Mouse Models for Drug Discovery: Methods and Protocols (Methods in Molecular Biology Vol 602)

ME T H O D S

IN

MO L E C U L A R BI O L O G Y

Series Editor John M. Walker School of Life Sciences University of Hertfordshire Hatfield, Hertfordshire, AL10 9AB, UK

For other titles published in this series, go to www.springer.com/series/7651

TM

Mouse Models for Drug Discovery Methods and Protocols

Edited by

Gabriele Proetzel and Michael V. Wiles The Jackson Laboratory, Bar Harbor, ME, USA

Editors Gabriele Proetzel The Jackson Laboratory 610 Main Street Bar Harbor ME 04609 USA [email protected]

Michael V. Wiles The Jackson Laboratory 600 Main Street Bar Harbor ME 04609 USA [email protected]

ISSN 1064-3745 e-ISSN 1940-6029 ISBN 978-1-60761-057-1 e-ISBN 978-1-60761-058-8 DOI 10.1007/978-1-60761-058-8 Library of Congress Control Number: 2009939644 © Humana Press, a part of Springer Science+Business Media, LLC 2010 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Humana Press, c/o Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. While the advice and information in this book are believed to be true and accurate at the date of going to press, neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect to the material contained herein. Printed on acid-free paper springer.com

Preface The drug discovery process has shifted from essentially a trial and error approach toward unraveling disease and underlying molecular mechanisms with the goal to specifically target pathways and molecules. A key for understanding and exploring disease mechanisms is the availability of good models, preferably in vivo models as these more closely reflect the complexity of life. Each organism is a highly complex integrated system and is considerably more than the sum of its parts being the net result of an evolving genome and its interactions with the environment. This volume, Mouse Models for Drug Discovery: Methods and Protocols, attempts to illustrate The Mouse as an exceptionally versatile and sophisticated platform which can meet this challenge. With the development of inbred strains which are genetically invariant within a strain, it has become possible to use genetically defined animals and highly reproducible systems. This has allowed The Mouse to rise from being a pest, to a cute collectable pet, to an advanced and well-established tool for genetic and molecular research, becoming the premium instrument for drug discovery, validation, preclinical, and toxicological studies. The reasons for the mouse’s rise in prominence are many. In brief, mice are small, require relatively little space, have simple nutritional needs, a short generation time, and few special needs to reproduce. However, it is in the last 20 years with the advent of genetic engineering and the ease of manipulating the mouse’s genome that has made the mouse the most versatile mammalian experimental system. Further, the sequencing of the human and mouse genomes has clearly demonstrated that mouse and humans have direct gene and functional homologies for more than 90 % of their genes. With the KOMP and other international programs, all mouse genes will be available as null alleles (i.e., knockouts) by approximately 2015. Development and distribution of these resources is aided by advances in Assisted Reproductive Technologies. With all of this in mind, it can now be truly stated that The Mouse has become a respected, indispensable tool in biomedical research and is the most commonly used animal research model. With thousands of mouse models available covering practically all disease areas, it is beyond this volume to cover the whole field of mouse applications in the drug discovery process. In this volume, we have selected chapters which cover general background as well as a few specific disease topics with the idea of introducing those less familiar with mice as experimental model platforms. The chapters by Festing and Wiles cover general aspects of experimental design, inbred vs. outbred mice, and how to manage the risks working with live animals. The chapter by McNeish highlights a pharma approach as how to use genetically engineered mouse models for target identification and validation. Koentgen et al. has given a general overview of the many approaches used to genetically engineer mice. Rando et al. show the power of combining novel imaging tools with genetically engineered mice for drug discovery. Representatives of the young and rapidly developing field of humanized mice are provided in the chapters by Roopenian and Shultz, highlighting also the possibility of engraftment of human tissue into mouse, for example, in regenerative biology and stem cell research.

v

vi

Preface

As examples for the wide field of specific disease areas and mouse models, we have included type 1 and 2 diabetes (Serreze and Baribault), cardiovascular disease (Howles), arthritis (Tak), skin disorders (Sundberg), cancer (Talmadge, Surguladze, and Li), the use of behavioral models for depression and anxiety (Kalueff), neurodegenerative diseases (Janus), neuromuscular diseases (Burgess), and infectious diseases (Medina). We hope that this volume will stimulate those not familiar with the power of the mouse and its potential for the drug discovery process, and further, that it will encourage the development of new models as well as new ways in utilizing existing models. We also hope to promote the development of more standardized models and assays such that results can be more easily compared and reproduced. We like to thank all the contributors, their discussions, and their patience in making this an important volume on mouse models and drug discovery. Gabriele Proetzel and Michael V. Wiles

Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

v

Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

ix

1.

Improving Toxicity Screening and Drug Development by Using Genetically Defined Strains . . . . . . . . . . . . . . . . . . . . . . . Michael F.W. Festing

1

2.

The Sophisticated Mouse: Protecting a Precious Reagent . . . . . . . . . . . . . Michael V. Wiles and Rob A. Taft

23

3.

Genetically Engineered Mouse Models in Drug Discovery Research . . . . . . . Rosalba Sacca, Sandra J. Engle, Wenning Qin, Jeffrey L. Stock, and John D. McNeish

37

4.

Engineering the Mouse Genome to Model Human Disease for Drug Discovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Frank Koentgen, Gabriele Suess, and Dieter Naf

55

Profiling of Drug Action Using Reporter Mice and Molecular Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Gianpaolo Rando, Andrea Biserni, Paolo Ciana, and Adriana Maggi

79

Human FcRn Transgenic Mice for Pharmacokinetic Evaluation of Therapeutic Antibodies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Derry C. Roopenian, Gregory J. Christianson, and Thomas J. Sproule

93

5.

6.

7.

Development of Novel Major Histocompatibility Complex Class I and Class II-Deficient NOD-SCID IL2R Gamma Chain Knockout Mice for Modeling Human Xenogeneic Graft-Versus-Host Disease . . . . . . . . . . . . 105 Steve Pino, Michael A. Brehm, Laurence Covassin-Barberis, Marie King, Bruce Gott, Thomas H. Chase, Jennifer Wagner, Lisa Burzenski, Oded Foreman, Dale L. Greiner, and Leonard D. Shultz

8.

Bridging Mice to Men: Using HLA Transgenic Mice to Enhance the Future Prediction and Prevention of Autoimmune Type 1 Diabetes in Humans . 119 David V. Serreze, Marijke Niens, John Kulik, and Teresa P. DiLorenzo

9.

Mouse Models of Type II Diabetes Mellitus in Drug Discovery . . . . . . . . . 135 Helene Baribault

10.

Cholesterol Absorption and Metabolism . . . . . . . . . . . . . . . . . . . . . 157 Philip N. Howles

11.

Collagen-Induced Arthritis in Mice . . . . . . . . . . . . . . . . . . . . . . . . 181 Lisette Bevaart, Margriet J. Vervoordeldonk, and Paul P. Tak

12.

Skin Diseases in Laboratory Mice: Approaches to Drug Target Identification and Efficacy Screening . . . . . . . . . . . . . . . . . . . . . . . 193 John P. Sundberg, Kathleen A. Silva, Caroline McPhee, and Lloyd E. King

vii

viii

Contents

13.

Models of Metastasis in Drug Discovery . . . . . . . . . . . . . . . . . . . . . 215 James E. Talmadge

14.

Methods for Evaluating Effects of an Irinotecan + 5Fluorouracil/Leucovorin (IFL) Regimen in an Orthotopic Metastatic Colorectal Cancer Model Utilizing In Vivo Bioluminescence Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235 David Surguladze, Philipp Steiner, Marie Prewett, and James R. Tonra

15.

CML Mouse Model in Translational Research . . . . . . . . . . . . . . . . . . 253 Cong Peng and Shaoguang Li

16.

Mouse Models for Studying Depression-Like States and Antidepressant Drugs . . 267 Carisa L. Bergner, Amanda N. Smolinsky, Peter C. Hart, Brett D. Dufour, Rupert J. Egan, Justin L. LaPorte, and Allan V. Kalueff

17.

Virus-Delivered RNA Interference in Mouse Brain to Study Addiction-Related Behaviors . . . . . . . . . . . . . . . . . . . . . . . . . . . 283 Amy W. Lasek and Nourredine Azouaou

18.

Experimental Models of Anxiety for Drug Discovery and Brain Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299 Peter C. Hart, Carisa L. Bergner, Amanda N. Smolinsky, Brett D. Dufour, Rupert J. Egan, Justin L. LaPorte, and Allan V. Kalueff

19.

Mouse Models of Neurodegenerative Diseases: Criteria and General Methodology 323 Christopher Janus and Hans Welzl

20.

Neuromuscular Disease Models and Analysis . . . . . . . . . . . . . . . . . . . 347 Robert W. Burgess, Gregory A. Cox, and Kevin L. Seburn

21.

Murine Model of Cutaneous Infection with Streptococcus pyogenes . . . . . . . . 395 Eva Medina

22.

Murine Model of Pneumococcal Pneumonia . . . . . . . . . . . . . . . . . . . 405 Eva Medina

23.

Murine Model of Polymicrobial Septic Peritonitis Using Cecal Ligation and Puncture (CLP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 411 Eva Medina

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 417

Contributors NOURREDINE AZOUAOU • The Ernest Gallo Clinic and Research Center, University of California at San Francisco, Emeryville, CA, USA HELENE BARIBAULT • Department of Metabolic Disorders, Amgen, South San Francisco, CA, USA CARISA L. BERGNER • Department of Physiology and Biophysics, Georgetown University Medical School, Washington, DC, USA LISETTE BEVAART • Division of Clinical Immunology and Rheumatology, Academic Medical Center/University of Amsterdam, Amsterdam, The Netherlands; Arthrogen B.V., Amsterdam, The Netherlands ANDREA BISERNI • TOP s.r.l., Lodi, Italy MICHAEL A. BREHM • Department of Medicine, The University of Massachusetts Medical School, Worcester, MA, USA ROBERT W. BURGESS • The Jackson Laboratory, Bar Harbor, ME, USA LISA BURZENSKI • The Jackson Laboratory, Bar Harbor, ME, USA THOMAS H. CHASE • The Jackson Laboratory, Bar Harbor, ME, USA GREGORY J. CHRISTIANSON • The Jackson Laboratory, Bar Harbor, ME, USA PAOLO CIANA • Center of Excellence on Neurodegenerative Diseases and Department of Pharmacological Sciences University of Milan, Milan, Italy LAURENCE COVASSIN-BARBERIS • Department of Medicine, The University of Massachusetts Medical School, Worcester, MA, USA GREGORY A. COX • The Jackson Laboratory, Bar Harbor, ME, USA TERESA P. DILORENZO • Department of Microbiology & Immunology and Department of Medicine, Division of Endocrinology, Albert Einstein College of Medicine, Bronx, NY, USA BRETT D. DUFOUR • Department of Animal Sciences, Purdue University, West Lafayette, IN, USA RUPERT J. EGAN • Department of Physiology and Biophysics, Georgetown University Medical School, Washington, DC, USA SANDRA J. ENGLE • Genetically Modified Models Center of Emphasis, Pfizer Global Research and Development, Pfizer Inc., Groton, CT, USA MICHAEL F.W. FESTING • Understanding Animal Research, London, UK ODED FOREMAN • The Jackson Laboratory, Bar Harbor, ME, USA

ix

x

Contributors

DALE L. GREINER • Department of Medicine, The University of Massachusetts Medical School, Worcester, MA, USA PETER C. HART • Department of Physiology and Biophysics, Georgetown University Medical School, Washington, DC, USA PHILIP N. HOWLES • Department of Pathology and Laboratory Medicine, Center for Lipid and Arteriosclerosis Studies, Genome Research Institute, University of Cincinnati College of Medicine, Cincinnati, OH, USA CHRISTOPHER JANUS • Department of Neuroscience, Mayo Clinic Jacksonville, Jacksonville, FL, USA ALLAN V. KALUEFF • Department of Physiology and Biophysics, Stress Physiology and Research Center (SPaRC), Georgetown University Medical School, Washington, DC, USA; Department of Pharmacology, Tulane University Medical Center, New Orleans, LA, USA LLOYD E. KING, JR. • The Skin Disease Research Center, Department of Medicine, Division of Dermatology, Vanderbilt Medical Center, Nashville, TN, USA MARIE KING • Department of Medicine, The University of Massachusetts Medical School, Worcester, MA, USA FRANK KOENTGEN • Ozgene Pty. Ltd., Bentley DC, Western Australia, Australia JOHN KULIK • The Jackson Laboratory, Bar Harbor, ME, USA JUSTIN L. LAPORTE • Department of Physiology and Biophysics, Stress Physiology and Research Center (SPaRC), Georgetown University Medical School, Washington, DC, USA AMY W. LASEK • The Ernest Gallo Clinic and Research Center, University of California at San Francisco, Emeryville, CA, USA SHAOGUANG LI • Division of Hematology & Oncology, Department of Medicine, University of Massachusetts Medical School, Worcester, MA, USA ADRIANA MAGGI • Center of Excellence on Neurodegenerative Diseases and Department of Pharmacological Sciences Universit`a degli Studi di Milano, via Balzaretti 9, 20133 Milan, Italy JOHN D. MCNEISH • Regenerative Medicine Unit, Pfizer Global Research and Development, Pfizer Inc., Cambridge, MA, USA CAROLINE MCPHEE • The Jackson Laboratory, Bar Harbor, ME, USA EVA MEDINA • Infection Immunology Research Group, Department of Microbial Pathogenesis, Helmholtz Centre for Infection Research, Braunschweig, Germany DIETER NAF • Ozgene Pty. Ltd., Bentley DC, Western Australia, Australia MARIJKE NIENS • The Jackson Laboratory, Bar Harbor, ME, USA CONG PENG – Division of Hematology & Oncology, Department of Medicine, University of Massachusetts Medical School, Worcester, MA, USA

Contributors

xi

STEVE PINO • Department of Medicine, The University of Massachusetts Medical School, Worcester, MA, USA MARIE PREWETT • Imclone Systems, a wholly-owned subsidiary of Eli Lilly & Company, New York, NY, USA WENNING QIN • Genetically Modified Models Center of Emphasis, Pfizer Global Research and Development, Pfizer Inc., Groton, CT, USA GIANPAOLO RANDO • Center of Excellence on Neurodegenerative Diseases and Department of Pharmacological Sciences University of Milan, Milan, Italy DERRY C. ROOPENIAN • The Jackson Laboratory, Bar Harbor, ME, USA ROSALBA SACCA • Genetically Modified Models Center of Emphasis, Pfizer Global Research and Development, Pfizer Inc., Groton, CT, USA KEVIN L. SEBURN • The Jackson Laboratory, Bar Harbor, ME, USA DAVID V. SERREZE • The Jackson Laboratory, Bar Harbor, ME, USA LEONARD D. SHULTZ • The Jackson Laboratory, Bar Harbor, ME, USA KATHLEEN A. SILVA • The Jackson Laboratory, Bar Harbor, ME, USA AMANDA N. SMOLINSKY • Department of Physiology and Biophysics, Georgetown University Medical School, Washington, DC, USA THOMAS J. SPROULE • The Jackson Laboratory, Bar Harbor, ME, USA PHILIPP STEINER • Imclone Systems, a wholly-owned subsidiary of Eli Lilly & Company, New York, NY, USA JEFFREY L. STOCK • Pfizer Global Research and Development, Pfizer Inc., Genetically Modified Models Center of Emphasis, Groton, CT, USA GABRIELE SUESS • Ozgene Pty. Ltd., Bentley DC, Western Australia, Australia JOHN P. SUNDBERG • The Jackson Laboratory, Bar Harbor, ME, USA; The Skin Disease Research Center, Department of Medicine, Division of Dermatology, Vanderbilt Medical Center, Nashville, TN, USA DAVID SURGULADZE • Imclone Systems, a wholly-owned subsidiary of Eli Lilly & Company, New York, NY, USA ROB A. TAFT • The Jackson Laboratory, Bar Harbor, ME, USA PAUL-PETER TAK • Division of Clinical Immunology and Rheumatology, Academic Medical Center/University of Amsterdam, Amsterdam, The Netherlands; Arthrogen B.V., Amsterdam, The Netherlands JAMES E. TALMADGE • University of Nebraska Medical Center, Omaha, NE, USA JAMES R. TONRA • Imclone Systems, a wholly-owned subsidiary of Eli Lilly & Company, New York, NY, USA

xii

Contributors

MARGRIET J. VERVOORDELDONK • Division of Clinical Immunology and Rheumatology, Academic Medical Center/University of Amsterdam, Amsterdam, The Netherlands; Arthrogen B.V., Amsterdam, The Netherlands JENNIFER WAGNER • The Jackson Laboratory, Bar Harbor, ME, USA HANS WELZL • Division of Neuroanatomy and Behavior, Institute of Anatomy, University of Z¨urich, Z¨urich, Switzerland MICHAEL V. WILES • The Jackson Laboratory, Bar Harbor, ME, USA

Chapter 1 Improving Toxicity Screening and Drug Development by Using Genetically Defined Strains Michael F.W. Festing Abstract According to the US Food and Drugs Administration (Food and Drug Administration (2004) Challenge and opportunity on the critical path to new medical products.) “The inability to better assess and predict product safety leads to failures during clinical development and, occasionally, after marketing”. This increases the cost of new drugs as clinical trials are even more expensive than pre-clinical testing. One relatively easy way of improving toxicity testing is to improve the design of animal experiments. A fundamental principle when designing an experiment is to control all variables except the one of interest: the treatment. Toxicologist and pharmacologists have widely ignored this principle by using genetically heterogeneous “outbred” rats and mice, increasing the chance of false-negative results. By using isogenic (inbred or F1 hybrid, see Note 1) rats and mice instead of outbred stocks the signal/noise ratio and the power of the experiments can be increased at little extra cost whilst using no more animals. Moreover, the power of the experiment can be further increased by using more than one strain, as this reduces the chance of selecting one which is resistant to the test chemical. This can also be done without increasing the total number of animals by using a factorial experimental design, e.g. if the ten outbred animals per treatment group in a 28-day toxicity test were replaced by two animals of each of five strains (still ten animals per treatment group) selected to be as genetically diverse as possible, this would increase the signal/noise ratio and power of the experiment. This would allow safety to be assessed using the most sensitive strain. Toxicologists should also consider making more use of the mouse instead of the rat. They are less costly to maintain, use less test substance, there are many inbred and genetically modified strains, and it is easier to identify gene loci controlling variation in response to xenobiotics in this species. We demonstrate here the advantage of using several inbred strains in two parallel studies of the haematological response to chloramphenicol at six dose levels with CD-1 outbred, or using four inbred strains of mice. Toxicity to the white blood cell lineage was easily detected using the inbred strains but not using the outbred stock, clearly showing the advantage of using the multi-inbred strain approach. Key words: Toxicity testing, pre-clinical development, inbred strain, drug development, factorial experimental designs, statistics, signal/noise ratio.

G. Proetzel, M.V. Wiles (eds.), Mouse Models for Drug Discovery, Methods in Molecular Biology 602, DOI 10.1007/978-1-60761-058-8 1, © Humana Press, a part of Springer Science+Business Media, LLC 2010

1

2

Festing

1. Introduction It is now 50 years since Russell and Burch, in their classical book on The Principles of Humane Experimental Technique (1), suggested that toxicologists should use small numbers of animals of several inbred strains rather than using outbred stocks in toxicological screening. The aim of this chapter is to explain how 1. by using genetically defined (inbred and F1 hybrid) laboratory mice (preferably) or rats, toxicologists and pharmacologists can immediately improve the power of their experiments and reduce false-negative results. 2. by using small numbers of animals of several inbred strains, without increasing the number of animals used, falsenegative results from the chance use of a resistant strain will be reduced. 3. by taking account of genetic variation, toxicologists can apply modern genetic research and develop a better understanding of toxicological mechanisms and the genes in humans and animals associated with susceptibility to toxic and pharmaceutically active compounds. 1.1. Need to Improve Toxicity Screening

There is an urgent need to improve methods of toxicity screening. According to the FDA 2004 “Critical path” white paper “The traditional tools used to assess product safety – animal toxicology and outcomes from human studies – have changed little over many decades and have largely not benefited from recent gains in scientific knowledge. The inability to better assess and predict product safety leads to failures during clinical development and, occasionally, after marketing” (2). According to one study (3) the attrition rate of new chemical entities (excluding “me-too” drugs) is about 96% including 27% rejected for toxicity and 46% rejected due to lack of efficacy. Clinical trials are considerably more expensive than pre-clinical testing. Therefore, if the number of misleading results could be minimised the cost of developing new drugs would be substantially reduced. The FDA “Critical path initiative” (4) and the European Union Innovative Medicines Initiative (IMI) both aim to improve methods for developing new drugs (5), so this is an opportune time to explain how the testing of potential new drugs could be improved and made more cost-effective by a relatively simple change in the type of animals which are used.

1.2. Use of Mice and Rats in Drug Development

When methods of toxicity testing were first developed most biochemical and physiological determinations required large samples of tissue. Traditionally, toxicologists have used the rat for this

Inbred Strains in Toxicity Testing

3

reason. However, the majority assays have now been miniaturised for work on smaller species such as Caenorhabditis elegans and Drosophila and the small size of the mouse is no longer a limitation, except in a few special situations. Many protocols are also available for studying mouse biology such as the 270 standard operating procedures for phenotyping the mouse developed by the EUMORPHIA consortium (6). Many of these can be used directly by toxicologists and pharmacologists when studying the effects of xenobiotics. There is also an extensive literature on the characteristics of the many inbred and genetically modified mouse strains (www.informatics.jax.org and http://jaxmice.jax.org/). Mice are also less expensive to maintain and use less of the test agent than rats. Even the use of techniques such as telemetry is now available for the mouse and new methods of whole body imaging such as SPECT and SPECT/CT (7) actually favour smaller animals, i.e. the balance is changing. According to one anonymous toxicologist in a major drug firm (8): “Although the rat genome is now available, the knowledge about differences between mouse strains, the huge access to mouse knockouts and the lower cost for maintaining mice will make mice preferred as an experimental species. However, there will be situations where it is very difficult to make measurements in mice, and sometimes this can be done in rats instead”.

The National Institute of Environmental Health Sciences (NIEHS) has recognised the value of isogenic mouse strains in identifying genes associated with response to xenobiotics. As part of their Host Susceptibility Program “. . . NTP (National Toxicology Program) scientists will take chemicals identified as toxicants in the research and testing program and evaluate them in multiple genetically diverse isogenic mouse strains to determine which strains are particularly sensitive or insensitive to the chemicals causing toxicity and associated disease” (http://ntp.niehs.nih.gov). They go on to state that “Ultimately, the NTP expects to learn more about the key genes and pathways involved in the toxic response and the etiology of disease mediated by substances in our environment. Such an understanding of genes and environment interactions will lead to more specific and targeted research with testing strategies for the NTP scientists to use for predicting the potential toxicity of substances in our environment and their presumptive risk to humans and disease susceptibility”. However, although there are many advantages in using mice, the main principle discussed here, namely taking control of genetic variation in the test animals, is equally applicable to both rats and mice.

4

Festing

1.3. Strains and Stocks of Rats and Mice

The three major classes of stocks which can be used in biomedical research are the following: 1. Outbred stocks (the term “stock” is used for outbred and “strain” for inbred colonies) are usually closed colonies (i.e. no new genetic material is normally introduced) in which each animal is genetically different and unique (9). These stocks may be designated by such names as Sprague–Dawley rats or Swiss mice, or a designation such as CD-1 or CFW. Although there are internationally recognised nomenclature rules, these are not always used consistently. 2. Inbred strains are like immortal clones of genetically identical individuals. They are produced by at least 20 generations of brother × sister mating with each individual being derived from a single breeding pair in the 20th or a subsequent generation, resulting in isogenicity (i.e. all animals being genetically identical) and homozygosity of practically all genetic loci (>98.6%). They are designated by a code such as LEW rats or C57BL mice. Different sub-strains (i.e. branches of a strain which are, or are presumed to be, slightly different) are indicated by further codes following a forward slash, e.g. C57BL/6J (www.informatics.jax.org). Nomenclature rules for inbred strains are well established and widely used (see www.informatics.jax.org/mgihome/nomen/). 3. Mutant and genetically modified strains either have a spontaneous or induced mutation or the deliberate incorporation of foreign DNA into the genome, sometimes in such a way as to disrupt a gene. This class of stock is of great importance in drug development and genetic research and is likely to be used increasingly in the study of toxic mechanisms, but a full discussion is beyond the scope of this chapter. However, the genetic modification may be maintained on either a genetically heterogeneous or an inbred genetic background. The former is not recommended because the mutation phenotype is often highly dependent upon the genetic background, and this may change rapidly in a heterogeneous stock unless it is maintained in large numbers preferably with a maximum avoidance of inbreeding rotational scheme (10). A brief summary of the properties of inbred strains and outbred stocks is given in Table 1.1. Inbred strains (or F1s made from them) have many advantages as experimental animals when compared with outbred stocks. They are uniform for most characteristics of toxicological and pharmaceutical interest because all animals within each inbred strain are genetically identical. In most cases this leads to less “noise” so either smaller numbers of animals are used or alternatively the experiments will be more powerful with less chance of false-negative results. Inbred strains are genetically much more

Inbred Strains in Toxicity Testing

5

Table 1.1 Brief summary of differences between inbred strains and outbred stocks Inbred strains

Outbred stocks

Isogenic All animals genetically identical (see Note 1) Homozygous Breed true. Parents and offspring genetically identical

Heterogenic Each individual genetically different and unique Heterozygous Do not breed true. May carry recessive genes. Parents and offspring genetically different

Phenotypically uniform Phenotypic variation essentially only due to environmental factors

Phenotypically variable Phenotypic variation is due to both genetic and environmental factors

Identifiable Each individual can be authenticated by genetic markers. Genetic quality control easy

Not identifiable There is no set of markers which can be used to authenticate an outbred stock so genetic quality control of individuals is impossible

Genetically stable Genetic drift slow and only due to mutation

Genetically unstable Genetic drift can be rapid due to changes in gene frequency caused by selection, random drift and mutation

Consistent data Extensive, reliable, data on genotype and phenotype of common strains

Variable data No data on genotype and reliability of data on phenotype questionable because of lack of strain authentication

Multi-strain experiments common. Differences between strains often the starting point for identifying sensitivity genes

Single stock experiments done Most investigators only use a single stock so differences between stocks are not commonly seen and the investigator is unaware of genetic variation in response

Universal Internationally distributed in academic and commercial organisations. Investigators in different countries can use genetically identical animals

Limited Source often limited to one or more commercial breeders. Each colony is unique so investigators in different countries have no access to a particular genotype

Consistent naming Genetic nomenclature well established with extensive lists of individual strains and their properties

Inconsistent naming Genetic nomenclature rules often ignored. No listings of stock characteristics.

stable than outbred stocks. Selective breeding (inadvert or otherwise) is ineffective in altering an inbred strain, whereas it can cause rapid change in an outbred stock. Many of the more widely used inbred strains were exchanged between investigators before they were fully inbred, so there are major sublines such as C57BL/6 and C57BL/10 which differed initially as a result of residual heterozygosity. Further, since then these strains have also gradually drifted apart as a result of the accumulation of new mutations

6

Festing

(11). This is generally relatively slow, although of some practical importance if a new mutation should affect a character of particular interest to an investigator. Even this source of variation can be greatly reduced by maintaining foundation banks of frozen embryos and restoring the strains from the freezer regularly (12). Recovery of a single sibling (brother–sister or parent– child) pair of animals is sufficient to recover the whole strain as each mouse has all the alleles present in the strain. Embryo freezing can also be used to preserve outbred stocks, but as each mouse is genetically different and as the genetic loci are not homozygous, large embryo stocks sizes are needed in order not to lose genetic variation. Each inbred strain can be easily identified using genetic markers. There is extensive data on the characteristics of most mouse and rat strains, and many genetic modifications (GM) and mutations are maintained on an inbred background so as to reduce genetic drift and increase the sensitivity when comparing GM versus wild-type animals. In short, inbred strains are the nearest representation of a “standard analytical grade reagent” that is possible when doing experiments with mammals. More details are given in www.isogenic.info. In contrast, outbred stocks are phenotypically more variable than inbred strains, with the possible exception of some aspects of reproduction, they are genetically labile because the frequency of alleles at any one locus within a colony can change rapidly as a result of selection or random genetic drift (13). Genetic quality control is impossible at the individual level. For example, it is not even easy to distinguish genetically between Wistar and Sprague– Dawley rats because there is no standard “Sprague–Dawley” rat or “CFW” mouse and colonies with the same name can be genetically quite different (9). In fact scientists have rarely made a scientific case for using them (14) except in special cases where, for example they want to use selective breeding or fine-mapping of a gene.

1.4. A Brief Review of the (Supposed) Logic for Continued Use of Outbred Stocks

With a few exceptions, such as the NTP Carcinogenesis bioassay, toxicologists have always used outbred stocks and innovation in toxicity testing has not been encouraged by the drug regulators, so there has been little incentive to change. Moreover, there seems to have been no serious discussion of the topic for at least 30 years. A brief review in 1987 noted a couple of committees which had recommended the use of outbred stocks, but these did not include mammalian geneticists and biostatisticians and their reasons for recommending the use of outbred stocks were unconvincing (15). The most common justification can be summed up by a quote from an anonymous toxicologist in 2005:

Inbred Strains in Toxicity Testing

7

“The variability of toxicity obtained in less well defined animals is a strength in itself, not a problem, when trying to predict safety margin in the non-isogenic human population”.

However, this simple logic is faulty. The fundamental assumption in toxicity testing is that if the compound causes toxicity in mice or rats, then it may cause similar toxicity in humans. Therefore we should be designing experiments which have the best possible chance of detecting adverse effects in our test animals. Every good scientist knows that the fundamental rule in designing an experiment is to control all variables except for the treatment. Toxicologists go to great pains to use animals of uniform weight, maintaining them free of disease in a controlled environment and feed them a standardised diet. The very last thing we should be doing in a toxicity test is trying to model human genetic variation with a handful of variable animals. Clinical trials need to be large because humans are genetically diverse. The advantage of using genetically defined rodents is that variability can be rigidly controlled, so that experiments can be much smaller or statistically more powerful. It makes no sense to leave crucial genetic variation uncontrolled when it can be controlled easily by using a group of isogenic strains. In any case, the genetic variation present in a few hundred outbred mice or rats from a single colony is not remotely comparable to the genetic variation found in many millions of humans of several different races. The UK Committee on Toxicity stated (incorrectly) that toxicologists use inbred strains (16) and went on to say that “A potential disadvantage of such tight controls of experimental conditions is that this approach reduces the chance of detecting an adverse effect that occurs only in a sub-group of the experimental animals. The use of larger groups of more outbred animals might increase the chances of detecting such groups, but this could not be guaranteed”.

This is a bad suggestion. Searching for genetic variation in response to a known toxic compound and testing whether a compound is toxic require different experimental designs. In the former case several hundred to a few thousand genetically heterogeneous animals are all treated with the toxic compound and sensitive and resistant animals are genotyped at a large number of polymorphic gene loci to see if any correlate with the sensitivity. In the latter case, animals are randomised into treatment and control groups and differences between group means and proportions are studied. Any attempt to combine the two objectives into a single experiment is likely to result in failure to achieve either objective. A more thoughtful claim is that . . . it is more correct to test on a random-bred stock on the grounds that it is more likely that at least a few individuals will respond to the administration of an active agent in a group which is genetically heterogeneous (17).

8

Festing

This refers particularly to discrete outcomes such as presence/absence of a tumour. However, although the variation within a single outbred stock is greater than that in a single inbred strain, outbred stocks do differ in mean response (18). A different conclusion could easily be reached if a resistant rather than a sensitive one is used. So if using outbreds it would be better to use small numbers of several stocks rather than larger numbers of one stock. However, a small collection of inbred strains will usually have an even wider range of susceptibility, particularly if unrelated strains are used. This considerably improves statistical power both for measurement and for discrete outcomes (19). Other objections raised, such as it being impractical to use several inbred strains or that the strains may not be available, are unconvincing. Geneticists commonly use many strains without difficulty, and many standard strains of both mice and rats are available commercially. Until a few years ago some toxicologists also claimed that they did not want to improve methods of toxicity testing because they were already discarding too many potentially useful compounds. However, most now recognize that it is far more cost-effective to discard risky compounds early, as clinical trials are extremely expensive and the withdrawal of drugs after marketing is even more so especially if linked with a lawsuit. In conclusion, modern toxicologists now have the opportunity to have serious discussions with geneticists and statisticians about the relative merits of using inbred strains versus outbred stocks in toxicity screening. Their past reasons for using outbred stocks which at first appear justified use incorrect logic and information. The solution for cost-effective drug screening is very simple. Control the within-strain genetic variation using isogenic strains and use small numbers of several strains to represent a wide range of sensitivity. This will result in more powerful experiments with fewer false negatives without using any more animals. Any differences between strains will give some indication of genetic variation in response. These strain differences can be also used to explore toxic mechanisms by studying the biology of response in sensitive and resistant strains which leads into personalized medicine, especially as many inbred strains have been sequenced.

2. Current Methods of Toxicity Testing

Currently, the pre-clinical testing of new drugs involves a number of formal experiments required by the regulatory authorities (16). These include 28-day, 90-day and 2-year studies in rodents usually involving four dose levels (including the control) and both

Inbred Strains in Toxicity Testing

9

sexes, with ten animals per group, or a total of 80 animals for the short-term tests. Many outcomes are measured from these tests. A 90-day study in a small numbers of dogs may also be required. For carcinogenesis 2-year studies are usually done in rats with both sexes, four dose levels and 50 animals per group, or a total of 400 animals. Data on one other species, often the mouse, are also required. More recently novel methods such as the use of transgenic strains are also being used, e.g. Trp53 knockout mouse (B6.129S2-Trp53tm1Tyj/J) for short-term carcinogenicity testing (20). Reproductive toxicity studies and in some cases multi-generation studies are increasingly required. A number of other tests such as for ocular toxicity, skin sensitisation and DNA damage may also be required. The tests actually used are tailored depending upon on individual circumstances and will also differ between pharmaceutical and industrial/environmental chemicals. 2.1. Study Design Using Inbred Strains

A small change which would substantially improve the existing experimental design for the 28-day and 90-day studies would be to replace, for example ten outbred animals in each experimental group either by three animals of each of three different inbred strains or two animals of each of five inbred strains (2 × 5). The choice being largely dependent on getting a balance between statistical power, which favours more strains, and practicality, which favours fewer strains. The existing 28-day experiment is already a factorial design with two sexes and four doses. Adding two or more strains would make it an elegant three-way factorial design (21). In order to make the experiment as simple as possible, if a three-strain study was chosen, it could be run as three separate “mini experiments” each involving a single strain with three male and three female animals at each dose level. This would involve a total of 24 animals, and would be repeated three times, once for each of the three strains. The data would then be combined into a single analysis with 3 (strains) × 4 (doses) × 2 (sexes) × 3 (animals per sub-group) involving all 72 animals. A similar procedure could be used if there were five strains, with each mini-experiment involving 16 animals, which is repeated once with each of the five strains. Several other variants and options become available once the principle of using several inbred strains is accepted.

2.1.1. Choice of Strains

Inbred strains could be chosen on the basis of availability, knowledge of strain characteristics and absence of any biological properties which would preclude use of the strain, such as autoimmune disease or cancer. Strains which are genetically dissimilar could be chosen in order to maximise the chance of choosing at least one susceptible strain. For example inbred mouse strains have been classified into seven major families, based on the analysis

10

Festing

of single nucleotide polymorphisms, so each strain could be chosen from a different family (22). F1 hybrids, the first-generation cross between two inbred strains, are isogenic (i.e. all animals are genetically identical), they are more vigorous than inbred strains and make a good choice in these assays. One possible objection is that they tend to be intermediate between the parental strains for many characteristics. So if the aim is to test the compound against strains which differ as much as possible it may be better to use pure inbred strains. However, the use of one or two F1 hybrids should certainly be considered. A multi-strain toxicity test will nearly always reveal strain differences in response to test substances and possibly also in basal levels, as is shown in the example given below. Most characteristics of interest to toxicologists have a polygenic mode of inheritance, and the strain differences will usually be due to quantitative trait loci (QTLs). Again another reason to use inbred strains because detecting QTLs in small samples of outbred stocks is virtually impossible. If the strain differences are small, then compound safety could be assessed according to the most sensitive strain. If there are large strain differences, then these should probably be investigated in more detail. They may be due to a major gene which may also be segregating in humans allowing potentially patient selection. 2.1.2. Dose Finding

This could normally be done in one strain provided the top dose used in the assay was well below the maximum tolerated dose, as this might be acutely toxic to a more sensitive strain. A lot of the increased sensitivity from using inbred strains comes from the reduced noise making it possible to detect subtle effects, not because the animals show more marked symptoms.

2.2. Use of Inbred Strains in Biomarker Development

Toxicologists are actively searching for better biomarkers of toxicity which give earlier and more subtle indications of tissue damage. A good biomarker should give repeatable results with a high signal/noise ratio. Unfortunately, most of this research is currently being done using outbred stocks, which may give unrepeatable results both over time, in different laboratories and even when nominally identical outbred animals come from different breeders. This is a waste of resources. Figure 1.1 shows the percent responders to a synthetic polypeptide in successive samples of about 30 outbred CD (Sprague–Dawley) rats from a single breeder over a 2-year period (23). Seven inbred strains gave consistent results (not shown), so it was the animals, not the assay which varied, but this would have been unclear if inbred strains had not also been used. It would be difficult to develop a new biomarker if the test animals vary in this sort of manner because it would not be clear whether the protocols or the animals were at fault.

Inbred Strains in Toxicity Testing

11

100 90

Percent responders

80 70 60 50 40 30 20 10 0 1

3

5

7

9

11 13 15 17 Sample number

19

21

23

25

Fig. 1.1. Percent responders in successive samples, each of about 30 outbred CD (Sprague–Dawley) rats from the same breeder, over a 2-year period. Re-drawn from Simonian et al. (24).

2.3. Identifying Susceptibility Loci: Towards Personalised Medicine

In the future it is thought that drugs will be administered to people according to their individual genotype for responses to a particular drug, i.e. personalized medicine. However, many of the genes controlling sensitivity to xenobiotics have yet to be discovered. Identifying these genes remains difficult in humans. If mild toxicity is seen in a few people in early clinical trials there is no assurance that it is due to genetic sensitivity, and it would be impossible to identify the susceptibility genes with such small sample sizes. Further, it is already too late once a drug is in phase III trials as it has cost many hundreds of millions of dollars by this time. A better strategy is to identify susceptibility genes in inbred rodents before starting the clinical trials. These are initially detected as differences between the strains. Using increasingly powerful genomic tools it is becoming possible to identify the genes involved, particularly if susceptibility depends on one or a few major loci, although this requires further experiments (24). If candidate genes can be identified in mice or rats the published literature could be checked to see if the same loci are associated with any adverse reactions in humans and these loci could be monitored in the early clinical trials. A full discussion of how individual susceptibility genes could be detected, starting with strain differences is beyond the scope of this chapter. However, a first step would usually be to characterise a wider range of inbred strains. If these fall into two groups of susceptible and resistant, then the difference may be due to a single locus. This may be mapped by determining whether any

12

Festing

genetic makers in F2 crosses between susceptible and resistant strains are associated with susceptibility. Combining this sort of information with knowledge of loci which have already been characterised, gene expression data, the location and type of adverse reactions and the known characteristics of the inbred strains often including their full DNA sequence may make it possible to identify the genes. These sorts of investigations are not trivial and would not always be worthwhile, but methods are developing rapidly so this likely to become easier in the future. Further details are available on the Complex Trait Consortium web site http://www.complextrait.org/.

3. Example: A Multi-strain Assay

This example demonstrates the increased power from using inbred strains compared with outbred stocks in obtaining relevant dose–response curves. The data in Table 1.2a show white blood cell (WBC) counts in four inbred strains of mice with two mice of each strain at each dose level (48 mice), treated with chloramphenicol succinate at six dose levels, these data were extracted from previously published data (25). Originally there were eight mice of each inbred strain at each dose, but in order to obtain two comparable experiments two mice of each of the four inbred strains were chosen at random at each dose level. Table 1.2b shows the WBC counts in 47 CD-1 mice, with seven to nine mice per dose level in a parallel experiment conducted at the same time and under the same conditions. Differences between two of the inbred mice of the same strain in the same treatment group (say the first two CBA mice) in Table 1.2a are due to non-genetic variation. This could be the

Table 1.2a WBC counts (× 109 /L) in four inbred strains of mice in response to chloramphenicol succinate (mg/kg) Dose

CBA

0

1.6

C3H 0.5

2.1

2.2

BALB/c

C57BL

2.3

1.9

2.2

Mean 2.6

1.925

500

1.1

1.2

1.2

2.1

1.8

1.8

3.5

1.1

1.725

1000

0.7

1.1

1.2

1.0

3.1

2.1

1.7

1.0

1.488

1500

0.9

1.1

1.1

0.6

2.1

3.0

2.1

1.3

1.525

2000

1.7

1.4

1.3

1.2

1.7

2.1

1.5

1.0

1.488

2500

1.3

1.4

0.4

0.4

0.8

1.1

0.4

0.2

0.750

Inbred Strains in Toxicity Testing

13

Table 1.2b WBC counts in CD-1 mice, in response to chloramphenicol succinate (mg/kg) Dose

Mean

0

3.0

1.7

1.5

2.0

3.8

0.9

2.6

2.3

500

2.8

1.9

1.5

1.7

1.6

1.3

1.9

1.6

1.788

1.7

2.075

1000

3.4

1.8

1.2

1.9

2.6

3.6

1.3

1500

2.3

2.3

2.1

2.4

1.6

1.7

2.5

2000

2.3

3.6

1.0

1.8

2.6

1.6

1.6

2500

1.9

1.9

3.5

1.2

2.3

1.0

1.3

1.7

2.167 2.257 2.071

1.6

1.838

result of environmental and developmental differences between mice and measurement error, i.e. experimental noise. Differences between two mice of different strains (e.g. a CBA and a C3H mouse) in the same treatment group are due to non-genetic plus genetic variation. Differences between two mice of the same strain in different treatment groups are due to a treatment effect plus non-genetic variation. Thus in this case it is possible to estimate non-genetic variation, genetic variation and treatment variation. Differences between any two CD-1 mice in the same treatment group in Table 1.2b represent non-genetic plus genetic variation. Differences between two mice in different treatment groups represents non-genetic, genetic and treatment variation. Thus when examining these data it is not possible to estimate either the non-genetic variation or the genetic variation! The two sources of variation are “confounded” or inextricably mixed. As the genetic and non-genetic variations are added together the total variation is increased. Moreover, the greater the genetic variation is, and therefore the greater its importance, the greater the noise will be and the less chance there will be of detecting a treatment effect. This is absolutely the opposite of what is wanted and needed in today’s toxicology studies. The inbred strain data in Table 1.2a can be statistically analysed using a two-way analysis of variance. The result is shown in Table 1.3. All statistical analyses shown below used the MINITAB statistical package (MINITAB Inc., 3081 Enterprise Drive, State College, PA 16801-3008, USA.). Other dedicated statistical packages will give similar results. Readers can repeat the analyses themselves using the data in Table 1.2 if they wish. Differences between strains and between doses are highly significant (p < 0.01 in both cases). The strain × dose interaction is approaching significance p = 0.081, implying that there may be strain differences in response. This could be indicated by a

Festing

Table 1.3 Analysis of variance for WBC counts in the inbred mice Source

DF

SS

MS

F

p

6.83

0.002 0.003

Strain

3

5.2817

1.7606

Dose

5

6.3442

1.2688

4.92

(Linear

1

4.9031

4.9031

19.07

Deviations

0.000)

4

1.4411

0.3603

1.40

0.283

Strain × dose

15

7.2708

0.4847

1.88

0.081

Error

24

6.1900

0.2579

Total

47

25.0867

The Linear Analysis is taken from Table 1.4.

significant linear strain × dose interaction term, not calculated here in order to avoid getting too complicated. Mean WBC counts for all five strains (including CD-1) at each dose are shown graphically in Fig. 1.2. For the inbred strains, the regression of WBC counts on dose can be estimated using all the inbred data (ignoring strain differences). This gives the regression equation WBC=1.9512−0.0003743 × dose. The ANOVA for the regression calculations is given in Table 1.4. The regression is significant at p = 0.002. This level of significance is adequate, but it is underestimated because strains differ in control mean WBC counts, as can be seen in Fig. 1.2, so variation about

BALB/c C3H C57BL CBA CD-1

2.5 2.0 WB Ccounts

14

1.5 1.0 0.5 0.0 0

500

1000 1500 Dose

2000

2500

Fig. 1.2. Dose–response relationships for each strain. The thick dotted line is the regression of WBC on dose averaged across the four inbred strains. This is statistically significantly different from 0 (p = 0.002). The thick dashed line is the corresponding line for CD-1. This is not significantly different from 0 (p = 0.63). Note that BALB/c has low basal counts and it and CD-1 show no significant dose–response. Strains CBA and C3H are most sensitive, with the response of C57BL being not significant, possibly due to non-linearity.

Inbred Strains in Toxicity Testing

15

Table 1.4 Analysis of variance for regression (four strains, strain differences ignored) Source Regression Residual

DF

SS

MS

F

p

11.17

0.002

1

4.9031

4.9031

46

20.1835

0.4388

47

25.0867

error Total

the regression line is much higher than if this were not the case. The MS for regression in Table 1.4 (4.9031) should be compared with the error MS in Table 1.3 (0.2579) to correct for this effect. This is shown as the “linear” effect in Table 1.3 and results in a much higher significance level. This linear effect could also be calculated using orthogonal contrasts, described in some textbooks, e.g. Snedecor and Cochran (26), though the calculations are not available in many statistical packages. It is probably easier just to estimate the regression of WBC counts on dose separately for each strain. These are summarised in Table 1.5. The R2 values are the percent of variance due to regression, the “S” is the standard deviation about the regression line and the p values indicate the probability that a slope of this magnitude could have arisen by chance. From these data only strains CBA and C3H show a statistically significant response (see Note 2). In all cases the 12 inbred mice give a better fit to the dose–response line than the 47 mice outbred stock mice. There are also highly significant differences in mean counts between strains ranging from 1.23 in C3H to 2.01 in C57BL, although here it is really only the response to the test compound rather than basal levels within each strain which is of interest. Some QTL associated with baseline WBC counts in mice have already been identified (27). Related to this observation there are also racial differences in WBC counts in humans, with African Americans having lower counts than European Americans, and a locus accounting for about 20% of the variation in WBC counts has been identified on chromosome 1q (28). There does not seem to be any information on human genetic variation in the WBC response to chloramphenicol although very rarely it can cause a fatal pernicious anaemia. The CD-1 data can be analysed by a one-way ANOVA. The p value for differences between doses is 0.792; this does not even approach statistical significance. The difference between the control and the highest dose level averaged across the four inbred strains is 1.175 counts and the standard deviation is 0.508 counts, so the signal/noise ratio is 1.175/0.508 = 2.314. In the CD-1 the difference is

16

Festing

Table 1.5 Summary of individual regression calculations for each strain

Strain

N

R2

S

Regression coefficient × 1000

p

CBA

12

48.1

0.70

−0.72

0.012

C3H

12

69.0

0.37

−0.58

0.001

BALB/c

12

16.6

0.34

+0.16

0.188

C57BL

12

23.3

0.61

−0.36

0.112

CD-1

47

1.0

0.73

−0.06

0.634

. R2 is the proportion of variance accounted for by regression. S is the standard deviation of deviations from regression. A lower value indicates a better fit of the regression line. The p values indicate the probability that a slope as great as this could have arisen by chance

only 0.329 counts and the standard deviation is 0.746 counts, giving a signal/noise ratio of 0.329/0.746 = 0.442. Using power analysis, the sample size that would be needed in some future experiment to detect a signal/noise ratio of 2.314 with a 90% power and a 5% significance level using a two-sided t test would be six treated and six control mice. However, in order to detect a signal/noise ratio of 0.442 found with CD-1 mice, it would require 109 animals per group, i.e. 18 times more resources. Altogether eight primary haematological parameters were measured, and the signal/noise ratios (i.e. difference between control and treated means divided by the standard deviation) of each of these are shown in Figs. 1.3 and 1.4. A comparison of these clearly shows the increased sensitivity of the multi-strain assay across all outcomes as well as showing which outcomes are responding and at which dose levels. In summary, the assay using four inbred strains (48 mice) is substantially more powerful than the one based on outbred CD-1 (47 mice), showing a highly significant regression of WBC counts on dose of chloramphenicol, averaged across four strains. However, this is restricted to only two of the strains. This clearly shows that there is genetic variation both in baseline WBC counts and in response to chloramphenicol and leads to some human literature on basal WBC levels. In contrast, the outbred stock CD-1 fails to show any dose-related response and provides no information on the genetic control of WBC counts. It is known that chloramphenicol causes a dose-related bone marrow suppression in humans (29) which is detected in the WBC lineage only by using

Inbred Strains in Toxicity Testing

HCT HGB Retics LYMP Neuts PLT RBC WBC

1 0 –1 Signal/noise ratio

17

–2 –3 –4 –5 –6 –7 –8 0

500

1000 1500 Dose

2000

2500

Fig. 1.3. Signal/noise ratios [(treated mean–control mean)]/SD for eight haematological traits in the CD-1 outbred stock. The dotted line represents the signal/noise ratio that should be detectable in a future experiment with a group size of 20 animals and a 90% power using a two-sided t-test and a 5% significance level.

1

HCT HGB Retics LYMP Neuts PLT RBC WBC

0 Signal/noise ratio

–1 –2 –3 –4 –5 –6 –7 –8 0

500

1000

1500

2000

2500

Dose Fig. 1.4. Signal/noise ratios [(treated mean–control mean)]/SD for eight haematological traits averaged across the four inbred strains. The dotted line represents the signal/noise ratio that should be detectable in a future experiment with a group size of 20 animals and a 90% power using a two-sided t-test and a 5% significance level. The last two points for the reticulocytes have been truncated at −7.5 so that the scale for Figs. 1.3 and 1.4 are identical. Note the increased sensitivity compared with the CD-1 mice in Fig. 1.3.

inbred strains. Similar conclusions are reached when all eight primary haematological characters are considered.

18

Festing

4. Conclusion A fundamental rule when designing an experiment is to control all variables except for the one being studied, namely the treatment. Failure to do this results in an increased chance of a false-negative result due to a low signal/noise ratio. Where the experimental subjects can also be classified into different groups, such as male and female, or strain A and strain B which may respond differently, this also needs to be taken into account. This can best be done by replicating the experiment within each class, so part of the experiment can be done on strain A males and part on strain B males and so on with the females. This of course assumes that the character being studied is present in all groups. Obviously we cannot study the effect of a xenobiotic on testis in females! What is not so widely understood is that using inbred animals in this type of design does not require four times as many animals. This is because the data can be combined into a single so-called factorial statistical analysis which separates out the effect of sex, strain and treatment and any interactions between them. Factorial designs are very powerful. R.A. Fisher, the great statistician and geneticist stated that “. . . in a wide class of cases (by using factorial designs) an experimental investigation, at the same time as it is made more comprehensive, may also be made more efficient if by more efficient we mean that more knowledge and a higher degree of precision are obtainable by the same number of observations.” (30). It is well established that different inbred strains and outbred stocks can respond differently to a xenobiotic. The important implication is that if a single strain or stock is used it may be relatively resistant to the test compound, giving a false-negative result. Moreover, if only a single strain or stock is used, the importance of strain/stock variation never becomes apparent. If it is possible to include more than one stock/strain in an experiment without the need to increase the total number of animals, and still with the same number of animals in each treatment group, it should be obvious that this would be a good strategy. Additionally, within each branch of such an experiment it makes sense to control the inter-individual variation by using isogenic strains. Indeed, the greater the genetic component of variation is, the more important it is to control it in this way if falsenegative results are to be avoided. There are, of course, several questions that need to be addressed in such a strategy. For example, how many strains should be used and how should they be chosen? The answer is that if the total number of animals (the main determinant of cost) is not increased, this sets a limit on the number of strains to be

Inbred Strains in Toxicity Testing

19

used. Strains should then be chosen to be as genetically different as is practical, given the availability of such strains. In the numerical example comparing a multi-strain study with one using a single outbred stock, group size was eight animals in both cases, so within this limit it was possible to use two mice of each of four inbred strains. The numerical example showed clearly that for the white blood cell lineage the CD-1 stock was both more resistant to the effects of the test chemical, chloramphenicol, and more variable in response. This resulted in a false-negative result. In contrast, two of the inbred strains responded strongly to the compound (a true positive result), also making it clear that there is genetic variation in response. Had the BALB/c strain been the only one used, then again there would have been a false-negative result; i.e. the use of a single inbred strain is not a good idea for a screening type of experiment. These data raise the question – So why don’t toxicologists use such a multi-strain assay of this sort? It is obviously more powerful and provides additional information? Perhaps this reflects a conflict between intuition and science. Intuitively, most scientists assume that as humans are genetically variable, they should use genetically variable animals as models. However, in a controlled experiment, genetic variation is just like any other source of variation, it needs to be controlled, if it is not it will confound the result. This is equally true in a clinical trial. If genetic variation was controlled using monozygous twins, then this would enormously increase the power of a clinical trial. Unfortunately this is not practical in clinical trials, but it is practical when using laboratory rodents. Also, unfortunately the idea that by using a factorial experimental design it is possible to get more information from the same number of animals is not well known. This seems to be due to a failure for most scientists to understand the principles of combining experimental design and statistics. Fortunately, research is becoming increasingly multi-disciplinary with statisticians becoming an important component of the research team. Toxicologists and pharmacologists only need to understand the broad principles. It is not essential that we all know exactly how to do the statistical analyses of this type of experiment. That can be left to the statisticians. The FDA in their “critical path” white paper have highlighted the need for improved methods of toxicity testing. Presented here is one obvious way ahead. By changing the type of animals used, the whole power of modern genetic techniques will become available to the pharmaceutical industry. The NIEHS is already starting down this path using 15 fully sequenced inbred strains of mice. With the dramatic advances made in the science of genetics in the last few decades, now is the time for toxicologists and others in drug development to start “thinking genetics”.

20

Festing

5. Notes 1. F1 hybrids, the first-generation cross between two inbred strains are isogenic but not homozygous and as such they do not breed true, i.e. F1’s between any pair of inbred strains are isogenic, however, if interbred lead to highly variable phenotypes. 2. Note that the standard deviation S, a measure of noise, is higher in the outbred CD-1 stock than in any of the inbred strains. References 1. Russell, W. M. S. and Burch, R. L. (1959) The principles of humane experimental technique, Universities Federation for Animal Welfare (UFAW), Potters Bar, Herts. 2. Food and Drug Administration (2004) Challenge and opportunity on the critical path to new medical products. http://www.fda. gov/oc/initiatives/criticalpath/whitepaper. html. 3. Caldwell, G. W., Ritchie, D. M., Masucci, J. A., Hageman, W. and Yan, Z. (2001) The new pre-preclinical paradigm: compound optimization in early and late phase drug discovery. Curr Top Med Chem 1, 353–366. 4. Food and Drug Administration (2008) The FDA Critical Path Initiative. http://www. fda.gov/oc/initiatives/criticalpath/ 5. The Innovative Medicines Initiative (2008) Innovative Medicines Initiativw. http://imi.europa.eu/docs/imi-gb-006v215022008-research-agenda˙en.pdf 6. Brown, S. D., Chambon, P. and de Angelis, M. H. (2005) EMPReSS: standardized phenotype screens for functional annotation of the mouse genome. Nat Genet 37, 1155. 7. Franc, B. L., Acton, P. D., Mari, C. and Hasegawa, B. H. (2008) Small-animal SPECT and SPECT/CT: important tools for preclinical investigation. J Nucl Med 49, 1651–1663. 8. Petit-Zeman, S. (2004) Rat genome sequence reignites preclinical model debate. Nat Rev Drug Discov 3, 287–288. 9. Chia, R., Achilli, F., Festing, M. F. and Fisher, E. M. (2005) The origins and uses of mouse outbred stocks. Nat Genet 37, 1181– 1186. 10. Festing, M. F. W. (2003) Laboratory animal genetics and genetic quality control, in

11.

12. 13. 14.

15. 16.

17. 18.

19.

Handbook of laboratory animal science: essential principles and practices (Hau, J. and Van Hoosier, G. L., Jr., eds.), 2nd ed. CRC Press, Boca Raton, London, New York, pp. 173– 204. Stevens, J. C., Banks, G. T., Festing, M. F. and Fisher, E. M. (2007) Quiet mutations in inbred strains of mice. Trends Mol Med 13, 512–519. Taft, R. A., Davisson, M. and Wiles, M. V. (2006) Know thy mouse. Trends Genet 22, 649–653. Papaioannou, V. E. and Festing, M. F. (1980) Genetic drift in a stock of laboratory mice. Lab Anim 14, 11–13. Festing, M. F. W. (1999) Warning: the use of genetically heterogeneous mice may seriously damage your research. Neurobiol Aging 20, 237–244. Festing, M. F. (1987) Genetic factors in toxicology: implications for toxicological screening. Crit Rev Toxicol 18, 1–26. Committee on Toxicity and Food Standards Agency (2007) Variability and Uncertainty in Toxicology of Chemicals in Food, Consumer Products and the Environment. cot.food.gov.uk/pdfs/cotstatementworkshop 200703.pdf Arcos, J. C., Argus, M. F. and Wolf, G. (1968) Chemical induction of cancer. Academic Press, Inc., New York. Kacew, S., Ruben, Z., McConnell, R. F. and MacPhail, R. C. (1995) Strain as a determinant factor in the differential responsiveness of rats to chemicals. Toxicol Pathol 23, 701– 715. Felton, R. P. and Gaylor, D. W. (1989) Multistrain experiments for screening toxic substances. J Toxicol Environ Health 26, 399– 411.

Inbred Strains in Toxicity Testing 20. Floyd, E., Mann, P., Long, G. and Ochoa, R. (2002) The Trp53 hemizygous mouse in pharmaceutical development: points to consider for pathologists. Toxicol Pathol 30, 147–156. 21. Montgomery, D. C. (2004) Design and analysis of experiments, 6th ed., John Wiley & Sons, Inc., Hoboken, NJ. 22. Petkov, P. M., Ding, Y., Cassell, M. A., Zhang, W., Wagner, G., Sargent, E. E., et al. (2004) An efficient SNP system for mouse genome scanning and elucidating strain relationships. Genome Res 14, 1806– 1811. 23. Simonian, S. J., Gill, T. J., 3rd and Gershoff, S. N. (1968) Studies on synthetic polypeptide antigens. XX. Genetic control of the antibody response in the rat to structurally different synthetic polypeptide antigens. J Immunol 101, 730–742. 24. Churchill, G. A., Airey, D. C., Allayee, H., Angel, J. M., Attie, A. D., Beatty, J., et al. (2004) The Collaborative Cross, a community resource for the genetic analysis of complex traits. Nat Genet 36, 1133–1137. 25. Festing, M. F., Diamanti, P. and Turton, J. A. (2001) Strain differences in haematologi-

26. 27.

28.

29.

30.

21

cal response to chloramphenicol succinate in mice: implications for toxicological research. Food Chem Toxicol 39, 375–383. Snedecor, G. W. and Cochran, W. G. (1980) Statistical methods, 7th ed. Iowa State University Press, Ames, Iowa. Peters, L. L., Zhang, W., Lambert, A. J., Brugnara, C., Churchill, G. A. and Platt, O. S. (2005) Quantitative trait loci for baseline white blood cell count, platelet count, and mean platelet volume. Mamm Genome 16, 749–763. Nalls, M. A., Wilson, J. G., Patterson, N. J., Tandon, A., Zmuda, J. M., Huntsman, S., et al. (2008) Admixture mapping of white cell count: genetic locus responsible for lower white blood cell count in the Health ABC and Jackson Heart studies. Am J Hum Genet 82, 81–87. Feder, H. M., Jr., Osier, C. and Maderazo, E. G. (1981) Chloramphenicol: a review of its use in clinical practice. Rev Infect Dis 3, 479–491. Fisher, R. A. (1960) The design of experiments, 7th ed. Hafner Publishing Company, New York.

Chapter 2 The Sophisticated Mouse: Protecting a Precious Reagent Michael V. Wiles and Rob A. Taft Abstract Definable, genetically and environmentally, the humble mouse has become a reagent with which to probe the human condition. The information thus gained is leading to a greater understanding of interindividual variation in drug responses and disease processes and is forming the basis for personalized medicine. Inbred mice are the tool of choice as each strain is essentially clonal in nature creating a defined, uniform setting where the effects of genetic background and modifications can be evaluated coherently. However, the creation and characterization of novel mouse strains remain expensive and time consuming. Further, the continual maintenance of these valuable animals as live colonies is financially draining and carries continual potential risks, including disastrous loss due to fire, flood, disease, etc. There are also other more insidious disasters including genetic contamination and genetic drift, either of which can go undiscovered until their effects ruin experiments. With this in mind, we strongly recommend that all mouse strains be cryopreserved as a matter of standard mouse management. Cryopreservation is a powerful colony management tool, assuring strains are available upon demand, for example, for regulatory requirements, re-initiation of projects, collaborations, re-evaluation of data etc. However, it is essential that any cryopreservation approach be cost-effective for both strain closure and strain recovery. In this chapter, we describe the variables which can afflict an inbred mouse’s genetic background (and hence phenotype), options to consider for strain archiving, and describe how to economically store and recover strains by sperm cryopreservation. Key words: Genetically modified mice, inbred mice, sperm cryopreservation, embryo cryopreservation, IVF, in vitro fertilization, genetic drift, genetic contamination. Abbreviations: IVF in vitro fertilization

1. Introduction 1.1. The Reagent Grade Mouse

Since the time of alchemy those engaged in research have continually needed to further refine and define their basic experimental reagents, their scientific tools of the trade. The resulting progress

G. Proetzel, M.V. Wiles (eds.), Mouse Models for Drug Discovery, Methods in Molecular Biology 602, DOI 10.1007/978-1-60761-058-8 2, © Humana Press, a part of Springer Science+Business Media, LLC 2010

23

24

Wiles and Taft

has allowed alchemy to become science and has driven the scientific revolutions of the 20th and 21st centuries. The term “reagent grade” is usually only applied to chemicals where impurities are either vanishingly absent or have been defined. In the field of experimental biomedical and genetic research, the evolution of the mouse as a reagent has been similar to other reagents whose precise definition and refinement has continually improved. Now, with the sequencing of the mouse genome and improvements in genetic monitoring of inbred mouse strains we believe that the term reagent grade can justifiably be applied to the mouse (1). This particular reagent is, however, one of the most sophisticated scientific tools available. It is a living creature (and as such deserving respect), a highly complex mammal capable of independent survival and is the net result of millions of years of evolution. So as with any reagent, its basic attributes must be understood if it is to be used successfully, including purity and/or quality, and how to maintain its full functionality. Two things make the reagent grade mouse invaluable as a model organism. First, it is possible to introduce precise genetic modifications into the mouse genome at will. These genetic modifications including gene ablation, addition, and modulation enable the rapid examination of the effects of these modifications in a complete living organism (2, 3). Second, comparisons of the mouse and human genomes reveal that mice and humans share a common ancestor which diverged about 75 million years ago, as such these two species have maintained many similarities in gene function. Comparisons between the two species show 99% shared gene function (4). Combined, these attributes make the mouse a reagent of exquisite subtlety and sophistication, enabling us to understand gene interactions within the complete organism, test their interplay with the environment, and extrapolate these data to human biology. However, “living reagents” require continual maintenance if they are not to be lost or degraded. It is here that mouse management by cryopreservation can be used, providing versatile archiving of these valuable resources. Lastly, the scientific method is based on the altruistic notion that others can take data and information, repeat it, and build upon it. Without repeatability there is no scientific advancement. As Sir Isaac Newton reportedly said, “If I have seen further it is by standing on the shoulders of Giants.” Treating the mouse as a living reagent will help ensure the creation of a sound foundation on which others can stand upon, build, and hopefully see further. 1.2. Mice Change and be Can Lost, If Not Cared for

Mice are a key element in many biological experimental designs, but their origin and quality are often overlooked. The variability introduced by using ill-defined mice, for example, randomly crossed strains, e.g., CD-1, or incomplete inbred

Sophisticated Mouse

25

congenics, increases experimental noise within and among experiments. If this is not understood and part of an experimental approach, this genetic variation will lead to a lack of reproducibility and difficulty in interpretation (see Chapter 1 by Festing this volume). Inbred mice, defined by at least 20 generations of brother– sister matings, are the most defined mammal available for experimental manipulation, with individual mice within each inbred strain being essentially clonal in nature (>99 % homozygosed at all loci). This allows precise experimental comparisons within strains, between multiple inbred strains, and between genetically modified versus non-modified mice of the same genetic background (5). However, all life has an innate biochemical, evolutionary capability to change and mutate, generating variation; mice are the selective result of this evolutionary past and their current environment (6). To work with these complex reagents successfully requires an appreciation of this. Inbred mice are generally maintained as a continually breeding colony requiring precise control of breeding. If done incorrectly their genetic background will change. There are two major sources of genetic change reported in live mouse colonies: (i) fast, disastrous genetic contamination (one breeding cycle) and (ii) insidious genetic drift. Until recently genetic drift was viewed as a slow process, however, recently the phenomena of copy number variation (CNV) has been discovered as a rapid source of genetic variation and hence drift. Although its impact is not fully understood, it is thought to cause very rapid genetic drift (within one generation) via duplication or deletion of one to thousands of kilobases of DNA, potentially containing entire genes. The resulting varying copy number effectively changes gene dosage which can result in phenotypic shifts (7, 8). Appropriate use of cryopreservation can forestall the cumulative adverse effects of genetic drift, including CNV, and allow rapid restoration of strains if genetic contamination, disease, phenotypic shifts, etc. occur (for a review ref. 1). Additionally, cryopreservation increases mouse management options, facilitating more cost-effective colony management. In regard to mainly custom genetically modified strains, although it is tempting to believe that these strains are safe during active experimental work, all vivariums carry risks including the possibility of disease, breeding cessation, genetic contamination, and other disasters. For example, in June 2001 the tropical storm Allison caused the flooding of vivariums at the Texas Medical Center killing more than 30,000 mice and rats, causing incalculable losses (9). Strain backup is therefore a prudent management step and also facilitates dynamic cost management allowing strains to be closed down and only upon demand, rapidly re-initiated.

26

Wiles and Taft

1.3. Mouse Archiving – Options

In the management of any resource a key consideration is “Return on Investment,” i.e., in this case, it is of little value to cheaply store/archive mouse strains in a format which makes their recovery prohibitively expensive and/or unpredictable (see Table 2.1 for summary of approaches). While at the same time it is not viable to invest large sums into the archiving of strains if the likelihood of them ever being wanted at a later date is small or totally unknown. Mouse strains can be archived as embryos (2–8 cell), gametes (sperm, oocytes), or as sources of gametes (spermatogonial stem cells, ovaries), see Table 2.1. When looking at costs to cryopreserve and recover mouse strains, sperm is in general the most logical choice based on the ease of collection and the sheer numbers of sperm available from a single male (∼30 × 106 sperm/male). However, until very recently recovery of live born from C57BL/6 sperm, the most commonly used background, was less than 5%, making routine recovery expensive and unpredictable. Females can provide naturally ∼6–8 oocytes or upon superovulation up to 50 oocytes. However, this is highly strain dependent, for example, C57Bl/6J give high numbers, while 129 strains give very few embryos. Complicating this there is also evidence suggesting that embryo quality falls with the high oocytes yields (10, 11). Thus upon comparing the economics of the two methods, embryos and sperm, it is very apparent that the freezing of embryos is considerably more expensive due to the need for more resources; for example, with a C57BL/6 background, to produce ∼250 two-cell embryos for cryopreservation by IVF requires >15 females. If strains are never recovered or only recovered once or twice, then the bulk of this expense remains forever frozen. In contrast, cryopreserving sperm has a low initial cost as only few animals (1–3 males) and relatively little labor and materials are required (12). It is upon recovery from sperm by in vitro fertilization (IVF) that animals and labor are used, but then only the required number animals per recovery are used as the IVF process can be scaled to produce the desired number of offspring. A major disadvantage of sperm cryopreservation of a strain is that only half the genome is stored by this approach (i.e., the donor sperm is haploid!). This does not represent a major problem when a genetic modification is on a readily available standard background, as high-quality female animals (oocyte donors) are readily available from reputable providers for most standard inbred mice. Although it should be appreciated that genetic drift still occurs unless the supplier has addressed this issue by restoring the breeder stock from cryopreserved pedigreed embryos every few generations (1).

Very simple Needs only1–3 carrier males ∼30×106 sperm/male Inexpensive

Simple Inexpensive

Moderately simple Needs only 1–3 carrier males

Ovary cryopreservation

Two cell embryo cryopreservation (heterozygotes embryos generated via IVF)

Pros

Sperm cryopreservation

Cryopreservation method/gametes

Cryopreservation

Only half the genome is preserved Needs IVF to make embryos Needs female (wild type) oocyte donors. Strain dependent 5–50 oocytes/female Only half the genome is persevered

Only half the genome is preserved. Needs multiple donor females, i.e., one female has only two ovaries

Due to potential rejection needs appropriate recipient animals. Moderate level of surgical skill to implant. Low yield of offspring/ovary Off-spring will be heterozygotes Can transmit disease Not scalable

Simple recovery into Low yield of offspring pseu-dopregnant animals Offspring will be Can be used to achieve heterozygotes strain rederivation. Not scalable (based in initial investment)

Represents female linage

Moderately simple Requires IVF to recover strain Can be used to achieve Requires appropriate oocyte strain rederivation donor strain Highly strain reproducible Some strains and mutaHighly scalable tions adversely affect IVF success. Offspring will be heterozygotes

Only half the genome is preserved Needs IVF quality control

Cons

Pros

Recovery Cons

Table 2.1 Summary of approaches to manage mouse strains

$$$

$$

$

(continued)

Cumulative cost/ animal to cryopreserve and recover

Sophisticated Mouse 27

Simple Inexpensive

ES cellsb None

Only half the genome is preserved

None

None

Simple to implant Can be used to achieve rederivation of strain

$$$$$

$$$$

Need to make germline $$$$$ transmitting chimeras Upon germline transmission only half the ES derived offspring will be heterozygotes ES cells carry tissue culture associated genetic damage Not scalable

Requires a high level of resources/skill to recover Pathogen transmission could occur Offspring will be heterozygotes Associated with genetic damage Not scalable

Expensive resource to restock Not scalable

Cons

Cumulative cost/ animal to cryopreserve and recover

Note: all the above approaches are also subject to strain effects and to possible deleterious effects of gene modification or addition. a Intracytoplasmic sperm injection, ICSI although not strictly necessary a cryopreservation method – this approach has been used as method to archive and restore mouse strains. b Embryonic Stem Cell, ES cell are often generated as part of the process for a targeted genetic modification. Although a strain can be recreated from the original targeted ES line it requires the re-creation of germline chimeras and their successful germline breeding before the strain is recovered. Further, ES cells are known the mutate in culture.

Very simple Inexpensive

ICSIa

Needs a large colony of strain to be cryopreserved (expense) to provide embryos

Pros

Cons

Pros

Simple Provides “homozygous” storage

Recovery

Cryopreservation

Two to eight cell embryo cryopreservation (flushed homozygous embryos)

Cryopreservation method/gametes

Table 2.1 (continued)

28 Wiles and Taft

Sophisticated Mouse

1.4. How Safe Is Gamete Cryopreservation

29

Any approach for archiving gametes has to provide long-term secure storage. Most methods for archiving mouse strains cryopreserve embryos, sperm, ovaries, etc., in vapor phase or liquid nitrogen. There are many papers discussing the longevity of gametes in cryogenic storage; however, it is generally accepted that once samples pass through the glass transition temperature of water, ∼−137◦ C all biological activities cease. Although gamma radiation can still cause accumulative damage, simulation studies suggest that this is insignificant over ∼2000 years under normal background radiation levels (13, 14). Of much greater concern with long-term cryogenic storage is temperature variation, where gametes are exposed to temperature fluctuations above −137◦ C. The most likely causes of temperature variation (increase) is improper handling of the frozen gametes (e.g., while “rummaging” in the liquid nitrogen storage tanks), failure to fill liquid nitrogen tanks, i.e., they run dry, destruction of the storage facility due to fire etc., or the physical failure of the tanks vacuum (15). As such, it is strongly recommended that cryopreserved gametes be stored physically in at least two liquid nitrogen storage tanks and, additionally, that tanks be in two or more separate facilities as one part of a comprehensive approach to repository operation (16).

2. Materials 2.1. Cryopreservation of Mouse Sperm

1. Distilled water (Invitrogen, cat # 15230-238)

2.1.1. Cryoprotective Medium

3. 3% w/v skim milk (BD Diagnostics cat # 232100).

2.1.2. Consumables

1. 0.25 mL French straws (IMV cat # AAA201)

2. 18% w/v raffinose (Sigma cat # R7630) 4. MTG: 447 ␮M monothioglycerol (Sigma cat # M6145)

2. Cassettes (Zander Medical Supplies, 145 mm 16980/0601) 3. Styrofoam box internal dimensions ∼35 cm × ∼30 cm, a Styrofoam float (piece should be approximately ∼2–3 cm thick and be cut to cover ∼80% of internal area of box) 4. Monoject insulin 1 mL syringe 2.1.3. Mice – Strain to be Cryopreserved

Two to three male mice, preferably 10–16 weeks old (see Note 1)

2.2. In Vitro Fertilization Method

1. Pregnant mare serum gonadotropin (PMSG)

2.2.1. Hormones for Superovulation

3. Sterile physiological saline

2. Human chorionic gonadotropin (hCG)

30

Wiles and Taft

2.2.2. Mice for Superovulation 2.2.3. In Vitro Fertilization

Five to ten female mice, 3 weeks or 6–12 weeks of age, depending on the strain (17) 1. MVF media: Research vitro Fert (K-RVFE-50) COOKS Mouse In Vitro Fert Fertilization medium (Cook MVF, Australia, see http://www.specialtyvet.net/ page/page/6095801.htm) or Human Tubal Fluid media (see 18): NaCl (FW 58.44, Sigma S-5886) KCl (FW 74.55, Sigma P-5405) MgSO4 . 7H2 O (FW 246.5, Sigma M-7774) KH2 PO4 (FW136.09, Sigma P5655) CaCl2 . 2H2 O (FW 147, Sigma C-7902) NaHCO3 (FW 84.01, Sigma S-5761) Glucose (FW180.16, Sigma G-6152) Na-pyruvate (FW110.0, Sigma P-4562) Na-lactate 60% syrup (FW 112.1, Sigma L-7900) Penicillin-G (FW 372.5, Sigma P-4687) Streptomycin sulfate (FW 1457.4, Sigma S-1277) Phenol red (5%) (Sigma P-0290) BSA (Equitech-Bio BAC62-0050) Fill with water-cell culture grade (Sigma 59900C)

5.9375 g 0.3496 g 0.0493 g 0.0504 g 0.3 g 2.1 g 0.5 g 0.0365 g 3.42 mL 0.075 g 0.05 g 0.20 mL 4.0 g

Weigh each component and dissolve in high quality water (cell culture grade, double glass distilled or reverse osmosis, and filtration, i.e., 18 M) in a 1 L volumetric flask, but withhold the BSA for addition later. Bring the volume up to 1 L. Measure the osmolarity (290 ± 5). Bubble gas (5% CO2 , 5% O2 , 90% N2 ) through the medium for ∼5 min, add the BSA to the media, mix gently to avoid frothing. Filter through a 0.2 ␮m filter into sterile bottles. Gas the medium with a mix of 5% CO2 , 5% O2 , 90% N2 to displace air above the medium, cap tightly, and store at 4◦ C for not more than 2 weeks. Repeat gassing after every use to maintain a pH of 7.2–7.4. 2. Mixed gas (5% O2 , 5% CO2 , balanced with N2 ) 3. One large 60 × 100 mm Falcon Petri dish for every three females (BD Biosciences) 4. One small 35 × 10 Falcon Petri dish for each male (BD Biosciences) 4. Embryo-tested mineral oil 5. Phosphate Buffered Saline(PBS)

Sophisticated Mouse

31

3. Methods It is critical that all media are carefully prepared and have the correct pH and temperature, as well as batch-tested reagents. Where possible we suggest buying reagents readymade. Further, it is helpful to pay attention to details and efficient laboratory setup, e.g., small incubators, heated stages, or other devices that ensure proper temperature and pH stability. 3.1. Sperm Cryopreservation 3.1.1. Preparation of Cryoprotective Media (CPM)

This method has been successfully used for many different mouse strains and backgrounds (12). 1. Place ∼80 mL of bottled distilled water in a beaker. 2. Heat for ∼40 sec in microwave to ∼60–80◦ C (do not boil). 3. Place beaker on heated stir plate, add 18 g of raffinose, and heat and stir till solution clears (see Note 2). 4. Add 3 g of skim milk to the raffinose mixture and heat and stir until dissolved (see Note 3). 5. Transfer solution to volumetric flask and bring to 100 mL with bottled distilled water. 6. Add MTG now or after thawing (see Note 4) 7. Mix well and divide the solution into two 50 mL centrifuge tubes. 8. Centrifuge at 13,000 × g for 15 min at room temperature (∼22◦ C). 9. Filter through a 0.22 ␮m cellulose filter (a prefilter may help the flow). 10. Verify that the osmolarity is in the range of 470– 490 mOsm. 11. Aliquot 10 mL of filtered cryoprotective media into labeled 15 mL conical tubes. 12. Cap and store at −20◦ C until ready for use (see Note 5).

3.1.2. Sperm Cryopreservation Setup

1. Thaw and warm CPM in 37.5◦ C water bath (see Note 6). While the media is warming. 2. Label and mark straws and affix to a 1 mL monoject syringe. 3. Fill Styrofoam box to a depth 6–9 cm of liquid nitrogen. 4. Place Styrofoam float into Styrofoam box. 5. Replace Styrofoam box lid to slow the evaporation of liquid nitrogen. 6. Place the lid from Petri dish on the warming tray and lean the bottom of a Petri dish against it so that one side of the

32

Wiles and Taft

Petri dish is elevated. This arrangement forces the CPM to collect on one side, making it easier to fill the straws. 7. If monothioglycerol was not added prior to freezing, add it now, to a final concentration of 477 ␮M. Add 1 mL of CPM to the dish for each male from which sperm will be collected. 3.1.3. Sperm collection

1. Euthanize the males (1–3) and remove the cauda epididymides and vas deferentia, carefully removing the testicular artery to avoid contaminating the sperm with blood. 2. Release sperm into the CPM by making several cuts through the epididymides and vas deferentia using a beveled hypodermic needle while holding the tissues with a pair of forceps. 3. Remove tissue from the CPM after 10 min.

3.1.4. Sperm cryopreservation

1. Aspirate a 4.5 cm column of CPM into a French straw followed by a 2 cm column of air. 2. Aspirate a 0.5 cm column of sperm into the French straw then aspirate additional air until the column of CPM without sperm contacts the PVA powder in the cotton plug. 3. Seal the end of the French straw with a brief pulse from an instantaneous heat sealer. 4. Repeat this process until the desired number of straws has been filled (we suggest minimum of 20/strain). 5. Place five straws into one cassette. Repeat until four cassettes have been filled. 6. Place the cassettes in the liquid nitrogen-filled box on the float (i.e., in vapor phase) so that they are not touching. 7. Put the lid on the box for 10–30 min. 8. Plunge the cassettes into the liquid nitrogen. 9. After at least 10 min in liquid nitrogen the cassettes can be removed and rapidly placed into storage in liquid nitrogen (see Note 7).

3.2. In Vitro Fertilization (IVF) with Frozen Sperm

Once a strain is frozen as sperm we strongly recommend that 1–2 straws be used to assess the quality of the sperm post-thaw. Although various devices exist to measure sperm motility etc., the only relevant test for sperm function is an actual IVF. Additionally, fertilization rates vary widely among commonly available inbred strains; also the introduction of mutations and genetic modifications into a strain can have indirect and unanticipated effects on the quantity and quality of oocytes and sperm produced, as well their performance during IVF (12, 17). IVF can be difficult to establish – the quality of the reagents is crucial, also during the IVF process it is essential for repro-

Sophisticated Mouse

33

ducible success to maintain media pH and temperature and that all practical steps are taken to keep precise control of the culture conditions. 3.2.1. Superovulation

1. Inject females with PMSG 44–48 h prior to injection with hCG (see Note 8). 2. Inject females with hCG 13 h prior to oocyte collection (see Note 8).

3.2.2. IVF Setup

1. Prepare oocyte collection dish by adding 2 mL of PBS to a 25 mm × 10 mm dish and keep at 37.5ºC in air. 2. Prepare IVF dish by placing a 250 ␮L drop of MVF medium (see Note 9) in the center of a 60 mm Petri dish. Place four additional 150 ␮L drops of MVF medium around the 250 ␮L drop. 3. Carefully add sufficient oil to cover the media and place in an incubator or sealed chamber filled with mixed gas (5% O2 , 5% CO2 , 90% N2 ) at least 1 h prior to IVF (see Note 10).

3.2.3. Thawing Sperm

1. Place the straw in a 37.5◦ C (clean) water bath. 2. Rapidly swirl the straw in the water until all ice has melted (about 30 sec). 3. Dry the straw with a paper towel. 4. Cut off the sealed end of the straw opposite the cotton plug. Using a metal rod, expel the sperm from the straw into the 250 ␮L IVF drop. 5. Allow sperm to incubate at 37◦ C for 1 h prior to adding oocytes.

3.2.4. Oocyte Collection and IVF

1. Euthanize 2–5 superovulated females approximately 13 h post-hCG (see Note 11). 2. Remove the ovary, oviduct, and a small portion of the uterine horn and place in the dish containing PBS from one female at a time (see Note 12). 3. Repeat for all females. 4. Working under low magnification, identify the ampulla. Cumulus enclosed oocytes should be easily visible within the ampulla of the oviduct. Using a beveled hypodermic needle, open the ampulla to release the cumulus-enclosed oocytes. 5. Repeat until all oocytes have been released. 6. Using a 1 mL pipette (or a wide bore pipette tip) transfer the cumulus enclosed oocytes to the dish containing MVF

34

Wiles and Taft

medium and thawed sperm (∼10 ␮L), transferring as little medium as possible. 7. Incubate at 37ºC for 4 h under mixed gas. 8. Using a finely drawn glass pipette with a diameter slightly larger than an oocyte. Wash the oocytes through the 150 ␮L media drops to remove cumulus cells and sperm. 9. Culture overnight at 37º C under mixed gas (see Note 13). 10. Count and evaluate embryos the following morning. Embryos can now be cultured, transferred to a pseudopregnant animal, or cryopreserved (see Notes 14 and 15).

4. Notes 1. Variations in sperm quality among individual males within a strain are common. 2. This is a nearly saturated solution, heating the solution makes it easier to get the raffinose into solution. 3. The solution will be opaque after the addition of the skim milk, centrifugation at room temperature (13,000 × g for 15 min) is recommended. 4. Addition of MTG is recommended immediately prior to use. Alternatively, it can be added in advance and the solution stored at −80ºC for up to 3 months. Solutions containing MTG should not be stored at 4ºC for more than a few days. MTG is viscous and needs to be pipetted carefully. Making an MTG stock solution that is added to the media helps reduce the likelihood of errors. MTG diluted stock solution should be used only on the day it is made. 5. This solution can be stored for at least 6 months at −80ºC without MTG or up to 3 months −80ºC with MTG. 6. Water baths are a common source of bacterial contamination and also liquid nitrogen is not sterile. Straws should be carefully wiped to remove any moisture from the outside of the straw prior to cutting the end off before dispensing sperm to reduce the risk of contaminating the IVF. 7. It is essential that during the transfer from liquid nitrogen to long-term storage to handle the cassettes rapidly thus preventing any warming. 8. Typical doses are in the range of 2.5–5 i.u. per mouse. Optimal dose varies by strain, age, and weight of the mouse. Extending the oocyte collection window beyond 14 h post-hCG may reduce fertilization rates and compromise embryo quality due to oocyte aging.

Sophisticated Mouse

35

9. COOKS Mouse Vitro Fert is similar to Human Tubal Fluid media reported by Quinn (18, 19). 10. Oil can be washed and filtered. Oil should be stored in a dark cool place. 11. The response to superovulation is highly strain dependent (see 17) and some strains appear to be entirely refractory to superovulation. 12. In order to reduce the risk of changes in temperature and pH, oocyte collection should take no more than 5 min from euthanasia to oocyte collection (i.e., practice). Typically cervical dislocation is used to prevent possible exposure to agents that may affect oocyte or embryo quality and to reduce the time from euthanasia to oocyte collection. 13. The use of a low O2 culture environment may not improve fertilization rate, but appears to improve embryo quality (20, 21). 14. The laboratory environment can have a significant effect on the outcome of IVF (22, 23). Materials that release volatile organic compounds (VOC), cleaning/sanitizing agents, such as bleach and floor waxes, should be avoided. 15. Prior to embryo transfer, embryos should be washed following the IETS protocol if the sperm or oocytes were collected from animals with an unknown or unacceptable health status (24).

References 1. Taft, R. A., Davisson, M. and Wiles, M. V. (2006) Know thy mouse. Trends Genet 22, 649–653. 2. Doetschman, T., Gregg, R. G., Maeda, N., Hooper, M. L., Melton, D. W., Thompson, S., et al. (1987) Targetted correction of a mutant HPRT gene in mouse embryonic stem cells. Nature 330, 576–578. 3. Thomas, K. R., Folger, K. R. and Capecchi, M. R. (1986) High frequency targeting of genes to specific sites in the mammalian genome. Cell 44, 419–428. 4. Waterston, R., Lindblad-Toh, K., Birney, E., Rogers, J., Abril, J., Agarwal, P., et al. (2002) Initial sequencing and comparative analysis of the mouse genome. Nature 420, 520–562. 5. Bogue, M. A., Grubb, S. C., Maddatu, T. P. and Bult, C. J. (2007) Mouse phenome database (MPD). Nucl Acids Res 35, D643–D649.

6. Darwin, C. (1859) On the origin of species by means of natural selection, 1st ed. John Murray, London. 7. Cook Jr, E. H. and Scherer, S. W. (2008) Copy-number variations associated with neuropsychiatric conditions. Nature 455, 919–923. 8. Iafrate, A. J., Feuk, L., Rivera, M. N., Listewnik, M. L., Donahoe, P. K., Qi, Y., et al. (2004) Detection of large-scale variation in the human genome. Nat Genet 36, 949–951. 9. Sincell, M. (2001) Houston flood: research toll is heavy in time and money. Science 293, 589. 10. Fortier, A. L., Lopes, F. L., Darri¨ carrEre, N., Martel, J. and Trasler, J. M. (2008) Superovulation alters the expression of imprinted genes in the midgestation mouse placenta. Hum Mol Genet 17, 1653–1665.

36

Wiles and Taft

11. Wang, Y., Ock, S. A. and Chian, R. C. (2006) Effect of gonadotropin stimulation on mouse oocyte quality and subsequent embryonic development in vitro. Reprod BioMed Online 12, 304–314. 12. Ostermeier, G. C., Wiles, M. V., Farley, J. S. and Taft, R. A. (2008) Conserving, distributing and managing genetically modified mouse lines by sperm cryopreservation. PLoS ONE 3. 13. Whittingham, D. G. (1986) Principles of embryo preservation, (Ashwood-Smith, M. J., Farrant, J., eds.). Low Temperature Preservation in Medicine and Biology. Pitman Medical, Tunbridge Wells, pp. 65–83. 14. Lyon, M. F. (1981) Sensitivity of various germ-cell stages to environmental mutagens. Mutat Res 87, 323–345. 15. Tomlinson, M. (2008) Risk management in cryopreservation associated with assisted reproduction. Cryo Lett 29, 165–174. 16. International Society for, B. and Environmental, R. (2008) Best practices for repositories: Collection, storage, distribution and retrieval of biological materials for research. Cell Preserv Technol 6, 3–58. 17. Byers, S. L., Payson, S. J. and Taft, R. A. (2006) Performance of ten inbred mouse strains following assisted reproductive technologies (ARTs). Theriogenology 65, 1716–1726. 18. Quinn, P., Kerin, J. F. and Warnes, G. M. (1985) Improved pregnancy rate in human in vitro fertilization with the use of a medium

19.

20.

21.

22.

23.

24.

based on the composition of human tubal fluid. Fertil Steril 44, 493–498. Quinn, P. (1995) Enhanced results in mouse and human embryo culture using a modified human tubal fluid medium lacking glucose and phosphate. J Assist Reprod Genet 12, 97–105. Adam, A. A., Takahashi, Y., Katagiri, S. and Nagano, M. (2004) Effects of oxygen tension in the gas atmosphere during in vitro maturation, in vitro fertilization and in vitro culture on the efficiency of in vitro production of mouse embryos. Jpn J Vet Res 52, 77–84. Dumoulin, J. C., Vanvuchelen, R. C., Land, J. A., Pieters, M. H., Geraedts, J. P. and Evers, J. L. (1995) Effect of oxygen concentration on in vitro fertilization and embryo culture in the human and the mouse. Fertil Steril 63, 115–119. Cohen, J., Gilligan, A., Esposito, W., Schimmel, T. and Dale, B. (1997) Ambient air and its potential effects on conception in vitro. Hum Reprod 12, 1742–1749. Hall, J., Gilligan, A., Schimmel, T., Cecchi, M. and Cohen, J. (1998) The origin, effects and control of air pollution in laboratories used for human embryo culture. Hum Reprod 13 Suppl 4, 146–155. Stringfellow, D. A. and Seidel, S. M. (eds.) (1998) Manual of the International Embryo Transfer Society: procedural guide and general information for the use of embryo transfer technology, emphasizing sanitary procedures, 3rd ed. International Embryo Transfer Society, Illinois.

Chapter 3 Genetically Engineered Mouse Models in Drug Discovery Research Rosalba Sacca, Sandra J. Engle, Wenning Qin, Jeffrey L. Stock, and John D. McNeish Abstract Genetically modified mouse models have been proven to be a powerful tool in drug discovery. The ability to genetically modify the mouse genome by removing or replacing a specific gene has enhanced our ability to identify and validate target genes of interest. In addition, many human diseases can be mimicked in the mouse and signaling pathways have been shown to be conserved. In spite of these advantages the technology has limitations. In transgenic animals there may be significant heterogeneity among different founders. In knock-out animals the predicted phenotypes are not always readily observed and occasionally a completely novel and unexpected phenotype emerges. To address the latter and ensure that a deep knowledge of the target of interest is obtained, we have developed a comprehensive phenotyping program which has identified novel phenotypes as well as any potential safety concerns which may be associated with a particular target. Finally we continue to explore innovative technologies as they become available such as RNAi for temporal and spatial gene knock-down and humanized models that may better simulate human disease states. Key words: Drug discovery, knock-out, transgenic, phenotype, humanized mice, knock-in.

1. Introduction The discovery of new drug candidates can be a daunting effort considering the high attrition rate which results in only ∼10% of the compounds tested in clinical trials translating into a successful drug. With the cost of launching a new medicine approaching $1 billion, investment in technologies that impact the assessment of targets and lead compounds early in the drug discovery process is essential in reducing the high attrition rate. G. Proetzel, M.V. Wiles (eds.), Mouse Models for Drug Discovery, Methods in Molecular Biology 602, DOI 10.1007/978-1-60761-058-8 3, © Humana Press, a part of Springer Science+Business Media, LLC 2010

37

38

Sacca et al.

Transgenic technologies have become an important tool in influencing decision making by providing resources that play a key role in target identification, validation of function, as well as in vivo models for providing confidence in the efficacy and safety of the novel therapeutic (genetic versus chemical selectivity and toxicity). The mouse is by far the most widely used genetically modified animal in drug discovery. The availability of extensive information on mouse genetics, advances in molecular biology allowing precise manipulation of the mouse genome, and miniaturization of phenotyping technologies to assess CNS behaviors as well as tracking metabolic changes (1–3) have clearly contributed to establishing genetically modified mice (GeMM) as the single most valuable in vivo resource to modern drug discovery. Many human diseases can be mimicked in the mouse and signaling pathways have been shown to be conserved. Loss of function has been the prominent approach to understanding gene function across numerous laboratory species. The ability to inactivate or modify mammalian genes by homologous recombination in mouse embryonic stem (mES) cells has become a central tool in understanding gene function in association with potential therapeutic targets. Retrospective evaluation of murine knock-out (KO) phenotypes was shown to have a close correlation with the therapeutic effects on the targets of the top 100 selling drugs (4) clearly demonstrating the value of applying KO mice in the validation of novel targets identified through the sequencing of the human genome. Knock-in (KI) technology allowing pre-planned replacement mutations in the endogenous murine alleles has also been used extensively. These replacement modifications range from the change of single base pairs to the replacement of the complete murine gene (5,6). KI technology has allowed the generation of better predictive “humanized” animal models to assess the effect of the compound on the human target and also serve as important in vivo models for screening lead chemical material that is highly species selective. This latter example is often the case with important drug-able target gene families such as G-protein coupled receptors (GPCR). In this chapter, we provide examples of how investigators at Pfizer Inc. are currently utilizing GeMM to advance drug discovery programs, with emphasis on the specific strengths and limitations of the various technologies. Here, we will also present our comprehensive phenotyping program by which KO mice are assessed across various disease areas as a way to maximize the value obtained from these in vivo models. Finally, our current efforts in the generation of human–mouse chimeras as tools to address compound specificity and safety will be described.

Genetically Engineered Mouse Models

2. Applications of Genetically Modified Animals

2.1. Transgenics

39

The use of GeMM by the pharmaceutical industry as a whole has increased substantially over the past two decades. Many factors have influenced this rise in GeMM use within Pfizer Inc. and other companies including a better understanding by scientists of how and when to apply GeMM to discovery programs. Technology improvements including the availability of genomic sequence, PCR genotyping, inbred mouse ES cell lines, karyotype analysis, tetraploid complementation (7,8), and laser-assisted injection of eight-cell embryos (9), all have contributed to increasing the quality of the GeMM models, while decreasing development timelines. Shorter timelines provide models more in sync with drug discovery timelines thus increasing their potential to impact decision making. Arguably, the advancement in molecular biology using Red/ET Recombineering (10) that allows for rapid genetic engineering of highly sophisticated transgenic DNA vectors has had a significant impact in the reduction of GeMM development timelines. Recombineering makes nearly any genetic modification a realistic undertaking by allowing the development of the most appropriate GeMM model based on the needs of a drug discovery program. Greater accessibility to GeMM models has also been a factor for their increased use in drug discovery. In addition to internal efforts, Pfizer Inc. also uses external resources to obtain GeMM models including academic institutions The Jackson Laboratories Induced Mutant Resource (Bar Harbor, ME.), contract research organizations, such as Xenogen Corporation (acquired by Caliper Life Sciences; Alameda, CA.), and private collaborations with companies such as Lexicon Pharmaceuticals, Inc. (The Woodlands, TX.) and Deltagen (San Mateo, CA.). Transgenic mice created by the microinjection of purified linear DNA into one-celled, pronuclear stage embryos were first described by Gordon and Ruddle in 1980 (11). By 1982, Brinster and Palmiter demonstrated that precise DNA constructs could be developed that allowed tissue-specific transgene expression (12) and ushered in the era of gain of function modification of the mouse genome. To this day, the techniques used in the development of pronuclear transgenic mice are similar to those described in 1980. The methodologies and technical improvements for the generation of transgenic mice have been previously reviewed (13). In the early 1990s, Pfizer Inc. scientists began to develop pronuclear transgenic mice not only to validate targets and pathways but also to develop in vivo models of human pathophysiology and disease that recapitulate clinical conditions for the testing of novel therapeutic strategies.

40

Sacca et al.

An early, yet innovative approach to validating therapeutically relevant pathways using pronuclear transgenic mice was the expression of human Glucose Transporter 4 (GLUT4) gene in an inbred model of diabetes (db/db). These results demonstrated that expression of the GLUT4 transgene resulted in the alleviation of insulin resistance and significant improvement in glycemic control (14). A key area of unmet medical need is the identification of new therapies for Alzheimer’s disease (AD). AD is characterized by the deposit of neurofibrillary tangles and amyloid plaques in the brain. However, discovery of new medicines for AD is hindered by the lack of in vivo models that spontaneously display these pathologic hallmarks of the disease. Probably no area of transgenic disease model development has been more rigorously pursued than AD models (15). An important example is the Tg2576 transgenic mouse (16), one of the most widely used Alzheimer’s disease model across the world. This transgenic model of AD develops amyloid plaques by 9–11 months. To shorten the timeframe for plaque development a second transgenic was created using the neuron-specific Thy-1 promoter expressing a human presenilin G384A mutation, associated with early onset AD (17). Mice harboring both the Tg2576 and the G384A transgenes develop plaque as early as 6 months. Another example of a GeMM transgenic model displaying clinical like conditions including cardiac hypertrophy, fibrosis, and heart failure is the ␣-myosin heavy chain promoter 11␤hydroxysteroid dehydrogenase type 2 transgenic mouse (18). Eplerenone (INSPRATM , Pfizer Inc.), a selective aldosterone blocker, was tested in this model and proved to ameliorate the phenotype. This model revealed the deleterious consequences of inappropriate mineralocorticoid receptor activation in the heart and supported the notion that aldosterone blockade may provide additional therapeutic benefit in the treatment of heart failure. Beyond their use as models of human disease, transgenics are widely used as GeMM “tool” lines. Tissue-specific and/or ubiquitous Cre recombinase transgenic lines are commonly used for conditional knock-outs and excision of floxed antibiotic selection cassettes. When temporal transgene expression is essential, the tetracycline-inducible expression system has been utilized to study the acute effects of transgene expression to circumvent developmental compensation or lethality due to transgene expression during development. The development and application of inducible and/or conditional transgenic mice have previously been reviewed (19,20). Transgenic models have a broad spectrum of utilities across drug discovery, but they also have their limitations. Because transgene integration is random, occasionally the presence of multiple insertion sites, varied expression levels, and tissue distribution

Genetically Engineered Mouse Models

41

results in significant heterogeneity among founder lines and subsequent generations. This variability results in the necessity to thoroughly evaluate and characterize multiple mouse lines and although rare, the transgene integration may also disrupt an endogenous gene resulting in insertional mutagenesis. In addition, the expression of transgenes is at times reduced or lost after many generations of breeding, presumably due to epigenetic modification of the inserted transgenic DNA sequences (21). This can be avoided by cryopreservation of important transgenic mouse lines in their earliest generations. 2.2. Knock-outs

Since the isolation of mES cells in 1985 (22) and the subsequent ability to direct pre-planned mutagenesis of the murine genome by homologous recombination in the mES cells (23–25), investigators have been using this seminal technology to address all aspects of mammalian biology. Over the last 20 years, thousands of KO mice have been developed worldwide providing drug discovery scientists’ access to this critical resource to investigate the role of potential drug targets. KO mice have become an invaluable tool for determining target function, selectivity, and potential toxicity liabilities (26). KO mice within Pfizer Inc. have been utilized numerous times to validate or nullify hypotheses regarding confidence in a novel gene’s rationale as a potential therapeutic target. To improve the decision making regarding target validation using KO mice, we developed mES cells from the DBA/1lacJ mouse strain, an inbred mouse model of collagen-induced arthritis. This DBA mES cell line enabled us to evaluate the effect of single gene deletions on the resulting inflammatory phenotype. For example, KO mice for the 5-lipoxygenase-activating protein (FLAP) gene resulted in marked reduction in the collagen-induced arthritis compared to inbred, control littermates (Fig. 3.1) and provided in vivo data implicating FLAP and the leukotriene pathway in inflammation (27). The KO mouse provides an in vivo model with 100% inhibition of the target to understand differences in selectivity between genetic and pharmacological inhibition. An excellent example of which is the phosphodiesterase 10a (Pde10a) KO mouse. PDE10a emerged as a potential target for the treatment of psychosis based on conserved mouse and human striatal brain expression data, as well as activity of inhibitors in rodent models predictive of clinical antipsychotic activity. Efficacy of these inhibitors was absent in Pde10a KO mice, confirming that PDE10a inhibition underlies the antipsychotic-like activity of these compounds (28,29). KO mice are useful as models for important clinical conditions such as is the case with the Apolipoprotein E (ApoE) KO, a robust atherosclerosis model (30,31). Not only has the ApoE KO mouse been extensively used as a single GeMM disease model but also

42

Sacca et al.

Fig. 3.1. Collagen-induced arthritis in FLAP-deficient mice. DBA/1 mice were immunized with chick type II collagen on day 0 and 21. IL-1 was administered subcutaneously on days 45 and 46 to trigger an arthritic flare. Disease severity was scored by observation of the paws for redness and swelling. Open circles, +/+ mice; closed triangles, +/− mice; open squares, −/− mice. Results are mean ± SEM, n = three experiments.

additional genetic modifications have been introduced into the ApoE null background for evaluating the role of these new genes in atherosclerosis pathophysiology. KO mice have also been used to investigate whether a toxicity finding is the result of the chemical lead or has a mechanistic basis. 5-Lipoxygenase KO mice were used to prove that a chemical series of leukotriene B4 antagonists was responsible for the induction of hepatic enzymes, not an associated phenotype of leukotriene-deficient KO mice (32). Although KO mice are extremely useful as in vivo models for drug discovery scientists, they are not without their limitations. Because the gene inactivation is produced in the zygote, embryonic lethality is a reality for genes essential in development. Redundancy or compensatory effects due to altered expression of unmodified gene family members may also occur, all of which can complicate interpretation of the KO phenotype. The use of a tissue-specific conditional KO is one way to circumvent these issues, and as discussed earlier, many tissue-specific Cre tool GeMM lines are available for generating this type of model. Another option is to create an inducible KO using the

Genetically Engineered Mouse Models

43

tamoxifen-inducible Cre-ER system (33,34) allowing for temporal control of gene knock-down in all tissues of adult animals. This approach provides an opportunity for drug discoverer’s to more robustly reproduce the effect of a therapeutic molecule. However, these multiple genetic modifications substantially increase the time and costs for GeMM model delivery, making these complex approaches a consideration when initiating a new project. 2.3. Knock-In

Just as KO mice have been utilized to address a variety of needs for drug discovery scientists, KI models have also had many different applications to drug discovery programs. KI modifications can range from single base changes to complete replacement of a murine gene with an orthologous human gene. The need for more predictive in vivo models to evaluate a lead compound’s efficacy, pharmacokinetic profile, and toxicological properties has made humanized KI mice an important tool. For example, chemokine receptor-1 (CCR1) humanized KI mice were used to assess the activity of human-specific CCR1 antagonists and their ability to modulate inflammatory responses (35). In this model the entire mouse CCR1 gene was replaced with the CCR1 human ortholog. A complete gene replacement is not always necessary for the “humanization” of a target gene as illustrated by a model generated for the Thrombopoietin receptor (TPOr) program. Pfizer Inc. identified agonists for the thrombopoietin receptor (TPOr) that were shown to have selective human receptor binding. In vitro mouse/human receptor constructs identified a region of the protein responsible for this species specificity. A humanized TPOr KI mouse was developed in which only mouse exons 8-10, the region responsible for the specificity, were replaced with human exons 8-10. When treated with TPOr agonists, TPOr KI mice demonstrated a dosedependent increase in platelet numbers as compared to their wildtype littermates (Fig. 3.2, manuscript in preparation). Knock-in models have also been used to introduce point mutations to create mouse models of human disease alleles (36). Point mutant KI mice can also be used to inactivate essential regions of a protein, such as a catalytic kinase domain, or to alter a drug-binding site within the target protein. The ␣2␦1 R217A KI mouse was used to demonstrate the mechanism of action for the analgesic effects of gabapentin and pregabalin, showing that it is mediated by an interaction with the ␣2␦1 subunit, specifically the arginine amino acid at position 217 (37). Another important application of KI mice is the creation of fluorescent and bioluminescent reporter models which have a broad spectrum of utility. These models can be developed to tag specific genes, cell lineages, and biological pathways. In this section, we have tried to review a broad crosssection of genetically engineered mouse models and some specific

44

Sacca et al.

Fig. 3.2. The graph compares TPOr knock-in and wild-type responses to PF-356316902. Mice were dosed with PF-3563169-02 s.c. for two consecutive days and blood was analyzed on day 5. Each group is represented by five mice.

applications aligned with real drug discovery programs from the hundreds of possible examples. Clearly, the technologies for manipulating the murine genome allow for highly innovative genomic modifications limited only by one’s imagination. Most importantly, understanding the needs of a drug discovery project will help determine the appropriate genetically modified mouse to develop, as no one model type is appropriate for every situation. For example, a standard KO mouse may provide relatively quick and valuable information about mechanisms and pathways for a novel target with little background literature. On the other hand, a conditional or inducible KO may be essential to understanding the biology of a more mature target with known embryonic lethality or complex, interdependent phenotypes. Humanized knock-in mice are now becoming essential for developing biotherapeutics. Once an understanding of the needs of the project is attained and the most appropriate GeMM model is developed, then the focus is on assessing its phenotype.

3. Comprehensive Phenotyping of Genetically Modified Mouse Models

The greatest challenge to understanding targets in drug discovery is assigning in vivo function to the genes of interest and to do so in an efficient and cost-effective manner. As discussed, KO mice provide a powerful approach for defining gene function in the context of mammalian physiology (4,38) and are usually generated in response to a particular hypothesis. When the KO model confirms the predicted hypothesis, there is often little incentive

Genetically Engineered Mouse Models

45

to evaluate the model in additional assays in the rush to progress a drug discovery program. When evaluation of the KO mouse does not confirm the predicted phenotype or the mouse has an unexpected or unclear phenotype, the model may be set aside. In our experience, this is relatively common phenomenon as almost half of all KO mice generated do not present with the expected phenotype. In the past, KO mice are rarely examined in diverse physiological assays to comprehensively define the function of the gene, therefore failing to capitalize on the significant investment made in the target or the generation and initial characterization of the KO model. Phenotype Pfinder was developed as a tool for systematically phenotyping KO mice to maximize on our investment in the KO model and underlying gene target. At its inception Phenotype Pfinder was envisioned to be a panel of assays that combined speed and efficiency to deliver high-quality assessment of a broad spectrum of physiological parameters that would promote drug discovery. Individual assays (Table 3.1) were required to meet several criteria for inclusion in the program. All assays were expected to be decision making for the respective therapeutic area such that when the model returns a statistically significant finding, the therapeutic area would followup on the finding. On the other hand, if the results showed no significant difference between wild-type (WT) and KO mice, scientists were expected to have enough confidence in the results to discontinue efforts in that area. Biomedical statisticians were engaged to ensure that each assay used enough mice to identify statistically significant differences between WT and KO mice and that assays used the appropriate age and gender of mice. To ensure that each assay can detect either improved or impaired function, assays are required to be validated with positive and negative controls using either pharmacological or genetic validation strategies. Additionally, assays were tested for reproducibility and robustness across the variety of mouse background strains (C57BL/6, DBA/1lacJ, 129.B6) commonly used to generate KO models. Only those assays that were independent of the background strain were included. Highly labor-intensive assays (for example, those requiring surgical intervention) or time-intensive assays (for example, aging paradigms or oncology assays) were excluded from the program in order to manage cost and time to delivery of data. To further manage cost and efficiency, all assays were consolidated into a single platform at a single site and performed by a dedicated group of in vivo biologists. Each assay was validated independently and followed by testing of small combinations of assays. Those combinations for which results differed significantly from na¨ıve control data were eliminated. Successful combinations were assembled into larger panels and revalidated. Panels were combined and rigorously repeated to ensure

46

Sacca et al.

Table 3.1 Assays evaluated in Phenotype Pfinder Immunology/inflammation

Thrombosis

B-cell proliferation

Collagen-induced platelet aggregation

CD4 T-cell proliferation

Ex vivo coagulation

Cytokine production by CD4 T cells

Tail transection bleeding time

Delayed-type hypersensitivity

Thrombin–antithrombin III complex

FACS analysis of splenocytes

Frailty and osteoporosis

In vivo antibody production

Body composition by DEXA

KRN serum-induced arthritis

Calcein labeling for bone histopathology

LPS induced inflammation

Excised bone DEXA

Pulmonary inflammation

Rotarod

Thioglycollate-induced monocyte infiltration

Selected muscle weights

Gastrointestinal function

Dermatology

Colon:body weight ratio

Water retention test for sebum production

Longitudinal body weight assessment

Pain

Diabetes and obesity

Formalin-induced pain

Body composition by DEXA

Hot plate

Corticosterone level

Von Frey

Food intake

Neurodegeneration

Free fatty acids

Neurodegeneration: excitotoxicity

Leptin level

Neurodegeneration: lactacystin

Long-term monitoring in comprehensive cage monitoring system for metabolism and activity

Neurodegeneration: neurite outgrowth

Pre- and post-high-fat diet

Neurodegeneration: vitamin K

Adiponectin levels

Psycotherapeutics

Cholesterol distribution

Elevated plus maze

Insulin level

Irwin behavior test

Oral glucose tolerance test

Open field

Total cholesterol level

Pentylenetetrazole-induced seizure

Phospholipids B level

Startle and pre-pulse inhibition

Body composition by DEXA

Tail suspension

Regional fat depot assessment by ␮CT

Sexual health

Weekly body weight

Female contact sexual behavior

Tissue glycogen level

Forced erection test

Cardiovascular function

Male contact sex behavior (continued)

Genetically Engineered Mouse Models

47

Table 3.1 (continued) Blood pressure

Male fertility

Heart rate

Testosterone

Angiotensin II and aldosterone levels

Cross-therapeutic assays

Bladder and renal function

Automated blood chemistries

Diuretic-induced stress test

Hematology

Urine chemistries

that upstream, less invasive assays did not interfere with or alter the response to the increasingly more invasive downstream assays. The comprehensive Phenotype Pfinder protocol evaluates 50 KO and 50 WT mice through a sequential combination of ∼50 assays in 15 disease areas in 17 weeks. If done individually and not at a centralized facility, these assays would consume more than 800 mice and require substantially more time and money for breeding, genotyping, and labor. To date, more than 100 unique KO mouse lines have been characterized in the Phenotype Pfinder protocol and have impacted numerous drug discovery projects. Raw data generated by the Phenotype Pfinder protocol is maintained in a dedicated database. The database automatically graphs the data, generates statistical analyses of assays, and flags those assays which show a statistically significant difference between WT and KO mice. The data can be searched and viewed by either specific KO line or assay. The database also offers a tool for analysis of historically accumulated control data sorted by genetic background. Information generated by Phenotype Pfinder has been used in a variety of ways. At its core, Phenotype Pfinder tests the hypothesis that a gene may be a previously unsuspected target in novel therapeutic areas and ensures that we would identify the serendipitous finding. For example, the previously mentioned Pde10a KO mice were generated at the request of the psychotherapeutics program. The KO line was evaluated in Phenotype Pfinder as part of a strategic decision to target specific gene families. Unexpectedly, the KO mice were completely resistant to weight gain on a highfat diet . Furthermore the KO mice showed increase basal energy expenditure and no change in food intake compared to WT mice. When WT mice on a high-fat diet were treated with a Pde10aspecific inhibitor, they consumed as much food as the vehicletreated cohort but failed to gain weight. KO mice treated with the Pde10a inhibitor also consumed as much food as the WT cohort and showed no additional weight loss thus strongly suggesting that the failure to gain weight on a high-fat diet was specifically

48

Sacca et al.

related to the loss of the Pde10a. By identifying a relevant phenotype in a different therapeutic area, Phenotype Pfinder allowed scientists to leverage the investment in knowledge, reagents, and tool compounds to rapidly explore a new drug discovery program. Conversely, the broad phenotypic picture afforded by Phenotype Pfinder can minimize investment in targets with previously unsuspected liabilities. HCN1 is a hyperpolarization-activated cation channel. It is known to play a role in the regulation of cell excitability and contributes to spontaneous rhythmic activity in both heart and brain. Data suggested that HCN1 may be a target for epilepsy and neuropathic pain. As hypothesized, Hcn1 KO mice display increased sensitivity to pentylenetetrazole (PTZ)-induced clonus seizures (39). Hcn1 KO mice also showed significant impairments in gait, body posture, coordination, and locomotion as well as changes in body composition, metabolism, bladder function, heart rate, and blood coagulation. The constellation and severity of the phenotypes reduced the attractiveness of the target. Phenotype Pfinder data can also build strategic information for hypothesis generation. G-protein coupled receptors (GPRs) are a class of protein that historically has been very amenable to drug activation or inhibition. Prior to the identification of a ligand for a GPR, so-called orphan status, it is often difficult to ascertain the GPR’s specific function. In 2005 when Gpr35 KO mice were evaluated in Phenotype Pfinder, the “orphan” GPR had been associated with expression in gastric cancer cells, a chromosomal deletion found in Albright hereditary osteodystrophy-like (AOH) syndrome and type II diabetes in a Mexican-American population (40–42). Male Gpr35 KO mice consumed less food per unit body weight, had higher leptin levels, a lower metabolic rate, and tended to be heavier than WT mice. These phenotypes support the association of GPR35 with an obesity and diabetes phenotype. No bone phenotypes were detected suggesting that deletion of GPR35 is not responsible for the bone phenotypes in AOH syndrome. Because Phenotype Pfinder is time constrained, no gastric cancers were detected. Subsequent work has identified kynurenic acid, an endogenous metabolite of tryptophan, and Zaprinast, a well-known cyclic guanosine monophosphate-specific phosphodiesterase inhibitor, as ligands for GPR35 (43,44). Recent work has shown that levels of kynurenic acid in rat intestine are sufficient to affect GPR35 thus supporting a link between GPR35, gastrointestinal function and a potential role in obesity and/or diabetes (45). Occasionally, Phenotype Pfinder can identify new pharmacological models. Map3K11 KO mice were reported to be healthy and viable (46). The only reported phenotype in the literature was a defect in the ability of embryonic fibroblasts from the

Genetically Engineered Mouse Models

49

Map3K11 KO to activate JNK in response to TNF␣. Studies using a Map3k11 inhibitor suggested that inhibition of JNK activity improved neuron survival and growth (47). Phenotype Pfinder studies showed no difference between WT and KO neurons in their response to glutamate or vitamin K-induced neurotoxicity. However, both before and after the Map3K11 KO mice were fed a high-fat diet, they showed statistically elevated systolic blood pressure compared to WT mice. When treated with an FDAapproved angiotensin I converting enzyme (ACE) inhibitor, the blood pressure of the Map3k11 KO mice fell to WT control levels (O. Buiakova, personal communication). This presents a new mouse model of high blood pressure which can be evaluated with pharmacological agents to determine clinical relevance. Systematically evaluating KO mice across a panel of diverse assays provides a cost-effective, time constrained method for maximizing the collection of information. It can lead to the identification of new phenotypes and therapeutic indications, confirm literature-reported phenotypes, and hypothesized effects, highlight potential target related safety effects so that they can be monitored or managed and draw attention to potential clinical biomarkers. Assembly of the data into an easily accessible, searchable, dedicated database allows the information to be continuously referenced and evaluated as new information becomes available. With the ever-increasing cost of drug discovery programs, it is essential to identify the most promising targets early, to terminate nonviable project quickly, and to exploit all available therapeutic applications around a single target. The identification of one high value, novel therapeutic application can justify the entire expense of a comprehensive KO mouse phenotyping program. It maximizes the investment in any one model and target and promotes better in vivo biology.

4. Future Directions of Genetically Engineered Models in Drug Discovery

With 99% of the mouse genes having homologues in humans (48), and the ability to readily manipulate the mouse genome, mice rightfully have been an important model organism to understand human physiology and diseases. However, the protein sequences of these homologues may differ from their human counterparts such that they do not bind compounds or biotherapeutic agents targeted against the human counterparts. In addition, mice differ from humans in many aspects, including reproduction, olfaction, behavior, metabolism, and immunity. Thus, to robustly model the human condition, it would be desirable to, at times, replace the mouse homologue with the entire human gene,

50

Sacca et al.

the entire gene family, or an entire organ/system. Some specific examples of variation of “humanization” will be described here. Deleting the mouse genes and replacing them with a human gene family residing over a million base pairs of genomic sequence has been accomplished and is exemplified by the creation of mouse models carrying the human immunoglobulin genes. The first fully human antibody directed against epidermal growth factor receptor (EGFR), panitumumab (VECTIBIX, Amgen Inc., Thousand Oaks, CA.), was generated by utilizing the XenoMouse of Abgenix Incorporated (Fremont, CA.) and recently won regulatory approval as treatment for advanced colorectal cancer (49). Several other important candidate gene families for humanization are the MHC loci for a human adaptive immune response for vaccine testing and the cytochorme P450 genes for drug metabolism. Single gene replacements from these gene families have been reported (50,51). In the last 20 years, naturally occurring mouse mutants were identified and genetically engineered mouse models created that allow the design and creation of sophisticated immunocompromised murine hosts that can support chimerism with cells from other species (52). Employing this new generation of immune compromised host mice (53), it is now routinely possible to achieve up to 70% chimerism of human CD45+ cells in the bone marrow and peripheral blood from these bone marrow engrafted models (54,55). Although these achievements are encouraging, major limitations must be overcome before the model can be used for testing and developing human vaccines. Specific technology improvements that are needed include humanized MHC genes to improve human T-cell selection and maturation in mouse thymus, retaining the Peyer’s patches and lymph nodes in these models to allow antigen presentation that normally occurs in these primary lymphoid organs and human cytokines critical for differentiation and maturation of human immune cells in a mouse host. These are daunting tasks but if successful will considerably extend the use of animal models for drug testing. In drug discovery, it is important to understand how a drug is metabolized in humans. In both the urokinase-type plasminogen activator (uPA) transgenic (56) and the fumarylacetoacetate hydrolase (Fah) knock-out models (57) bred onto an immunocompromised mouse background, liver engraftment in the resulting chimera can be achieved up to 90% with human hepatocytes. In addition to human hepatocytes becoming integrated into the recipient liver, they also express human liver-specific genes (CYP1A2 and 3A4) and respond to prototypical inducers of the cytochrome P450 genes. However, these models currently rely on organ donor or cadaver for the supply of human hepatocytes. This presents a major limitation in reproducibility in these human liver

Genetically Engineered Mouse Models

51

chimera models, as the donor hepatocytes likely differ in genetic background, age, environmental exposure during the course of life, and whether isolated fresh or revived from cryopreservation (57). In this respect, human embryonic stem cells or induced pluripotent stem (iPS) cells hold the promise to derive an unlimited supply of genetically homogeneous population of hepatocytes for engraftment (58). The goal for cell therapy is to restore or replace damaged cells by mobilizing host cells or by providing terminally differentiated cell types for diseases such as type 1 diabetes, Parkinson’s disease, Alzheimer’s disease, myocardial infarction, or muscle dystrophy. In each case, it would be desirable to have a relevant animal model to follow the mobilization or engraftment, migration, and differentiation of the human cells in the target organs in the host. This scheme will likely necessitate an immunodeficient background to avoid xenograft rejection, labeling of the donor cells to allow tracking and an in vivo environment to allow structural and functional integration of these cells. Genetically engineered models, combined with cell and tissue transplantation, will undoubtedly generate the next generation of humanized mice that promises to translate into better preclinical models for drug discovery and regenerative medicine research and development.

5. Conclusion In the decade of the 1980s, a number of powerful and reproducible technologies were described which allowed the precise manipulation of the murine genome. These methods resulted in the evaluation of gene function by classical approaches of gain and loss of function mutagenesis. In addition, homologous recombination technologies allowed for targeted replacement or humanized knock-ins into the murine genome, resulting in the development of human alleles in mouse models. Although, these technologies are of great importance to basic research, drug discoverers soon recognized that the application of transgenic technologies could deliver novel approaches to in vivo target validation and model development aimed at better decisions for the discovery of new medicines. In this chapter, we have selected a few of the many examples of pronuclear transgenic mice and mES cell-derived knock-out and knock-in mice that have been developed at Pfizer Inc. or by other researchers for specific applications from therapeutic idea generation to hypothesis testing. The ability to derive or obtain GeMM has become routine for most academic or applied laboratories; however, the ability to fastidiously evaluate the phenotype of the GeMM is less well refined.

52

Sacca et al.

Here, our efforts to ensure the careful evaluation of the primary KO mouse phenotypes and also maximize the return of the KO mouse investment by comprehensively evaluating for unsuspected phenotypes was described. The central aim of this chapter was to illustrate the types of problem statements and challenges that face investigators, and the application of innovative approaches using transgenic technologies that assist in our decision making. Improvements in technologies have always guided our application of GeMM to drug discovery. In the future, new technologies such as cellular reprogramming may lead to pluripotent, germline competent stem cells that allow genetic modification in other lab species and the expression of RNAi to introduce temporal or spatial gene knock-down that more closely resembles the effect of a drug. References 1. van der Staay, F. J. and Steckler, T. (2001) Behavioural phenotyping of mouse mutants. Behav Brain Res 125, 3–12. 2. Crawley, J. N. (1999) Behavioral phenotyping of transgenic and knockout mice: experimental design and evaluation of general health, sensory functions, motor abilities, and specific behavioral tests. Brain Res 835, 18–26. 3. Doevendans, P. A., Daemen, M. J., de Muinck, E. D. and Smits, J. F. (1998) Cardiovascular phenotyping in mice. Cardiovasc Res 39, 34–49. 4. Zambrowicz, B. P. and Sands, A. T. (2003) Knockouts model the 100 best-selling drugs– will they model the next 100? Nat Rev Drug Discov 2, 38–51. 5. Forlino, A., Porter, F. D., Lee, E. J., Westphal, H. and Marini, J. C. (1999) Use of the Cre/lox recombination system to develop a non-lethal knock-in murine model for osteogenesis imperfecta with an alpha1(I) G349C substitution. Variability in phenotype in BrtlIV mice. J Biol Chem 274, 37923– 37931. 6. Reichardt, H. M., Kaestner, K. H., Tuckermann, J., Kretz, O., Wessely, O., Bock, R., et al. (1998) DNA binding of the glucocorticoid receptor is not essential for survival. Cell 93, 531–541. 7. Nagy, A., Rossant, J., Nagy, R., AbramowNewerly, W. and Roder, J. C. (1993) Derivation of completely cell culture-derived mice from early-passage embryonic stem cells. Proc Natl Acad Sci U S A 90, 8424–8428. 8. Nagy, A., Gocza, E., Diaz, E. M., Prideaux, V. R., Ivanyi, E., Markkula, M., et al. (1990) Embryonic stem cells alone are able to sup-

9.

10.

11.

12.

13. 14.

15.

16.

port fetal development in the mouse. Development 110, 815–821. Poueymirou, W. T., Auerbach, W., Frendewey, D., Hickey, J. F., Escaravage, J. M., Esau, L., et al. (2007) F0 generation mice fully derived from gene-targeted embryonic stem cells allowing immediate phenotypic analyses. Nat Biotechnol 25, 91–99. Zhang, Y., Muyrers, J. P., Testa, G. and Stewart, A. F. (2000) DNA cloning by homologous recombination in Escherichia coli. Nat Biotechnol 18, 1314–1317. Gordon, J. W., Scangos, G. A., Plotkin, D. J., Barbosa, J. A. and Ruddle, F. H. (1980) Genetic transformation of mouse embryos by microinjection of purified DNA. Proc Natl Acad Sci U S A 77, 7380–7384. Palmiter, R. D., Chen, H. Y. and Brinster, R. L. (1982) Differential regulation of metallothionein-thymidine kinase fusion genes in transgenic mice and their offspring. Cell 29, 701–710. Wall, R. J. (2001) Pronuclear microinjection. Cloning Stem Cells 3, 209–220. Gibbs, E. M., Stock, J. L., McCoid, S. C., Stukenbrok, H. A., Pessin, J. E., Stevenson, R. W., et al. (1995) Glycemic improvement in diabetic db/db mice by overexpression of the human insulin-regulatable glucose transporter (GLUT4). J Clin Invest 95, 1512–1518. Higgins, G. A. and Jacobsen, H. (2003) Transgenic mouse models of Alzheimer  s disease: phenotype and application. Behav Pharmacol 14, 419–438. Hsiao, K., Chapman, P., Nilsen, S., Eckman, C., Harigaya, Y., Younkin, S., et al. (1996) Correlative memory deficits, Abeta elevation,

Genetically Engineered Mouse Models

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

and amyloid plaques in transgenic mice. Science 274, 99–102. Cruts, M., Backhovens, H., Wang, S. Y., Van Gassen, G., Theuns, J., De Jonghe, C. D., et al. (1995) Molecular genetic analysis of familial early-onset Alzheimer s disease linked to chromosome 14q24.3. Hum Mol Genet 4, 2363–2371. Qin, W., Rudolph, A. E., Bond, B. R., Rocha, R., Blomme, E. A., Goellner, J. J., et al. (2003) Transgenic model of aldosteronedriven cardiac hypertrophy and heart failure. Circ Res 93, 69–76. Belteki, G., Haigh, J., Kabacs, N., Haigh, K., Sison, K., Costantini, F., et al. (2005) Conditional and inducible transgene expression in mice through the combinatorial use of Cre-mediated recombination and tetracycline induction. Nucleic Acids Res 33, e51. Bockamp, E., Maringer, M., Spangenberg, C., Fees, S., Fraser, S., Eshkind, L., et al. (2002) Of mice and models: improved animal models for biomedical research. Physiol Genomics 11, 115–132. Sutherland, H. G., Kearns, M., Morgan, H. D., Headley, A. P., Morris, C., Martin, D. I., et al. (2000) Reactivation of heritably silenced gene expression in mice. Mamm Genome 11, 347–355. Doetschman, T. C., Eistetter, H., Katz, M., Schmidt, W. and Kemler, R. (1985) The in vitro development of blastocyst-derived embryonic stem cell lines: formation of visceral yolk sac, blood islands and myocardium. J Embryol Exp Morphol 87, 27–45. Koller, B. H., Hagemann, L. J., Doetschman, T., Hagaman, J. R., Huang, S., Williams, P. J., et al. (1989) Germ-line transmission of a planned alteration made in a hypoxanthine phosphoribosyltransferase gene by homologous recombination in embryonic stem cells. Proc Natl Acad Sci U S A 86, 8927–8931. Thomas, K. R. and Capecchi, M. R. (1987) Site-directed mutagenesis by gene targeting in mouse embryo-derived stem cells. Cell 51, 503–512. Doetschman, T., Maeda, N. and Smithies, O. (1988) Targeted mutation of the Hprt gene in mouse embryonic stem cells. Proc Natl Acad Sci U S A 85, 8583–8587. Harris, S. (2001) Transgenic knockouts as part of high-throughput, evidence-based target selection and validation strategies. Drug Discov Today 6, 628–636. Griffiths, R. J., Smith, M. A., Roach, M. L., Stock, J. L., Stam, E. J., Milici, A. J., et al. (1997) Collagen-induced arthritis is reduced in 5-lipoxygenase-activating protein-deficient mice. J Exp Med 185, 1123–1129.

53

28. Siuciak, J. A., Chapin, D. S., Harms, J. F., Lebel, L. A., McCarthy, S. A., Chambers, L., et al. (2006) Inhibition of the striatum-enriched phosphodiesterase PDE10A: a novel approach to the treatment of psychosis. Neuropharmacology 51, 386–396. 29. Siuciak, J. A., McCarthy, S. A., Chapin, D. S., Fujiwara, R. A., James, L. C., Williams, R. D., et al. (2006) Genetic deletion of the striatum-enriched phosphodiesterase PDE10A: evidence for altered striatal function. Neuropharmacology 51, 374–385. 30. Plump, A. S., Smith, J. D., Hayek, T., AaltoSetala, K., Walsh, A., Verstuyft, J. G., et al. (1992) Severe hypercholesterolemia and atherosclerosis in apolipoprotein E-deficient mice created by homologous recombination in ES cells. Cell 71, 343–353. 31. Zhang, S. H., Reddick, R. L., Piedrahita, J. A. and Maeda, N. (1992) Spontaneous hypercholesterolemia and arterial lesions in mice lacking apolipoprotein E. Science 258, 468–471. 32. Beierschmitt, W. P., McNeish, J. D., Griffiths, R. J., Nagahisa, A., Nakane, M. and Amacher, D. E. (2001) Induction of hepatic microsomal drug-metabolizing enzymes by inhibitors of 5-lipoxygenase (5-LO): studies in rats and 5-LO knockout mice. Toxicol Sci 63, 15–21. 33. Seibler, J., Zevnik, B., Kuter-Luks, B., Andreas, S., Kern, H., Hennek, T., et al. (2003) Rapid generation of inducible mouse mutants. Nucl Acids Res 31, e12. 34. Hayashi, S. and McMahon, A. P. (2002) Efficient recombination in diverse tissues by a tamoxifen-inducible form of Cre: a tool for temporally regulated gene activation/inactivation in the mouse. Dev Biol 244, 305–318. 35. Gladue, R. P., Cole, S. H., Roach, M. L., Tylaska, L. A., Nelson, R. T., Shepard, R. M., et al. (2006) The human specific CCR1 antagonist CP-481,715 inhibits cell infiltration and inflammatory responses in human CCR1 transgenic mice. J Immunol 176, 3141–3148. 36. Guo, Q., Fu, W., Sopher, B. L., Miller, M. W., Ware, C. B., Martin, G. M., et al. (1999) Increased vulnerability of hippocampal neurons to excitotoxic necrosis in presenilin-1 mutant knock-in mice. Nat Med 5, 101–106. 37. Bian, F., Li, Z., Offord, J., Davis, M. D., McCormick, J., Taylor, C. P., et al. (2006) Calcium channel alpha2-delta type 1 subunit is the major binding protein for pregabalin in neocortex, hippocampus, amygdala, and spinal cord: an ex vivo autoradiographic study

54

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

Sacca et al. in alpha2-delta type 1 genetically modified mice. Brain Res 1075, 68–80. Zambrowicz, B. P., Turner, C. A. and Sands, A. T. (2003) Predicting drug efficacy: knockouts model pipeline drugs of the pharmaceutical industry. Curr Opin Pharmacol 3, 563–570. Zong, X., Eckert, C., Yuan, H., WahlSchott, C., Abicht, H., Fang, L., et al. (2005) A novel mechanism of modulation of hyperpolarization-activated cyclic nucleotidegated channels by Src kinase. J Biol Chem 280, 34224–34232. Vander Molen, J., Frisse, L. M., Fullerton, S. M., Qian, Y., Del Bosque-Plata, L., Hudson, R. R., et al. (2005) Population genetics of CAPN10 and GPR35: implications for the evolution of type 2 diabetes variants. Am J Hum Genet 76, 548–560. Okumura, S., Baba, H., Kumada, T., Nanmoku, K., Nakajima, H., Nakane, Y., et al. (2004) Cloning of a G-protein-coupled receptor that shows an activity to transform NIH3T3 cells and is expressed in gastric cancer cells. Cancer Sci 95, 131–135. Shrimpton, A. E., Braddock, B. R., Thomson, L. L., Stein, C. K. and Hoo, J. J. (2004) Molecular delineation of deletions on 2q37.3 in three cases with an Albright hereditary osteodystrophy-like phenotype. Clin Genet 66, 537–544. Wang, J., Simonavicius, N., Wu, X., Swaminath, G., Reagan, J., Tian, H., et al. (2006) Kynurenic acid as a ligand for orphan G protein-coupled receptor GPR35. J Biol Chem 281, 22021–22028. Taniguchi, Y., Tonai-Kachi, H. and Shinjo, K. (2006) Zaprinast, a well-known cyclic guanosine monophosphate-specific phosphodiesterase inhibitor, is an agonist for GPR35. FEBS Lett 580, 5003–5008. Kuc, D., Zgrajka, W., Parada-Turska, J., Urbanik-Sypniewska, T. and Turski, W. A. (2008) Micromolar concentration of kynurenic acid in rat small intestine. Amino Acids. Brancho, D., Ventura, J. J., Jaeschke, A., Doran, B., Flavell, R. A. and Davis, R. J. (2005) Role of MLK3 in the regulation of mitogen-activated protein kinase signaling cascades. Mol Cell Biol 25, 3670–3681. Handley, M. E., Rasaiyaah, J., Chain, B. M. and Katz, D. R. (2007) Mixed lineage kinases (MLKs): a role in dendritic cells, inflammation and immunity? Int J Exp Pathol 88, 111–126.

48. Waterston, R. H., Lindblad-Toh, K., Birney, E., Rogers, J., Abril, J. F., Agarwal, P., et al. (2002) Initial sequencing and comparative analysis of the mouse genome. Nature 420, 520–562. 49. Jakobovits, A., Amado, R. G., Yang, X., Roskos, L. and Schwab, G. (2007) From XenoMouse technology to panitumumab, the first fully human antibody product from transgenic mice. Nat Biotechnol 25, 1134–1143. 50. Gonzalez, F. J. and Yu, A. M. (2006) Cytochrome P450 and xenobiotic receptor humanized mice. Annu Rev Pharmacol Toxicol 46, 41–64. 51. Pascolo, S. (2005) HLA class I transgenic mice: development, utilisation and improvement. Expert Opin Biol Ther 5, 919–938. 52. Shultz, L. D., Ishikawa, F. and Greiner, D. L. (2007) Humanized mice in translational biomedical research. Nat Rev Immunol 7, 118–130. 53. Ito, M., Hiramatsu, H., Kobayashi, K., Suzue, K., Kawahata, M., Hioki, K., et al. (2002) NOD/SCID/gamma(c)(null) mouse: an excellent recipient mouse model for engraftment of human cells. Blood 100, 3175–3182. 54. Traggiai, E., Chicha, L., Mazzucchelli, L., Bronz, L., Piffaretti, J. C., Lanzavecchia, A., et al. (2004) Development of a human adaptive immune system in cord blood celltransplanted mice. Science 304, 104–107. 55. Ishikawa, F., Yasukawa, M., Lyons, B., Yoshida, S., Miyamoto, T., Yoshimoto, G., et al. (2005) Development of functional human blood and immune systems in NOD/SCID/IL2 receptor {gamma} chain(null) mice. Blood 106, 1565–1573. 56. Tateno, C., Yoshizane, Y., Saito, N., Kataoka, M., Utoh, R., Yamasaki, C., et al. (2004) Near completely humanized liver in mice shows human-type metabolic responses to drugs. Am J Pathol 165, 901–912. 57. Azuma, H., Paulk, N., Ranade, A., Dorrell, C., Al-Dhalimy, M., Ellis, E., et al. (2007) Robust expansion of human hepatocytes in Fah-/-/Rag2-/-/Il2rg-/- mice. Nat Biotechnol 25, 903–910. 58. Hay, D. C., Zhao, D., Fletcher, J., Hewitt, Z. A., McLean, D., Urruticoechea-Uriguen, A., et al. (2008) Efficient differentiation of hepatocytes from human embryonic stem cells exhibiting markers recapitulating liver development in vivo. Stem Cells 26, 894–902.

Chapter 4 Engineering the Mouse Genome to Model Human Disease for Drug Discovery Frank Koentgen, Gabriele Suess, and Dieter Naf Abstract Genetically engineered mice (GEM) have become invaluable tools for human disease modeling and drug development. Completion of the mouse genome sequence in combination with transgenesis and gene targeting in embryonal stem cells have opened up unprecedented opportunities. Advanced technologies for derivation of GEM models will be introduced and discussed. Key words: Transgenic mice, gene targeting, homologous recombination, humanized mice, knockout, reporter gene.

1. Introduction The mouse (Mus musculus) has in the course of a mere century completed a remarkable journey; the metamorphosis from ancient agricultural pest – the most plausible cause for domestication of the cat by despairing early farmers in the Fertile Crescent (1) – into the premier experimental model of human genetics and disease (2). Studies on mouse coat color inheritance first provided evidence for the applicability of Mendel’s laws to mammals (3). Clarence Little, a student of William Ernest Castle’s at the Bussey Institute of Harvard University, recognized the need for genetically homogeneous lines of mice and derived the first inbred strains, using them to demonstrate the genetic basis of cancer. Little established The Jackson Laboratory in Bar Harbor, Maine, and his inbred mice became essential tools for various fields of biomedical research. Cancer biology is among the many fields G. Proetzel, M.V. Wiles (eds.), Mouse Models for Drug Discovery, Methods in Molecular Biology 602, DOI 10.1007/978-1-60761-058-8 4, © Humana Press, a part of Springer Science+Business Media, LLC 2010

55

56

Koentgen, Suess, and Naf

that have greatly benefited from disease modeling in the mouse (4). Mice are attractive research models not only because they are relatively easy and inexpensive to breed but mainly because technologies of genetic engineering have become available that now enable us to manipulate the mouse genome virtually at will. In this introductory chapter, we will review and discuss state-of-the-art methodologies used in the production of genetically engineered mice (GEM) and explain some of the terminology used in the field.

2. Transgenic Mice 2.1. Pronuclear Transgenesis

The well-established field of mouse genetics underwent a revolutionary transformation in the winter of 1980–1981 when the powerful new tools of molecular biology were introduced (5). Five near-simultaneous publications demonstrated that cloned foreign genes could be inserted into the mouse genome by injection of DNA into the male pronucleus of fertilized oocytes, resulting in live mice that passed the foreign gene on to their offspring (6–10). Most notably, Wagner et al. (10) showed that such transgenic mice harboring a rabbit beta-globin gene faithfully expressed the foreign protein in erythrocytes. The ramifications were monumental; it became apparent that transgenic mice could be used to experimentally elucidate the function of human genes. The term “transgenic” broadly refers to any organism carrying a foreign gene in a stable, heritable form. Pronuclear DNA injection continues to be employed in the generation of transgenic mice, and the method has been adapted to many other mammals, including laboratory rats and agricultural livestock (11). The typical makeup of a transgenic mouse is exemplified by the TRAMP strain, which was designed to model human prostate cancer (12). With the aim of directing expression of a potent oncogene to epithelial cells of the prostate, the coding sequence of simian virus 40 (SV40) T antigen was placed under control of regulatory sequences borrowed from the prostate-specific promoter of the rat probasin gene (Fig. 4.1A). Pronuclear injection of the DNA construct resulted in transgenic mice that exhibited the expected phenotype; males developed epithelial prostatic neoplasias that progressed to adenocarcinomas by 3 months and culminated in metastatic cancer with a frequency approaching 100% (13). The histopathology of TRAMP mice closely resembled human prostate cancer (14), validating their use as preclinical models for the development of new combination immunotherapies (15) that have since progressed to phase II clinical trials (N. Greenberg, personal communication).

GEM Models for Drug Discovery

57

Fig. 4.1. Pronuclear and lentiviral transgenesis. (A) Scheme of a pronuclear transgene construct and pronuclear DNA microinjection. Basic pronuclear transgenesis constructs are plasmids containing a mammalian or viral promoter (gray arrow) followed by the coding sequence of the gene to be expressed (black box) from initiation to termination codon (black and white lollipop, respectively). Most pronuclear transgenes include an intron as splicing enhances translation of mRNAs. A polyadenylation signal (pA, hatched box) is used to terminate transcription. Linearized plasmid DNA is injected directly into one of the pronuclei of a fertilized oocyte. (B) Scheme of a lentiviral construct and perivitelline injection of virions. Lentiviral constructs are based on a rudimentary version of the HIV-1 genome that consists of the 5 and 3 long terminal repeats (LTR, stippled boxes) and the viral packaging signal (). A promoter and the coding sequence of the gene of interest are inserted as for pronuclear transgenes. Introns cannot be included as these would be spliced out in the packaging cell line after transcription of the DNA template into viral genomic RNA. Instead, a posttranscriptional regulatory element from woodchuck hepatitis virus (WPRE, light gray box) may be used to enhance transgene expression. Transcription is terminated by a viral polyadenylation signal in the 3 LTR. Lentiviral plasmid constructs are transiently transfected into packaging cell lines from which infectious virions are harvested. These are injected into the perivitelline space of fertilized oocytes.

The major advantage of pronuclear transgenesis is that the technology imposes virtually no limits on transgene size (16). Transgenic mice have been obtained even with yeast artificial chromosomes (YAC) carrying inserts as large as 1.3 Mb (17). Smith et al. (18) used a panel of YAC transgenic mice covering a total of 2 Mb of human chromosome 21q22.2 to model neurological aspects of Down syndrome. As valuable as the technology has proven to be, pronuclear transgenesis is plagued with significant drawbacks. Transgene integration into the genome is an uncontrolled, stochastic event, and transgenes tend to insert as head-to-tail concatemers rather than single copies (8, 19). One serious limitation of the technology is tissue specificity of transgene expression or – more often – the lack thereof. Host sequences flanking the insertion site can severely affect transgene expression patterns, a phenomenon known as the chromosome position effect (20). The sword cuts both ways; because most transgenes contain strong transcriptional regulatory elements, such as promoters and enhancers, they in turn may cause dysregulation of host genes in the vicinity of the

58

Koentgen, Suess, and Naf

integration site, sometimes (in 5–10% of the cases studied) resulting in homozygous lethality (21). Some transgenic strains have been observed to lose transgene expression over time, either gradually or catastrophically, as a consequence of epigenetic silencing (e.g., DNA methylation) or even loss of transgene copies. Another caveat follows from the random nature of transgene insertion: each transgenic founder (i.e., each mouse derived from an individual injected oocyte) is genetically unique. It is therefore necessary to breed and characterize multiple transgenic lines for each construct independently in order to control for phenotypic distortions caused by integration site and copy effects. Even individual progeny from the same founder may show variation in transgene expression (22). Finally, not all transgenic animals will transmit the transgene. 2.2. Lentiviral Transgenesis

An alternative to pronuclear transgenesis is the use of lentiviral vectors for transgene delivery. These vector systems were originally developed as tools of gene therapy for human disease. Lentiviruses are a clade of retroviruses that have the capacity to productively infect non-dividing cells (23). The most extensively studied member is human immunodeficiency virus 1 (HIV-1), from which many of the currently used vectors are derived. For a detailed account of lentiviral biology and the characteristics of lentiviral vector systems we refer the reader to a recent review by Buchschacher and Wong-Staal (24). For the purpose of the present discussion it is sufficient to know that lentiviral vectors are, in essence, rudimentary versions of the HIV genome from which all genes required for viral replication have been removed. The vector retains only the cis-acting regulatory sequences for transcription into the viral RNA genome and its encapsidation into viral particles, as well as a cloning site for insertion of the transgene. Virus is produced by transient transfection of the construct into packaging cell lines, which are engineered to provide the essential viral proteins for assembly of infectious particles (25). Virions are then harvested from the cell culture medium. Importantly, these are capable of completing one infectious cycle only and cannot replicate because they contain no viral genes. The types of transgenes used in lentiviral transgenesis are not dissimilar to the pronuclear ones described above. They usually contain a mammalian or viral promoter linked to the coding sequence of the gene of interest (Fig. 4.1B). There is, however, a natural limit to transgene size as virions can only encapsidate RNA molecules up to approximately 10 kb. Many retroviral vectors include a posttranscriptional regulatory element from woodchuck hepatitis virus to boost transgene expression (26). The preferred method for delivery of lentiviral vectors is microinjection into the perivitelline space of single cell or cleavage-stage embryos, which is a far less invasive procedure

GEM Models for Drug Discovery

59

than pronuclear DNA injection and hence results in higher survival rates, sometimes approaching 90% (27). Most lentiviral vectors are pseudotyped with the surface protein VSV-G of vesicular stomatitis virus, which enables them to infect a wide variety of cell types, including mouse embryos. Infected cells reverse transcribe the lentiviral RNA into DNA (provirus), which then inserts into the genome. Unlike pronuclear transgenes, proviruses integrate as single copies, but there can be multiple, independent integrations into different chromosomal sites. Proviruses integrate preferentially into actively transcribed loci (28, 29) and hence may disrupt host genes. Lentiviral transgenes appear to be expressed more predictably than pronuclear ones and are less prone to epigenetic silencing (27, 30, 31). Many of the caveats discussed above for pronuclear transgenes apply to lentiviral transgenesis as well (Table 4.2). Retroviral integration into transcribed regions of the genome may cause dysregulation or disruption of unknown host genes, and effects of the integration site on transgene expression have also been documented (32). Multiple transgenic founders need to be produced and characterized because the lentivirus will integrate into different loci in each embryo. In addition, it may be difficult to establish stable transgenic lines because transgenes that integrated into multiple sites on different chromosomes will segregate during breeding. A further complication has been observed with constructs carrying strong, ubiquitously active promoters; the transgene may be expressed at high levels in the packaging cell line, and its product may be toxic to the cells or interfere with virion assembly. Thus, it is advisable to use inducible or conditional systems wherever possible.

3. Embryonal Stem Cell Technology and Gene Targeting

In 2007, Mario Capecchi, Sir Martin Evans, and Oliver Smithies were awarded the Nobel Prize in Physiology or Medicine in recognition of “their discoveries of principles for introducing specific gene modifications in mice by the use of embryonic stem cells.” The advent of embryonic stem (ES) cell technology and gene targeting in the late 1980s (33) squarely placed the mouse in pole position as the model of choice for preclinical drug discovery. ES cells are derived from the inner cell mass of blastocysts (34, 35). This cell population normally gives rise to the embryo proper. Under appropriate conditions, ES cells can be cultured in vitro and will retain their pluripotency and undifferentiated state (36). When injected into a host blastocyst and returned to a foster mother, ES cells will repopulate the embryo and give rise to

60

Koentgen, Suess, and Naf

chimeric mice in which part of each organ system, including the germline, can be derived from the ES cells (37). It follows that any genetic modification engineered in vitro into the genome of ES cells can be propagated through the chimera’s germline (germline transmission) and passed on to its offspring, provided that the mutation does not adversely affect pluripotency and the cells’ ability to contribute to the germline. Doetschman et al. (38) first demonstrated that homologous recombination could be exploited for targeted modification of a specific gene in ES cells. Since those pioneering days, thousands of genes have been modified in the mouse through gene targeting. In the following, we will discuss the targeting strategies most commonly used today and technical refinements that have been made to enhance the technology’s versatility. 3.1. Gene Knockouts

Every gene targeting project starts with design and construction of a targeting vector, usually a plasmid that contains parts of the gene of interest and the mutation(s) to be introduced. This is a critically important step as poor vector design will inevitably result in disaster. Figure 4.2A shows the structure of a basic targeting vector. It consists of three essential components: two stretches of mouse genomic DNA that are identical in sequence to the gene to be targeted (the 5 and 3 “homology arms”) and a mini-transgene that serves as a positive selection marker (“selection cassette”). Upon transfection into ES cells, the vector’s homology arms guide it to the matching position in the mouse genome and drive homologous recombination, which copies vector sequences, including the (positive) selection cassette, into the endogenous gene. The resulting targeted allele is a constitutive genetic “knockout” if the selection marker is positioned to replace a functionally important segment of the gene (e.g., a crucial exon). Positive selection is necessary because transfection of ES cells and homologous recombination are frustratingly inefficient. The transfection method of choice is electroporation due to a high degree of reproducibility, but its efficiency is low (12 h) in water and process for paraffin embedding (dehydrate through at ethanol series and then to xylene, do not allow nervous system samples to sit longer than necessary in ethanol or white matter tracts will look like Swiss cheese). 4. Embed the samples for cross-section (horizontal section). 5. Stain 4–5 ␮m sections using standard hematoxylin and eosin protocols (H&E) or with cresyl violet/luxol fast blue (CV/LFB). H&E stains protein rich cells (eosin) and counterstains nuclei (hemotoxylin). CV/LFB also stains protein,

366

Burgess, Cox, and Seburn

but also stains myelin. If available, an automated tissue processor is recommended for consistency of staining. 6. Interpretation: dying cells in the ventral horn of the spinal cord may be distinguished by loss of Nissl substance in the cell bodies, by pyknotic nuclei, and by pink cytoplasm in cells that are early in the degeneration process (see Note 1). 3.1.2. Assessing Motor Neuron Number (Loss)

In the spinal cord, motor neuron number can be assessed in cross-sections as described above. However, there are two primary challenges. First, motor neuron cell bodies need to be accurately identified in the ventral horn. Markers such as HB9-GFP, Thy1.2-YFP, or ChAT can be useful to positively identify motor neurons, but these are often incompatible with strong fixatives, paraffin embedding, or histological counterstains. A second challenge is being sure that comparisons are made at exactly the same level of the spinal cord, which is difficult considering the cord does not fill the full rostral/caudal extent of the vertebral column and the number of motor neurons per cross-section varies greatly with level. An easier approach that we prefer is to count axons in the ventral roots as they exit the spinal cord (Fig. 20.3). The ventral roots can be dissected, embedded, cross-sectioned, and myelinated axons can be counted after staining. A decrease in axon number should reflect a loss in motor neurons in the ventral horn. 1. Perfuse an animal transcardially with strong fixative (2% paraformaldehyde, 2% glutaraldehyde in 0.1 M cacodylate buffer, standard electron microscopy fixative). 2. Expose the sciatic nerve dorsally adjacent to the femur and follow it proximally to the sciatic notch where it disappears under the pelvis. 3. Using a pair fine blunt-point scissors carefully split pelvis in the direction followed by the nerve and gently separate the split bone. Continue to follow the nerve until branches into L6, the L5, and L4 spinal nerves are evident (NB: L6 is most distal and smallest). 4. Perform a hemilaminectomy (removing the dorsal bone of the vertebrae) in the lumbar region using a pair of No. 2 forceps to remove small pieces of bone until spinal cord is exposed from the dorsal midline laterally to the entry point of the spinal nerves. 5. Grasping each spinal nerve, follow it to its bifurcation into the ventral (motor) and dorsal (sensory) roots (see Note 2). 6. Cut the ventral root free at the bifurcation point and as close to the spinal cord as possible (NB: ventral roots enter cord proximal to spinal nerve entry point). 7. Process for plastic embedding (as for electron microscopy).

Neuromuscular Disease Models and Analysis

367

Fig. 20.3. Dorsal and ventral root dissection. The anatomy of the ventral and dorsal roots during dissection is shown. Mouse is prone (rostral is up) and the left sciatic nerve (Sci) has been exposed along the femur and to its origin at the lumbar spinal nerves. The spinal nerves are labeled (lumbar 3 through 6). The ischium (Isch) has been removed to follow the sciatic nerve. The inset shows the L4 root separated into its ventral and dorsal roots, note the DRG associated with the dorsal root. The scale bar is 6 mm. (Color figure is available online).

8. Cut 500 nm cross-sections near the middle of the sample to avoid dissection damage and stain with Toluidine Blue. The most straightforward interpretation is that decreased axon number will reflect death of motor neurons in the spinal cord. See Fig. 20.4 for an example of this analysis. This interpretation is strengthened if combined with spinal cord histology described above. Degenerating axon profiles may also be seen, which may indicate an axonopathy in the absence of evidence for dying cells in the ventral horn. An often-invoked complication to this analysis is that motor axons may branch and this would mask the loss in axon number. Motor axons typically branch only after they have entered the muscle. While a pathological state may cause them to branch at the level of the ventral root, we have not encountered any examples of this, although it is a formal possibility and a caveat to this analysis. Furthermore, if axon loss is seen,

368

Burgess, Cox, and Seburn

Fig. 20.4. An example of motor neuron loss reflected in ventral root axon number. The L4 ventral root stained for myelin is shown in cross-section in (A) a control mouse, (B) an nmd mouse (Ighmbp2 mutant), (C) an nmd mouse carrying a modifier locus from CAST, and (D) an nmd mouse with transgenic rescue of the Ighmbp2 gene driven in the nervous system. For details, see (39).

the worst-case scenario is that the number is an underestimate due to branching. Degenerating axons in the peripheral nerves can also be examined as described below, and electrophysiological estimates of motor unit number could be informative. Motor units are defined as a single motor neuron (axon) and the muscle fibers its terminals innervate. In motor neuron diseases, the number of motor units is anticipated to decrease as motor neurons die, but the size of motor units may increase with compensatory sprouting and reinnervation (e.g., (1)). Interpretation can be further confounded by factors such as a change in muscle fiber number or innervation of muscle fibers by multiple motor axons. Therefore, the best interpretation results from corroborating evidence from the spinal cord, nerve, synapse, and the muscle. 3.2. Peripheral Neuropathies

The femoral nerve provides an excellent system for evaluating peripheral neuropathy, provided there is hind limb involvement. In combination with counts of ventral roots, femoral axon counts can be used to distinguish peripheral neuropathy from motor neuron death. The nerve has a primarily motor branch that innervates the quadriceps, and a primarily sensory branch that becomes the saphenous nerve more distally (Fig. 20.5). Each branch can be easily dissected free (the animals can be transcardially perfused before, or the nerves can be fixed by immersion after dissection if care is taken to be sure they are extended at full-length when immersed). 1. Nerves should be plastic embedded and cross-sectioned as above for the ventral roots. 2. Axons can be counted from Toluidine Blue-stained sections. 3. The distribution of axon diameters, myelin thickness, and Gratios (inner/outer diameters, the inner being the axoplasm, the outer including the myelin) can also be determined.

Neuromuscular Disease Models and Analysis

369

Fig. 20.5. Femoral nerve dissection. The motor and sensory branches of the femoral nerve are exposed. The mouse is supine and the right hip is shown (forceps are retracting the abdominal wall, A, P, M, L are anterior, posterior, medial, and lateral respectively, H is hamstring muscles). Some adipose tissue has been removed for clarity. The motor branch of the femoral nerve innervates the quadriceps (Q). The sensory branch becomes the saphenous nerve, which runs adjacent to the saphenous vein (Saph) on the medial side of the thigh. Dissecting the nerve where the tick marks provides a reasonable length of nerve to work with. Note the sensory branch sometimes runs as two fascicles and both should be taken to get reproducible counts. The scale bar is 2 mm. (Color figure is available online).

This may be done most accurately by low magnification (4000–6000×) transmission electron microscopy. 4. The assessment of axon diameters may reveal general axonal atrophy, or a missing class, such as large diameter, fast motor axons. 5. The assessment of myelin layering and myelin thickness may reveal a demyelinating or hypomyelinating neuropathy. 6. The G-ratio may indicate abnormal reciprocal signaling between the axon and the myelinating Schwann cell or may highlight thin myelin or conversely, thin axons, since there is normally a rough correlation between axon diameter and myelin thickness.

370

Burgess, Cox, and Seburn

7. Other peripheral nerve pathologies such as onion bulbs (indicating rounds of demyelination/remyelination), Schmidt–Lanterman Incisures (indicating abnormal myelin packing), myelinating Schwann cells wrapping multiple axons, and bundles of regenerating axons can also be seen in these sections. For examples of these pathologies see Fig. 20.6 and ref. (2, 3).

Fig. 20.6. Peripheral neuropathy phenotypes. A cross-section of the motor branch of the femoral nerve in a control mouse (A) and a Gars Nmf249/+ model of Charcot-Marie-Tooth 2D (B) are shown. The insets highlight the varied axon diameters in a control nerve and the almost complete absence of large diameter axons in the mutant nerve. (C) The same mutation examined by transmission electron microscopy demonstrates degenerating axon profiles. Note the irregularities in the myelin are fixation artifact and not pathology, highlighting the need to always process control samples in parallel. (D) A myelinating Schwann cell that has ensheathed multiple axons, probably representing a failure in radial sorting during early postnatal development. Note, this is different from a Remak bundle of small sensory axons, in which a nonmyelinating Schwann cell enwraps a number of axons in a single basal lamina (see the bottom right corner of (C). (E) Nodes of Ranvier can be examined by light microscopy in teased nerve preparations. The nucleus of the Schwann cell (black arrow) is typically midway between the nodes (white arrows). (E) An example of hypomyelination in the ventral root of a Lama2 dy/dy mouse. Normally, 100% of the ventral root axons are myelinated. In this mutation, bundles of large but unmyelinated axons are evident.

3.2.1. Internodal Distance Assessment

In addition to axon loss/atrophy and defects in myelination, the internodal distance can also affect nerve conduction velocities. 1. To determine internodal distance, dissect a 1–2 cm segment of peripheral nerve such as the femoral or sciatic and fix as above (see Section 3.1.2). 2. Tease the nerve longitudinally to individual fibers using No. 5 forceps or a 30-G needle. Keeping the nerve immersed in a drop of PBS, tease the nerve directly on a microscope slide. 3. Coverslip the teased nerves and view using Nomarski-DIC optics.

Neuromuscular Disease Models and Analysis

371

4. Measure internodal distances and correlate them with axon diameters. 5. Again, there should be a rough correlation, with larger axons having longer internodal distances. This analysis requires software calibrated for digital image analysis to determine the distances ( Fig. 23.6 ). 3.2.2. Nerve Conduction Velocities

Nerve conduction velocities are used diagnostically to distinguish type 1 (demyelinating) and type 2 (axonal) neuropathies in humans. Type 1 neuropathies typically have pathologically reduced NCVs (below 30 m/s compared to normal values near 50 m/s in humans). Nerve conduction velocities can also be measured in the mouse as described below. However, significant decreases in axon diameter or internodal distance can also contribute to decreased NCV (3, 4). Furthermore, NCVs record the fastest (largest) axons present and may therefore miss axonal pathologies that do not cause a marked decrease in these neurons. Therefore, this functional measure should again be combined with an examination of axon diameters, myelination, and internodal distance to determine the underlying mechanism.

3.3. Myasthenias

Myasthenias are diseases of the NMJ. This synapse is highly accessible, highly stereotyped in its morphology, and easily visualized by light or electron microscopy. Evaluation of the morphology of the junctions is generally very well correlated with function, although electrophysiology may be required to fully assess synaptic transmission and to determine if defects are pre- or postsynaptic.

3.3.1. Staining NMJs for Light Microscopy

Neuromuscular junctions can be visualized by light microscopy following labeling of the presynaptic nerve terminal and the postsynaptic acetylcholine receptors. For best results, muscles should be prepared for longitudinal sections, and NMJs in an en face orientation can be imaged. In almost all muscles, the end plate band is near the middle of the muscle and this represents the region of interest. 1. Muscles should be lightly fixed in buffered 2% paraformaldehyde (2–4 h on ice) (see Note 3). 2. The muscles can be prepared for staining in a number of ways, but should be oriented for longitudinal sections. 3. Cutting thick, frozen sections using a cryostat (20–40 ␮m thick sections) or using a vibratome to cut 50 ␮m sections of unfrozen tissue (remove the tendons to facilitate sectioning) gives good results. Muscle fibers can also be teased directly on slides to obtain individual fibers.

372

Burgess, Cox, and Seburn

4. Samples are then stained using standard immunocytochemistry procedures. 5. Presynaptic antigens that work well include a cocktail of anti-neurofilament (for instance the monoclonal antibody 2H3) and anti-SV2 to fully label both the axon and the nerve terminal. 6. The postsynaptic receptors are brightly and specifically labeled using fluorescent conjugates of alphabungarotoxin. 7. Standard immuno staining techniques involving application of the cocktail of primary antibodies followed by washes, then application of a cocktail of the secondary antibody and ␣-bungarotoxin followed by washes, work well provide patience is used (for instance, primary antibodies should be applied overnight followed by several hours of washing the next day). 8. Standard antibody dilution buffers such as PBS with BSA or normal goat serum as blockering agents can be used, provided generous detergent (0.5–1% triton X-100) is also included. 9. Samples can be viewed on a standard fluorescence scope, but given the large size and three-dimensional nature of the samples, a confocal Z-series usually gives better results. 10. A number of defects can be readily observed, including partially innervated or completely denervated postsynaptic receptor sites, fragmented or shrunken postsynaptic receptors, atrophied axons or terminals, and swollen or dystrophic axons or terminals. More subtle defects include sprouting nerve terminals and multiple innervation or synaptic sites (single innervation is normally present after 2 weeks of age). Normal NMJs have a contiguous pretzellike morphology, the AChR staining has a zebra stripe appearance (the junctional folds), and the nerve terminals completely overlap the receptors. Examples are shown in Fig. 20.7. Caveats include minor differences in size and shape from muscle to muscle, and fixation artifacts that can eliminate staining, especially presynaptically. In general, defects in the presynaptic terminal, such as partial retraction, are reflected less-precise definition in the postsynaptic receptors. Additional analyses that can be informative include staining with anti-S100 to visualize Schwann cells (the terminal Schwann cells play an important role in guiding terminal sprouting and reinnervation (5, 6)) and histochemical stains to visualize acetylcholine esterase (7).

Neuromuscular Disease Models and Analysis

373

Fig. 20.7. Neuromuscular junctions, light microscopy. In a control neuromuscular junction (A), the nerve (stained green using anti-neurofilament plus anti-SV2, or using a Thy1-YFP transgenic strain) completely overlaps the postsynaptic AChRs (stained red using fluorescent -bungarotoxin conjugates). (B) An example of partial innervation, where an atrophied motor axon and terminal fails to completely cover the AChRs on the muscle. (C) An example of frank denervation, where a site of postsynaptic AChRs with no associated nerve (arrowhead) is observed near a site of partial innervation (double arrowhead). (B) and (C) are examples from the Gars Nmf249/+ mouse model of Charcot-Marie-Tooth 2D (3). Other NMJ pathologies are evident in an example from an unpublished spontaneous mutation. The axons and terminals have irregular diameters and varicosities (arrowheads) and postsynaptic sites are fragments (double arrowheads). Such changes are predictive of eventual denervation. These pathologies are also observed in very old mice (greater than 15 months), but are evident in this mutant by three months of age. (Color figure is available online).

3.3.2. Electron Microscopy of the NMJ

NMJs can also be visualized by transmission electron microscopy for a more detailed look at pre- and postsynaptic anatomies. 1. Muscles should be prepared for electron microscopy using standard techniques, including rapid fixation (preferably by perfusion) with glutaraldehyde-based fixatives. 2. The muscle should be dissected free and trimmed for crosssections at the point where the nerve enters the muscle. The end plate band is only a narrow region near the middle of the muscle (see Note 4).

374

Burgess, Cox, and Seburn

3. Samples should be postfixed, osmicated, and embedded in plastic, cross-sectioned, and mounted on EM grids using standard procedures (for example, (8)). The challenge to viewing sections is finding NMJs. They can be spotted by scanning the grids at 10–12 K magnification and concentrating on areas where axons, fat, or blood vessels are also present. NMJs are rarely found in areas where the muscle fibers tightly stacked. Detailed images can be obtained at 30–60 K magnification. Normal NMJs have a generally polarized nerve terminal with accumulations of 40–50 nm small clear vesicles near the presynaptic membrane and mitochondria located farther away (Fig. 20.8). In mice, the terminal Schwann cell capping the nerve terminal can be difficult to resolve. The postsynaptic membrane has a series of junctional folds invaginating into the muscle fiber. At the mouth (crest) of each fold, the membrane appears electron dense because of the accumulation of AChRs. The synaptic cleft is pronounced and contains a visible basal lamina.

Fig. 20.8. Neuromuscular junctions analyzed by transmission electron microscopy. (A) In wild-type mice, the motor nerve terminal (MN) is depressed into the muscle fiber surface. The terminal is polarized, with small clear vesicles near the presynaptic membrane and mitochondria in the more proximal portion of the terminal. The postsynaptic membrane has deep convolutions (junctional folds, JF) and the membrane near the tops of these folds is very electron dense because of the high density of acetylcholine receptors (arrowheads). (B) In some myasthenias where the nerve sprouts but remains in contact with the muscle, terminals with mitochondria and vesicles are observed in the absence of any postsynaptic specialization. Presumably these are sprouting terminals that have not established a functional connection. (C) Partial innervation of postsynaptic sites is evident as elaborate junctional folds in the muscle membrane with no overlying nerve terminal. In these examples, the interpretations were aided by light microscopy examination of other samples as described in Fig. 20.8 in parallel with electron microscopy. The mutation shown in (B, C) is an unpublished ENU-induced allele of agrin.

Pathological deviations include an absence of junctional folds, partial innervation (folds without an overlying nerve terminal), and vacuolated mitochondria. Assessing more subtle defects, such as changes in vesicle number, requires a statistical analysis on many junctions. 3.4. Muscular Dystrophies 3.4.1. Histological Analysis of Muscular Dystrophies

Muscle pathology can be accurately assessed by histology, focusing on similar hallmarks to those used in the diagnosis of human muscular dystrophies (Fig. 20.9). Muscle weight to body weight ratios can be used to indicate pathology in the muscle that is not

Neuromuscular Disease Models and Analysis

375

simply proportional to decreased body size. Mice with muscular dystrophy or atrophy will typically have lower body weights, but an even greater reduction in muscle weight:body weight. 1. Muscles can be fixed by perfusion or immersion in Bouin’s fixative, dehydrated, and embedded in paraffin (record body weights and muscle weights following dissection, prior to processing). 2. Cross-sections should be cut using a microtome and stained using H&E. All of these techniques are standard histological protocols. Care should be taken to collect sections from standardized regions of the muscle (such as the belly) to avoid differences in fiber number, fiber size, or composition that can vary along the longitudinal axis in limb muscles. 3. Samples should be evaluated for muscle fiber diameters (and the uniformity of fiber sizes), centrally located myonuclei (a sign of a regenerated muscle fiber), fibrosis, fatty infiltration, and atrophied muscle fibers. 4. Interpretation: muscular dystrophies can be distinguished from other neuromuscular conditions based on the loss of muscle fibers, fibrosis, and signs of degeneration/regeneration. Atrophy resulting from denervation may have a similar appearance, but the affected fibers are typically scattered throughout the muscle (reflecting the anatomy of motor units and the pattern of innervation by motor neurons), while dystrophies tend to affect most of the fibers in a particular region of the muscle. 3.4.2. Muscle Fiber Integrity, Evan’s Blue Staining

Many muscular dystrophies result in the loss of sarcolemmal integrity. This is true of dystrophies affecting the Dystrophin/Glycoprotein Complex (DGC), including defects in dystrophin, the sarcoglycans, and dystroglycan, or in dysferlin, which is involved in membrane repair. Other mutations, such as those in titin, do not result in a loss of muscle fiber integrity. Therefore, particularly for the mechanistic evaluation of new models, it is important to determine if membrane integrity is compromised. This is easily done using Evan’s blue, an Azo dye that binds albumin and is normally excluded from healthy cells, but infiltrates into the cytoplasm of compromised cells (Fig. 20.9). 1. Evan’s blue dye should be dissolved in sterile saline at 10 mg/ml. 2. This solution can then be injected IP using 0.1 ml per 10 g of body weight (100 mg/kg). 3. Within a few hours of injection, exposed skin, such as the feet, tail, and ears, should have a pronounced blue tinge. 4. After 12–18 h (overnight), dissect muscles of interest and view.

376

Burgess, Cox, and Seburn

Fig. 20.9. Dystrophy phenotypes. (A) In histological staining of wild-type muscle (H&E), the muscle fibers are closely packed, regular in size, and the nuclei are near the cortex of the muscle fiber. (B) In a dystrophic muscle (Lama2dy/dy is shown), the fibers are variable in diameter, some are atrophied, fibrotic cells are replacing muscle fibers, and nuclei in the center of fibers indicate regenerated fibers. (C, D) Mice lacking dysferlin, which is involved in membrane repair, have a progressive dystrophy. Muscle is histologically normal (C) until approximately 8 months of age, by 14 months (D), the muscle is severely dystrophic, with a great deal of fatty infiltration. (E) The rmd mouse mutation also has a severe dystrophy with fatty infiltration into the muscle, but the phenotype is much more severe in the hind limbs (shown) than the forelimbs. (F) Antibody staining for myosin isoforms (fast myosin is shown) can determine if certain fiber types are selectively sensitive. The myd mouse is shown, in which both fast and slow fibers show central nuclei and signs of dystrophy, indicated both fiber types are affected. (G, H) Fiber-type staining can also identify grouping, as seen with slow myosin stains of control (G) and GarsNmf249/+ muscle (H). Such grouping is indicative of denervation and reinnervation and not a dystrophy intrinsic to the muscle. (I, J) Muscle fiber integrity can be assessed by Evan’s blue dye infiltration. In control muscle (I), Evan’s blue administered intraperitoneally is excluded from the muscle fibers but stains the membranes and connective tissue. In dystrophic muscles in which the sarcolemmal membrane integrity is compromised (mdx is shown), the dye stains the entire fiber (J). (Color figure is available online).

5. All tissue will have a bluish cast, but dystrophic muscle will have marked streaks of blue due to the compromised fibers. 6. This can be most readily assessed by embedding the tissue and cutting cryostat cross-sections. 7. Under fluorescence (rhodamine filters), the Evan’s blue fluoresces red. Healthy muscle will have positively labeled membranes, as well as blood vessels, but muscles with compromised fiber integrity will contain fibers in which the entire cross-sectioned cytoplasm is strongly fluorescent. This technique is simple and very sensitive. Intraperitoneal injection is straightforward and as effective as intravenous injection. Positive muscle fibers can also be the result of necrosis or injury in the muscle, but this usually results in a few, widely scattered

Neuromuscular Disease Models and Analysis

377

fibers instead of a large number of fibers grouped in a specific region of the muscle. 3.4.3. Caveats

As with all neuromuscular diseases, there can be tremendous variation muscle to muscle, and even within a muscle (this is true for neuropathies as well, see, for example, (9)). For instance, in some dystrophy models the diaphragm is severely affected while in others, it is spared. Therefore, it is very important to evaluate the same muscle in each mouse and to evaluate more than one muscle. A picture of the anatomy of the lower hind limb of the mouse is shown in Fig. 20.10. The underlying basis of these differences in not clear and does not seem to be as obvious as differences in fiber type composition, activity levels, or force generated. Within a model, the pattern of pathology is usually fairly reproducible, but there can again be animal-to-animal variability that complicates quantification of results.

Fig. 20.10. The muscles of the lower hind limb in cross-section. In this image, anterior is down and medial is left. Abbreviations are as follows: MG, medial gastrocnemius; LG, lateral gastrocnemius; Plant, plantaris; PN, plantar nerve; Sol, soleus; Fib, fibula; EDL, extensor digitorum longii; TA, tibialis anterior; Tib, tibia. The mouse muscles are predominantly fast muscle fibers, but the soleus is valuable for its high percentage of slow fibers. Note, the darker mass on the posterior portion of the leg is a lymph node that provides a convenient landmark when sectioning to establish that reproducible sections are examined in the proximal/distal axis. Also, the peripheral muscles in the section are the hamstrings, which insert along the tibia in the lower leg in the mouse. (Color figure is available online).

Other techniques such as serum levels of muscle creatine kinase can also be used to assess dystrophy, but this requires great care in mouse handling. Creatine kinase is released from damaged

378

Burgess, Cox, and Seburn

muscle fibers. If the mouse struggles significantly while the sample is collected (more than a few seconds), values will be artificially high and variable. In addition, hemolysis in the sample will also reduce accuracy. Therefore, while this measurement is outwardly straightforward and valuable, obtaining consistent results can be challenging. 3.5. Techniques Generally Applicable to Neuromuscular Disease Models 3.5.1. Gross Motor Performance

3.5.2. Grip Strength

The disease models described above, motor neuron diseases, neuropathies, myasthenias, and muscular dystrophies, can all benefit from an analysis of gross motor performance. The rotarod is often used to assess neuromuscular function and other movement disorders and has been successfully used to describe defects in many mouse models (e.g., (10–12)). The widespread use of the device is due in part to the apparent simplicity of the testing procedures. However, recent work emphasizes that, as with any behavioral test, there are a number of potential confounding factors (13, 14). In our experience data collection is more efficient using gait analysis (treadmill). In addition raw video data is often useful on its own and the extensive and varied measurements that are possible offer significant flexibility. Grip strength measures also provide a more direct assessment of muscle force generation that can be combined with electrophysiological techniques described below, and are only minimally confounded by other parameters affecting motor performance such as coordination or balance. Therefore, our preferred methods are grip strength and gait analysis. Measure of grip strength gives a good indicator of muscle strength and is analogous to grip strength measures in humans, e.g. (15), which reveal weakness as a major presenting symptom of neuromuscular dysfunction. To measure grip strength, the mouse is prompted to grab a bar connected to a force transducer with either its hind or its fore paws (or less often all paws using a grid). Once the mouse achieves a grip the tester typically pulls the mouse horizontally away from the bar until the animal is no longer able to maintain its grip. The peak force registered by the transducer is recorded. This approach, recommended by equipment suppliers, requires careful attention to the rate and direction of force applied by the tester to “break” the animals grip. If this approach is used, reliability should be confirmed for each individual performing the test and between testers if multiple individuals are collecting data. However, to improve reproducibility for forelimb grip strength we recommend orienting the force transducer vertically (Columbus Instruments, Columbus, OH, USA) and modifying the test procedure as follows. A weight (100 g) is attached to the base of the mouse’s tail using a small plastic clip. The mouse is held by the scruff of the neck with the tail and weight in the

Neuromuscular Disease Models and Analysis

379

tester’s other palm. The mouse is advanced toward the bar until it instinctively grasps the bar with both paws. The mouse is slowly lowered to a vertical position and both animal and weight are released. The transducer records the peak force generated by the animal just prior to grip loss. This method insures consistent force and acceleration away from the grip bar and is not any more traumatic for the mouse. Typically measures from three repeat trials are averaged to represent the value for each animal. This approach can be used to track changes in longitudinal studies and has been used effectively for most standard inbred strains (http://phenome.jax.org). 3.5.3. Gait Analysis

Analysis of gait has a long history, and because it is also routinely used in humans, establishing clinical relevance for murine disease models is facilitated. The simplest method for quantification of gait in mice is to paint the feet of the mice and to motivate them to walk across a piece of paper. Manual measurements of footprints can then be made to derive stride length and stance width. This method is particularly useful for quantification in mice with overt, observable movement defects, but may also be sufficient for more subtle phenotypes (e.g., (16)). Practically, the greatest difficulty with this method is in motivating the mice to walk, and investigators have addressed this with a variety of strategies. For example, placing the mice in a lighted “corridor” facing toward a darkened space will both restrict wandering and improve motivation, as mice will seek to “escape” to the dark enclosure. Two additional considerations for this method are that only a limited number of parameters can be measured and that the speed of locomotion cannot be controlled. Thus, common time domain parameters that divide a stride (step) into component phases of stance and swing are not possible and measurements that are derived need to consider locomotory speed in the interpretation. We expanded the utility of the basic footprint analysis by employing video recordings of mice walking on a treadmill (Fig. 20.11). The mice are placed in an enclosure on a treadmill with a clear plastic tread (Columbus Instruments, Columbus, OH). The ventral surface of the mouse is reflected in a mirror placed at 45◦ under the tread and is recorded by an adjacent digital video camera (Basler, Inc). Typically a video clip of ∼10 s provides sufficient data for valid measurement. Interactive analysis software (Clever Sys Inc, Reston VA) is used to track body position and paw placement of each paw during locomotion at a fixed speed. The software derives the standard time domain gait parameters for each paw (e.g., stride and stance, swing phase times) as well as a variety of additional measures (body angle, foot placement angles, inter-limb phase ratios, etc.). The variety of measures that can be derived with such a system allow detailed description of gait in many different models and

380

Burgess, Cox, and Seburn

Fig. 20.11. Gait analysis device. (A) The gait analysis device is shown. The mouse is place on the treadmill in the green box. The mouse would be facing left and the loop at right in the chamber bumps the mouse’s tail if they lag on the treadmill. The white box in front contains the digital camera that videos the mouse using an angled mirror. (B) The automated data analysis is trained to recognize the mouse’s paws in the video and gait parameters are calculated, as well as body axis, toe spread, and other data.

offers the significant advantage of comparing animals at the same speed of locomotion and may be more sensitive to some subtle neuromuscular changes prior to overt movement deficits (17). However, the treadmill represents a novel context for the mice,

Neuromuscular Disease Models and Analysis

381

and we have noted changes in treadmill gait with repeated trials and age both of which can complicate characterization of progressive changes. Also, different mouse strains vary in their willingness to walk consistently on the treadmill, and compliance may be further reduced with repeated exposures (for details see (18)). These factors need to be considered carefully for each model in order to optimize experimental design. 3.5.4. Electrophysiology

We will describe basic electrophysiological analysis of neuromuscular function, similar to evaluation using electromyography by neurologists. More specialized techniques such as two electrode voltage clamp for monitoring synaptic transmission may also be useful, but are beyond the basic nature of this chapter. We describe briefly techniques for assessment of muscle contractile function including electromyography and nerve conduction velocity. These techniques provide measures of muscle force output and fatigability and insight into contractile dynamics related to energy supply, calcium handling, and excitation–contraction coupling. In addition, motor unit number can be estimated and significant deficits in synaptic function may be detectable.

3.5.5. Muscle Contractile Function

Contractile properties in rodents can be measured either in vitro in a dissected muscle or in vivo in an intact preparation with an anesthetized animal (e.g., (19–20)). Measurements made under isometric conditions are perhaps most common and use the most straightforward setup. The addition of servomotors for dynamic control of muscle length allows simulation of dynamic conditions (eccentric, isotonic, etc.) that may be modified by disease or other processes (22, 23). For in vitro studies, the muscle is anchored by ligating the tendon (origin) to a support, for in vivo studies, the bone (femur) is clamped to prevent movement. The other tendon (insertion) is then coupled to a force transducer. In both cases, a recording electrode is also placed in contact with the muscle to record the compound action potential, and a stimulating electrode is used to stimulate the nerve or the muscle, as described below. It should be noted that both in vitro and in vivo preparations present considerable technical challenges and are best developed/established in consultation with an experienced laboratory. In general, an in vitro preparation is somewhat less demanding because surgical preparation is less demanding and extended life support/maintenance of the animal during experimentation is not required. Muscle force can be elicited by nerve stimulation to test both muscle function and the integrity of the nerve–muscle connection or by direct muscle stimulation to evaluate only muscle contractile independent of the synapse. The latter measure reflects the total force that the muscle is able to generate, if it is significantly

382

Burgess, Cox, and Seburn

greater than that obtained by nerve stimulation, it implies denervation, or a failure in conduction or synaptic transmission in the nerves. Two general types of stimuli are typically used, a single brief (100–200 ␮s) stimulus of intensity sufficient to activate all functional connections, referred to as maximal twitch force (Pt), and trains of stimuli delivered at a frequency (100–200 pps) which induces tetanic fusion of individual contractions to produce maximum tetanic force (P0). 3.5.6. Contractile Measures

Measurement of single twitches provides insight into contraction/relaxation dynamics of the muscle. For example, reduced AChE or altered Ca2+ handling in the muscle causes a slower relaxation phase, whereas shifts toward “faster” ATPase isoforms will reduce time to reach peak twitch force (Fig. 20.12).

Fig. 20.12. Muscle twitch force. The force generated by a control muscle in response to a single 200 ␮s stimulus (black trace) compared to two mutants showing exaggerated force (red) or slower relaxation time (orange) is shown. This is an unpublished mutation that we interpret as having decreased AChE at the synapse, possible as a secondary compensation for impaired presynaptic neurotransmitter release. (Color figure is available online).

Stimulus trains of varying lengths (0.5–1.2 s) and frequencies of 10–200 pps can provide additional information. As stimulus frequency is increased, there is greater tetanic fusion and higher peak forces are produced. A force–frequency curve (F–F) can also be generated to evaluate factors that affect muscle contractile speed (time-to-peak, half-relaxation). If, for example, disease processes lead to slower contraction/relaxation time, more tetanization occurs at lower frequencies and the shape of the F–F curve will be shifted compared to normal muscle. A variety of protocols exist for measurement of muscle fatigue in mice. In general a series of stimulus trains are repeated at a set frequency for several minutes. The protocol selected depends on the muscle being tested and can vary depending on the model and the experimental objective. Different laboratories have established unique protocols that can be implemented (24–28). The basis for the difference in protocols is not always evident, different stim-

Neuromuscular Disease Models and Analysis

383

ulus paradigms may have been selected on a theoretical basis or determined empirically for the purposes of a given laboratory. Long stimulus trains will also induce “tetanic fade,” observed as a decrease in force during a single contraction. Fade is measured as the ratio of final force to initial peak force. Diseased muscle/nerve may show a lower ratio. Accompanying measurement of EMG can reveal if fade is due to failure of the muscle or the nerve. In purely muscle defects (muscle dystrophy) force will drop without any change in EMG whereas reduced force accompanied by reduced EMG suggests a defect in transmission (NMJ) (e.g., (3)) or excitation–contraction coupling. 3.5.7. Motor Unit Number Estimates (MUNE)

Delivering brief nerve stimuli with gradually increasing amplitudes will evoke an incremental increase in Pt as additional motor units are recruited. If step-wise increases in twitch force or electromyogram amplitudes are recorded the distinct increments can be counted to provide an estimate of motor unit number or the number functional motor axons innervating a given muscle (Fig. 20.13). There are a variety of MUNE techniques (29) but no universally accepted standard has emerged. Nonetheless, the method, originally used in humans (30), has been used successfully in other species including mice (e.g., (1)) and recently developed methods can facilitate implementation (31).

3.5.8. Nerve Conduction Velocities (NCV)

In contractile experiments the time from the stimulation of the nerve to the CMAP recorded in muscle provides an estimate of NCV. The length of the nerve from the stimulating electrode to the muscle can simply be measured and divided by the time. However, the time recorded in this way includes the delay for synaptic transmission, which may be increased in models with synaptic defects. If this is a concern or if the only parameter desired is NCV, then the measurement can be obtained non-invasively with a relatively simple setup (e.g., (3)). Using the sciatic nerve, NCV can be calculated by measuring the latency of compound motor action potentials recorded in the muscle of a rear paw. Action potentials are produced by subcutaneous stimulation at two separate sites: proximal stimulation at the sciatic notch and distally at the ankle. NCV is then calculated by using the two latencies and conduction distance. Decreases in nerve conduction velocity most often reflect defects in myelination, but may also be the result of changes in internodal distance, decreased axon diameters, or altered excitability.

384

Burgess, Cox, and Seburn

4. Conclusion The pathophysiology of neuromuscular diseases is highly integrated, reflecting the extensive reciprocal signaling and functional interdependency of the peripheral nervous system and the skeletal muscles. As a result, changes in the nerves result in changes in the muscles and vice versa, making it a challenge to distinguish primary and secondary effects. This is particularly true when new models are being examined and it is unknown where the causative gene normally functions. As a result, the functions of genes and the mutant phenotypes must be examined comprehensively, and the results must be taken as a whole in order to understand how the final phenotype arises. In this chapter, we have given examples of mouse models used in neuromuscular disease research,

Fig. 20.13. Motor unit estimation. Stimuli of gradually increasing intensity are applied to the nerve and the force generated by the muscle is recorded. The number of steps of force produced approximately reflects the number of motor units in the nerve.

Neuromuscular Disease Models and Analysis

385

and provided a battery of analyses that can be applied to these models to understand the underlying disease mechanisms. While many of these diseases are now well understood, many remain equally mysterious, and continued efforts to understand both the genetics and physiology are needed to eventually have a complete understanding of the neuromuscular system.

5. Notes 1. Purple condensed cells in H&E can be “Dark Neuron Artifact” caused by deforming the tissue before it was well fixed (32–34). 2. The dorsal root has an enlargement, the dorsal root ganglion containing the sensory cell bodies near the point of bifurcation. 3. The paraformaldehyde should be of high quality (such as that supplied by Electron Microscopy Services) and should be prepared fresh each day. Over fixation or low-grade fixatives will dramatically reduce or eliminate the antigenicity of presynaptic proteins. 4. It is better to lose some NMJs than to waste a lot of effort sectioning regions of the muscle where there are no synapses. 5. There are a numberof other useful online resources and reference texts that further explain the genetics, physiology, and clinical features of neuromuscular diseases in both humans and mice. The database “Online Mendelian Inheritance in Man (OMIM)” is particularly useful as a source of genetic information related to all forms of heritable human diseases (http://www.ncbi.nlm.nih.gov/sites/entrez?db=OMIM& itool=toolbar). For a resource of mouse genetics, including mapping data, alleles, molecular characterization, and expression information, the Mouse Genome Informatics web site provides a comprehensive and current database (http://www.informatics.jax.org/). Other online resources related to specific human diseases also exist, such as the Inherited Peripheral Neuropathies website, providing genetic and molecular information pertaining to inherited diseases of the peripheral nervous system (http://www.molgen.ua.ac.be/CMTMutations/). Useful reference texts include Myology (A. G. Engel, C FranziniArmstrong, Eds.), Peripheral Neuropathy (P. J. Dyck and P. K. Thomas, Eds.), Pathology of Peripheral Nerves (J. M. Schroder, Ed.), and Neuromuscular Disorders: Clinical and Molecular Genetics (A. E. H. Emery, Ed.). See ref. (35–38).

386

Burgess, Cox, and Seburn

References 1. Hegedus, J., Putman, C. T. and Gordon, T. (2007) Time course of preferential motor unit loss in the SOD1 G93A mouse model of amyotrophic lateral sclerosis. Neurobiol Dis 28, 154–164. 2. Grohmann, K., Schuelke, M., Diers, A., Hoffmann, K., Lucke, B., Adams, C., et al. (2001) Mutations in the gene encoding immunoglobulin mu-binding protein 2 cause spinal muscular atrophy with respiratory distress type 1. Nat Genet 29, 75–77. 3. Seburn, K. L., Nangle, L. A., Cox, G. A., Schimmel, P. and Burgess, R. W. (2006) An active dominant mutation of glycyl-tRNA synthetase causes neuropathy in a CharcotMarie-Tooth 2D mouse model. Neuron 51, 715–726. 4. Court, F. A., Sherman, D. L., Pratt, T., Garry, E. M., Ribchester, R. R., Cottrell, D. F., et al. (2004) Restricted growth of Schwann cells lacking Cajal bands slows conduction in myelinated nerves. Nature 431, 191–195. 5. Son, Y. J. and Thompson, W. J. (1995) Nerve sprouting in muscle is induced and guided by processes extended by Schwann cells. Neuron 14, 133–141. 6. Son, Y. J. and Thompson, W. J. (1995) Schwann cell processes guide regeneration of peripheral axons. Neuron 14, 125–132. 7. Enomoto, H., Araki, T., Jackman, A., Heuckeroth, R. O., Snider, W. D., Johnson, E. M., Jr., et al. (1998) GFR alpha1-deficient mice have deficits in the enteric nervous system and kidneys. Neuron 21, 317–324. 8. Patton, B. L., Cunningham, J. M., Thyboll, J., Kortesmaa, J., Westerblad, H., Edstrom, L., et al. (2001) Properly formed but improperly localized synaptic specializations in the absence of laminin alpha4. Nat Neurosci 4, 597–604. 9. Pun, S., Santos, A. F., Saxena, S., Xu, L. and Caroni, P. (2006) Selective vulnerability and pruning of phasic motoneuron axons in motoneuron disease alleviated by CNTF. Nat Neurosci 9, 408–419. 10. Morgan, D., Munireddy, S., Alamed, J., Deleon, J., Diamond, D. M., Bickford, P., et al. (2008) Apparent behavioral benefits of tau overexpression in P301L tau transgenic mice. J Alzheimers Dis 15, 605–614. 11. Stam, N. C., Nithianantharajah, J., Howard, M. L., Atkin, J. D., Cheema, S. S. and Hannan, A. J. (2008) Sex-specific behavioural effects of environmental enrichment in a transgenic mouse model of amyotrophic lat-

eral sclerosis. Eur J Neurosci 28, 717–723. 12. Turgeman, T., Hagai, Y., Huebner, K., Jassal, D. S., Anderson, J. E., Genin, O., et al. (2008) Prevention of muscle fibrosis and improvement in muscle performance in the mdx mouse by halofuginone. Neuromuscul Disord 18, 857–868. 13. Bohlen, M., Cameron, A., Metten, P., Crabbe, J. C. and Wahlsten, D. (2009) Calibration of rotational acceleration for the rotarod test of rodent motor coordination. J Neurosci Methods 178, 10–14. 14. Rustay, N. R., Wahlsten, D. and Crabbe, J. C. (2003) Influence of task parameters on rotarod performance and sensitivity to ethanol in mice. Behav Brain Res 141, 237–249. 15. Bautmans, I. and Mets, T. (2005) A fatigue resistance test for elderly persons based on grip strength: reliability and comparison with healthy young subjects. Aging Clin Exp Res 17, 217–222. 16. Fernagut, P. O., Diguet, E., Labattu, B. and Tison, F. (2002) A simple method to measure stride length as an index of nigrostriatal dysfunction in mice. J Neurosci Methods 113, 123–130. 17. Wooley, C. M., Sher, R. B., Kale, A., Frankel, W. N., Cox, G. A. and Seburn, K. L. (2005) Gait analysis detects early changes in transgenic SOD1(G93A) mice. Muscle Nerve 32, 43–50. 18. Wooley, C. M., Xing, S., Burgess, R. W., Cox, G. A. and Seburn, K. L. (2009) Age, experience and genetic background influence treadmill walking in mice. Physiol Behav 96, 350–361. 19. Brooks, S. V., Faulkner, J. A. and McCubbrey, D. A. (1990) Power outputs of slow and fast skeletal muscles of mice. J Appl Physiol 68, 1282–1285. 20. Seburn, K. L. and Gardiner, P. (1995) Adaptations of rat lateral gastrocnemius motor units in response to voluntary running. J Appl Physiol 78, 1673–1678. 21. Lynch, G. S., Hinkle, R. T., Chamberlain, J. S., Brooks, S. V. and Faulkner, J. A. (2001) Force and power output of fast and slow skeletal muscles from mdx mice 6-28 months old. J Physiol 535, 591–600. 22. Brooks, S. V. and Faulkner, J. A. (1988) Contractile properties of skeletal muscles from young, adult and aged mice. J Physiol 404, 71–82. 23. Grange, R. W., Gainer, T. G., Marschner, K. M., Talmadge, R. J. and Stull, J. T. (2002) Fast-twitch skeletal muscles of dys-

Neuromuscular Disease Models and Analysis

24.

25.

26.

27.

28.

29.

30.

31.

32.

33. 34.

35.

trophic mouse pups are resistant to injury from acute mechanical stress. Am J Physiol Cell Physiol 283, C1090–1101. Belluardo, N., Westerblad, H., Mudo, G., Casabona, A., Bruton, J., Caniglia, G., et al. (2001) Neuromuscular junction disassembly and muscle fatigue in mice lacking neurotrophin-4. Mol Cell Neurosci 18, 56–67. Danieli-Betto, D., Esposito, A., Germinario, E., Sandona, D., Martinello, T., Jakubiec-Puka, A., et al. (2005) Deficiency of alpha-sarcoglycan differently affects fastand slow-twitch skeletal muscles. Am J Physiol Regul Integr Comp Physiol 289, R1328–1337. Grange, R. W., Meeson, A., Chin, E., Lau, K. S., Stull, J. T., Shelton, J. M., et al. (2001) Functional and molecular adaptations in skeletal muscle of myoglobinmutant mice. Am J Physiol Cell Physiol 281, C1487–1494. Parry, D. J. and Desypris, G. (1985) Fatiguability and oxidative capacity of forelimb and hind limb muscles of dystrophic mice. Exp Neurol 87, 358–368. Swallow, J. G., Garland, T., Jr., Carter, P. A., Zhan, W. Z. and Sieck, G. C. (1998) Effects of voluntary activity and genetic selection on aerobic capacity in house mice (Mus domesticus). J Appl Physiol 84, 69–76. Shefner, J. M. (2001) Motor unit number estimation in human neurological diseases and animal models. Clin Neurophysiol 112, 955–964. McComas, A. J., Fawcett, P. R., Campbell, M. J. and Sica, R. E. (1971) Electrophysiological estimation of the number of motor units within a human muscle. J Neurol Neurosurg Psychiatry 34, 121–131. Major, L. A., Hegedus, J., Weber, D. J., Gordon, T. and Jones, K. E. (2007) Method for counting motor units in mice and validation using a mathematical model. J Neurophysiol 97, 1846–1856. Ebels, E. J. (1975) Dark neurons: a significant artifact: the influence of the maturational state of neurons on the occurrence of the phenomenon. Acta Neuropathol 33, 271–273. Garman, R. H. (1990) Artifacts in routinely immersion fixed nervous tissue. Toxicol Pathol 18, 149–153. Jortner, B. S. (2006) The return of the dark neuron. A histological artifact complicating contemporary neurotoxicologic evaluation. Neurotoxicology 27, 628–634. Schr`eoder, J. M. (2001) Pathology of peripheral nerves: an atlas of structural and

36. 37. 38.

39.

40.

41.

42.

43.

44.

45.

46.

387

molecular pathological changes, Springer, Berlin/New York. Emery, A. E. H. (1998) Neuromuscular disorders: clinical and molecular genetics, John Wiley, Chichester/New York. Dyck, P. J. and Thomas, P. K. (2005) Peripheral neuropathy, 4th ed. Elsevier Saunders, Philadelphia, PA. Engel, A. and Franzini-Armstrong, C. (2004) Myology: basic and clinical, 3rd ed. McGraw-Hill, Medical Pub. Division,INTbreak; New York. Maddatu, T. P., Garvey, S. M., Schroeder, D. G., Hampton, T. G. and Cox, G. A. (2004) Transgenic rescue of neurogenic atrophy in the nmd mouse reveals a role for Ighmbp2 in dilated cardiomyopathy. Hum Mol Genet 13, 1105–1115. Hershkovitz, E., Rozin, I., Limony, Y., Golan, H., Hadad, N., Gorodischer, R., et al. (2007) Hypoparathyroidism, retardation, and dysmorphism syndrome: impaired early growth and increased susceptibility to severe infections due to hyposplenism and impaired polymorphonuclear cell functions. Pediatr Res 62, 505–509. Noakes, P. G., Miner, J. H., Gautam, M., Cunningham, J. M., Sanes, J. R. and Merlie, J. P. (1995) The renal glomerulus of mice lacking s-laminin/laminin b2: nephrosis despite molecular compensation by laminin b1. Nat Genet 10, 400–406. Zhai, R. G., Cao, Y., Hiesinger, P. R., Zhou, Y., Mehta, S. Q., Schulze, K. L., et al. (2006) Drosophila NMNAT maintains neural integrity independent of its NAD synthesis activity. PLoS Biol 4, e416. Zhai, R. G., Zhang, F., Hiesinger, P. R., Cao, Y., Haueter, C. M. and Bellen, H. J. (2008) NAD synthase NMNAT acts as a chaperone to protect against neurodegeneration. Nature 452, 887–891. Press, C. and Milbrandt, J. (2008) Nmnat delays axonal degeneration caused by mitochondrial and oxidative stress. J Neurosci 28, 4861–4871. Maddatu, T. P., Garvey, S. M., Schroeder, D. G., Zhang, W., Kim, S. Y., et al. (2005) Dilated cardiomyopathy in the nmd mouse: transgenic rescue and QTLs that improve cardiac function and survival. Hum Mol Genet 14, 3179–3189. Grohmann, K., Schuelke, M., Diers, A., Hoffmann, K., Lucke, B., Adams, C., et al. (2001) Mutations in the gene encoding immunoglobulin ␮-binding protein 2 cause spinal muscular atrophy with respiratory distress type 1. Nat Genet 29, 75–77.

388

Burgess, Cox, and Seburn

47. Cox, G. A., Mahaffey, C. L. and Frankel, W. N. (1998) Identification of the mouse neuromuscular degeneration gene and mapping of a second site suppressor allele. Neuron 21, 1327–1337. 48. Monani, U. R., Pastore, M. T., Gavrilina, T. O., Jablonka, S., Le, T. T., Andreassi, C., et al. (2003) A transgene carrying an A2G missense mutation in the SMN gene modulates phenotypic severity in mice with severe (type I) spinal muscular atrophy. J Cell Biol 160, 41–52. 49. Le, T. T., Pham, L. T., Butchbach, M. E., Zhang, H. L., Monani, U. R., Coovert, D. D., et al. (2005) SMN Delta 7, the major product of the centromeric survival motor neuron (SMN2) gene, extends survival in mice with spinal muscular atrophy and associates with full-length SMN. Hum Mol Genet 14, 845–857. 50. Monani, U. R., Sendtner, M., Coovert, D. D., Parsons, D. W., Andreassi, C., Le, T. T., et al. (2000) The human centromeric survival motor neuron gene (SMN2) rescues embryonic lethality in Smn(-/-) mice and results in a mouse with spinal muscular atrophy. Hum Mol Genet 9, 333–339. 51. Schrank, B., Gotz, R., Gunnersen, J. M., Ure, J. M., Toyka, K. V., Smith, A. G., et al. (1997) Inactivation of the survival motor neuron gene, a candidate gene for human spinal muscular atrophy, leads to massive cell death in early mouse embryos. Proc Natl Acad Sci U S A 94, 9920–9925. 52. Martin, N., Jaubert, J., Gounon, P., Salido, E., Haase, G., Szatanik, M., et al. (2002) A missense mutation in Tbce causes progressive motor neuronopathy in mice. Nat Genet 32, 443–447. 53. Bommel, H., Xie, G., Rossoll, W., Wiese, S., Jablonka, S., Boehm, T., et al. (2002) Missense mutation in the tubulin-specific chaperone E (Tbce) gene in the mouse mutant progressive motor neuronopathy, a model of human motoneuron disease. J Cell Biol 159, 563–569. 54. Puls, I., Jonnakuty, C., LaMonte, B. H., Holzbaur, E. L., Tokito, M., Mann, E., et al. (2003) Mutant dynactin in motor neuron disease. Nat Genet 33, 455–456. 55. Chevalier-Larsen, E. S., Wallace, K. E., Pennise, C. R. and Holzbaur, E. L. (2008) Lysosomal proliferation and distal degeneration in motor neurons expressing the G59S mutation in the p150Glued subunit of dynactin. Hum Mol Genet 17, 1946–1955. 56. Lai, C., Lin, X., Chandran, J., Shim, H., Yang, W. J. and Cai, H. (2007) The G59S mutation in p150(glued) causes dysfunc-

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

tion of dynactin in mice. J Neurosci 27, 13982–13990. LaMonte, B. H., Wallace, K. E., Holloway, B. A., Shelly, S. S., Ascano, J., Tokito, M., et al. (2002) Disruption of dynein/dynactin inhibits axonal transport in motor neurons causing late-onset progressive degeneration. Neuron 34, 715–727. Ripps, M. E., Huntley, G. W., Hof, P. R., Morrison, J. H. and Gordon, J. W. (1995) Transgenic mice expressing an altered murine superoxide dismutase gene provide an animal model of amyotrophic lateral sclerosis. Proc Natl Acad Sci U S A 92, 689–693. Kunst, C. B., Messer, L., Gordon, J., Haines, J. and Patterson, D. (2000) Genetic mapping of a mouse modifier gene that can prevent ALS onset. Genomics 70, 181–189. Bruijn, L. I., Becher, M. W., Lee, M. K., Anderson, K. L., Jenkins, N. A., Copeland, N. G., et al. (1997) ALS-linked SOD1 mutant G85R mediates damage to astrocytes and promotes rapidly progressive disease with SOD1-containing inclusions. Neuron 18, 327–338. Wang, J., Xu, G., Gonzales, V., Coonfield, M., Fromholt, D., Copeland, N. G., et al. (2002) Fibrillar inclusions and motor neuron degeneration in transgenic mice expressing superoxide dismutase 1 with a disrupted copper-binding site. Neurobiol Dis 10, 128–138. Wong, P. C., Pardo, C. A., Borchelt, D. R., Lee, M. K., Copeland, N. G., Jenkins, N. A., et al. (1995) An adverse property of a familial ALS-linked SOD1 mutation causes motor neuron disease characterized by vacuolar degeneration of mitochondria. Neuron 14, 1105–1116. Gurney, M. E., Pu, H., Chiu, A. Y., Dal Canto, M. C., Polchow, C. Y., Alexander, D. D., et al. (1994) Motor neuron degeneration in mice that express a human Cu,Zn superoxide dismutase mutation. Science 264, 1772–1775. Heiman-Patterson, T. D., Deitch, J. S., Blankenhorn, E. P., Erwin, K. L., Perreault, M. J., Alexander, B. K., et al. (2005) Background and gender effects on survival in the TgN(SOD1-G93A)1Gur mouse model of ALS. J Neurol Sci 236, 1–7. Messer, A. and Flaherty, L. (1986) Autosomal dominance in a late-onset motor neuron disease in the mouse. J Neurogenet 3, 345–355. Bronson, R. T., Lake, B. D., Cook, S., Taylor, S. and Davisson, M. T. (1993) Motor neuron degeneration of mice is a model of neuronal

Neuromuscular Disease Models and Analysis

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

ceroid lipofuscinosis (Batten’s disease). Ann Neurol 33, 381–385. Ranta, S., Zhang, Y., Ross, B., Lonka, L., Takkunen, E., Messer, A., et al. (1999) The neuronal ceroid lipofuscinoses in human EPMR and mnd mutant mice are associated with mutations in CLN8. Nat Genet 23, 233–236. Hafezparast, M., Klocke, R., Ruhrberg, C., Marquardt, A., Ahmad-Annuar, A., Bowen, S., et al. (2003) Mutations in dynein link motor neuron degeneration to defects in retrograde transport. Science 300, 808–812. Chen, X. J., Levedakou, E. N., Millen, K. J., Wollmann, R. L., Soliven, B. and Popko, B. (2007) Proprioceptive sensory neuropathy in mice with a mutation in the cytoplasmic Dynein heavy chain 1 gene. J Neurosci 27, 14515–14524. Ilieva, H. S., Yamanaka, K., Malkmus, S., Kakinohana, O., Yaksh, T., Marsala, M., et al. (2008) Mutant dynein (Loa) triggers proprioceptive axon loss that extends survival only in the SOD1 ALS model with highest motor neuron death. Proc Natl Acad Sci U S A 105, 12599–12604. Chow, C. Y., Landers, J. E., Bergren, S. K., Sapp, P. C., Grant, A. E., Jones, J. M., et al. (2009) Deleterious variants of FIG4, a phosphoinositide phosphatase, in patients with ALS. Am J Hum Genet 84, 85–88. Chow, C. Y., Zhang, Y., Dowling, J. J., Jin, N., Adamska, M., Shiga, K., et al. (2007) Mutation of FIG4 causes neurodegeneration in the pale tremor mouse and patients with CMT4J. Nature 448, 68–72. Zhang, X., Chow, C. Y., Sahenk, Z., Shy, M. E., Meisler, M. H. and Li, J. (2008) Mutation of FIG4 causes a rapidly progressive, asymmetric neuronal degeneration. Brain 131, 1990–2001. Schmitt-John, T., Drepper, C., Mussmann, A., Hahn, P., Kuhlmann, M., Thiel, C., et al. (2005) Mutation of Vps54 causes motor neuron disease and defective spermiogenesis in the wobbler mouse. Nat Genet 37, 1213–1215. Suter, U., Welcher, A. A., Ozcelik, T., Snipes, G. J., Kosaras, B., Francke, U., et al. (1992) Trembler mouse carries a point mutation in a myelin gene. Nature 356, 241–244. Giese, K. P., Martini, R., Lemke, G., Soriano, P. and Schachner, M. (1992) Mouse P0 gene disruption leads to hypomyelination, abnormal expression of recognition molecules, and degeneration of myelin and axons. Cell 71, 565–576. Wrabetz, L., Feltri, M. L., Quattrini, A., Imperiale, D., Previtali, S., D’Antonio,

78.

79.

80.

81.

82.

83.

84.

85.

86.

87.

389

M., et al. (2000) P(0) glycoprotein overexpression causes congenital hypomyelination of peripheral nerves. J Cell Biol 148, 1021–1034. Wrabetz, L., D’Antonio, M., Pennuto, M., Dati, G., Tinelli, E., Fratta, P., et al. (2006) Different intracellular pathomechanisms produce diverse Myelin Protein Zero neuropathies in transgenic mice. J Neurosci 26, 2358–2368. Le, N., Nagarajan, R., Wang, J. Y., Araki, T., Schmidt, R. E. and Milbrandt, J. (2005) Analysis of congenital hypomyelinating Egr2Lo/Lo nerves identifies Sox2 as an inhibitor of Schwann cell differentiation and myelination. Proc Natl Acad Sci U S A 102, 2596–2601. Topilko, P., Schneider-Maunoury, S., Levi, G., Baron-Van Evercooren, A., Chennoufi, A. B., Seitanidou, T., et al. (1994) Krox-20 controls myelination in the peripheral nervous system. Nature 371, 796–799. Nelles, E., Butzler, C., Jung, D., Temme, A., Gabriel, H. D., Dahl, U., et al. (1996) Defective propagation of signals generated by sympathetic nerve stimulation in the liver of connexin32-deficient mice. Proc Natl Acad Sci U S A 93, 9565–9570. Anzini, P., Neuberg, D. H., Schachner, M., Nelles, E., Willecke, K., Zielasek, J., et al. (1997) Structural abnormalities and deficient maintenance of peripheral nerve myelin in mice lacking the gap junction protein connexin 32. J Neurosci 17, 4545–4551. Jeng, L. J., Balice-Gordon, R. J., Messing, A., Fischbeck, K. H. and Scherer, S. S. (2006) The effects of a dominant connexin 32 mutant in myelinating Schwann cells. Mol Cell Neurosci 32, 283–298. Detmer, S. A., Vande Velde, C., Cleveland, D. W. and Chan, D. C. (2008) Hindlimb gait defects due to motor axon loss and reduced distal muscles in a transgenic mouse model of Charcot-Marie-Tooth type 2A. Hum Mol Genet 17, 367–375. Lee, M. K., Marszalek, J. R. and Cleveland, D. W. (1994) A mutant neurofilament subunit causes massive, selective motor neuron death: implications for the pathogenesis of human motor neuron disease. Neuron 13, 975–988. Zhu, Q., Couillard-Despres, S. and Julien, J. P. (1997) Delayed maturation of regenerating myelinated axons in mice lacking neurofilaments. Exp Neurol 148, 299–316. Huang, L., Min, J. N., Masters, S., Mivechi, N. F. and Moskophidis, D. (2007) Insights

390

88.

89.

90.

91.

92.

93.

94.

95.

96.

97.

Burgess, Cox, and Seburn into function and regulation of small heat shock protein 25 (HSPB1) in a mouse model with targeted gene disruption. Genesis 45, 487–501. Gillespie, C. S., Sherman, D. L., FleetwoodWalker, S. M., Cottrell, D. F., Tait, S., Garry, E. M., et al. (2000) Peripheral demyelination and neuropathic pain behavior in periaxindeficient mice. Neuron 26, 523–531. Brandon, E. P., Lin, W., D’Amour, K. A., Pizzo, D. P., Dominguez, B., Sugiura, Y., et al. (2003) Aberrant patterning of neuromuscular synapses in choline acetyltransferase-deficient mice. J Neurosci 23, 539–549. Misgeld, T., Burgess, R. W., Lewis, R. M., Cunningham, J. M., Lichtman, J. W. and Sanes, J. R. (2002) Roles of neurotransmitter in synapse formation. Development of neuromuscular junctions lacking choline acetyltransferase. Neuron 36, 635–648. Buffelli, M., Burgess, R. W., Feng, G., Lobe, C. G., Lichtman, J. W. and Sanes, J. R. (2003) Genetic evidence that relative synaptic efficacy biases the outcome of synaptic competition. Nature 424, 430–434. Prado, V. F., Martins-Silva, C., de Castro, B. M., Lima, R. F., Barros, D. M., Amaral, E., et al. (2006) Mice deficient for the vesicular acetylcholine transporter are myasthenic and have deficits in object and social recognition. Neuron 51, 601–612. Feng, G., Krejci, E., Molgo, J., Cunningham, J. M., Massoulie, J. and Sanes, J. R. (1999) Genetic analysis of collagen Q: roles in acetylcholinesterase and butyrylcholinesterase assembly and in synaptic structure and function. J Cell Biol 144, 1349–1360. Adler, M., Manley, H. A., Purcell, A. L., Deshpande, S. S., Hamilton, T. A., Kan, R. K., et al. (2004) Reduced acetylcholine receptor density, morphological remodeling, and butyrylcholinesterase activity can sustain muscle function in acetylcholinesterase knockout mice. Muscle Nerve 30, 317–327. Mesulam, M. M., Guillozet, A., Shaw, P., Levey, A., Duysen, E. G. and Lockridge, O. (2002) Acetylcholinesterase knockouts establish central cholinergic pathways and can use butyrylcholinesterase to hydrolyze acetylcholine. Neuroscience 110, 627–639. Burgess, R. W., Nguyen, Q. T., Son, Y. J., Lichtman, J. W. and Sanes, J. R. (1999) Alternatively spliced isoforms of nerve- and muscle-derived agrin: their roles at the neuromuscular junction. Neuron 23, 33–44. Burgess, R. W., Skarnes, W. C. and Sanes, J. R. (2000) Agrin isoforms with distinct amino

98.

99.

100.

101.

102.

103.

104.

105.

106.

107.

108.

termini: differential expression, localization, and function. J Cell Biol 151, 41–52. Lin, W., Burgess, R. W., Dominguez, B., Pfaff, S. L., Sanes, J. R. and Lee, K. F. (2001) Distinct roles of nerve and muscle in postsynaptic differentiation of the neuromuscular synapse. Nature 410, 1057–1064. Harvey, S. J., Jarad, G., Cunningham, J., Rops, A. L., van der Vlag, J., Berden, J. H., et al. (2007) Disruption of glomerular basement membrane charge through podocytespecific mutation of agrin does not alter glomerular permselectivity. Am J Pathol 171, 139–152. Gautam, M., Noakes, P. G., Moscoso, L., Rupp, F., Scheller, R. H., Merlie, J. P., et al. (1996) Defective neuromuscular synaptogenesis in agrin-deficient mutant mice. Cell 85, 525–535. DeChiara, T. M., Bowen, D. C., Valenzuela, D. M., Simmons, M. V., Poueymirou, W. T., Thomas, S., et al. (1996) The receptor tyrosine kinase MuSK is required for neuromuscular junction formation in vivo. Cell 85, 501–512. Weatherbee, S. D., Anderson, K. V. and Niswander, L. A. (2006) LDL-receptorrelated protein 4 is crucial for formation of the neuromuscular junction. Development 133, 4993–5000. Gautam, M., Noakes, P. G., Mudd, J., Nichol, M., Chu, G. C., Sanes, J. R.,et al. (1995) Failure of postsynaptic specialization to develop at neuromuscular junctions of rapsyn-deficient mice. Nature 377, 232–236. Okada, K., Inoue, A., Okada, M., Murata, Y., Kakuta, S., Jigami, T., et al. (2006) The muscle protein Dok-7 is essential for neuromuscular synaptogenesis. Science 312, 1802–1805. Noakes, P. G., Gautam, M., Mudd, J., Sanes, J. R. and Merlie, J. P. (1995) Aberrant differentiation of neuromuscular junctions in mice lacking s-laminin/laminin beta 2. Nature 374, 258–262. Friese, M. B., Blagden, C. S. and Burden, S. J. (2007) Synaptic differentiation is defective in mice lacking acetylcholine receptor betasubunit tyrosine phosphorylation. Development 134, 4167–4176. Missias, A. C., Mudd, J., Cunningham, J. M., Steinbach, J. H., Merlie, J. P. and Sanes, J. R. (1997) Deficient development and maintenance of postsynaptic specializations in mutant mice lacking an ’adult’ acetylcholine receptor subunit. Development 124, 5075–5086. Witzemann, V., Schwarz, H., Koenen, M., Berberich, C., Villarroel, A., Wernig, A.,

Neuromuscular Disease Models and Analysis

109.

110.

111.

112.

113.

114.

115.

116.

117.

118.

et al. (1996) Acetylcholine receptor epsilonsubunit deletion causes muscle weakness and atrophy in juvenile and adult mice. Proc Natl Acad Sci USA 93, 13286–13291. Takahashi, M., Kubo, T., Mizoguchi, A., Carlson, C. G., Endo, K. and Ohnishi, K. (2002) Spontaneous muscle action potentials fail to develop without fetal-type acetylcholine receptors. EMBO Rep 3, 674–681. Koenen, M., Peter, C., Villarroel, A., Witzemann, V. and Sakmann, B. (2005) Acetylcholine receptor channel subtype directs the innervation pattern of skeletal muscle. EMBO Rep 6, 570–576. Liu, Y., Padgett, D., Takahashi, M., Li, H., Sayeed, A., Teichert, R. W., et al. (2008) Essential roles of the acetylcholine receptor gamma-subunit in neuromuscular synaptic patterning. Development 135, 1957–1967. Sicinski, P., Geng, Y., Ryder Cook, A. S., Barnard, E. A., Darlison, M. G. and Barnard, P. J. (1989) The molecular basis of muscular dystrophy in the mdx mouse: a point mutation. Science 244, 1578–1580. Chapman, V. M., Miller, D. R., Armstrong, D. and Caskey, C. T. (1989) Recovery of induced mutations for X chromosome-linked muscular dystrophy in mice. Proc Natl Acad Sci U S Am 86, 1292–1296. Danko, I., Chapman, V. and Wolff, J. A. (1992) The frequency of revertants in mdx mouse genetic models for Duchenne muscular dystrophy. Pediatr Res 32, 128–131. Cox, G. A., Phelps, S. F., Chapman, V. M. and Chamberlain, J. S. (1993) New mdx mutation disrupts expression of muscle and nonmuscle isoforms of dystrophin. Nature Genetics 4, 87–93. Im, W. B., Phelps, S. F., Copen, E. H., Adams, E. G., Slightom, J. L. and Chamberlain, J. S. (1996) Differential expression of dystrophin isoforms in strains of mdx mice with different mutations. Hum Mol Genet 5, 1149–1153. Araki, E., Nakamura, K., Nakao, K., Kameya, S., Kobayashi, O., Nonaka, I., et al. (1997) Targeted disruption of exon 52 in the mouse dystrophin gene induced muscle degeneration similar to that observed in Duchenne muscular dystrophy. Biochem Biophys Res Commun 238, 492–497. Kudoh, H., Ikeda, H., Kakitani, M., Ueda, A., Hayasaka, M., Tomizuka, K., et al. (2005) A new model mouse for Duchenne muscular dystrophy produced by 2.4 Mb deletion of dystrophin gene using Cre-loxP recombination system. Biochem Biophys Res Commun 328, 507–516.

391

119. Grewal, P. K., Holzfeind, P. J., Bittner, R. E. and Hewitt, J. E. (2001) Mutant glycosyltransferase and altered glycosylation of alphadystroglycan in the myodystrophy mouse. Nat Genet 28, 151–154. 120. Lee, Y., Kameya, S., Cox, G. A., Hsu, J., Hicks, W., Maddatu, T. P., et al. (2005) Ocular abnormalities in Large(myd) and Large(vls) mice, spontaneous models for muscle, eye, and brain diseases. Mol Cell Neurosci 30, 160–172. 121. Kuang, W., Xu, H., Vachon, P. H., Liu, L., Loechel, F., Wewer, U. M., et al. (1998) Merosin-deficient congenital muscular dystrophy. Partial genetic correction in two mouse models. J Clin Invest 102, 844–852. 122. Michelson, A. M., Russell, E. S. and Harman, P. J. (1955) Dystrophia muscularis: a hereditary primary myopathy in the house mouse. Proc Natl Acad Sci U S A 12, 1079–1084. 123. Miyagoe, Y., Hanaoka, K., Nonaka, I., Hayasaka, M., Nabeshima, Y., Arahata, K., et al. (1997) Laminin alpha2 chain-null mutant mice by targeted disruption of the Lama2 gene: a new model of merosin (laminin 2)-deficient congenital muscular dystrophy. FEBS Lett 415, 33–39. 124. Patton, B. L., Wang, B., Tarumi, Y. S., Seburn, K. L. and Burgess, R. W. (2008) A single point mutation in the LN domain of LAMA2 causes muscular dystrophy and peripheral amyelination. J Cell Sci 121, 1593–1604. 125. Sunada, Y., Bernier, S. M., Utani, A., Yamada, Y. and Campbell, K. P. (1995) Identification of a novel mutant transcript of laminin a2 chain gene responsible for muscular dystrophy and dysmyelination in dy2J mice. Hum Mol Genet 4, 1055–1061. 126. Bittner, R. E., Anderson, L. V., Burkhardt, E., Bashir, R., Vafiadaki, E., Ivanova, S., et al. (1999) Dysferlin deletion in SJL mice (SJLDysf) defines a natural model for limb girdle muscular dystrophy 2B. Nat Genet 23, 141–142. 127. Bansal, D., Miyake, K., Vogel, S. S., Groh, S., Chen, C. C., Williamson, R., et al. (2003) Defective membrane repair in dysferlindeficient muscular dystrophy. Nature 423, 168–172. 128. Ho, M., Post, C. M., Donahue, L. R., Lidov, H. G., Bronson, R. T., Goolsby, H., et al. (2004) Disruption of muscle membrane and phenotype divergence in two novel mouse models of dysferlin deficiency. Hum Mol Genet 13, 1999–2010. 129. Grady, R. M., Grange, R. W., Lau, K. S., Maimone, M. M., Nichol, M. C., Stull, J. T., et al. (1999) Role for alpha-dystrobrevin

392

130.

131.

132.

133.

134.

135.

136.

137.

138.

139.

Burgess, Cox, and Seburn in the pathogenesis of dystrophin-dependent muscular dystrophies. Nat Cell Biol 1, 215–220. Bonaldo, P., Braghetta, P., Zanetti, M., Piccolo, S., Volpin, D. and Bressan, G. M. (1998) Collagen VI deficiency induces early onset myopathy in the mouse: an animal model for Bethlem myopathy. Hum Mol Genet 7, 2135–2140. Irwin, W. A., Bergamin, N., Sabatelli, P., Reggiani, C., Megighian, A., Merlini, L., et al. (2003) Mitochondrial dysfunction and apoptosis in myopathic mice with collagen VI deficiency. Nat Genet 35, 367–371. Eklund, L., Piuhola, J., Komulainen, J., Sormunen, R., Ongvarrasopone, C., Fassler, R., et al. (2001) Lack of type XV collagen causes a skeletal myopathy and cardiovascular defects in mice. Proc Natl Acad Sci U S A 98, 1194–1199. Galbiati, F., Engelman, J. A., Volonte, D., Zhang, X. L., Minetti, C., Li, M., et al. (2001) Caveolin-3 null mice show a loss of caveolae, changes in the microdomain distribution of the dystrophin-glycoprotein complex, and t-tubule abnormalities. J Biol Chem 276, 21425–21433. Crawford, K., Flick, R., Close, L., Shelly, D., Paul, R., Bove, K., et al. (2002) Mice lacking skeletal muscle actin show reduced muscle strength and growth deficits and die during the neonatal period. Mol Cell Biol 22, 5887–5896. Ilkovski, B., Cooper, S. T., Nowak, K., Ryan, M. M., Yang, N., Schnell, C., et al. (2001) Nemaline myopathy caused by mutations in the muscle alpha-skeletal-actin gene. Am J Hum Genet 68, 1333–1343. Garvey, S. M., Rajan, C., Lerner, A. P., Frankel, W. N. and Cox, G. A. (2002) The muscular dystrophy with myositis (mdm) mouse mutation disrupts a skeletal musclespecific domain of titin. Genomics 79, 146–149. Gotthardt, M., Hammer, R. E., Hubner, N., Monti, J., Witt, C. C., McNabb, M., et al. (2003) Conditional expression of mutant Mline titins results in cardiomyopathy with altered sarcomere structure. J Biol Chem 278, 6059–6065. May, S. R., Stewart, N. J., Chang, W. and Peterson, A. S. (2004) A Titin mutation defines roles for circulation in endothelial morphogenesis. Dev Biol 270, 31–46. Brady, J. P., Garland, D. L., Green, D. E., Tamm, E. R., Giblin, F. J. and Wawrousek, E. F. (2001) AlphaB-crystallin in lens development and muscle integrity: a gene knock-

140.

141.

142.

143.

144.

145.

146.

147.

148.

149.

150.

out approach. Invest Ophthalmol Vis Sci 42, 2924–2934. Homma, S., Iwasaki, M., Shelton, G. D., Engvall, E., Reed, J. C. and Takayama, S. (2006) BAG3 deficiency results in fulminant myopathy and early lethality. Am J Pathol 169, 761–773. Selcen, D., Muntoni, F., Burton, B. K., Pegoraro, E., Sewry, C., Bite, A. V., et al. (2009) Mutation in BAG3 causes severe dominant childhood muscular dystrophy. Ann Neurol 65, 83–89. Li, Z., Colucci-Guyon, E., Pincon-Raymond, M., Mericskay, M., Pournin, S., Paulin, D., et al. (1996) Cardiovascular lesions and skeletal myopathy in mice lacking desmin. Dev Biol 175, 362–366. Milner, D. J., Weitzer, G., Tran, D., Bradley, A. and Capetanaki, Y. (1996) Disruption of muscle architecture and myocardial degeneration in mice lacking desmin. J Cell Biol 134, 1255–1270. Thornell, L., Carlsson, L., Li, Z., Mericskay, M. and Paulin, D. (1997) Null mutation in the desmin gene gives rise to a cardiomyopathy. J Mol Cell Cardiol 29, 2107–2124. Moza, M., Mologni, L., Trokovic, R., Faulkner, G., Partanen, J. and Carpen, O. (2007) Targeted deletion of the muscular dystrophy gene myotilin does not perturb muscle structure or function in mice. Mol Cell Biol 27, 244–252. Garvey, S. M., Miller, S. E., Claflin, D. R., Faulkner, J. A. and Hauser, M. A. (2006) Transgenic mice expressing the myotilin T57I mutation unite the pathology associated with LGMD1A and MFM. Hum Mol Genet 15, 2348–2362. Sher, R. B., Aoyama, C., Huebsch, K. A., Ji, S., Kerner, J., Yang, Y., et al. (2006) A rostrocaudal muscular dystrophy caused by a defect in choline kinase beta, the first enzyme in phosphatidylcholine biosynthesis. J Biol Chem 281, 4938–4948. Buj-Bello, A., Laugel, V., Messaddeq, N., Zahreddine, H., Laporte, J., Pellissier, J. F., et al. (2002) The lipid phosphatase myotubularin is essential for skeletal muscle maintenance but not for myogenesis in mice. Proc Natl Acad Sci U S A 99, 15060–15065. Sullivan, T., Escalante-Alcalde, D., Bhatt, H., Anver, M., Bhat, N., Nagashima, K., et al. (1999) Loss of A-type lamin expression compromises nuclear envelope integrity leading to muscular dystrophy. J Cell Biol 147, 913–920. Arimura, T., Helbling-Leclerc, A., Massart, C., Varnous, S., Niel, F., Lacene, E., et al. (2005) Mouse model carrying H222P-Lmna

Neuromuscular Disease Models and Analysis

151.

152.

153.

154.

155.

156.

157.

158.

159.

160.

mutation develops muscular dystrophy and dilated cardiomyopathy similar to human striated muscle laminopathies. Hum Mol Genet 14, 155–169. Diaz, F., Thomas, C. K., Garcia, S., Hernandez, D. and Moraes, C. T. (2005) Mice lacking COX10 in skeletal muscle recapitulate the phenotype of progressive mitochondrial myopathies associated with cytochrome c oxidase deficiency. Hum Mol Genet 14, 2737–2748. Mankodi, A., Logigian, E., Callahan, L., McClain, C., White, R., Henderson, D., et al. (2000) Myotonic dystrophy in transgenic mice expressing an expanded CUG repeat. Science 289, 1769–1773. Seznec, H., Agbulut, O., Sergeant, N., Savouret, C., Ghestem, A., Tabti, N., et al. (2001) Mice transgenic for the human myotonic dystrophy region with expanded CTG repeats display muscular and brain abnormalities. Hum Mol Genet 10, 2717–2726. Orengo, J. P., Chambon, P., Metzger, D., Mosier, D. R., Snipes, G. J. and Cooper, T. A. (2008) Expanded CTG repeats within the DMPK 3’ UTR causes severe skeletal muscle wasting in an inducible mouse model for myotonic dystrophy. Proc Natl Acad Sci U S A 105, 2646–2651. Reddy, S., Smith, D. B., Rich, M. M., Leferovich, J. M., Reilly, P., Davis, B. M., et al. (1996) Mice lacking the myotonic dystrophy protein kinase develop a late onset progressive myopathy. Nat Genet 13, 325–335. Kanadia, R. N., Johnstone, K. A., Mankodi, A., Lungu, C., Thornton, C. A., Esson, D., et al. (2003) A muscleblind knockout model for myotonic dystrophy. Science 302, 1978–1980. Malicdan, M. C., Noguchi, S., Nonaka, I., Hayashi, Y. K. and Nishino, I. (2007) A Gne knockout mouse expressing human GNE D176V mutation develops features similar to distal myopathy with rimmed vacuoles or hereditary inclusion body myopathy. Hum Mol Genet 16, 2669–2682. Tronche, F., Kellendonk, C., Kretz, O., Gass, P., Anlag, K., Orban, P. C., et al. (1999) Disruption of the glucocorticoid receptor gene in the nervous system results in reduced anxiety. Nat Genet 23, 99–103. Yang, X., Arber, S., William, C., Li, L., Tanabe, Y., Jessell, T. M., et al. (2001) Patterning of muscle acetylcholine receptor gene expression in the absence of motor innervation. Neuron 30, 399–410. Arber, S., Han, B., Mendelsohn, M., Smith, M., Jessell, T. M. and Sockanathan, S. (1999) Requirement for the homeobox gene Hb9 in

161.

162.

163.

164.

165.

166.

167.

168.

169.

170.

171.

393

the consolidation of motor neuron identity. Neuron 23, 659–674. Schwander, M., Leu, M., Stumm, M., Dorchies, O. M., Ruegg, U. T., Schittny, J., et al. (2003) Beta1 integrins regulate myoblast fusion and sarcomere assembly. Dev Cell 4, 673–685. Bruning, J. C., Michael, M. D., Winnay, J. N., Hayashi, T., Horsch, D., Accili, D., et al. (1998) A muscle-specific insulin receptor knockout exhibits features of the metabolic syndrome of NIDDM without altering glucose tolerance. Mol Cell 2, 559–569. Tallquist, M. D., Weismann, K. E., Hellstrom, M. and Soriano, P. (2000) Early myotome specification regulates PDGFA expression and axial skeleton development. Development 127, 5059–5070. Feltri, M. L., D’Antonio, M., Previtali, S., Fasolini, M., Messing, A. and Wrabetz, L. (1999) P0-Cre transgenic mice for inactivation of adhesion molecules in Schwann cells. Ann N Y Acad Sci 883, 116–123. Young, P., Qiu, L., Wang, D., Zhao, S., Gross, J. and Feng, G. (2008) Single-neuron labeling with inducible Cre-mediated knockout in transgenic mice. Nat Neurosci 11, 721–728. Feng, G., Mellor, R. H., Bernstein, M., Keller-Peck, C., Nguyen, Q. T., Wallace, M., et al. (2000) Imaging neuronal subsets in transgenic mice expressing multiple spectral variants of GFP. Neuron 28, 41–51. Misgeld, T., Kerschensteiner, M., Bareyre, F. M., Burgess, R. W. and Lichtman, J. W. (2007) Imaging axonal transport of mitochondria in vivo. Nat Methods 4, 559–561. Zuo, Y., Lubischer, J. L., Kang, H., Tian, L., Mikesh, M., Marks, A., et al. (2004) Fluorescent proteins expressed in mouse transgenic lines mark subsets of glia, neurons, macrophages, and dendritic cells for vital examination. J Neurosci 24, 10999–11009. Wichterle, H., Lieberam, I., Porter, J. A. and Jessell, T. M. (2002) Directed differentiation of embryonic stem cells into motor neurons. Cell 110, 385–397. Livet, J., Weissman, T. A., Kang, H., Draft, R. W., Lu, J., Bennis, R. A., et al. (2007) Transgenic strategies for combinatorial expression of fluorescent proteins in the nervous system. Nature 450, 56–62. Coleman, M. P., Conforti, L., Buckmaster, E. A., Tarlton, A., Ewing, R. M., Brown, M. C., et al. (1998) An 85-kb tandem triplication in the slow Wallerian degeneration (Wlds) mouse. Proc Natl Acad Sci U S A 95, 9985–9990.

Chapter 21 Murine Model of Cutaneous Infection with Streptococcus pyogenes Eva Medina Abstract Despite the medical advances achieved during the last century to fight against bacteria, viruses, fungi and parasites, infectious diseases are still a major cause of death, disability, and social and economic upheaval for millions around the world. Challenges remain in countering microorganisms even where antibiotics and vaccines are available. Much remains to be learned about basic aspects of the host–pathogen relationship and the complexity of the immune response to infection. Animal models represent a powerful tool to dissect the host response to infection, as well as the pathogenesis of the microbe. One of the advantages of using animal models is that both genetic and environmental factors that may influence the course of an infection can be controlled, allowing a precise cause–effect analysis of the host–pathogen interactions. In addition, there are no real alternatives to whole animal models in the study of integrative physiology and dynamic pathophysiologic alterations. The use of animal models has also proven invaluable for testing the efficacy of experimental antimicrobial agents and their therapeutic regimes. The mouse model is the most widely used for many reasons, including its cost effectiveness, the high number of immunological reagents available for this species, and the relative ease of biocontainment. Mouse strains with specific properties such as transgenic mouse strains with gene insertion or targeted mutation (knock-out) are very effective tools for studying the role of specific genes controlling the immune response to infectious pathogens. Murine models will remain the most appropriate tool for evaluating new therapeutic strategies for the treatment of various diseases. The closer the model is adapted to the human disease, the more reliable will be the results. In this chapter, the experimental procedures required to establish a mouse model of cutaneous and soft tissue infection are detailed. This model has provided invaluable insights into the pathogenicity of the agent for the human host. Key words: Subcutaneous inoculation, Streptococcus pyogenes, skin and soft tissue infection.

1. Introduction Skin and soft tissue infections are defined as infections of the epidermis, dermis, or subcutaneous tissue. They are among the most common human bacterial infections and are frequently observed G. Proetzel, M.V. Wiles (eds.), Mouse Models for Drug Discovery, Methods in Molecular Biology 602, DOI 10.1007/978-1-60761-058-8 21, © Humana Press, a part of Springer Science+Business Media, LLC 2010

395

396

Medina

in clinical practice (1). Most skin and soft tissue infections are caused by Gram-positive bacteria, primarily Staphylococcus aureus and Streptococcus pyogenes (2). Although these infections can be successfully treated using empirical antimicrobial therapy, the increasing prevalence of antibiotic resistance among some bacterial strains implies that new treatment options are required (3). Experimental mouse models of skin and soft tissue infections have played a critical role in providing detailed data about the efficacy of specific antimicrobial agents, pharmacokinetics, and disease pathogenesis. In this chapter, an experimental mouse model of skin and soft tissue infection induced by Streptococcus pyogenes is described. Rapidly progressing soft tissue infections due to group A beta-hemolytic streptococcus, or Streptococcus pyogenes can present with erysipelas, cellulitis, or with necrotizing fasciitis (4). Erysipelas is an acute inflammation of the superficial skin and cutaneous lymphatic vessels, whereas cellulitis is an infection of the lower dermis and subcutaneous tissue (4). The most severe soft tissue infection caused by S. pyogenes is necrotizing fasciitis, also known as the “Flesh-eating disease.” Necrotizing fasciitis is a deep-seated infection of the subcutaneous tissue that results in the rapidly progressive destruction of fascia and fat (4). Despite prompt antibiotic therapy and surgical debridement, streptococcal necrotizing fasciitis is associated with high death rates ranging from 20 to 60% (4). The mouse model described here mimics many of the features of severe S. pyogenes skin infection in humans and involves the subcutaneous inoculation of S. pyogenes resulting in severe local infection that lead to systemic bacterial dissemination, multiorgan failure, and death. This model of streptococcal skin infection provides a very useful tool for testing drugs with potential effect on bacterial dissemination. In this case, drug therapy should start soon (1–6 h) after bacterial inoculation. Also drugs that can stop progressive bacterial growth can be tested in this infection models and the therapy should start after the systemic dissemination of S. pyogenes (approximately at 24 h after bacterial inoculation).

2. Materials 1. Phosphate-buffered saline (PBS): 8 g NaCl, 0.2 g KCl, 1.44 g Na2 HPO4 , and 0.24 g KH2 PO4 in 800 ml of distilled H2 O. Adjust the pH to 7.4 with HCl. Add H2 O to 1 l. Sterilize by autoclave.

Murine Model of Cutaneous Infection

397

2. 70% (v /v) ethanol. R System PT 4000 (KINE3. Homogenator: POLYTRON MATICA AG, Littau/Luzern, Switzerland).

4. EasyClean Dispersing aggregates (KINEMATICA AG, Littau/Luzern, Switzerland). 5. 1–3 ml syringes and several sizes of needles. 6. Heat lamp. 7. Equipment for restraint of the animal. 8. Centrifuge (RCF: 3345 × g; capacity: 4 × 400 ml). 9. THY broth: Todd–Hewitt broth (Beckton Dickisnson, TM MD, USA) supplemented with 1% Bacto yeast extract (Beckton Dickisnson, MD, USA). 10. Blood agar plates (Beckton Dickisnson, MD, USA). 11. Glycerol (Roth, Karlsruhe, Germany Cat. Nr. 4043.3). 12. Isoflurane (Baxter, Germany). 13. Gauze or cotton. 14. Closed glass container with raised floor. 15. Caliper. 16. Electric shaver for small animals. 17. Bacteria (see Note 1): different strains of S. pyogenes have been used in mouse models of streptococcal skin infection including the commercially available S. pyogenes strains DSM 2071, which can be obtained from the German Culture Collection or from The American Type Culture Collection (ATCC 2105). 18. Mice (see Note 2): inbred 7- to 8-week-old female C57BL/6NHsd mice (Harlan-Winkelmann, Borchen, Germany) or C57BL/6 J mice (The Jackson Laboratory, Bar Harbor, USA) have been used in this infection model (see Note 3). Mice should be housed in microisolator cages. Other mouse strains such as outbred CD1 mice, inbred BALB/c, C3H/HeN as well as hairless crl:SKH1(hrhr) Br mice have also been used in models of streptococcal skin infection (see Note 4). Although outbred mice have been used in this model of infection, the use of inbred mouse provide a genetically better-defined model with the advantage of high reproducibility not available with standard outbred mouse strains. In addition, the phenotypic uniformity of an inbred mouse strain means that sample size can be reduced in comparison with the use of outbred stocks.

398

Medina

3. Methods 3.1. Inoculum Preparation

1. Stock bacteria cultures are maintained in THY broth plus 20% glycerol at –70◦ C. 2. Plate a small amount of stock onto blood agar the day before infection and incubate overnight at 37◦ C. 3. Inoculate a fresh colony from a blood agar plate into THY medium. 4. Incubate at 37◦ C until midlog phase (OD600 = 0.5–0.6). 5. Harvest streptococci by centrifugation and wash twice with sterile PBS. 6. Adjust to a concentration of 5×108 colony-forming units (CFU)/ml in sterile PBS (see Note 5). For standardization of streptococci inoculum, a regression curve should be constructed for the selected strain by plotting the log number of serial bacterial dilutions against the percentage transmittance of the suspensions read at 600 nm wavelength in a specR trophotometer (Novospec II, Pharmacia). Transmittance of uninoculated THY broth from the same batch should be used as blank. 7. The inoculum concentration should be verified by plating 10-fold serial dilutions onto blood agar and counting after incubating for 24 h at 37◦ C (see Note 6).

3.2. Subcutaneous Infection Procedure

1. Anesthetize mice by inhalation of isoflurane (see Note 7): 1.1 Pour isoflurane on gauze or cotton placed in the bottom of a closed glass container. 1.2 Place a raised floor over the soaked material to allow isoflurane to vaporize without impregnating the mouse fur. 1.3 Place the mouse in the container and close the lid. 1.4 Keep the animal inside until reaching slow and regular breathing. 1.5 Remove the animal from the container to perform infection. 2. Shave the back of mice using an electric shaver for small animals. 3. Inoculate streptococci on the shaved back of the mouse (alternative sites for the initiation of subcutaneous infections are the flanks and the scruff of the neck) by subcutaneous injection: 3.1. Fill 1 ml syringe carrying a 22 G needle with 100 ␮l of a suspension containing 107 CFU of bacteria.

Murine Model of Cutaneous Infection

399

3.2. Place anesthetized animal on a clean paper towel. 3.3. Insert the needle under the skin of the intercapsular area and inject slowly with moderate pressure. 3.4. Inoculate control mice with sterile PBS. 3.3. Key Parameters to Monitor Infection

The main parameters commonly used for monitoring the course of an experimental skin and soft tissue infection are the death or survival of the animal, the development and progression of the local skin lesion, and the removal of organs for determination of bacterial loads, immune response analysis, or histological evaluation. Survival curves can be constructed by monitoring the time of death of individual mice after bacterial inoculation (see Note 8). 1. Lesion development: the development of a lesion at the site of infection can be monitored by measuring daily the area of dermonecrosis with a caliper and calculated by using the formula: (L × W ) A= 2 where L is length and W is width. 2. Quantification of bacterial loads: groups of mice are killed at selected intervals to count the numbers of bacteria at the local skin and systemic organs (see Note 9): 2.1 Euthanize the infected animal by CO2 asphyxiation. 2.2 Place animal on absorbent paper tissue over the dissection board and fix the upper and lower extremities with dissection pins. 2.3 Remove the infected lesion by using a sterile scalpel or scissors and place in a 10 ml tube containing 4.5 ml of sterile PBS. 2.4 Remove carefully the spleen, liver, kidneys, and lungs and place them in a 10 ml homogenization tubes containing 4.5 ml of sterile PBS. 2.5 Homogenize the skin and organs. 2.6 Perform 10-fold serial dilutions of tissue homogenate by mixing 100 ␮l of tissue homogenate and 900 ␮l of sterile PBS. 2.7 Apply 100 ␮l of each dilution to a blood agar plate and gently swirl the plate. 2.8 Incubate plates at 37◦ C for 24 h. 2.9 Determine the bacterial load within the organ using the following formula: C × Vt CFU/organ = Vp × DF

400

Medina

where C is the colony counts, Vt is the total volume of the organ homogenate, VP is the volume plated, and DF is the dilution factor of the sample. 3.4. Blood Collection

Sampling of blood from mice can be performed using several methods depending upon the purpose and volume of the blood sample.

3.4.1. Blood Collection from the Tail Vein

From the tail vein, small amounts of blood for monitoring systemic bacterial dissemination from the local site of infection can be collected at different times of infection without the need to kill the animal (see Note 10): 1. Restrain the animal. 2. Warm the tail with a heat lamp to dilate the vessels. 3. Extend the tail and make a small incision with a sterile scalpel blade in the lateral vein at the distal one-third of the tail. 4. Collect the blood flowing from the incision with a pipette. 5. Plate blood immediately onto blood agar after making 10-fold serial dilutions using sterile PBS. 6. Incubate plates at 37◦ C for 24 h. 7. Count colonies and express as CFU/ml of blood: CFU/ml =

C Vp × DF

where C is the colony counts, VP is the volume plated, and DF is the dilution factor of the sample. 3.4.2. Blood Collection by Cardiac Puncture

By cardiac puncture, when animal survival is not required, exsanguinations by cardiac puncture yields the maximal volume of blood: 1. Anesthetize the mouse and place on its back over clean paper towels. 2. Insert a 20 G needle attached to a 1 or 3 ml syringe just below and slightly to the left of the xiphoid cartilage at the base of the sternum, at a 20◦ angle. 3. Apply very slight negative pressure on the barrel of the syringe and aspirate gently until blood flow comes to an end (see Notes 11 and 12). 4. Euthanize the animal by CO2 asphyxiation.

3.5. Quantification of Cytokines in Serum Samples

1. Allow blood to clot. 2. Separate serum by centrifugation at 4◦ C and 450×g for 10 min.

Murine Model of Cutaneous Infection

401

3. Determine cytokines levels by Enzyme-Linked ImmunoSorbent Assay (ELISA) (see Note 13). 4. Serum samples can also be stored at –70◦ C until required for analysis.

4. Notes 1. S. pyogenes is considered to be a potential hazard to personnel, and Biosafety Level 2 practices and facilities should be used when working with these pathogens. All work with this microorganism should have prior approval of the institutional Biological Safety Department and should strictly follow the guidelines and regulations for the handling of this pathogen. 2. All protocols using live animals must first be reviewed and approved by the corresponding Governmental and Institutional Animal Care and Use Committee and must conform to the regulations regarding the care and use of laboratory animals. 3. Female mice are recommended since male mice are extremely susceptible to S. pyogenes infection (5). 4. Different inbred mouse strains have been shown to strongly differ in their level of susceptibility to S. pyogenes infection (5). Thus, CBA and C3H/HeN have been reported to be very susceptible, whereas BALB/c and DBA/2 are much more resistant (5). The reader is encouraged to consult the relevant literature for details relating to the levels of susceptibility of different mouse strains to S. pyogenes infection. 5. Different strains of S. pyogenes vary strongly in their levels of virulence. It is usually a good idea to determine a 50% lethal dose (LD50 ) for the selected strain of S. pyogenes. For this purpose, groups of 10 mice are inoculated with increasing numbers (e.g., 104 , 106 , 107 , 108 CFU) of bacteria. The LD50 is determined by using the Reed–Muench formula (6): ⎤ Dilution ⎥ ⎢ % mortality > 50%−50 coefficient of − ⎥ anti Lg10 ⎢ ⎦ ⎣ % mortality > 50% − % mortatity < 50% % of mortality > 50% ⎡

The selection of a lethal (leading to 100% mortality) or sublethal (leading to >50% survival) dose

402

Medina

should be performed according to the purpose of the experiment. 6. The inoculum should be used immediately after preparation since the viability of S. pyogenes in PBS decreases over time. 7. All procedures using isoflurane should be conducted in a fume hood that continuously exhausts anesthetic gases away from personnel. 8. In some countries, for ethical reasons, the mice are not allowed to die from infection and they must be euthanized or processed for sampling when they develop signs of severe infection (e.g., 20% of weight loss, lethargy, piloerection, stop taking food, and water). 9. The number of mice per group should be adequate to allow statistical analysis of the data. Because of the phenotypic uniformity, experimental infections using inbred mouse strains will require smaller number of animals to reach statistical significance than those employing outbred strains. 10. No more than 10% of the blood volume should be removed at any one time and maximum once a week. The maximum volume of a single sample should not exceed 0.25 ml since the average blood volume of an adult mouse is 2.5 ml. 11. Advancing or retracting the needle may be necessary to obtain a maxilla volume. 12. To prevent hemolysis, remove needle from the syringe before transferring blood to collection tube. 13. The simplest method for determination of cytokines by ELISA is to use commercially available kits with matching antibody pairs and standards for the specific cytokine. The reader can consult web sites such as http://www.biocompare.com/ to select from a wide range of commercially available kits with the corresponding manufactures instructions for use.

References 1. Vinh, D. C. and Embil, J. M. (2005) Rapidly progressive soft tissue infections. Lancet Infect Dis. 5, 501–513. 2. Moet, G. J., Jones, R. N., Biedenbach, D. J., Stilwell, M. G., and Fritsche, T. R. (2007) Contemporary causes of skin and soft tissue infections in North America, Latin America, and Europe: Report from the SENTRY

Antimicrobial Surveillance Program (1998– 2004). Diagn Microbiol Infect Dis. 57, 7–13. 3. Segreti, J. (2005) Efficacy of current agents used in the treatment of Grampositive infections and the consequences of resistance. Clin Microbiol Infect. 11 (suppl 3), 29–35.

Murine Model of Cutaneous Infection 4. Bisno, A. L. and Stevens, D. L. (1996) Streptococcal infections of skin and soft tissues. N Engl J Med. 334, 240–245. 5. Medina, E., Goldmann, O., Rohde, M., Lengeling, A., and Chhatwal, G. S. (2001)

403

Genetic control of susceptibility to group A streptococcal infection in mice. J Infect Dis. 184, 846–852. 6. Reed, L. J. and Muench,H. A. (1938) A simple method of stimating fifty percent end points. Am J Hyg. 27, 493–497.

Chapter 22 Murine Model of Pneumococcal Pneumonia Eva Medina Abstract Respiratory tract infections remain among the most common clinical problems worldwide. Pneumonia or inflammation of the lungs can be caused by infection with bacteria, viruses, and other organisms. Pneumonia management has been challenged by the widespread distribution of antibiotic-resistant strains of Streptococcus pneumoniae, the commonest cause of community acquired pneumonia. Experimental models of pneumonia have played a crucial role for testing the efficacy of antimicrobial agents as well as for gaining a better understanding of the disease pathogenesis. These models have also received increased attention as tools for deriving pharmacodynamic data and for determining the clinical significance of drug resistance. Key words: Respiratory infection, Streptococcus pneumoniae , pneumonia, intranasal inoculation.

1. Introduction Streptococcus pneumoniae is a major human pathogen that colonizes the upper respiratory tract and causes both lifethreatening diseases such as pneumonia, sepsis, and meningitis and milder but common diseases, such as sinusitis and otitis media (1). S. pneumoniae is the leading cause of bacterial pneumonia. The burden of pneumococcal infections is particularly large among children and the elderly and is exacerbated by the rising numbers of isolates resistant to antibiotics (2). Mouse models of pneumonia have been used to characterize the course of infection by determining animal survival after infection, the bacterial loads in lungs and other organs, levels of inflammation, and histology of lung tissue (3). In addition, studies in the mouse model of streptococcal pneumoniae have provided critical information on antimicrobial efficacy that is directly relevant to the treatment G. Proetzel, M.V. Wiles (eds.), Mouse Models for Drug Discovery, Methods in Molecular Biology 602, DOI 10.1007/978-1-60761-058-8 22, © Humana Press, a part of Springer Science+Business Media, LLC 2010

405

406

Medina

of human infection (4, 5). Murine models of S. pneumoniae have been also essential for testing strategies of vaccination against this pathogen (6). In the mouse model described here, pneumonia is induced after the intranasal instillation of S. pneumoniae. When using this model for evaluation of antimicrobial agents, therapeutic treatment should start after the infection has already been established in the lungs (6–12 h after bacterial inoculation).

2. Materials 1. Phosphate-buffered saline (PBS): 8 g NaCl, 0.2 g KCl, 1.44 g Na2 HPO4 , and 0.24 g KH2 PO4 in 800 ml of distilled H2 O. Adjust the pH to 7.4 with HCl. Add H2 O to 1 l. Sterilize by autoclave. 2. 70% (v /v) ethanol. R 3. Homogenator: POLYTRON System PT 4000 (KINEMATICA AG, Littau/Luzern, Switzerland).

4. EasyClean Dispersing aggregates (KINEMATICA AG, Littau/Luzern, Switzerland). 5. 1 ml syringes and 25 G needles. 6. THY broth: Todd–Hewitt broth (Beckton Dickinson, MD, USA) supplemented with 1% BactoTM yeast extract (Beckton Dickinson, MD, USA). 7. Blood agar plates (Beckton Dickinson, MD, USA; Cat. Nr. 221165). 8. Glycerol (Roth, Karlsruhe, Germany). 9. Isoflurane (Baxter, Germany). 10. Gauze or cotton. 11. Closed glass container with raised floor. 12. 10% neutral buffered formalin: 10% Formaldehyde in PBS (see Note 1). R , Ravensburg, Germany). 13. Ketamine (Narketan

14. Xylene (J.T. Baker Chemical Co, Phillipsburg, USA). 15. Bacteria (see Note 2): the majority of the published studies have used the serotype 2 S. pneumoniae (NCTC 7466) that can be obtained from the National Collection of Type Cultures, London, United Kingdom. Other strains suitable for the mouse model are the serotype 3 S. pneumoniae (ATCC 6303) and serotype 1 S. pneumoniae (ATCC 6301), both can be obtained from the American Type Culture Collection.

Murine Model of Pneumococcal Pneumonia

407

16. Mice (see Note 3): several inbred strains of mice have been used in this infection model including BALB/c and CBA/Ca among others (see Note 4).

3. Methods 3.1. Inoculum Preparation

1. Stock bacteria cultures are maintained in THY broth plus 20% glycerol at –70◦ C. 2. Plate a small amount of stock onto blood agar the day before infection and incubate overnight at 37◦ C in 5% CO2 . 3. Inoculate a fresh colony from a blood agar plate into THY medium. 4. Incubate at 37◦ C until mid-log phase (OD600 = 0.3–0.4). 5. Harvest pneumococci by centrifugation and wash twice with sterile PBS. 6. Adjust the inoculum to the desired concentration in sterile PBS (see Note 5). For standardization of S. pneumoniae inoculum, a regression curve should be constructed for the selected strain by plotting the log number of serial bacterial dilutions against the percentage optical density of the suspensions read at 600 nm wavelength in a spectrophotomeR ter (Novospec II, Pharmacia). Optical density of uninoculated THY from the same batch is used as blank. 7. The inoculum concentration should be verified by plating 10-fold serial dilutions onto blood agar and counting after incubating for 24 h at 37◦ C (see Note 6).

3.2. Intranasal Infection Procedure

1. Anesthetize mice by inhalation of isoflurane in order to facilitate aspiration (see Note 7): 1.1 Pour isoflurane on gauze or cotton placed in the bottom of a closed glass container. 1.2 Place a raised floor over the soaked material to allow isoflurane to vaporize without impregnating the mouse fur. 1.3 Place the mouse in the container and close the lid. 1.4 Keep the animal inside until reaching slow and regular breathing. 1.5 Remove the animal from the container to perform infection.

408

Medina

2. Holding the mice vertically, deliver a 40 ␮l volume of bacterial inoculum to the nostrils to induce aspiration pneumonia. Inoculate control mice with sterile PBS. 3.3. Bacterial Counts in the Lungs as Parameter to Monitor Infection

1. Euthanize mice at selected time intervals following infection by cervical dislocation. Remove the lungs aseptically and place them in a 10 ml homogenization tube containing 4.5 ml of sterile saline. 2. Homogenize the lungs and perform 10-fold serial dilutions of lung homogenate in sterile PBS. 3. Apply 100 ␮l of each dilution to a blood agar plate and gently swirl the plate. 4. Incubate plates at 37◦ C for 24 h. 5. Determine the bacterial load within the lungs using the following formula:

CFU/lungs =

C × Vt V p × DF

where C is the colony counts, Vt is the total volume of the lung homogenate, VP is the volume plated, and DF is the dilution factor of the sample. 3.4. Histological Examination of Lung Tissue to Monitor Infection

1. At chosen intervals after bacterial inoculation, mice are euthanized by an overdose of ketamine (see Note 8). 2. Remove lungs and fix them with neutral buffered formalin for 48 h at 4◦ C. 3. Wash lungs with dH2 O and dehydrate the tissue by serial immersion in ethanol (50, 70, 95, and 100%) 1 h each, finishing with incubation for 1 h in 100% xylene. 4. Embed lung tissue in paraffin and cut 5 ␮m thickness sections. 5. Dewax tissue section starting with 100% xylene for 1 h and continuing with serial immersion in ethanol (100, 95, 70, and 50%) 1 h each finishing with dH2 O. 6. Stain sections and mount (see Note 9). 7. Examine by light microscopy.

3.5. Leukocyte Recruitment into the Lungs as Parameter to Monitor Infection

The recruitment of leukocytes into the lungs following infection with S. pneumoniae is analyzed by broncho-alveolar lavage (BAL) fluid counts: 1. Euthanize mice by an overdose of ketamine (see Note 8). 2. Disinfect the neck region with 70% ethanol.

Murine Model of Pneumococcal Pneumonia

409

3. Make a middle incision in the neck area and retract the skin with forceps. 4. Cut away the muscles overlying the trachea carefully to not affect the aorta. 5. Make a small incision in the upper part of the trachea and insert a polyethylene catheter attached to a 25 G needle on a 1 ml tuberculin syringe (see Note 10). 6. Slowly inject 0.5 ml of sterile PBS through the catheter into the mouse lungs recovering the resultant BAL. 7. Determine total leukocyte counts under optical microscope using a Neubauer chamber. 8. Determine differential cell counts on cytospin smears of lung washes (3 min at 100×g) stained by the Wright–Giemsa method. 3.6. Quantification of Inflammatory Cytokines in Broncho-alveolar Lavage (BAL): As Parameter to Monitor Infection

1. Centrifuge lavage fluids for 10 min at 800×g. 2. Remove supernatant with a pipette and use for quantification of cytokines by ELISA (see Note 11).

4. Notes 1. Work with 10% buffered neutral formalin is recommended to be executed in a well-ventilated area, wearing goggles, gloves, and lab coat. Storage areas should have appropriate ventilation systems. 2. S. pneumoniae is considered to be a potential hazard to personnel, and Biosafety Level 2 practices and facilities should be used when working with this pathogen. All work with this microorganism should have prior approval of the institutional Biological Safety Department and should strictly follow the guidelines and regulations for the handling of this pathogen. 3. All protocols using live animals must first be reviewed and approved by the corresponding Governmental and Institutional Animal Care and Use Committee and must conform to the regulations regarding the care and use of laboratory animals. 4. Different mouse strains strongly differ in their degree of susceptibility to S. pneumoniae (7). While CBA/Ca and SJL are highly susceptible, BALB/c is highly resistant and C3H/He, FVB/n, NIH, AKR, C57BL/6, and DBA/2 exhibit intermediate resistance.

410

Medina

5. Concentrations ranging between 5×105 and 107 CFU can be used. 6. Due to propensity of S. pneumoniae to autolysis, the inoculum should never be vortexed or subjected to drastic shaking. 7. All procedures using isoflurane should be conducted in a fume hood that continuously exhausts anesthetic gases away from personnel. 8. CO2 asphyxiation is not recommended since it can cause perivascular edema in the lungs or alveolar hemorrhage. 9. The staining technique will depend on the type of cell of interest. Hematoxylin and eosin (H&E) stain is the most widely used method in histology. Using this technique, basophilic white blood cells stain dark blue, eosinophilic white blood cells stain bright red, and neutrophils stain a neutral pink. 10. The catheter should not be insert too far into the trachea since this will result in the dispensing of PBS into only one lobe of the lungs. 11. The simplest method for determination of cytokines by ELISA is to use commercially available kits with matching antibody pairs and standards for the specific cytokine. The reader can consult web sites such as http://www. biocompare.com/ to select form a wide range of commercially available kits with the corresponding manufactures instructions for use. References 1. Austrian, R. (1999) The pneumococcus at the millennium: not down, not out. J Infect Dis. 179 Suppl 2, S338–S341. 2. Ortqvist, A., Hedlund, J., and Kalin, M. (2005) Streptococcus pneumoniae: epidemiology, risk factors, and clinical features. Semin Respir Crit Care Med. 26, 563–574. 3. Chiavolini, D., Pozzi, G., and Ricci, S. 2008. Animal models of Streptococcus pneumoniaedisease. Clin Microbiol Rev. 21, 666–685. 4. Craig, W. A. (1998) Pharmacokinetic/ pharmacodynamic parameters: rationale for antibacterial dosing of mice and men. Clin Infect Dis. 26, 1–10. 5. Moine, P., Vall´ee, E., Azoulay-Dupuis, E., Bourget, P., B´edos, J. P., Bauchet, J.,

and Pocidalo, J. J. (1994) In vivo efficacy of a broad-spectrum cephalosporin, ceftriaxone, against penicillin-susceptible and -resistant strains of Streptococcus pneumoniae in a mouse pneumonia model. Antimicrob Agents Chemother. 38, 1953– 1958. 6. Steinhoff, M.C. (2007) Pneumococcal vaccine animal model consensus group. Animal models for protein pneumococcal vaccine evaluation: a summary, Vaccine 25, 2465– 2470. 7. Kadioglu, A. and Andrew, P. W. (2005) Susceptibility and resistance to pneumococcal disease in mice. Brief Funct Genomic Proteomic. 4, 241–247.

Chapter 23 Murine Model of Polymicrobial Septic Peritonitis Using Cecal Ligation and Puncture (CLP) Eva Medina Abstract Although a number of animal models such as endotoxic shock and bacteremia have been used to study the pathogenesis of sepsis, cecal ligation and puncture (CLP) represents a peritonitis model with clinical features of polymicrobial infection comparable with those of peritonitis in humans. The CLP consists in the surgical perforation of the legated cecum of mice that results in immediate and constant drainage of cecal bacteria into the peritoneal cavity. The severity of the diseases depends on the diameter of the needle used for the perforation as well as on the number of cecal punctures. The CLP model of sepsis in mice is the most commonly used for studying the process of septic peritonitis and can be used as a preclinical model to test the efficacy of pharmacological agents for the treatment of sepsis. Key words: Septic peritonitis, cecal ligation and puncture, polymicrobial sepsis.

1. Introduction Sepsis is a complex clinical syndrome characterized by a severe infection in the body and bloodstream and patients with septic peritonitis have a particular high mortality rate of 60–80% (1, 2). Bacterial invasion of the peritoneal cavity due to intestinal leakage after major abdominal surgery is the most frequent cause of septic peritonitis. Due to the large number of microorganisms in the bowel, this infection is by nature polymicrobial. Septic peritonitis is characterized by massive infiltration of neutrophils and macrophages into the peritoneum where these cells are the first line of defense for clearing invading microorganisms. However, once they fail to restrict microbes to the peritoneal cavity, microbes may reach the blood stream, resulting in an G. Proetzel, M.V. Wiles (eds.), Mouse Models for Drug Discovery, Methods in Molecular Biology 602, DOI 10.1007/978-1-60761-058-8 23, © Humana Press, a part of Springer Science+Business Media, LLC 2010

411

412

Medina

overwhelming systemic immune response via the production of proinflammatory mediators such as cytokines frequently leading to multi-organ failure, septic shock, and death (2). Although several mouse models of sepsis have been developed such as endotoxic shock or bacteremia (3, 4), the cecal and ligation puncture (CLP) model mirrors more closely the clinical course of human abdominal sepsis (5). The CLP surgery model takes into account the impact of pathogenic infection typically observed in sepsis since it involves trauma (perforation) to the bowel, thereby permitting the introduction of multiple bacterial strains into the peritoneal cavity. Furthermore, the magnitude of the septic challenge can be controlled by changing the size of the needle used to puncture the ligated cecum. The mouse peritonitis model has been extensively used for the evaluation of numerous antimicrobial compounds and it has been pivotal for investigating fundamental issues in the relationship between infecting pathogen and antibiotic therapy.

2. Materials 1. Columbia blood agar (Beckton Dickinson, MD, USA). 2. Surgical staples. 3. 3-0 silk suture. 4. 4-0 braided absorbable suture. R , Ravensburg, Germany). 5. Ketamine (Narketan R , Leverkusen, Germany). 6. Xylazine (Rompun

7. Mice (see Notes 1 and 2): any variety of mice can be used. Pathogen-free CD1 and NMRI mice from either sex have also been commonly used. As mentioned above, we recommend the use of inbred mouse strains since they are genetically well defined and uniform providing the advantage of high reproducibility not available with standard outbred mouse strains. In addition, due to the phenotypic uniformity of an inbred mouse strain, the sample size can be smaller than when using outbred stocks. 8. Wright–Giemsa stain. 9. Cytospin apparatus (Shandon, Runcorn, UK). 10. Neubauer chamber.

3. Methods 3.1. Surgical Procedure

1. Anesthetize the animal by intraperitoneal injection of a mixture of ketamine (80 mg/kg/body weight) and xylazine (16 mg/kg/body weight) in 0.2 ml of sterile PBS.

Murine Model of Polymicrobial Septic Peritonitis

413

2. Place anesthetized animal on a clean paper towel, shave the anterior abdominal wall, and disinfect with 70% ethanol. 3. Make an approximately 1 cm longitudinal incision in the left-lower quadrant of the abdomen using a scalpel to open the peritoneal cavity (see Note 3). 4. Bring out the cecum (see Note 4). 5. Ligate one third of the distal cecum with a 3-0 silk immediately below the ileocecal valve. 6. The ligated cecum is then punctured once or twice using a hypodermic needle (see Note 5). 7. Press gently on the tied segment to ensure that a small amount of feces is extruded on to the surface of the bowel. 8. Return the cecum to the peritoneal cavity. 9. Close the wound using 4-0 braided absorbable suture for the muscle layer and the skin with surgical staples. 10. In sham controls, the cecum is exposed but not ligated or punctured, then returned to the abdominal cavity. 11. Administer 1 ml of sterile PBS intraperitoneally as a fluid resuscitation measure, immediately following surgery and place the animals on a heating pad until they recover from the anesthetic. 3.2. Quantification of Bacterial Loads in the Peritoneal Cavity

1. Euthanize the mice by CO2 asphyxiation (see Note 6). 2. Cut open the skin of the abdomen in the midline after thorough disinfection and without injury to the muscle. 3. Using a sterile 5 ml syringe and 20 G needle, inject 5 ml of sterile PBS into, and aspirate out of, the peritoneal cavity twice. 4. Make 10-fold serial dilutions of peritoneal lavage samples and plate on Columbia blood agar. 5. Incubate plates for 18 h at 37◦ C. 6. Count colonies and express as CFU/ml of lavage fluid.

3.3. Quantification of Bacterial Loads in the Blood

1. Euthanize the mice by CO2 asphyxiation. 2. Collect blood by cardiac puncture. 3. Plate blood immediately onto Columbia blood agar after making 10-fold serial dilutions using sterile PBS. 4. Count colonies and express as CFU/ml of blood: CFU/ml =

C Vp × DF

where C is the colony counts, VP is the volume plated, and DF is the dilution factor of the sample.

414

Medina

3.4. Quantification of Bacterial Loads in Systemic Organs

1. Euthanize the mice by CO2 asphyxiation. 2. Remove carefully the spleen, liver, kidneys, and lungs and place them in a 10-ml homogenization tube containing 4.5 ml of sterile saline. 3. Homogenize organs and perform 10-fold serial dilutions of tissue homogenate by mixing 100 ␮l of tissue homogenate and 900 ␮l of sterile PBS. 4. Apply 100 ␮l of each dilution to a Columbia blood agar plate and gently swirl the plate. 5. Incubate plates at 37◦ C for 24 h. 6. Determine the bacterial load within the organ using the following formula:

CFU/organ =

C × Vt Vp × DF

where C is the colony counts, Vt is the total volume of the organ homogenate, VP is the volume plated, and DF is the dilution factor of the sample. 3.5. Quantification of Cytokines and Chemokines in the Peritoneal Cavity

1. Perform peritoneal lavage as described for the quantification of bacterial loads in the peritoneal cavity. 2. Clarify peritoneal fluids by centrifugation for 10 min at 1000 rpm. 3. Remove the supernatant and determine the level of cytokines and chemokines by ELISA (see Note 7).

3.6. Determination of Peritoneal Cell Counts

1. Perform peritoneal lavage as described for the quantification of bacterial loads in the peritoneal cavity. 2. Determine total leukocyte counts under optical microscope using a Neubauer chamber. 3. Determine differential cell counts on cytospin smears of peritoneal fluids (100×g for 3 min) stained by the Wright– Giemsa method.

4. Notes 1. The animals should be fasted overnight prior to surgery. 2. All protocols using live animals must first be reviewed and approved by the corresponding Governmental and Institutional Animal Care and Use Committee and must conform

Murine Model of Polymicrobial Septic Peritonitis

415

to the regulations regarding the care and use of laboratory animals. 3. There are no major blood vessels in the region, but care should be taken to avoid any muscular damage, a source of subcutaneous hematoma. 4. The mouse cecum is located on the left lateral side of the abdomen, curved on itself, and filling the space between the stomach and the liver lobes. 5. The gauge of the needle can vary from 18 to 26 G, depending on the desired mortality to be induced. 6. The number of mice per group should be adequate to allow statistical analysis of the data. Because of the phenotypic uniformity, experimental infections using inbred mouse strains will require smaller number of animals to reach statistical significance than those employing outbread strains. 7. The simplest method for determination of cytokines by ELISA is to use commercially available kits with matching antibody pairs and standards for the specific cytokine. The reader can consult web sites such as http://www. biocompare.com/ to select from a wide range of commercially available kits with the corresponding manufactures instructions for use. References 1. Cohen, J. (2002) The immunopathogenesis of sepsis. Nature 420, 885–891. 2. Hotchkiss, R. S. and Karl, I. E. (2003) The pathophysiology and treatment of sepsis. N Engl J Med 348, 138–150. 3. Fink, M. P. and Heard S. O. (1990) Laboratory models of sepsis and septic shock. J Surg Res 49, 186–196.

4. Deitch, E. A. (1998) Animal models of sepsis and shock: a review and lessons learned. Skock 9, 1–11. 5. Wang, P. and Chaudry, I. H. (1998) A single model of polymicrobial sepsis: cecal ligation and puncture. Sepsis 2, 227– 233.

INDEX

A

serum, 94–95, 97, 99–100, 140, 142, 147, 154, 189, 257, 400 tail vein, 103, 142–144, 146, 217, 220–221, 259–260, 400 Body composition, 46–48, 140, 149–151 Bone marrow, 16–17, 50, 108–111, 128, 217, 253, 255–257, 259–260, 262, 264–265 Bowel, 411–413 Brain, 40–41, 48, 72, 81, 87–89, 275, 283–298, 299–319, 324–325, 327, 333 Brain imaging, 87–89 Burying test, 303, 307–308, 315

Acyl coenzyme A:cholesterol acyltransferase (ACAT), 158–159 Adipocytes, 136, 150 Albumin pharmacokinetics, 102 Alopecia areata, 194, 196, 208 Amyotropic Lateral Sclerosis (ALS), 348–352 Alzheimer’s disease, 40, 51, 324, 326 Amyloid plaques, 40, 325–326 Anesthesia, 81, 83, 87, 143, 149, 201–202, 245–247, 250–251 Angiogenesis, 218, 228–230 Anhedonia, 267, 272, 274 ANOVA, 14–15, 249, 272, 305–306, 335, 337 Antibody, 46, 50, 68, 93–94, 96–97, 102, 128, 184, 191, 197, 204, 261, 364, 402, 415 See also Therapeutic antibody, monoclonal antibodies Antidepressant, 267–279, 301, 305, 311 Anxiety, 268, 271–272, 276, 280, 299–319, 332, 336–337 Anxiogenic drugs, 306–310, 312, 316–317 Anxiolytic drugs, 306–307, 311, 318 Apolipoprotein E, 41–42 Apoptosis, 203–205, 230, 254 Arthritis, 41–42, 46, 68, 181–192 Atherosclerosis, 41–42, 158, 172 Atopic dermatitis, 195 Autoantigen, 125–127, 129 Autochthonous tumors, 226 Autoimmune, 9, 106, 119–130, 182, 194, 208, 349, 356 Autoimmunity, see Autoimmune

C C57BL/6, 5, 26, 98, 111, 143, 170, 182–183, 187, 191, 260, 315, 397, 409 Cancer BCR-ABL, 253, 264 chemotherapeutic, 228, 237 chronic myeloid leukemia, 253–265 colorectal cancer, 50, 235–251 5-fluorouracil, 235–251 irinotecan, 235–251 leucovorin, 235–251 leukemia, 253, 264 lymph node metastasis, 236 melanoma, 216 metastasis, 215–231, 236 metastatic colorectal cancer, 235–251 oncogene, 56, 225, 253 Philadelphia chromosome, 253 staging, 236 Trp53 knockout mouse, 9 tumor, 215, 216, 225–226, 228, 235–236 growth measurement, 247 model, 215, 225–228 xenograft, 226, 235–236 Candidate gene, 50 Carboxyl ester lipase (CEL), 158 Cardiac hypertrophy, 40 Cardiac puncture, 143, 400, 413 Cardiovascular disease (CVD), 157–159, 161, 172, 175 CD-1, 4, 12–17, 19 Cecal ligation and puncture, 411–415 CEL, see Carboxyl ester lipase (CEL) Cellulitis, 396 Charcot-Marie-Tooth diseases, 349, 353 Chimera, 50–51, 128 Chloramphenicol, 12–13, 15–16, 19 Cholesterol, 46–47, 141, 157–176 Chronic myeloid leukemia (CML), 253–265

B BALB/c, 12, 14, 16, 19, 107–108, 250, 255, 268, 397, 401, 407, 409 Barbering, 278 BCR-ABL, 253–255, 257–263, 265 Behavior, 46–47, 49, 152, 231, 271, 274–275, 277–279, 300, 305, 307, 316, 333 Behavioral response, 312 Behavioral test, 378 Beta-amyloid peptide, 324, 326 Beta-galactosidase, 69–70 Beta-2-microglobulin, 106 Bile acids, 164, 171, 173–175 Bioluminescence, 80, 86–87, 90, 228, 235–251 Biomarker, 10–11, 49 Blister, 199 Blood collection, 97, 103, 140–144, 147–150, 400 cardiac puncture, 143, 400, 413 pressure, 47, 49, 141, 333

G. Proetzel, M.V. Wiles (eds.), Mouse Models for Drug Discovery, Methods in Molecular Biology 602, c Humana Press, a part of Springer Science+Business Media, LLC 2010 DOI 10.1007/978-1-60761-058-8, 

417

MOUSE MODELS FOR DRUG DISCOVERY

418 Index

Chronic proliferative dermatitis, 204–205, 208 Chronic stress, 277 CIA, see Collagen-induced arthritis (CIA) Clinical score, 188, 191 CML, see Chronic myeloid leukemia (CML) CNV, see Copy Number Variation (CNV) Collagen, 41–42, 46–47, 181–192, 359–360 Collagen-induced arthritis (CIA), 41–42, 181–191 Colorectal cancer, 50, 235–251 Conditional gene targeting, 61–63, 68, 72 Conditional knockout, 4 Congenital Myasthenic Syndromes, 349 Copy Number Variation (CNV), 25 Cre-ER, 43, 63 Cre-ERT2, 63 Cre/loxp, 40, 42, 61–63, 65, 67, 72, 362, 363 Cre recombinase, 40, 62 Cryopreservation embryos, 26–29 sperm, 26–29, 31–34 Cryoprotective, 29, 31 CVD, see Cardiovascular disease (CVD) Cytokine, 46, 50, 107, 111, 120, 182, 190, 208, 259, 400–402, 409–410, 412, 415

D Data analysis, 91, 169, 174, 269, 272, 305–306, 334–335, 380 DBA, 41–42, 45, 137, 182, 184–185, 187–188, 190–191, 279, 284, 401, 409 Deficiency, 109, 113, 278, 335, 352, 356–357, 361 Dementia, 324–327, 330, 338 Demyelination, 349, 352, 355, 369 Depression, 267–280, 301, 315, 325 Dermatitis, 195, 204–205, 208 Dermis, 199, 363, 395–396 Despair, 271, 274, 276 DEXA, 46, 149 Diabetes, 40, 46, 48, 115, 119–130, 135–154 See also Idd; Type 1 diabetes (T1D); Type II diabetes mellitus (T2DM) Diet chow, 165–166 high fat, 46–47, 49, 141, 148, 151, 153 powder, 166 repelleted, 166 semi-purified, 163, 165, 167, 170, 174, 176 western diet, 158 Diet-induced obesity (DIO), 137–138 Dose-response, 12, 14–15, 84–86, 200 Drug addiction, 284 administration, 198, 271–272, 304–305, 311 development, 1–19, 65, 159, 175, 181–182 discovery, 37–52, 55–72, 135–154, 162, 215–231, 268, 299–318, 338 Dual-label, 168, 170 Duchenne’s disease, 349, 359

E Electrophysiology, 371, 381 Elevated-Plus Maze (EPM), 302, 306, 314 ELISA, 97–101, 140, 144–145, 184, 190–192, 286, 401–402, 409, 415 Embryo, 6, 26–28, 30, 34, 59, 123

Embryonic stem (ES) cells, 51, 59, 107, 225 Epidermis, 199–200, 203–204, 395–396 Epilepsy, 48, 352

F Fluorescent-activated cell sorting (FACS), 46, 97, 99, 257–258, 260–265 Fasting, 136, 146, 148–149, 154, 166, 170 Fat, 46–47, 49, 137–138, 149, 151, 171, 200, 374, 396 Fat pad, 224–225 Fatty acids, 46, 141, 145–146, 171 Fc fusion proteins, 96 FcRn, see Neonatal Fc receptor (FcRn) FDA, 2, 19, 196 Fear, 306, 331, 333 Fecal sterols, 164 Fibrosis, 40, 375 Filaggrin, 195 Fixative, 197, 207–208, 227, 362, 364, 366, 373, 375, 385 Flow cytometry, 109–110, 227, 257–258, 261–264 Flp/Frt, 61–63, 65–66, 126, 353, 355 Flp recombinase, 62 See also Flp/Frt Fluorescence, 372, 376 Fluorescent-activated cell sorting (FACS), 46, 97, 99, 257–258, 260, 262, 264–265 Food intake, 46–47, 136, 146, 152–153 Footpad, 222–223 Forced Swim Test (FST), 269–271, 274–275, 278 5-FU (5-fluorouracil), 235–251, 257 Full thickness skin grafts, 196, 208

G Gait analysis, 378–381 Gene replacement, 43, 50 See also Gene targeting; Knockout; Knock-in (KI) Gene targeting, 59–71, 72, 98, 157 GEMMs/GEMs, 37–52, 55–73, 216, 225–226, 230 Genetically engineered mouse models, see GEMMs/GEMs Genetically modified mice, 38 Genetic background, 4, 25, 47, 51, 98, 162, 216, 226, 264, 314, 328, 337, 350 Genetic contamination, 25, 328 Genetic disorder, 72, 106, 255, 299–300, 327, 385 Genetic drift, 5, 6, 25–26, 328 Gene trap, 69–70 Genome, 3, 4, 24, 26–28, 38–39, 41, 51, 55–73, 107, 268, 276–277, 279, 310, 317, 331, 385 Genotyping, 39, 47, 70, 99 GFP, see Green fluorescent protein (GFP) Glucose, 30, 40, 46, 136, 139–140, 144, 146–147, 154 Glucose-stimulated insulin secretion (GSIS), 139–140, 147–148 Glucose tolerance test (GTT), 46, 136, 146–147, 153–154 Graft, 105–115 Graft-versus-host disease (GVHD), 105–115 Green fluorescent protein (GFP), 68, 257–258, 262–263, 284, 362–363, 366 Grip strength, 378–379 Grooming Analysis, 303, 308–309 GVHD, see Graft-versus-host disease (GVHD)

MOUSE MODELS FOR DRUG DISCOVERY 419 Index H Hair, 195, 197, 202, 206, 208, 224, 279, 285 Hair cycle, 195, 200–201, 203, 205, 207 Hawthorne Effect, 334 HDL, see High-density lipoprotein (HDL) Hearing loss, 329 Heart failure, 40 Heart rate, 47–48 Hematopoietic stem cell (HSC), 106, 108, 128–129 Hepatocytes, 50–51, 112, 136, 150 Hereditary Motor and/or Sensory Neuropathies (HSMNS), 349 High-density lipoprotein (HDL), 158, 161–162, 167, 175 Hippocampus, 88, 325, 330–333 Histology, 112, 140–141, 150–151, 154, 226, 230, 362, 367, 374, 410 Histopathology, 46, 56, 197, 202–207, 218, 228, 260 HLA transgenics, 119–130 Hole Board Test, 304, 312, 318 Homologous recombination, 38, 41, 51, 60–61, 65, 68–71, 120–121, 157 See also Gene targeting Housing, 103, 143, 153, 269, 277, 300–301, 312, 314, 334, 336 See also Husbandry HPRT1, see Hypoxanthine phosphoribosyltransferase 1 (HPRT1) HSC, see Hematopoietic stem cell (HSC) HSMNS, see Hereditary Motor and/or Sensory Neuropathies Humanized mice, 51, 68–69, 108 See also Humanized knock-in mice Husbandry, 103, 269, 314–315, 336–337 Hyperglycemia, 137–138 Hyperinsulinemia, 138, 151 Hyperleptinemia, 137 Hypoxanthine phosphoribosyltransferase 1 (HPRT1), 69

I Ichthyosis, 194–195 Idd, see Insulin-dependent diabetes (Idd) IL2 receptor gamma chain, 107 Imaging, 3, 68, 79–92, 149, 228, 235–251, 325 Immunization, 114 Immunocompromised, 50 Immunodeficient, 51, 102, 106–107, 113–114, 196, 235–236, 254 Implant, 27–28 Implantation, 227–228, 236, 245–248 Inbred mouse, 9–10, 24, 39, 41, 176, 255, 264, 278, 317, 329, 397, 402, 415 strain, 4–6, 8, 12–13, 19, 25, 296 Inducible, 40, 42–44, 59, 63–65, 70, 72, 363 Infection bacteria, 395 bacterial load, 399, 405, 408, 413–414 cutaneous, 395–402 lentivirus, 58–59, 71–72, 284, 286 See also Lentiviral local, 396 lung, 399, 406, 408 peritonitis, 411–415 pneumonia, 405–410 polymicrobial, 411–415

respiratory, 405 respiratory tract, 405 retrovirus, 58, 69, 240–241, 255 sepsis, 405, 411–412 skin, 396–397 soft-tissue, 395–396, 399 Staphylococcus aureus, 396 Streptococcus pyogenes, 395–402 viral, 70, 193, 284 Infectious disease, 193 Inflammation, 41, 46, 54, 141 Inflammatory disease, 181 Inhibitor, 47–49, 159–161, 261–263 Injection, 28, 33, 39, 56, 68, 70, 80, 83–84, 87, 147, 162, 199, 219–221, 227, 246, 251, 271, 295, 297, 333, 398, 412 Inoculation, 396, 399, 406, 408 Insulin, 29, 40, 120, 125, 127–128, 136, 141, 144–145, 147–148, 152, 154 Insulin-dependent diabetes (Idd), 120, 122 Insulin secretion, 136–137, 139–140, 147–148 Insulin sensitivity, 136–137, 148–149 Insulin tolerance test (ITT), 148, 154 IRES, see Internal ribosome entry site (IRES) Internal ribosome entry site (IRES), 66–68 Intestine, 48, 150, 159–160, 167, 172, 174–175 Intranasal inoculation, 406 In vitro fertilization (IVF), 26–27, 32–35 In vivo imaging, 68, 80–81, 83–86, 91, 238 Irinotecan, 235–251 Isogenic strain, 7–8, 18 IVF, see In vitro fertilization (IVF)

K Keratinocytes, 203–205 Knock-in (KI), 38, 43–44, 51, 65–68, 225, 351, 354, 357, 359, 361 Knock-down, 43, 52, 71–72 Knockout, 3, 9, 60–61, 63–65, 105–115, 122, 158–160, 182, 284, 329, 353–354

L LDL, 157–159, 161–162, 172, 175, 356 Learning, 275–276, 314, 329–330, 332–333, 335–337 Lentiviral, 57–59, 70–72, 284 Leptin, 46, 48, 137 Leucovorin, 235–251 Leukemia, 106, 114, 254, 260–261, 263–264 Light-Dark Box Test, 303, 311, 318 Lipopolysaccharides (LPS), 46 Lipoprotein, 41, 159, 162, 175–176 Liver, 50, 87, 89, 112, 145, 154, 158, 162, 167, 175, 229, 399, 415 Locomotor, 305, 332, 335 LPS, see Lipopolysaccharides (LPS) Luciferase, 79–82, 86–87, 90, 237, 239–242, 251 Lung, 218, 221–222, 262, 405, 408–409 Lymph node, 222, 224, 236, 249, 250, 377 Lymphoma, 111–112

M Mab, see Monoclonal antibodies (Mab) Magnetic Resonance Imaging, 149 Melanoma, 216, 221–222

MOUSE MODELS FOR DRUG DISCOVERY

420 Index

Memory, 275–276, 315, 317, 324, 330–331, 335, 337 Metastasis, 215–231, 236, 248, 250 Metastatic colorectal cancer, 235–251 MHC class I, 94, 108–114, 120–121 MHC class II, 108–109, 111–121 Microscopy, 197, 206–207, 364, 366, 370–374, 408 Monoclonal antibodies (Mab), 96–97, 101, 103 Motor neuron, 348–349, 351, 353, 362–363, 366–368 Motor neuron diseases, 348–349, 364–368, 378 Muscle, 46, 51, 199, 203, 259, 348–349, 351–352, 360, 363, 376, 381, 383, 413 Muscular dystrophies, 349, 374–378 Mutation, 4, 5, 60, 107, 129, 202, 208, 328, 334, 350, 370, 374, 382 Myasthenias, 349, 371–374, 378

N NEFA, see Nonesterified free fatty acids (NEFA) Neonatal Fc receptor ( FcRn), 93–103 Neuroanatomy, 330 Neurodegenerative disease, 327, 332 Neuromuscular disease, 347–385 Neuromuscular junction, 349, 373 Neuropathic pain, 48, 355 Neuropathology, 325–327 NOD, 105–115, 120, 122–130, 254 NOD-scid IL2rγ null , 107–108, 109–114, 128 NONcNZO10/LtJ, 138 Nonesterified free fatty acids (NEFA), 140–141, 144–146 Novel Object Test, 304, 313 Nu/Nu mice, 250

O Obesity, 46, 48, 137–138, 160, 172, 336 Oncogene, 56, 225, 253–256 Oocyte, 26–27, 33–35, 56–57, 123, 328 Open Field, 46, 302, 304, 306–307, 312, 332 Orthotopic tumors, 225 Osmotic pumps, 142, 196, 199–200, 272, 304 Outbred, 2, 4–8, 10, 15, 18, 250, 264, 314, 328, 397 Ovary, 27, 33

P Pancreas, 136, 150–151 Pathogenesis, 106, 108, 114, 122, 396 Pathology, 46, 56, 196, 202–207, 260, 325–327, 338, 350, 359–360, 374, 385 Paw swelling, 182, 187–188 PCR, see Polymerase chain reaction (PCR) Peripheral neuropathies, 348, 368–370, 385 Peritonitis, 411–415 Pharmacodynamics, 86, 159 Pharmacokinetics (PK), 91, 96, 101–103, 152, 159, 231, 396 Phenotype, 4–5, 38, 40, 42, 44, 47, 51, 62, 122, 153, 205, 277, 300, 314, 316, 325, 328, 349, 353, 358–360, 370, 379, 384 Phenotyping, 3, 38, 44–49, 276, 328, 332, 338 Philadelphia chromosome (Ph+chromosome), 254, 259–261, 267 Phytosterols, 16, 169, 172–173, 176 Pneumonia, 405–410 Polymerase chain reaction (PCR), 39, 61, 99 Pooling fallacy, 334

PPAR-γ , 136, 141 PPRE-Luc, 80, 84 Preclinical, 51, 56, 59, 65, 68, 89, 96, 98, 102, 194, 228–229, 235–236 Presenilin 1, 326 scid Prkdc , 106, 109 See also Scid Promoter, 40, 56–57, 66–67, 70–71, 329, 362–363 Psoriasis, 194, 208 Pulmonary tumor, 221–222

R Rat, 2–3, 6, 48, 96, 170, 184, 271, 280, 304, 312–313, 318 Rat Exposure Test, 304, 312–313 Regenerative medicine, 51 Replacement, 38, 43, 50–51, 96, 328–329 See also Gene replacement Reporter gene, 69, 79 See also Beta-galactosidase; Luciferase; Green fluorescent protein (GFP) Reporter mouse, 79–92 Respiratory infection, 405–406 See also Infection Retinal degeneration, 202, 329 Retroviral, 58–59, 228, 237, 239–240, 250, 255–256, 259, 264 Retrovirus, 240–241, 255, 260, 265 Reverse cholesterol transport, 161–162, 172–175 Rheumatoid arthritis (RA), 181 RNA interference (RNAi), 52, 71–73, 283–298 ROSA26, 67, 69–71 Rosiglitazone, 136, 138–139, 141–142, 152

S Scid, 103, 105–115, 128–129, 227, 254 Screening, 1–20, 38, 61, 80, 85–86, 92, 193–209, 271, 274, 300, 315, 334 Selection Cassette, 40, 60–63 See also Selection Marker Selection Marker antibiotic, 40 diphtheria toxin, 61–62 ganciclovir, 62 hygromycin, 62 negative, 61–62, 128 neomycin, 61, 241 positive, 61–62 puromycin, 62 thymidine kinase, 62 Sepsis, 405, 412 Septic peritonitis, 411–415 Serum half-life, 95, 100 ShRNA, 71–73, 284, 287 Signal/noise ratio, 10, 16–18 SiRNA, 71–72, 142 Skin, 9, 152, 186, 189, 193–209, 225, 279, 395, 397, 399 Skin disorder, 193–209 Skin infection, 396–397 Social Interaction Test, 303, 309–310, 317 Soft tissue infection, 395, 396, 399 Sperm, 26–29, 31–35 Spinal cord, 347–348, 364–366, 368 Spinal Muscular Atrophy (SMA), 348, 350–351, 355 Spontaneous metastasis, 216–218, 222–230, 236

MOUSE MODELS FOR DRUG DISCOVERY 421 Index Staphylococcus aureus, 396 Startle Response, 302–303, 309, 316 Statin, 157 Statistics, 19, 335 Stem cell, 26, 28, 51–52, 59–71, 106–108, 114, 225, 253, 255, 257, 262–263 See also Embryonic stem (ES) cells; Hematopoietic stem cell (HSC) Streptococcus pneumoniae, 405 Streptococcus pyogenes, 395–402 Streptozotocin (STZ), 138 Stress-Induced Hyperthermia (SIH), 304, 311–312 STZ, see Streptozotocin (STZ) Subcutaneous injection (s.c.), 199 Subcutaneous inoculation, 396 Subcutaneous tissue, 395–396 Sucrose, 152, 171, 269, 272–273, 275–277 Suok Test, 303, 310–311 Superovulation, 26, 29–30, 33, 35, 130 Surgery, 224–225, 246, 285–287, 289–296, 411, 413–414 Survival curves, 399

Therapeutic antibody, monoclonal antibodies, 68, 96 Thymus, 50, 112, 121 Tissue collection, 140–141, 149–151 Toxicity testing, 2, 6–12, 19 Transgene, 39–40, 57–58, 60, 66, 69–71, 98, 122, 124–227, 255, 328–330, 332, 363 Transgenic, 9, 39–40, 51, 56–59, 66, 70, 93–103, 119–130, 142, 226, 255, 268, 284, 334, 350 Translational research, 253–265, 327 Transplantation, 51, 106, 229, 253, 255, 259–260, 263, 265 Trp53 knockout mouse, 9 Tumor cells, 217–218, 223, 227–228, 231, 236, 246 Tumor Growth Measurement, 247 Tumor models, 225–230 Type 1 diabetes (T1D), 119–122, 124, 127, 129–130 Type II diabetes mellitus (T2DM), 136–138, 141, 147, 150–151

T

W

Tail Suspension Test (TST), 270–271, 274, 278 Tamoxifen, 43, 90 Tauopathy, 325–327, 338 T cells, 46, 63, 106, 108, 113, 120–121, 125–128, 182 T1D, see Type 1 diabetes (T1D) T2DM, see Type II diabetes mellitus (T2DM) Tetracycline, 40, 72, 226 Tet-ON system, 72 See also Tetracycline

Welfare, 152, 201, 301

V Virus, see Retrovirus

X Xenograft, 51, 226, 229, 235, 254

Y YAC, see Yeast artificial chromosomes (YAC) Yeast artificial chromosomes (YAC), 57