A, Herskowitz A, Rose NR. Role of IL-1 and tumor necrosis factor in coxsackie virus-induced autoimmune myocarditis. J Immunol 1993;151(3):1682-1690. Smith SC, Allen PM. Neutralization of endogenous tumor necrosis factor ameliorates the severity of myosin-induced myocarditis. Circ Res 1992;70(4): 856-863. Bachmaier K, Pummerer C, Kozieradzki I, Pfeffer K, Mak TW, Neu N, Penninger JM. Low-molecularweight tumor necrosis factor receptor p55 controls

induction of autoimmune heart disease. Circulation 1997 ;95:655--661. 96. Neumann DA, Lane JR, Allen GS, Herskowitz A, Rose NR. Viral myocarditis leading to cardiomyopathy: do cytokines contribute to pathogenesis? Clin Immunol Immunopathol 1993;68:181-190. 97. Lim B K, Choe SC, Shin JO, Ho SH, Kim JM, Yu SS, Kim S, Jeon ES. Local expression of interleukin-1 receptor antagonist by plasmid DNA improves mortality and decreases myocardial inflammation in experimental coxsackieviral myocarditis. Circulation 2002; 105( 11): 1278-1281. 98. Eriksson U, Kurrer MO, Sonderegger I, Iezzi G, Tafuri A, Hunziker L, Suzuki S, Bachmaier K, Bingisser RM, Penninger JM, Kopf M. Activation of dendritic cells through the interleukin 1 receptor 1 is critical for the induction of autoimmune myocarditis. J Exp Med 2003;197(3):323-331. 99. Eriksson U, Kurrer MO, Schmitz N, Marsch SC, Fontana A, Eugster HP, Kopf M. Interleukin-6-deficient mice resist development of autoimmune myocarditis associated with impaired upregulation of complement C3. Circulation 2003; 107(2):320-325. 100. Kanda T, McManus JE, Nagai R, Imai S, Suzuki T, Yang D, McManus BM, Kobayashi I. Modification of viral myocarditis in mice by interleukin-6. Circ Res

1996;78(5):848-856. 101. Tanaka T, Kanda T, McManus BM, Kanai H, Akiyama H, Sekiguchi K, Yokoyama T, Kurabayashi M. Overexpression of interleukin-6 aggravates viral myocarditis: impaired increase in tumor necrosis factor-alpha. J Mol Cell Cardio12001;33(9):1627-1635. 102. Mann DL. Interleukin-6 and viral myocarditis: the YinYang of cardiac innate immune responses. J Mol Cell Cardio12001;33(9):1551-1553. 103. Gately MK, Renzetti LM, Magram J, Stem AS, Adorini L, Gubler U, Presky DH. The interleukin12/interleukin- 12-receptor system: role in normal and pathologic immune responses. Annu Rev Immunol 1998;16:495-521. 104. Cua DJ, Sherlock J, Chen Y, Murphy CA, Joyce B, Seymour B, Lucian L, To W, Kwan S, Churakova T, Zurawski S, Wiekowski M, Lira SA, Gorman D, Kastelein RA, Sedgwick JD. Interleukin-23 rather than interleukin-12 is the critical cytokine for autoimmune inflammation of the brain. Nature 2003;421(6924): 744-748. 105. Gor DO, Rose NR, Greenspan NS. TH1-TH2: a procrustean paradigm. Nat Immuno12003;4(6):503-505. 106. Eriksson U, Kurrer MO, Sebald W, Brombacher F, Kopf M. Dual role of the IL-12/IFN-gamma axis in the development of autoimmune myocarditis: induction by IL-12 and protection by IFN-gamma. J Immunol

2001; 167(9):5464-5469. 107. Afanasyeva M, Wang Y, Kaya Z, Stafford EA, Dohmen KM, Sadighi Akha AA, Rose NR. Interleukin-12 receptor/STAT4 signaling is required for the development of autoimmune myocarditis in mice by an interferon-gamma-independent pathway. Circulation 2001; 104(25):3145-3151. 108. Okura Y, Takeda K, Honda S, Hanawa H, Watanabe H, Kodama M, Izumi T, Aizawa Y, Seki S, Abo T. Recombinant murine interleukin-12 facilitates induction of cardiac myosin-specific type 1 helper T cells in rats. Circ Res 1998;82(10):1035-1042. 109. Eriksson U, Kurrer MO, Bingisser R, Eugster HP, Saremaslani P, Follath F, Marsch S, Widmer U. Lethal autoimmune myocarditis in interferon-gamma receptordeficient mice: enhanced disease severity by impaired inducible nitric oxide synthase induction. Circulation 2001;103(1):18-21. 110. Shioi T, Matsumori A, Nishio R, Ono K, Kakio T, Sasayama S. Protective role of interleukin-12 in viral myocarditis. J Mol Cell Cardiol 1997;29(9):23272334. 111. Horwitz MS, La Cava A, Fine C, Rodriguez E, Ilic A, Sarvetnick N. Pancreatic expression of interferongamma protects mice from lethal coxsackievirus B3 infection and subsequent myocarditis. Nat Med 2000;6(6):693-697. 112. Yamamoto N, Shibamori M, Ogura M, Seko Y, Kikuchi M. Effects of intranasal administration of recombinant murine interferon-gamma on murine acute myocarditis caused by encephalomyocarditis virus. Circulation 1998;97(10): 1017-1023. 113. Chomarat P, Banchereau J. An update on interleukin-4 and its receptor. Eur Cytokine Netw 1997;8(4): 333-344. 114. Moore KW, de Waal MR, Coffman RL, O'Garra A. Interleukin- 10 and the interleukin- 10 receptor. Annu Rev Immunol 2001; 19:683-765. 115. Kaya Z, Dohmen KM, Wang Y, Schlichting J, Afanasyeva M, Leuschner F, Rose NR. Cutting edge: a critical role for IL-10 in induction of nasal tolerance in experimental autoimmune myocarditis. J Immunol 2002; 168(4): 1552-1556. 116. Wang Y, Afanasyeva M, Hill SL, Kaya Z, Rose NR. Nasal administration of cardiac myosin suppresses autoimmune myocarditis in mice. J Am Coll Cardiol 2000;36(6): 1992-1999. 117. Watanabe K, Nakazawa M, Fuse K, Hanawa H, Kodama M, Aizawa Y, Ohnuki T, Gejyo E Maruyama H, Miyazaki J. Protection against autoimmune myocarditis by gene transfer of intedeukin-10 by electroporation. Circulation 2001 ;104(10): 1098-1100. 118. Nishio R, Matsumori A, Shioi T, Ishida H, Sasayama

S. Treatment of experimental viral myocarditis with interleukin- 10. Circulation 1999; 100(10): 1102-1108. 119. Felker GM, Boehmer JP, Hruban RH, Hutchins GM, Kasper EK, Baughman KL, Hare JM. Echocardiographic findings in fulminant and acute myocarditis. J Am Coll Cardiol 2000;36(1):227-232. 120. James KB, Lee K, Thomas JD, Hobbs RE, Rincon G, Bott-Silverman C, Ratliff N, Marchant K, Klein AL. Left ventricular diastolic dysfunction in lymphocytic myocarditis as assessed by Doppler echocardiography. Am J Cardiol 1994;73(4):282-285. 121. Little WC, Warner JG Jr, Rankfn KM, Kitzman DW, Cheng CE Evaluation of left ventricular diastolic function from the pattern of left ventricular filling. Clin Cardiol 1998;21(1):5-9. 122. Georgakopoulos D, Mitzner WA, Chen CH, Byme BJ, Millar HD, Hare JM, Kass DA. In vivo murine left ventricular pressure-volume relations by miniaturized conductance micromanometry. Am J Physiol 1998;274(4 Pt 2):H1416-H1422. 123. Nishio R, Sasayama S, Matsumori A. Left ventricular pressure-volume relationship in a murine model of congestive heart failure due to acute viral myocarditis. J Am Coil Cardio12002;40(8): 1506-1514. 124. Bers DM. Excitation-contraction coupling and cardiac contractile force. 2nd Ed. Dordrecht: Kluwer Academic, 2002. 125. Less H, Shilkrut M, Rubinstein I, Berke G, Binah O. Cardiac dysfunction in murine autoimmune myocarditis. J Autoimmun 1999;12(3):209-220. 126. Tomaselli GF, Beuckelmann DJ, Calkins HG, Berger RD, Kessler PD, Lawrence JH, Kass D, Feldman AM, Marban E. Sudden cardiac death in heart failure. The role of abnormal repolarization. Circulation 1994;90(5): 2534-2539. 127. Saito J, Niwano S, Niwano H, Inomata T, Yumoto Y, Ikeda K, Inuo K, Kojima J, Horie M, Izumi T. Electrical remodeling of the ventricular myocardium in myocarditis: studies of rat experimental autoimmune myocarditis. Circ J 2002;66(1):97-103. 128. Piacentino V, HI, Weber CR, Chen X, Weisser-Thomas J, Margulies KB, Bers DM, Houser SR. Cellular basis of abnormal calcium transients of failing human ventricular myocytes. Circ Res 2003;92(6):651-658. 129. Hasenfuss G, Reinecke H, Studer R, Meyer M, Pieske B, Holtz J, Holubarsch C, Posival H, Just H, Drexler H. Relation between myocardial function and expression of sarcoplasmic reticulum Ca(2+)-ATPase in failing and nonfailing human myocardium. Circ Res 1994;75(3):434-442. 130. Lehnart SE, Schillinger W, Pieske B, Prestle J, Just H, Hasenfuss G. Sarcoplasmic reticulum proteins in heart failure. Ann NY Acad Sci 1998;853:220-30.

317

131. Stull LB, Matteo RG, Sweet WE, Damron DS, Schomisch MC. Changes in calcium cycling precede cardiac dysfunction during autoimmune myocarditis in mice. J Mol Cell Cardio12001;33(3):449-460. 132. Badorff C, Lee GH, Lamphear BJ, Martone ME, Campbell KP, Rhoads RE, Knowlton KU. Enteroviral protease 2A cleaves dystrophin: evidence of cytoskeletal disruption in an acquired cardiomyopathy. Nat Med 1999;5(3):320-326. 133. Feng J, Yan J, Buzin CH, Towbin JA, Sommer SS. Mutations in the dystrophin gene are associated with sporadic dilated cardiomyopathy. Mol Genet Metab 2002;77(1-2): 119-126. 134. Yokoyama T, Vaca L, Rossen RD, Durante W, Hazafika P, Mann DL. Cellular basis for the negative inotropic effects of tumor necrosis factor-alpha in the adult mammalian heart. J Clin Invest 1993;92(5):2303-2312. 135. Kubota T, McTieman CE Frye CS, Slawson SE, Lemster BH, Koretsky AP, Demetris AJ, Feldman AM. Dilated cardiomyopathy in transgenic mice with cardiac-specific overexpression of tumor necrosis factoralpha. Circ Res 1997;81(4):627-635. 136. Bryant D, Becker L, Richardson J, Shelton J, Franco E Peshock R, Thompson M, Giroir B. Cardiac failure in transgenic mice with myocardial expression of tumor necrosis factor-alpha. Circulation 1998;97( 14): 1375-1381. 137. Diwan A, Tran T, Misra A, Mann DL. Inflamma-

318

tory mediators and the failing heart: a translational approach. Curr Mol Med 2003;3(2): 161-182. 138. Poon BY, Ward CA, Giles WR, Kubes P. Emigrated neutrophils regulate ventricular contractility via alpha4 integrin. Circ Res 1999;84(11):1245-1251. 139. Poon BY, Ward CA, Cooper CB, Giles WR, Bums AR, Kubes P. alpha(4)-integrin mediates neutrophil-induced free radical injury to cardiac myocytes. J Cell Biol 2001 ;152(5):857-866. 140. Li J, Schwimmbeck PL, Tschope C, Leschka S, Husmann L, Rutschow S, Reichenbach F, Noutsias M, Kobalz U, Poller W, Spillmann F, Zeichhardt H, Schultheiss HP, Pauschinger M. Collagen degradation in a murine myocarditis model: relevance of matrix metalloproteinase in association with inflammatory induction. Cardiovasc Res 2002;56(2):235-247. 141. Krisher K, Cunningham MW. Myosin: a link between streptococci and heart. Science 1985;227(4685):413415. 142. Gauntt CJ, Higdon AL, Arizpe HM, Tamayo MR, Crawley R, Henkel RD, Pereira ME, Tracy SM, Cunningham MW. Epitopes shared between coxsackievirus B3 (CVB3) and normal heart tissue contribute to CVB3-induced murine myocarditis. Clin Immunol Immunopathol 1993;68(2): 129-134. 143. Rose NR, Afanasyeva M. From infection to autoimmunity: the adjuvant effect. ASM News 2003;69: 132-137.


Rheumatic Fever: How Streptococcal Throat Infection Triggers an Autoimmune Disease Luiza Guilherme 1,2 and Jorge KaliP ,z,3

tHeart Institute - lnCor, University of Sao Paulo, School of Medicine; 2iii-lnstitute for Immunology Investigation, Millenium Institute; 3Clinical Immunology and Allergy, Department of Clinical Medicine University of Sao Paulo, School of Medicine, S~to Paulo, Brazil

1. I N T R O D U C T I O N Rheumatic Fever (RF) is a sequel of throat infection by group A streptococci (GAS), affecting 3-4% of untreated children. Rheumatic Heart Disease (RHD) develops 4-8 weeks or later after GAS infection in 30 to 45% of individuals with RF. It remains a major cause of morbidity and mortality in developing countries. Data from World Health Organization showed that 25 to 40% of cardiovascular diseases in these countries are due to RF. In Brazil, the damage to heart valves as a consequence of RF is responsible for 90% of children heart surgeries.

[2]. Schematic representation of S. pyogenes is in Fig. 1. Over 100 different serotypes of group A streptococci have been described [2] and it has been consistently found that some serotypes are more frequently associated with rheumatic fever while others are more often associated with acute glomerulonephritis. These serotypes or strains are called rheumatogenic and nephritogenic, respectively [3, 4]. The M protein is the most important antigenic structure and shares structural homology with alpha helical coiled-coil human proteins like cardiac myosin, tropomyosin, keratin, laminin, vimentin and several valvular proteins [5-9].

2. S T R E P T O C O C C U S PYOGENES Studies done by Rebecca Lancefield [1] in 1941 classified streptococci groups based on the cell wall polysaccharides (groups A, B, C, F and G). The S. Pyogenes (group A streptococcus) is characterized by carbohydrates composed of N-acetyl [3 D-glucosamine and rhamnose. The group A streptococci (GAS) contains the M, T and R surface proteins and the lipoteichoic acid (LTA) involved in the bacterial adherence to the throat epithelial cells. The M protein extends from the cell wall and it is composed by approximately 450 amino acid residues with antigenic variations but high homology on aminoterminal (N-terminal) portion, except for the 11 first amino acid residues that define the different serotypes. The carboxi-terminal (C-terminal) half end contains multiple repeat regions and is conserved

3. GENETIC MARKERS Determination of a genetic pattern of susceptibility to RF and RHD was pointed out by Cheadle more than a century ago [10]. To define the pattern of inheritance of RF some researchers have assumed an autossomic recessive model [11 ], whereas others, a mendelian pattern of inheritance [12]. Observation of RF or RHD in identical twins suggested that if a mendelian pattern is present, penetrance must be incomplete [13]. Correlation with blood groups or secretor status of patients with RF was observed with a higher incidence of a nonsecretor pattern in affected subjects as well as a reduction of blood group O frequency in rheumatic children [14]. Patarroyo described the presence of an alloantigen

321

M protein LTA

T,R .'-~ +--... I

/

N-terminal portion

9N . N , I li,.~,-N+~ -+'~+N,.~'..t+N'.'

.Nit_*V.P,i-~/*,'

.+.++,..

+/] i C-terminal portion

Outer hyaloronic acid capsule Figure 1. Schematic representation of S. pyogenes. Group A streptococcal cell is covered by an outer hyalorunic acid capsule and is characterized by the group A carbohydrates composed of N-acetyl [] D-glucosamine and rhamnose. M, T and R are surface proteins; LTA: lipoteichoic acid- involved in the bacterial adherence to the throat epithelial cells; N: amino-terminal portion that contains A and B regions; A region defines the serotypes of streptococci strains; C: carboxi-terminal portion contains C and D regions that are highly conserved among the streptococci strains. on the surface of B cells designated 883 present in more than 70% of RF patients from Bogota and New York [15]. However indirect evidence suggested that 883 alloantigen could be related to the HLAclass II molecules [ 16, 17]. A monoclonal antibody was produced against 883 alloantigen [18] called D8/17 that identifies a B cell antigen with enhanced expression in 90-100% of RF patients [ 17]. No consistent association with HLA class I antigens and RF/RHD was found, however, association with different HLA class II antigens has been indicated in several populations (Table 1). The HLA DR4, DR7 and DR9 antigens are in linkage disequilibrium with HLA-DR53. Interestingly, in American Caucasian and Arabian patients an association with HLA-DR4 and rheumatic fever was found [19-21] whereas in Brazilian and Egyptian patients HLA-DR7 was associated with the disease [23-26] (Table 1). HLA class II antigens play an important role in the antigen presentation to the T cell receptor (TCR). The divergence of HLA class II molecules associated with the disease in different

322

countries is probably due to the capability of these molecules to present strain-specific streptococcal epitopes present in more than 80 streptococcal serotypes [30], some of t h e m - including the rheumatogenic s t r a i n s - with peculiar geographic distribution.

4. PATHOGENESIS The pathogenic mechanisms involved in the development of RF/RHD are not fully understood. It is considered that the molecular mimicry mechanism is responsible for the cross reactive reactions between streptococcal antigens and human tissue proteins, mainly heart tissue proteins in susceptible individuals. Nowadays it is clear that the disease is mediated by both humoral and cellular immune responses and that the cellular branch of the immune response is more involved with the development of rheumatic heart disease (RHD).

Table 1. HLA class II antigens and rheumatic fever HLA

Country

Reference

Population ,,

DR4, DR9 DR2 DR4 DR3 DQW2,D8/17 DR7, DR53 Allogenotope TaqI DRbeta 13.81 kb DR7, DQ2 DR7 DR1, DR6 DR11 DR 1

[19, 20]

Saudi Arabia India

American Caucasian American Black Arabian Indian

Brazil Brazil Egypt Brazil South Africa Turkey Brazil

Mulatto Mulatto Egyptian Caucasian African Turkish Mulatto

[23] [24] [25] [26] [27] [28] [29]

USA

[21] [22]

Several HLA antigens are associated with RF/RHD in different countries. HLA-DR4, DR7 and DR9 were associated with HLA-DR53. HLA-DR4, DR9 were found in american caucasian and arabian patients; DR7 in Brazilian (mulatto) and egyptian patients.

4.1. Humoral Immune Response Streptococcal antibodies react with streptococcal antigens and several human tissues including heart, skin, brain, glomerular basement membrane, striated and smooth muscles [31]. Heart-reactive antibodies were described early [32] and Kaplan's studies demonstrated the presence of rabbit and human heart cross reactive antibodies with components of group A streptococci (GAS) in the sera of animals immunized with streptococcal antigens and sera from RF and RHD patients, as well as bound immunoglobulins and complement in the myocardium of ARF patients [33, 34]. After these works several studies were done analyzing sera from both animals and humans or using monoclonal antibodies demonstrating the presence of cross reactive antibodies to streptococcal and human proteins. Cardiac myosin seems to be one of the major cross reactive antigen (reviewed by M. Cunningham) [35]. Recently, we have analyzed the humoral response against overlapping peptides of N-terminal portion of M5 protein and we could identify several immunodominant epitopes recognized by mild RHD patients, most of them with Sydenham's chorea associated. Antibodies from severe RHD patients recognized few N-terminal epitopes (Fig. 2), (manuscript in preparation).

On the other hand, we have also tested the humoral reactivity of sera from RF and RHD patients against heart tissue proteins isolated by molecular weight and isoelectric point. Using this approach we could identify a very large number of heart tissue proteins (316 proteins derived from myocardium and 78 from mitral valve) recognized by sera from these patients. Interestingly, we did not find reactivity against cardiac myosin, but we could characterize six major proteins by peptide mass analysis, one of them with very high homology with vimentin (manuscript in preparation). Although, the presence of human and animal antibodies against streptococcal antigens and human tissue proteins have been described for more than 50 years, their role in the development of the disease remain unclear. One possibility to explain the presence of antibodies in the heart tissue was suggested by the work done by Roberts et al [36] that showed an increased expression of VCAM-1, an adhesion molecule, in the vascular endothelium that was activated after an inflammatory reaction started by antimyosin and N-acetyl-glucosamine. The VCAM-1 molecules interact with VLA-4, another adhesion molecule expressed on CD4 § T lymphocytes. This could be one way to recruit CD4 +T lymphocytes to the heart valves.

323

Residues

Peptide Sequences

11-25"

QRAKEALDKYELENH

Mild RHD/ chorea __~atients

ELENHDLKTKNEGLKTENEG

21-40 41--60

LKTENEGLKTENEGLKTEKK

81-96"

DKLKQQRDTLSTQKET

81-103" 101-120

LKQQRDTLSTQKE,TLEREVQN NGDLTKELNKTRQELANKQQ

111-130 121-140

TRQELANKQQESKENEKALN ESKENEKALNELLEKTVKDK

131-150

ELLEKTVKDKIAKEQENKET

141-160 163-177"

Severe RHD patients

IAKEQENKETIGTLKKILDE iiiiiiiiiiiii~il-iiii!!'i!!!:;.:i!~iiiiiii~!iiiii~ii~i!iiiiiii~i~i

ETIGTLKKILDETVK

iii!!i!ii!iii!~!iiii!iiiiiiiili!iiii!i!iii!!~ii!!! ~

181-200

KILDETVKDKLAKEQKSKQN

183-201 * 191-210

LDETVKDKLAKEQKSKQNI

iiiiiii!iii

i

iiiiiiiiiii!i i

LAKEQKSKQNIGALKQELAK

Figure 2. Immunodominant epitopes of streptococcal N-terminal region from M5 protein recognized by antibodies of RHD and Sydenham's chorea patients. Humoral reactivity against overlapping peptides was tested by ELISA immunoassay. The immunodominat regions were determined by comparing the reactivity of sera from RHD patients with sera from healthy individuals. P values < 0.05 were considered significant. The peptides preferentially recognized are represented as gray for mild RHD patients with or without Sydenham's chorea and dark gray for severe RHD patients. * The sequences of M5 peptides were based on the sequence of the M5 protein published by Manjula et al [51]. The other M5 peptide sequences were based on the sequence of the M5 protein published by Robinson et al [53]. Overlapping peptides are aligned or underlined.

4.2. Cellular I m m u n e Response The studies of cellular branch of immune response began around 1970. In favor of the important role of T cells in RF, some studies have been performed in tonsils and human peripheral blood showing that CD4 + T cells were increased [37, 38]. It was also demonstrated that T cells were able to recognize streptococcal cell wall and tissue antigens [39-44]. A cytotoxic activity towards immortalized human heart cells was also described [45, 46]. The first evidence that CD4 § T cells were

324

involved in RHD lesions was described 20 years ago [47]. The isolation of T cells from heart valves led Yoshinaga et al [48] to compare the reactivity of PHA stimulated T cell lines derived from heart valves specimens and peripheral blood lymphocytes of RF patients and showed that, although these cells recognized cell wall and membrane streptococcal antigens, they failed to react with M protein, myosin or other mammalian cytoskeletal proteins. The functional activity of heart-infiltrating CD4 § T cell clones was directly demonstrated by our group. We defined, for the first time, the presence

Table 2. Immunodominant T cell epitopes of N-terminal portion of streptococcal M5 protein M5 epitope

Sequence

Nature of T lymphocytes

Ref.

1-25" 81-96 83-103 163-177

TVTRGTISDPQRAKEALDKYELENH DKLKOORDTLSTQKET LKQQRDTLSTQKETLE.REVQN ETIGTLKKILDETVK

Human intralesional T cell clones from RHD patientsb

[9]

Peripheral blood of RF/RHD patients

[50]

40-58 (NT4) 59-76 (NT5) 72-89 (NT6) 137-154 (B 1B2) 150-167 (B2) 163-180 (B2B3A) (B3A)

GLKTENEGLKTENEGLKTE KKEHEAENDKLKQQRDTL QRDTLSTQKETLEREVQN VKDKIAKEQENKETIGTL TIGTLKKILDETVKDKIA KDKIAKEQENKETIGTLK IGTLKKILDETVKDKLAK

Murine Lymph node cells

[49]

The sequence of M5 protein (Refs. [9, 50]) was taken from sequence published by Manjula et al [51] and M5 protein used in Ref. [49] by sequence published by Miller et al [52]. aM5(1-25) align with M5(1-35) described by Robinson et al [53]. Underlined, Bold type- shared sequences of M5 peptides. bPeptides recognized by human T cell clones presented cross reaction with human valvar proteins [9]. of intralesional cross reactive T cell clones and we established the significance ofT-cell molecular mimicry in the pathogenesis of RHD. We mapped the Nterminal reactivity of intralesional T cell clones and this study led us to identify three immunodominant regions: 1-25, 81-103 and 163-177 residues within the streptococcal M protein and cross reactive with several heart tissue protein fractions, mainly those derived from valvular tissue with molecular mass of 95-150 kDa, 43-63 kDa and 30--43 kDa [9]. Myosin / M5 protein cross reactive T cell epitopes were also investigated in mice immunized with intact cardiac myosin [49]. Lymph node T cells were tested against overlapping M5 peptides named NT5/6/7 and B 1B2/B2 and B2B3A/B3A align with the M5 regions identified by us, the M5(81-96) and M5(163-177) respectively (Table 2). Robinson et al [53] obtained lymph node T cell clones from mice immunized with recombinant M5 protein that were able to recognize M5 epitopes. Among the M5 epitopes recognized by mice T cell clones, only the M5(1-35) epitope align with the M5(1-25) region recognized by the human infiltrating T cell clones (Table 2). T cells from peripheral blood of RF/RHD disease patients recognized several M5 peptides. Interestingly, the immunodominant peptide M5(81-96) was

preferentially recognized by DR7 § DR53 § severe RHD patients [50], suggesting that HLA DR7 DR53 molecules could be more involved with the selection of streptococcal peptides and their presentation to the T cell receptor (TCR). Several heart protein fractions were also recognized in the periphery by severe RHD patients. In order to better characterize the heart tissue proteins, we recently identified several heart-derived proteins isolated by molecular weight (MW) and isolelectric point (pI). Several valve-derived proteins were recognized by peripheral blood and intralesional T cell clones from severe RHD patients. Among them, we identified vimentin (MW 53 kDa/pI 5.12) and other cytoskeleton proteins as candidates for being the targets of the valvular lesions in RHD (manuscript in preparation). In line with these results, previous work showed the recognition of 50-54 kDa myocardial derived protein by peripheral T lymphocytes from RHD patients [54]. M protein has an important role in anti-streptococcal immune response of the host and was considered with superantigenic properties by some researchers. Superantigens are proteins that polyclonally activate T cells by an MHC class-II dependent, but haplotype-unresctricted mechanism. Proliferative responses to superantigens are limited to T cells

325

Table 3. Degeneracy of antigen recognition by intralesional CD4§T cell clones T cell Clone Antigens Identification Recognized Lu 3.1.8

BV Family CDR3(N-D-N) Sequences

35 kDa/pI 8.84 BV 13

BJ Family

SGRQGRYEQY BJ 2S7 (10aa)

LMM 28 (1647-1664) LMM 28B (1660-1677) LMM 32 (1699-1716) Lu 3.1.29

56-53 kDa/ pI 6.76

BV 13

SGRQGRYEQY BJ 2S7 (10aa)

AV Family CDR3 (N-D-N) Sequences

AJ Family

AV 2

MRTPVTSSI (9aa)

AJ- NT

AV 3

TDPITGTASKLTAJ 44 (12aa)

AV 2

MRTPVTSSI (9aa)

AV 3

TDPITGTASKLTAJ 44 (12aa)

AJ- NT

NT-not tested; LMM-light meromyosin peptides: LMM28-SLQSLLKDTQIQLDDAVR; LMM28B-DDAVRANDDLKENIAIVE; LMM31-LEELRAVVEQTERSRKL; LMM32-RSRKLAEQELIETSERV. Underlined - shared sequences. Adapted of Fa6 et al, Mol Immunol, 2004 (in press).

expressing a particular TCR-BV gene but independent of antigen specificity. In humans pepM5 preparations (pepsin cleaved fragment) were found to be superantigenic for human T cells expressing TCR-BV2, BV4, and BV8 [55-60], however, using recombinant M5 protein or recombinant pep M5 no evidence of superantigenicity was found [61]. It was also reported that the superantigenicity of pepM 1 and pepM5 were due to contamination with pyrogenic exotoxins that had a potent superantigen effect on BV2-bearing human T cells [62, 63]. Our studies on TCR B V usage in the PBMC of severe RHD patients and infiltrating T cell lines derived from myocardium and/or mitral valve showed expansion of several B V families with oligoclonal profiles mainly in infiltrating T cell lines. These results are in favor of no superantigenicity of M proteins in RHD patients. Some major oligoclonal BV expansions were shared between mitral valve and left atrium T cell lines but an indepth analysis of BJ segments usage in these shared expansions, as well as the nucleotide sequencing of the CDR3 regions suggested that different antigenic peptides could be predominantly recognized in the mitral valve and the myocardium [64]. The high frequency and the persistence of T cell oligoclonal expansions in the damaged heart valves seem to be

326

associated with the progression of the disease [65], probably related to the spreading of autoantigen epitope recognition. In agreement with these data, it has been described that it is possible to detect some T cell expansions in the damaged heart valves even 20 years after acute rheumatic fever episode [66]. The TCR analysis of intralesional T cell clones showed a degenerate pattern of reactivity. Several mitral valve-derived T cell clones recognizing different antigens, presented the same TCR B VBJ and CDR3 sequences. They expressed two alpha chains at the RNA level with same AVAJ segments (Table 3) indicating that intralesional T cell clones with common TCR usage can recognize several epitopes that probably amplify the deleterious immune reaction [67].

4.3. Cytokine Profile The evaluation of proinflammatory cytol,dnes levels produced by peripheral blood and tonsillar mononuclear cells after streptococcal antigen and pokeweed mitogen stimulations from RF/RHD patients without congestive heart failure showed a different pattern on PBMC and tonsillar cells. TNF alpha, IL-1 and IL-2 were overproduced by PBMC and decreased by tonsillar mononuclear cells [68].

In acute rheumatic fever (ARF) patients and patients with active RHD increased production of IL-2 and elevated numbers of CD4 § and CD25 § cells were observed suggesting the expansion of activated T CD4 § cells in the peripheral blood during the active phase of the disease [69]. These results were confirmed by other authors and increased plasma levels of TNF alpha in RF/RHD patients was also showed [70--72]. The Aschoff nodule is considered as the pathognomonic lesion of RF. It is composed of an agglomerate of cells having characteristics of monocytic and macrophage cells [73, 74] and probably function as antigen-presenting cells (APC) to the T cells. In the valve lesions of ARF patients, the production of IL-1, TNF alpha and IL-2 correlated with progression of Aschoff nodules as follows: stages 1 and 2, IL-1 and TNF alpha were secreted by monocytes/macrophages and stage 3, IL-2 by T lymphocytes [75]. We analyzed the cytokine pattern of infiltrating mononuclear cells in the mitral valve and myocardium tissue of ARF and chronic RHD patients (manuscript submitted). Our results showed a predominantly T1 type of cytokine produced mainly by CD4 § T cells that could mediate Rheumatic Heart Disease heart lesions.

5. ANIMAL MODEL Different protocols to reproduce rheumatic lesions in animal models were attempted for more than 60 years, however, at this time the immune reactions involved with the development of rheumatic lesions were not known. In the 1960's W.J. Cromartie and J.H Schwab developed a murine model injecting bacterial cell wall into mice and they observed that portions of the injected material were taken up by macrophages but not digested. The non degraded substance remained in the phagocytic cells and was able to produce cardiac inflammatory responses [76]. It was demonstrated later that macrophage cells of mice infected with extracts of GAS were able to induce the appearance of heart lesions when transferred to syngeneic receptors and, in vitro, a specific response to syngeneic heart extracts. This model suggested, at the time, that macrophage displayed a key role by selecting determinants for antigen presentation to the TCR [77]. By using M protein synthetic pep-

tides it was shown that some peptides were capable to induce inflammatory heart disease in mice [78] whereas other peptides after imunization induced strong proliferative response, but not heart lesions [49]. An epitope from M protein that contains a repeated residue sequences, named NT4, has been described.This epitope presents shared sequences with cardiac myosin protein region and was able to induce myocarditis in mice [79]. Recently, it was demonstrated that the injection of M6 recombinant protein in Lewis rat induced a myocarditis and inflammatory valvular heart lesions similar to those seen in rheumatic heart disease. A lymph node CD4 § T cell line obtained from immunized rat recognized M6 recombinant protein and cardiac myosin [80]. This cross reactive T cell line produces IL-2 and IFN), when stimulated with M5 peptides from B region of the M protein, which is implicated with heart-cross reactive reactions [79]. Following the experiments in animal models, the same group differentiated segments of cardiac myosin able to induce myocarditis from those inducing valvulitis [81]. The construction of these experimental models of myocarditis and valvulitis induced by streptococcal antigens and cardiac myosin certainly will contribute to better understand the pathogenesis of rheumatic heart disease and could be useful for vaccine designs.

6. CONCLUSIONS Altogether, all the results presented here delineate RF/RHD as a complex autoimmune disease mediated by both humoral and cellular immune response and point out the major role of CD4 § T cells in the development of rheumatic heart lesions. The autoimmune reaction in the heart probably is mediated by a network of immune reactions, involving the recognition of several auto-antigens triggered on the periphery by an immunodominant streptococcal antigen that expands several T cell clones by epitope spreading. These T cell clones migrate to the heart, the local production of inflammatory T-1 cytokines trigger the activation of autoreactive infiltrating T cells, that were able to recognize several auto-antigens with conformational or sequence homologies. New and important data shown here is the degeneracy of antigen recognition suggesting a new type

327

H L A class II molecules DR7 + DR53 +

9 Immunodominant peptides M 5 ( 8 1 - 9 6 ) and M5(83-103) / several heart-tissue derived proteins 9 -

-

-

TCR several oligoclonal expansions in the site of the heart lesions degeneracy of antigen recognition intramolecular epitope spreading

9 Cytokine profile Predominantly T 1-type in the site o f the lesions Figure 3. Model of T cell recognition for RHD patients. After group A streptococcal throat infection, untreated susceptible individuals (5-18 years old) developed RF/RHD. Humoral and cellular immune response against S. pyogenes, mediated mainly by proinflammatory cytokines (Th-1 type) lead to an autoimmune attack to human tissues. Here we propose a model of T cell recognition based on our results. The autoimmune reaction is initiated in the periphery where T cells recognize immunodominant M5 peptides as M5(81-96) and M5(83-103) presented by APC (antigen presenting cells) (macrophage/monocytes) in the context of HLA class II DR7 DR53 molecules. After, the activated T cell clones expanded migrate to the heart (myocardium and valvular tissue), several heart tissue proteins are recognized by molecular mimicry. T cell clones display degenerate TCR (T cell receptor) capable to recognize different antigens. These cells amplify the cross reactivity by intramolecular degenerate reactivity. In the heart tissue the cytokines produced are also predominantly Th-1 type.

of intramolecular epitope spreading that probably amplify the autoimmune reaction. Fig. 3 is a model of RHD development based on our results. One remaining question is the role of these degenerated T cell clones in establishing disease as pathological autoreactive T cells or regulating disease progression as physiological autoreactive T cells.

2. 3.

4. 5.

6. R

E

F

E

R

E

N

C

E

S

Lancefield RC. Specific relationships of cell composition to biologic activity of hemolytic streptococci. Harvey Lectures 1941 ;36:251-265.

328

7.

Fishetti V. Streptococcal M protein. Sci Am 1991;264 (6):32-39. Wannamaker LW. Differences between streptococcal infections of the throat and skin. (In two parts). N Engl J Med 1970;282:23-31, 78-85. Stollerman GH. The changing face of rheumatic fever in the 20th century. J Med Microbiol 1998;47:1-3. Shikhman AR, Cunningham MW. Immunological mimicry between N-acetyl-l]-D-glucosamine and cytokeratin peptides. J Immunol 1994;152:4375--4387. Shikhman AR, Greenspan NS, Cunningham MW. Cytokeratin peptide SFGSGFGGGY mimics N-acetyl-13-Dglucosamine in reaction with antibodies and lectins and induces in vivo anti-carbohydrate antibody response. J Immunol 1994;153:5593-5606. Manjula BN, Fischetti VA. Tropomyosin-like 7-

8.

9.

10.

11.

12. 13.

14.

15.

16.

17.

18.

19.

20.

residue periodicity in three immunologically distinct streptococcal M proteins and its implicatiopn for the antiphagocytic property of the molecule. J Exp Med 1980;151:695-708. Manjula B N, Trus B L, Fischetti VA. Presence of two distinct regions in the coiled-coil structure of the streptococcal PepM5 protein: Relationship to mammalian coiled-coil proteins and implications to its biological properties. Proc Natl Acad Sci (USA) 1985;82:10641068. Guilherme L, Cunha-Neto E, Coelho V, Snitcowsky R, Pomerantzeff PMA, Assis RV, Pedra F, Neumann J, Goldberg A, Patarroyo ME, Pillegi F, Kalil J. Humaninfiltrating T cell clones from rheumatic heart disease patients recognize both streptococcal and cardiac proteins. Circulation 1995;92:415--420. Cheadle WR. Harveian lectures on the various manifestations of the rheumatic state as exemplified in childhood and early life. Lancet 1889;371:821-827. Wilson MG, Schweitzer M. Pattern of hereditary susceptibility in rheumatic fever. Circulation 1954;10: 699-704. Uchida IA. Possible genectic factors in the etiology of rheumatic fever. Am J Hum Genet 1953;5:61-69. Taranta A, Torosdag S, Metrakos JD, Jegier W, Uchida I. Rheumatic fever in monozigotic and dizigotic twins (abstract). Circulation 1959;20:788. Glynn IE, Holborow EJ. Realation between blood groups, secretor status and susceptibility to rheumatic fever. Arthritis Rheum 1961 ;4:203-207. Patarroyo ME, Winchester RJ, Vejerano A, Gibofsky A, Chalem F, Zabriskie JB Kunkel HG. Association of a Bcell alloantigen with susceptibility to rheumatic fever. Nature 1979;278:173-174. Ayoub EM. The search for host determinants of susceptibility to rheumatic fever: the missing link. Circulation 1984 ;69:197-201. Gibofsky A, Khanna A, Suh E, Zabriskie JB. The genetics of rheumatic fever: relationship to streptococcal infection and autoimmune disease. J Rheumatol 1991;18 (suppl 30):1-13. Zabriskie JB, Lavenchy D, Willians Jr RC, Fu SM, Yeadon CA, Fotino M, Braun DG. Rheumatic fever-associated B cell alloantigens as identified by monoclonal antibodies. Arthritis Rheum 1985;28 (9): 1047-1051. Ayoub EM, Barrett DJ, MacLaren NK, Krischen JP. Association of class II histocompatibility leukocyte antigens with rheumatic fever. J Clin Invest 1986;77: 2019-2026. Anastasiou-Nana M, Anderson JL, Carquist JF, Nanas JN. HLA DR typing and lymphocyte subset evaluation in rheumatic heart disease: a search for immune

response factors. Am Heart J 1986;112 (5):992-997. 21. Rajapakse CNA, Halim K, A1-Orainey L, AI-Nozha M, A1-Aska AK. A genetic marker for rheumatic heart disease. Br Heart J 1987;58:659-662. 22. Jhinghan B, Mehra NK, Reddy KS, Taneja V, Vaidya MC, Bhatia ML. HLA, blood groups and secretor status in patients with established rheumatic fever and rheumatic heart disease. Tissue Antigens 1986;27:172-178. 23. Guilherme L, Weidebach W, Kiss MH, Snitcowsky R, Kalil J. Association of human leukocyte class II antigens with rheumatic fever or rheumatic heart disease in a Brazilian population. Circulation 1991;83: 1995-1998. 24. Weidebach W, Goldberg AC, Chiarella J, Guilherme L, Snitcowsky R, Pileggi F, Kalil J. HLA class II antigens in Rheumatic Fever: analysis of the DR locus by RFLP and Oligotyping. Hum Immunol 1994;40:253-258. 25. Gu6dez Y, Kotby A, E1-Demellaway M, Galal A, Thomson G, Zaher S, Kassem S, Kotb M. HLA class II associations with rheumatic heart disease are more evident and consistent among clinically homogeneous patients. Circulation 1999;99:2784-2790. 26. Visentainer JE, Pereira FC, Dalalio MM, Tsuneto, LT, Donadio PR, Moliterno RA. Association of HLA-DR7 with rheumatic fever in the Brazilian population. J Rheumato1200027(6): 1518-20. 27. Maharaj B, Hammond MG, Appadoo B, Leary WP, Pudifin DJ. HLA-A, B, DR and DQ antigens in black patients with severe chronic rheumatic heart disease, Circulation 1987;76:259-261. 28. Olmez U, Turgay M, Ozenirler S, Tutkak H, Duzgun N, Duman M, Tokgoz G. Association of HLA class I and class II antigens with rheumatic fever in a Turkish population. Rheumatol 1992;22 (2):49-52. 29. Donadi EA, Smith AG, Louzada-Junior P, Voltarelli JC, Nepom GT. HLA class I and class II profiles of patients presenting RF with Sydenham's chorea. J Neurol 2000;247 (2): 122-128. 30. Goldberg AC, Kalil J, Kotb Met al. HLA and rheumatic fever: 12th International Histocompatibility Workshop study. In: Charron D, ed. HLA: Genectic Diversity of HLA; Functional and Medical Implication. EDK, 1997;413-418. 31. Stollerman GH. Rheumatogenic streptococci and autoimmunity. Clin Immunol Immunopathol 1991;61: 113-142. 32. Calveti PA. Autoantibodies in rheumatic fever. Proc Soc Exp Biol Med 1945;60:379-381. 33. Kaplan MH, Suchy ML. Immunological relation of streptococcal and tissue antigens II. Cross-reactions antisera to mammalian heart tissue with cell wall constituent of certain strains of group A streptococci. J Exp Med 1964;119:643-650.

329

34. Kaplan MH, Svec KH. Immunologic relation of streptococcal antibody cross-reactive with heart tissue: Association with streptococcal infection, rheumatic fever and glomerulonephritis. J Exp Med 1964;119: 651-666. 35. Cunningham MW. Pathogenesis of group A streptococcal infections. Clin Microbiol Rev 2000;470-511. 36. Roberts S, Kosanke S, Dunn TS, Jankelow D, Duran CMG, Cunningham MW. Pathogenic Mechanism in Rheumatic Carditis: Focus on Valvular Endothelium. J Infect Diseases 2001 ;183:507-511. 37. Bathia R, Narula J, Reddy KS, Koicha M, Malaviya AN, Pothineni RB, Tandon R, Bathia ML. Lymphocyte subsets in acute rheumatic fever and rheumatic heart disease. Clin Cardiol 1989; 12:34-38. 38. Bhatnagar PK, Nijhawan R, Prakash K. T cell subsets in acute rheumatic fever, rheumatic heart disease and acute glomerulonephritis cases. Immunol Letters 1987;15:217-219. 39. Mclaughin JF, Paterson PY, Hartz RS, Embury SH. Rheumatic carditis: in vitro responses of peripheral blood leukocytes to heart and streptococcal antigens. Arthritis Rheum 1972;15:60(O608. 40. Read SE, Fischetti VA, Utermohlen V, Falk RE, Zabriskie JB. Cellular reactivity studies to streptococcal antigens. Migration inhibition studies in patients with streptococcal infections and rheumatic fever. J Clin Invest 1974;54:439--450. 41. Meric N, Berkel AI. Cellular immunity in children with acute rheumatic fever and rheumatic carditis. Pediatr Res 1979;13:16-20. 42. Read SE, Reid HFM, Fischetti VA, Poon-King T, Ramkissoon R, McDowell M, Zabriskie JB. Serial studies on the cellular immune response to streptococcal antigens in acute and convalescent rheumatic fever patients in Trinidad. J Clin Immunol 1986;6(6):433-441. 43. Gray ED, Wannamaker LW, Ayoub E, E1 Kholy A. Cellular immune responses to extracellular streptococcal products in rheumatic heart disease. J Clin Invest 1981;68:665-671. 44. Gross WL, Schlaak M. Modulation of Human lymphocyte functions by group A streptococci. Clin Immunol Immunopathol 1984;32:234-247. 45. Dale JB, Simpson WA, Ofek I, Beachey E. Blastogenic responses of human lymphocytes to structurally defined polypeptide fragments of streptococcal M protein. J Immunol 1981;126:1499-1505. 46. Kotb M, Courtney HS, Dale JB, Beachey EH. Cellular and biochemical responses of human T lymphocytes stimulated with streptococcal M proteins. J Immunol 1989;142:966-970. 47. Raizada V, Williams RC Jr, Chopra P, Gopinath N, Prakash K, Sharma KB, Cherian KM, Panday S, Arora

330

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

R, Nigam M, Zabriskie JB, Husby G. Tissue distribution of lymphocytes in rheumatic heart valves as defined by monoclonal anti-T cells antibodies. Am J Med 1983;74: 225-237. Yoshinaga M, Figueiroa F, Wahid MR, Marcus RH, Suh E, Zabriskie JB. Antigenic specificity of lymphocytes isolated from valvular specimens of rheumatic fever patients. J Autoimmun 1995;8:601-613. Cunningham MW, Antone SM, Smart M, Liu R, Kosanke S. Molecular analysis of human cardiac myosin-cross-reactive B and T-cell epitopes of the group A streptococcal M5 protein. Infect Immun 1997 ;65(9):3913-3923. Guilherme L, Oshiro SE, Fae KC, Cunha-Neto E, Renesto G, Goldberg AC, Tanaka AC, Pomerantzeff P, Kiss MH, Silva C, Guzman F, Patarroyo ME, Southwood S, Sette A, Kalil J. T cell reactivty against streptococcal antigens in the periphery mirrors reactivity of heart infiltrating T lymphocytes in rheumatic heart disease patients. Infect Immunity 2001;69(9):5345-535. Manjula BN, Acharya AS, Mische MS, Fairwell T, Fischetti VA. The complete amino acid sequence of a biologically active 197-residue fragment of M protein isolated from type 5 group A streptococci. J Biol Chem 1984 ;259:3686-3693. Miller LC, Gray ED, Beachey EH, Kehoe MA. Antigenic variation among group A streptococcal M proteins: nucleotide sequence of the serotype 5 M protein gene and its relationship with genes encoding types 6 and 24 M proteins. J Biol Chem 1988;263:5668-5673. Robinson JH, Atherton MC, Goodacre J,~, Pinkney M, Weightman H, Kehoe MA. Mapping T-cell epitopes in group A streptococcal type 5 M protein. Infect Immun 1991 ;59(12):4324-4331. E1-Demellawy M, E1-Ridi R, Guirguis NI, Alim MA, Kotby A, Kotb M. Preferential recognition of human myocardial antigens by T lymphocytes from rheumatic heart disease patients. Infect Immun 1997;65(6):21972205. Kotb M, Majumdar G, Tomai M, Beachey EH. Accessory cell-independent stimulation of human T cells by streptococcal M protein superantigen. J Immunol 1990; 145:1332-1336. Tomai M, Kotb M, Majumdar G, Beachey EH. Superantigenicity of streptococcal M protein. J Exp Med 1990; 172:359-362. Tomai MA, Schlievert PM, Kotb M. Distinct T-cell receptor VI3 gene usage by human T lymphocytes stimulated with the streptococcal pyrogenic exotoxins and pep M5 protein. Infect Immun 1992;60:701-705. Watanabe-Ohnishi R, Aelion J, Legros L, Tomai MA, Sokurenko EV, Newton D, Takahara J, Irino S, Rashed S, Kotb M. Characterization of unique human TCR Vb

59.

60.

61.

62.

63.

64.

65.

66.

67.

68.

69.

specificities for a family of streptococcal superantigens represented by rheumatogenic serotypes of M protein. J Immunol 1994; 152:2066-2073. Tomai MA, Aelion J, Dockter ME, Majumdar G, Spinella DG, Kotb M. T cell receptor V gene usage by human T cells stimulated with the superantigen streptococcal M protein. J Exp Med 1991;174:285-288. Wang B, Schlievert PM, Gaber AO, Kotb M. Localization of an immunologically functional region of the streptococcal superantigen pepsin-extracted fragment of type 5 M protein. J Immunol 1993; 151:1419. Degnan B, Taylor J, hawkes C, O'Shea U, Smith J, Robinson JH, Kehoe MA, Boylston A, Goodacre J,~. Streptococcus pyogenes type 5M protein is an antigen, not a superantigen, for human T cells. Hum Immunol 1997;53:206--215. Fleischer B, Schmidt KH, Gedach D, Kohler W. Separation of T cell stimulating activity from streptococcal M protein. Infect Immun 1992; 1:767-772. Li PLL, Tiedemann RE, Moffat LS, Fraser JD. The superantigen streptococcal pyrogenic exotoxin C (SPE-C) exhibits a novel mode of action. J Exp Med 1997; 186(3):375-391. Guilherme L, Dulphy N, Douay C, Coelho V, CunhaNeto E, Oshiro SE, Assis RV, Tanaka AC, Pomerantzeff PM, Chan'on D, Toubert A, Kalil J. Molecular evidence for antigen-driven immune responses in cardiac lesions of rheumatic heart disease patients. Int Immunol 2000; 12:1063-1074. Guilherme L, Cunha-Neto E, Tanaka AC, Dulphy N, Toubert A, Kalil J. Heart-directed autoimmunity: the case of rheumatic fever. J Autoimmun 2001; 16: 363-367. Figueroa F, Gonzalez M, Carrion F, Lobos C, Turner F, Lasagna N, Valdes E Restriction in the usage of variable beta regions in T-cells infiltrating valvular tissue from rheumatic heart disease patients. J Autoirnmun 2002;19:233-240. Fa6 K, Kalil J, Toubert A, Guilherme L. Heart infiltrating T-cell clones from a rheumatic heart disease patient display a common TCR usage and a Degenerate antigen recognition pattern. Mol Immunol 2004;40(14), in press. Miller LC, Gray ED, Mansour M, Abdin ZH, Kamel R, Zaher S, Regelmann WE. Cytokines and immunoglobulin in rheumatic heart disease: production by blood and tonsillar mononuclear cells. J Rheumatol 1989; 16: 1436--1442. Morris K, Mohan C, Wahi PL, Anand IS, Ganguly NK. Enhancement of IL-1, 11-2 production and 11-2 receptor generation in patients with acute rheumatic fever and active rheumatic heart disease: a prospective study. Clin

Exp Immunol 1993;91:429-436. 70. Narin N, Ktittikqtiler N, Ozytirek R, Bakiler AR, Parlar A, Arcasoy M. Lymphocyte subsets and plasma IL-1 ix, IL-2, and TNF-oc concentrations in acute rheumatic fever and chronic rheumatic heart disease. Clin Immunol Immunopathol 1995;77:172-176. 71. Samsonov MY, Tilz GP, Pisklakov VP, Reibnegger G, Nassonov EL, Nassonova VA, Wachter H, Fuchs D. Serum-soluble receptors for tumor necrosis factor-o~ and interleukin-2 and neopterin in acute rheumatic fever. Clin Immunol Immunopathol 1995;74:31-34. 72. Yegin O, Coskun M, Ertug H. Cytokines in acute rheumatic fever. Eur J Pediatr 1997;156:25-29. 73. Kemeny E, Grieve T, Marcus R, Sareli P, Zabriskie JB. Identification of mononuclear cells and T cell subsets in rheumatic valvulitis. Clin Immunol Immunopathol 1989;52:225-237. 74. Chopra P, Narula J, Kumar SA, Sachdeva S, Bathia ML. Immunohistochemical characterization of Aschoff nodules and endomyocardial inflammatory infiltrates in left atrial appendages from patients with chronic rheumatic heart disease. Int J Cardiol 1988;20:99-105. 75. Fraser WJ, Haffejee Z, Jankelow D, Wadee A, Cooper K. Rheumatic Aschoff nodules revisited. II. Cytokine expression corroborates recently proposed sequential stages. Histopathology 1997;31:460-464. 76. Williams RC Jr. Host factors in rheumatic fever and heart disease. Hosp Practice 1982; 125-138. 77. Dos Reis GA, Gaspar MIC, Barcinski MA. Immune recognition in the streptococcal carditis of mice: the role of macrophages in the generation of heart-reactive lymphocytes. J Immunol 1982; 128:1514-1521. 78. Huber AS, Cunningham MW. Streptococcal M protein peptide with similarity to myosin induces CD4+ T celldependent myocarditis in MRL/++ mice and induces partial tolerance against coxsackieviral myocarditis. J Immunol 1996; 156:3528-3534. 79. Cunningham MW. Molecular mimicry between group A streptococci and myosin in the pathogenesis of acute rheumatic fever. In: Jagat Narula, Renu Virmani, K Srinath Reddy and Rajendra Tandon, eds. Rheumatic Fever. Eds American Registry of Pathology, 1999;135165. 80. Quinn A, Kosanke SD, Fischetti VA, Factor SM, Cunningham M. Induction of autoimmune valvular heart disease by recombinant streptococcal M protein. Infect Immun 2001, 69:4072-4078. 81. Galvin JE, Hemric ME, Kosanke SD, Factor SM, Quinn A and Cunningham M. Induction of myocarditis and valvulitis in Lewis rats by different epitopes of cardiac myosin and its implications in rheumatic carditis. Am J Patho12002; 160 (1):297-306.

331

Published by Elsevier B. V 2004. Infection and Autoimmunity Y. Shoenfeld and N.R. Rose, editors

Pediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal Infections (PANDAS) Susan E. Swedo, Lisa A. Snider and Marjorie A. Garvey

Pediatrics & Developmental Neuropsychiatry Branch, National Institute of Mental Health, National Institutes of Health, Department of Health and Human Services, Bethesda, MD, USA

1. INTRODUCTION The relationship between obsessive-compulsive symptoms and Sydenham's chorea (SC) was first noted by Sir William Osier when he described "bizarre" and "perseverative behaviors" of children with "chorea minor" [ 1]. Three subsequent clinical reports noted an association between SC and OCD among children with rheumatic chorea, and adult psychiatric patients with a history of the disorder [24]. A series of studies conducted at the National Institute of Mental Health (NIMH) demonstrated that obsessive-compulsive symptoms were more problematic for children with SC, than for those with rheumatic carditis [5]; and that obsessions and compulsions affected more than 70% of the children in the weeks surrounding the onset of their chorea [6, 7]. The NIMH findings were subsequently replicated and extended by Asbahr and colleagues in Sao Paulo, Brazil, who demonstrated that not only does OCD affect approximately 2/3's of children presenting with an initial episode of SC, but also that the frequency of OCD increases with repeated bouts of chorea, affecting 100% of children experiencing three or more recrudescences [8]. The obsessive-compulsive symptoms in SC are indistinguishable from those of children with primary OCD, and include contamination fears, fear of harm coming to self or others, doubting, symmetry concerns, and other obsessional worries, as well as compulsive washing, checking, ordering, arranging, and hoarding rituals. These clinical similarities, and observations that the OCD in SC begins several weeks prior to the

manifestation of the adventitious movements, led to speculation that post-streptococcal obsessive-compulsive symptoms might occur in the absence of frank chorea [5, 6]. Longitudinal observations of a large cohort of pediatric patients with OCD [9, 10] provided support for this postulate, as a subgroup of patients had an acute symptom onset, an episodic course characterized by periods of complete symptom remission interrupted by abrupt, dramatic symptom exacerbations, and a close temporal relationship between these relapses and preceding Group A beta-hemolytic streptococcal (GAS) infections, (scarlet fever or streptococcal pharyngitis) [11, 12]. The subgroup was labeled with the acronym, "PANDAS", to indicate the shared clinical and presumed etiopathogenic features: Pediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal infections [12]. In this chapter, we will review the clinical features of the PANDAS subgroup, and explore the factors postulated to participate in etiopathogenesis.

2. CLINICAL FEATURES OF THE PANDAS SUBGROUP The clinical features of the first 50 children meeting criteria for the PANDAS subgroup were published in 1998 [12]. Those criteria are: 1) The presence of a tic disorder and/or obsessive-compulsive disorder. 2) Prepubertal age at onset, usually between 3 and 12 years of age.

333

3) Abrupt symptom onset and/or episodic course of symptom severity. 4) Temporal association between symptom exacerbations and streptococcal infections. 5) Presence of neurological abnormalities during periods of symptom exacerbation. Several unique characteristics of the PANDAS subgroup become apparent when the children are compared to unselected patients with childhoodonset OCD and tic disorders [12-16]. The average age at symptom onset in the PANDAS subgroup is nearly three years younger than that previously reported for childhood-onset OCD [9, 17] and up to two years younger than the average age of onset for tic disorders [ 18]. Further, comparisons of the age and sex-distribution of the PANDAS subgroup with that of other OCD patient groups suggests a bimodal distribution [9, 12], (consistent with the postulate that the PANDAS subgroup is truly distinct from other patient groups); however, this cannot be confirmed without large-scale commumunity-based epidemiologic investigations, or the demonstration of a unique etiopathogenesis for the PANDAS subgroup. The clinical course of the PANDAS subgroup differs markedly from that of other OCD patients [9, 12]. Symptom exacerbations in the PANDAS subgroup are sudden and severe, with parents describing the onset of symptoms as occurring "overnight" or "out of the blue." The symptoms remain at peak severity for a period of several weeks or longer, and then gradually subside in severity, often remitting completely, with patients remaining asymptomatic until they're infected again with GAS. This relapsing-remitting course is in striking contrast to the gradual onset and persistent symptoms typically seen in childhood-onset OCD [9, 10] and also differs substantially from the waxing and waning course of tic disorders [18]. Emotional lability, attentional difficulties, separation anxiety, and motoric hyperactivity frequently accompany the OCD/tics exacerbations in the PANDAS subgroup [12]; these clinical features are shared with Sydenham's chorea. Enuresis and daytime urinary frequency are also common [ 12, 16]. Deteriorations in handwriting also have been noted during the symptom exacerbations in the PANDAS subgroup, and may prove useful as an objective means of tracking

334

symptom severity [ 19]. Although the etiology of the handwriting changes isn't known, they appear to parallel the appearance of choreiform movements of the hands and fingers. The presence of choreiform movements during neuropsychiatric symptom exacerbations may prove to be one of the most reliable means of identifying children in the PANDAS subgroup [20]. These mild adventitious movements can be elicited during structured neurological examinations, such as the PANESS [21], and were found to be present in 25 of 26 children in the original cohort who were examined during an exacerbation [12]. These findings were replicated recently in a sample of 17 children in the PANDAS subgroup, all of whom demonstrated choreiform movements on the PANESS examination performed during a symptom exacerbation. The choreiform movements are thought to arise from dysfunction of the basal ganglia of the brain, particularly within the caudate nucleus and putamen. These structures are also implicated in OCD, where symptoms are postulated to result from dysfunction of the corticostriato-thalamocortical circuitry [22]. In Sydenham's chorea, functional imaging studies provide evidence of basal ganglia dysfunction during acute chorea [23, 24], and volumetric abnormalities of the caudate, putamen, and globus pallidus were demonstrated in a cohort of 24 children with Sydenham chorea through the use of structural MRI scans [25]. A volumetric MRI study of 34 children in the PANDAS subgroup also revealed enlargements of the caudate, putamen, and globus pallidus [26]. In some patients, the size of the basal ganglia structures has been found to normalize following successful immunomodulatory therapy with IVIG or plasma exchange (example shown in Fig. 1) [27].

3. MODEL OF ETIOPATHOGENESIS

The etiology of the neuropsychiatric symptoms in the PANDAS subgroup is postulated to be similar to that of Sydenham's chorea, the neurologic manifestation of rheumatic fever. Thus, the etiopathogenesis is hypothesized to occur when "rheumatogenic" GAS bacteria infect a susceptible host and induce an abnormal immune response. As shown in Fig. 2, the

Figure 2. MRI scans of a 14-year old male in the PANDAS subgroup pre-/post-treatment.

proposed model not only provides a framework for understanding the etiology of OCD and tic disor-

ders, but also for the development of novel intervention and prevention strategies.

335

4. THE ROLE OF S T R E P T O C O C C A L INFECTIONS IN PANDAS For rheumatic fever, the etiologic role of GAS infections was demonstrated indirectly, through three lines of research: 1) epidemiologic investigations which demonstrated a close temporal relationship between scarlet fever epidemics and subsequent outbreaks of rheumatic fever; 2) the prevention of rheumatic fever recrudescences by penicillin prophylaxis, and 3) demonstration of declining rates of rheumatic fever following the widespread application of antibiotic treatment for GAS pharyngitis [28]. Coburn [29] and Collis [30] are credited with establishing the relationship between GAS infections and rheumatic fever by demonstrating a temporal relationship between epidemics of streptococcal infections (scarlet fever and strep, pharyngitis), and subsequent outbreaks of rheumatic fever. Coburn and Young [31] extended these findings by demonstrating that each time the incidence of scarlet fever increased, it was followed three weeks later by a rise in the rate of rheumatic fever cases. Although epidemiological studies are not usually sufficient to establish causality, the clarity of the relationship in these investigations has been accepted as evidence that GAS infections are the etiologic trigger in rheumatic fever [32]. The associative strategy employed for rheumatic fever can be applied to the question of the relationship between GAS infections and OCD/tic symptoms in the PANDAS subgroup. By demonstrating that each neuropsychiatric symptom exacerbations occurs concurrently with, or is preceded by a GAS infection, it will be possible to demonstrate a temporal association, and possibly, evidence of causality. The PANDAS subgroup is defined by a "temporal association between neuropsychiatric symptom exacerbations and GAS infections" [12]. The "gold standard" for demonstrating such a temporal association is prospective, longitudinal assessments. As shown in Fig. 3, these prospective observations can be quite effective in demonstrating differences between a prototypical child in the PANDAS subgroup (A) and one whose symptoms show no temporal relationship to GAS infections (B). Systematic longitudinal studies are required to establish this relationship, and several investigative groups have such prospective studies underway.

336

One caveat in evaluating the relationship between streptococcal infections and neuropsychiattic symptoms is that the disorders are so common that co-occurrence can be a random coincidence, rather than a clinically significant finding. Obsessive-compulsive disorder occurs in 1-2% of schoolage children, and transient motor tics in as many as 10-25% of early elementary students [33, 34]. Further, during regional streptococcal epidemics, the majority of children will be infected at least once during the outbreak [35]. Thus, a single positive throat culture or elevated antistreptococcal antibody titer is not sufficient to determine that a child's neuropsychiatric symptoms are associated with streptococcal infections [12, 34, 36]. The determination that a child fits the PANDAS profile is made through prospective evaluation and documentation of the presence of streptococcal infections in conjunction with at least two episodes of neuropsychiatric symptoms, as well as demonstrating negative throat culture or stable titers during times of neuropsychiatric symptom remission [15]. A child who has multiple symptom exacerbations without evidence of streptococcal infection would not be considered as part of the PANDAS subgroup, nor would a child who has numerous streptococcal infections without subsequent symptom exacerbations. The reduction of rheumatic fever (RF) recurrences by antibiotic prophylaxis against GAS infections was a key factor in determining that GAS played an etiologic role in RF. This was particularly true for Sydenham's chorea, in which evidence of an inciting GAS infection was often unobtainable [28]. Antibiotic prophylaxis not only prevented recrudescences, but also improved the long-term prognosis of RF sufferers, by preventing additional scarring of the cardiac valves [37]. The same goal may apply to OCD. A recent report from Sao Paulo, Brazil demonstrated that the frequency and severity of obsessive--compulsive symptoms increased with repeated bouts of Sydenham's chorea [8]. During the initial choreic episode, approximately 65% of the patients had obsessive--compulsive symptoms, which were reported to be "mild" and non-impairing. If the child had two or more recrudescences, the risk of OCD increased to 100%, and all children reported clinically significant symptomatology [8]. If this pattern applies to the PANDAS subgroup, then secondary prophylaxis against GAS infections

A: P A N D A S

30 =L _

25

-=|

20

800

li---=---I

__/." \ 15-

600

.__--',

4,.I ,m.

400 P O

E o 10

~

200
12% of DRpositive cells reactive) in over 90% of rheumatic fever patients, but only 5-10% of healthy controls [43, 44]. Given the large separation between patients and controls, and the knowledge that rheumatic fever affects only 5-7% of the world's population, it was postulated that the D8/17 marker could serve as a trait marker of rheumatic fever susceptibility. Pilot data from the NIMH cohort of patients with SC was confirmatory, with over 90% of patients, and only 10% of age-/sex-matched controls identified as D8/17 positive [6]. Similar results were obtained in the PANDAS subgroup (85% positive among 27 patients vs. 17% in 24 controls; OR 28.8, p

Infection and Autoimmunity

Cancer and Autoimmunity (Autoanitbody Library)

Atherosclerosis and Autoimmunity

Autoimmunity: Methods and Protocols

Molecular Autoimmunity

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Hcv Infection and Cryoglobulinemia

Aging, Immunity, and Infection

Aging, Immunity, and Infection

Infection: Microbiology and Management

Infection and Autoimmunity

Cancer and Autoimmunity (Autoanitbody Library)

Atherosclerosis and Autoimmunity

Autoimmunity: Methods and Protocols

Molecular Autoimmunity

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Infection

Hcv Infection and Cryoglobulinemia

Aging, Immunity, and Infection

Aging, Immunity, and Infection

Infection: Microbiology and Management

Recommend Documents