Nod Mice and Related Strains: Research Applications in Diabetes, AIDS, Cancer, And Other Diseases (Molecular Biology Intelligence Unit)

MEDICAL I N T E L L I G E N C E U N I T

2

Edward Leiter • Mark Atkinson

NOD Mice and Related Strains: Research Applications in Diabetes, AIDS, Cancer and Other Diseases

R.G. LANDES C O M P A N Y

MEDICAL INTELLIGENCE UNIT 2

NOD Mice and Related Strains: Research Applications in Diabetes, AIDS, Cancer and Other Diseases Edward Leiter The Jackson Laboratory Bar Harbor, Maine, U.S.A.

Mark Atkinson University of Florida Gainesville, Florida, U.S.A.

R.G. LANDES COMPANY AUSTIN, TEXAS U.S.A.

MEDICAL INTELLIGENCE UNIT 2 NOD Mice and Related Strains: Research Applications in Diabetes, AIDS, Cancer and Other Diseases R.G. LANDES COMPANY Austin, Texas, U.S.A. Copyright © 1998 R.G. Landes Company All rights reserved. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Printed in the U.S.A. Please address all inquiries to the Publishers: R.G. Landes Company, 810 South Church Street, Georgetown, Texas, U.S.A. 78626 Phone: 512/ 863 7762; FAX: 512/ 863 0081

ISBN: 1-57059-466-X

While the authors, editors and publisher believe that drug selection and dosage and the specifications and usage of equipment and devices, as set forth in this book, are in accord with current recommendations and practice at the time of publication, they make no warranty, expressed or implied, with respect to material described in this book. In view of the ongoing research, equipment development, changes in governmental regulations and the rapid accumulation of information relating to the biomedical sciences, the reader is urged to carefully review and evaluate the information provided herein.

Library of Congress Cataloging-in-Publication Data

NOD mice and related strains: research applications in diabetes, AIDS, cancer, and other diseases / edited by Edward Leiter, Mark Atkinson. p. cm. -- (Medical intelligence unit) ISBN 1-57059-466-X (alk. paper) 1. Diabetes--Animal models. 2. Mice as laboratory animals. I. Leiter, Edward. II. Atkinson, Mark, 1961- . III. Series. [DNLM: 1. Mice, Inbred NOD. 2. Diabetes Mellitus, Insulin-Dependent. 3. Disease Models, Animal. QY 60.R6 N761 1997] RC660.N63 1997 619'.93--dc21 97-31480 CIP

MEDICAL INTELLIGENCE UNIT 2 PUBLISHER’S NOTE

NOD Mice andproduces Related Strains: R. G. Landes Company books in six Intelligence Unit series: Medical, Molecular Biology, Neuroscience, TissueE ngineering, Biotechnology and in Environmental. The Research Applications Diabetes, authors of our books are acknowledged leaders in their fields. Topics are unique; almostand withoutOther exception, Diseases no similar books AIDS, Cancer exist on these topics. Our goal is to publish books in important and rapidly changing areas of bioscience for sophisticated researchers and clinicians. To achieve this goal, we have accelerated our publishing program conform to the fast pace at which ThetoJackson Laboratory information grows in bioscience. of our books are Bar Harbor, Maine,Most U.S.A. published within 90 to 120 days of receipt of the manuscript. We would like to thank our readers for their continuing interest and welcome any comments or suggestions they may have for University of Florida future books.

Edward Leiter

Mark Atkinson

Gainesville, Florida, U.S.A.

Judith Kemper Production Manager R.G. Landes Company

R.G. LANDES COMPANY AUSTIN, TEXAS U.S.A.

CONTENTS 1. NOD Mice and Related Strains: Origins, Husbandry and Biology Introduction ........................................................... 1 Edward H. Leiter Development of the NOD and Related Strains ......................... 2 NOD Distribution and the Politics of International Competition ................................................. 4 NOD Strain Characteristics ........................................................ 5 Spontaneous Development of Insulin Dependent Diabetes Mellitus ..................................................................... 6 Differences in Characteristics of NOD Substrains .................... 7 Clinical and Pathophysiologic Features of IDDM in NOD Mice ........................................................................... 8 T- Lymphoaccumulation and the Possible Origins of Insulitis ................................................................. 10 Other NOD Strain-Characteristic Immunodeficiencies ......... 12 Leukocytic Infiltrates in Non-Pancreatic Glandular Tissues .................................................................. 15 Miscellaneous NOD Strain Characteristics ............................. 17 Reproductive and Developmental Biology .............................. 18 Characteristics of NOD-Related Strains .................................. 19 Appropriate Controls for NOD Mice ...................................... 26 2. Genetics and Immunogenetics of NOD Mice and Related Strains .................................................................... 37 Edward H. Leiter Introduction .............................................................................. 37 NOD and the Genetics of Autoimmunity ............................... 39 The H2g7 Haplotype: A Scaffolding for Initiation of Insulitis and Diabetes ........................................................ 43 Rare Allelles Do Not Necessarily Equate with Diabetogenic Allelles: the TAP Gene Imbroglio ......................................... 47 Genetic Analysis of Insulitis: IDDM Susceptibility Inherited as a Threshold Liability ......................................................... 49 Genetic Segregation Analysis as a First Step in Identification of Non-MHC Idd LOCI ........................................................ 50 Speed Cogenics .......................................................................... 52 Congenic and Subcogenic Stocks Indicate the Presence of Multiple Non-MHC Susceptibility LOCI on Numerous Chromosomes ............................................... 53 Connecting Idd Loci with Immunophenotypes ...................... 56 Do NOD Islets Express Strain-Specific Idd Genes? ................. 58 Lessons From Genetics of IDDM in NOD Mice for the Genetic Prediction of IDDM in Humans ............................. 59

3. The Identity and Ontogenic Origins of Autoreactive TLymphocytes in NOD Mice ................................................... 71 David V. Serreze Introduction .............................................................................. 71 Overview of T cell Ontogeny and Selection ............................. 71 The Diabetogenic Role of CD4+ Versus CD8+ T cells in NOD Mice ......................................................................... 75 TCR Gene Rearrangements Associated with β cell Autoreactivity in NOD Mice ................................................. 78 Pancreatic β cell Autoantigens Targeted by Diabetogenic T Lymphocytes in NOD Mice ............................................... 83 Mechanistic Basis for the Development of Autoreactive T cells in NOD Mice .............................................................. 86 Conclusions ............................................................................... 90 4. The Immunopathogenic Roles of Antigen Presenting Cells in the NOD Mouse ......................................................... 101 Michael Clare-Salzler Introduction ............................................................................ 101 The Contribution of Hematopoietically Derived APC to the Development of Diabetes Susceptibility in the NOD Mouse .............................................................. 102 The NOD H2g7 .............................................................................................................................. 102 APC Co-Stimulatory Molecules, CD80 and CD86 ............... 104 APC Subpopulations ............................................................... 106 B Lymphocytes ........................................................................ 107 Macrophages ............................................................................ 107 Dendritic Cells ......................................................................... 112 Conclusions ............................................................................. 113 5. The Natural History of Islet-Specific Autoimmunity in NOD Mice ........................................................................... 121 Jean-François Bach The Multi-Faceted Islet Specific Response ............................ 121 T Cell Repertoire ..................................................................... 126 The Nature of the β Cell Lesion .............................................. 128 Hypotheses on the Triggering of the β Cell Specific Response ................................................................. 130 Transient Early Protection from Diabetes by CD4 Immunoregulatory T Cells .................................... 130 The Search for Early Thymus and Bone Marrow Anomalies: An NK1+ T Cell Defect ........................................................ 132 Conclusions ............................................................................. 135

6. NOD Mice as a Model for Therapeutic Interventions in Human Insulin Dependent Diabetes Mellitus .................. 145 Mark A. Atkinson The Prevention of IDDM in Humans; Purpose and Historical Perspective ................................................... 145 The NOD Mouse as a Model for Prevention of IDDM in Humans ........................................................... 147 Genetic and Environmental Factors: Effect on Interpretation of Outcome Measures ........................... 148 Therapeutic Interventions in NOD Mice .............................. 150 Immunosuppression ............................................................... 152 Tolerance ................................................................................. 153 Immunostimulation ................................................................ 156 Reducing β Cell Metabolic Activity ........................................ 157 Dietary/Hormonal Manipulation .......................................... 158 Anti-inflammatory Agents ...................................................... 159 Human Trials for IDDM Prevention ..................................... 160 Conclusions ............................................................................. 162 7. The Use of NOD/LtSz-scid/scid Mice in Biomedical Research ........................................................... 173 Dale L. Greiner, Leonard D. Shultz Phenotypic Characteristics of scid Mice ................................. 173 Defective DNA Repair ............................................................. 174 Robust Innate Immunity ........................................................ 175 Care and Management of scid Mouse Colonies .................... 175 Utility of C.B-17-scid Mice as Research Tools ....................... 176 Characteristics of NOD/Lt/Sz-scid Mice ................................ 178 NOD/LtSz-scid Mice in Diabetes Research ............................ 181 Induction of Chemical Diabetes in NOD/LtSz-scid Mice ..... 184 NOD/LtSz-scid Mice as Islet Graft Recipients ....................... 185 Human Cell Engraftment in NOD/LtSz-scid Mice ............... 186 Engraftment of Hemopoietic Stem Cells ............................... 187 Epstein Barr Virus (EBV)-Related Tumors ........................... 189 Human Immunodificiency Virus-1 (HIV-1) Infection of Hu-PBL-NOD-scid Mice ................................................ 190 NOD/LtSz-scid Mice as Models for Human Tumor Grafts ....................................................................... 191 Use of NOD/LtSz-scid Mice in Parasitic Research ................ 191 New Models of NOD/LtSz-Immunodeficient Mice .............. 193 Conclusions ............................................................................. 194 Index ......................................................................................... 205

EDITORS Dr. Edward Leiter Dr. Leiter is a Senior Staff Scientist at The Jackson Laboratory in Bar Harbor, Maine. He earned a B.S. in Biology from Princeton University, and both M.S. and Ph.D. degrees in Biology/Cell Biology from Emory University. His research interests include the genetics and pathogenesis of insulin dependent and non-insulin dependent diabetes, and β cell metabolism. Chapters 1, 2 Dr. Mark Atkinson Dr. Atkinson is an Associate Professor of Pathology and Director for The Center for Immunology and Transplantation at The University of Florida College of Medicine in Gainesville, Florida. He earned a B.A. in Microbiology at The University of Michigan and a Ph.D. in Pathology from The University of Florida College of Medicine. His research interests are primarily focused on insulin dependent diabetes and include studies of cellular immunity, environmental factors associated with variances in the incidence of the disease, and designing interventions for disease prevention. Chapter 6

CONTRIBUTORS Dr. Jean-Francois Bach Dr. Bach is a Professor and Director of the Autoimmunity Section at the Hospital Neker in Paris, France. His research interests include many aspects of immunology (especially those relating to lymphocyte development and function), autoimmunity, genetic susceptibility, and interventions aimed at preventing insulin dependent diabetes. Chapter 5 Dr. Michael Clare-Salzler Dr. Clare-Salzer is an Associate Professor of Pathology and Medicine at The University of Florida College of Medicine in Gainesville, Florida. He earned a B.S. in Chemistry from The University of Notre Dame and an M.D. from The State University of New York (SUNY) School of Medicine. His research interests include various aspects of the pathogenesis of insulin dependent diabetes, macrophage function, and sepsis. Chapter 4

Dr. Dale Greiner Dr. Greiner is a Professor of Medicine in the Diabetes Division at The University of Massachusetts Medical School in Worcester, Massachusetts. In addition, he holds a joint appointment with The Department of Pathology at The University of Connecticut Health Center. He earned a B.A. in Biology and a Ph.D. in Microbiology and Immunology from The University of Iowa. His research interests include lymphocyte development and function, autoimmunity, and AIDS. Chapter 7 Dr. David Serreze Dr. Serreze is an Associate Staff Scientist at The Jackson Laboratory in Bar Harbor, Maine. While at The University of Maine, Dr. Serreze earned a B.S. in Biology, an M.S. in Microbiology, and a Ph.D. in Microbiology. His research interests include immunoregulation, function of class I major histocompatibility molecules, and the genetics of insulin dependent diabetes. Chapter 3 Dr. Leonard Shultz Dr. Shultz is a Senior Staff Scientist at The Jackson Laboratory in Bar Harbor, Maine. In addition, he holds joint appointments as a Graduate Faculty Member in Zoology at The University of Maine and a Research Professor at The University of Massachusetts Medical School. He earned a B.A. in Biology from Northeastern University and a Ph.D. in Pathogenic Bacteriology at The University of Massachusetts Medical School. His research interests include the development and regulation of the immune system, AIDS, and tumor immunology. Chapter 7

PREFACE Serendipity - making a fortunate or unexpected discovery by accident.

D

iscovery of a female mouse with autoimmune, insulin dependent diabetes mellitus (IDDM) in an incipient inbred line initially selected for normoglycemia was surely a serendipitous event. The subsequent selection for IDDM in the progeny of this female produced the current Nonobese Diabetic (NOD) strain. Prior to development of the NOD strain, the only animal model of autoimmune IDDM predictably developing the disease at high frequency was the BioBreeding (BB) rat. The advent of the NOD mouse provided researchers working in the field of autoimmunity with the needed perspective to assess findings in BB rats relative to what is known about pathogenic processes in humans. There are common etiopathogenic features observed in all three genera, as exemplified by involvement of susceptibility conferring major histocompatibility complex (MHC) haplotypes at the genetic level and of autoreactive T cells at the effector level. At the same time, there are sufficient distinctions between the rat and mouse models to remind the clinical investigator that rodents can only model various aspects of a human disease syndrome, while at the same time exhibiting speciesunique features. For example, resistance to development of severe ketoacidosis—despite severe hyperglycemia—is a genus-specific feature peculiar to the mouse model. Many generations of strict inbreeding have produced unusual phenotypes in both rodent models that distinguish them from each other, and probably from most humans destined to develop IDDM. The diabetes-prone BB rat is severely T-lymphocytopenic in peripheral lymphoid organs whereas the NOD mouse reflects the other extreme: T-lymphoaccumulation. Yet both BB rats and NOD mice develop a veritable “Pandora’s box” of organ-specific lymphocytic infiltrations, showing that immunoregyulatory pathways are compromised in both models, albeit in different ways. In-depth analysis of as many animal models as possible enhances an investigator’s appreciation of the etiologic complexity of IDDM in humans. Given the differences as well as the similarities between IDDM development in BB rats and NOD mice, a diabetes prophylactic treatment effective in both models becomes a potential treatment to prevent IDDM in humans. The Diabetes Prevention Trial using prophylactic insulin treatment discussed in chapter 6 is an illustration of how promising results obtained in several animal models can be translated into human clinical trials.

As detailed in chapter 5, the NOD mouse has been especially instructive for exploration the relationship between insulitis developNOD Mice and of Related Strains: Research Applications ment and the in ultimate expression of clinical disease. Insulitis in NOD Diabetes, AIDS, Cancer and Other Diseases mice entails the selective destruction of β cells following infiltration of the pancreatic islets by leukocytes (principally macrophages, T cells, and B lymphocytes). Certain of the genetic defects permitting development of destructive insulitis have been traced to bone marrow derived antigen presenting cells. The fact that IDDM susceptibility “tracks” with NOD hemopoietic stem cells makes the NOD mouse an attractive model to apply gene therapy approaches. Defects in T cell repertoire selection, discussed in chapters 3 and 4, have been correlated with specific immunophenotypic defects. If the only contribution of the NOD mouse were to the advancement of our understanding of the genetic and pathophysiologic basis for autoimmune endocrinopathies such as insulitis, thyroiditis, lupus and sialoadenitis, it would be a major contribution indeed. However, as pointed out in chapter 1 of this volume, the contributions of the NOD mouse to immunologic research comprise but a small aspect of its value to science. A genetically well-characterized inbred strain that is also phenotypically well-studied provides researchers with an important tool for dissecting how genes interact with the environment to control not only autoimmune phenotypes, but also many ISBN: 1-57059-466-X others. As described in the first several chapters of this volume, the genome of the NOD mouse has been intensively studied, and as described in chapter 6, this strain’s physiologic, endocrinologic and immunologic responses to multiple aspects of the physical environment have been published. The astounding breeding performance of the NOD mouse, coupled with the approximately 50% polymorphism in genomic simple sequence repeats when compared to those of other inbred strains, should bring this strain to the attention of all mouse geneticists doing physical mapping projects. MHC-congenic stocks of NOD mice are available that are IDDM-resistant, but which retain the high reproductive potential of standard NOD mice. These mice are ideal for outcross with other inbred strains carrying single gene mutations that can be physically mapped NOD mice andpositionally related strains: research in diabetes, (and thence cloned) in applications genetic segregation analyses. The AIDS, cancer, and by Edward Leiter, Mark Atkinson. NOD genome willother also diseases be useful/ edited for genetic control of multigenic disorcm. -- (Medical intelligence unit) ders,p.including deafness and inflammatory bowel disease. ISBN 1-57059-466-X (alk. paper) The deficiencies in immunoregulation discussed in chapters 3 and 1. Diabetes--Animal models.mouse 2. Micetoasthe laboratory animals. 4 should bring the NOD attention of cancer researchers. I. Leiter, Edward. Atkinson, Mark, 1961- . III. Possibly due to aII. dysfunctional population ofSeries. natural killer cells, NOD 1. Mice, NOD. 2.of Diabetes Insulin-Depenmice[DNLM: are subject to Inbred development a wideMellitus, spectrum of neoplasms. dent. 3. Disease Models, Animal.quite QY 60.R6 1997]strains, are relatively Among these, osteosarcomas, rare N761 in inbred RC660.N63 common in1997 NOD/Lt mice. Thymic lymphomas affect 100% of NOD619'.93--dc21 97-31480 scid/scid mice, but this development is circumvented in NOD-Rag/Rag CIPdescribed mice. The defective innate immune functions in NOD mice,

in chapters 2 through 4 of this volume combined with single gene mutations eliminating adaptive immune function (e.g., scid and Rag) is proving to be one of the most important biomedical applications of the NOD EDICAL genome. As described in chapter 7, these doubly immunocompromised NTELLIGENCE stocks of mice are being further genetically modified to abet the growth of normal and neoplastic human cells and tisNIT mice are dubbed “Hu-SCID” for humanized severe comsues. These bined immunodeficiency mice. Recent investigations have demonstrated the value of these mice for the study of human infectious diseases, including AIDS, filiariasis and malaria. The genomes of these NOD HuSCID mice are continually being modified by transgenic insertion of human genes to promote the growth of human hematopoietic cells, coupled with simultaneous elimination of selected murine genes that impair human cell development and survival. Continued modification of these mice may facilitate identification of children with pre-autoimmune diabetes by tracking adoptively transferred peripheral blood T cells to the host islets. The writers of the component chapters recognize that, because of the intense research activity involving NOD mice, the literature described will soon be outdated. What the authors have attempted, however, is to present, in an overview format, the literature current at the time of writing. Hence, information in this volume should permit the reader to incorporate the most recent published findings into a conceptual framework. When a conceptual is presented, some bias is inherTheframework Jackson Laboratory ent. The views expressed are based upon each author’s personal experiBar Harbor, Maine, U.S.A. ence working with the NOD model. As scientific knowledge increases, concepts will evolve. We acknowledge and gratefully thank our families who have supported us even when theUniversity NOD mouse distracted us from spending of has Florida more time with them. The staff at Landes Gainesville, Florida, Bioscience U.S.A. are also thanked for their patience and forbearance when deadlines were not met. Finally, the people who comprise The Juvenile Diabetes Foundation International “family” deserves a special “thank you” since many, if not most, of the pilot studies performed on the NOD mouse outside of Japan were supported by the generosity of this very special Foundation.

M I U 2

NOD Mice and Related Strains: Research Applications in Diabetes, AIDS, Cancer and Other Diseases Edward Leiter

Mark Atkinson

Edward Leiter, Ph.D. Mark Atkinson, Ph.D.

R.G. LANDES COMPANY AUSTIN, TEXAS U.S.A.

ABBREVIATIONS NOD Mice and Related Strains: Research Applications in Diabetes, AIDS, Cancer and Other Diseases

AIDS Acquired immunodeficiency syndrome AMLR Autologous mixed lymphocyte reaction APC Antigen presenting cell BCG Bacille Calmette-Guerin BSA Bovine serum albumin CFA Complete Freund’s adjuvant CTS Cataract Shionogi DTH Delayed type hypersensitivity EBV Epstein Barr virus GABA Gamma aminobutyric acid GAD Glutamic acid decarboxylase Hsp Heat shock protein HIV Human immunodeficiency virus IFN Interferon Ig Immunoglobulin IL Interleukin Idd Insulin dependent diabetes susceptibility loci, mouse ISBN: 1-57059-466-X IDDM Insulin dependent diabetes mellitus LPS Lipopolysaccharide NK Natural killer MHC Major histocompatibility complex NOD Nonobese diabetic NON Nonobese nondiabetic NOR Nonobese resistant RBC Red blood cell RT PCR Reverse transcriptase polymerase chain reaction PBMC Peripheral blood mononuclear cell PGE2 Prostaglandin E-2 PKC Protein kinase C scidmice and related Severe combined immunodeficiency NOD strains: research applications in diabetes, SMLR mixed leukocyte reaction AIDS, cancer, andSyngeneic other diseases / edited by Edward Leiter, Mark Atkinson. STZ Streptozocin p. cm. -- (Medical intelligence unit) SPF1-57059-466-X Specific pathogen free ISBN (alk. paper) TCR T cell receptor 1. Diabetes--Animal models. 2. Mice as laboratory animals. Th Edward. II. T helper I. Leiter, Atkinson, Mark, 1961- . III. Series. TNF Tumor necrosis [DNLM: 1. Mice, Inbred NOD.factor 2. Diabetes Mellitus, Insulin-DepenVDJC Variable, Diverse, Joining, Constant dent. 3. Disease Models, Animal. QY 60.R6 N761 1997] regions RC660.N63 1997 619'.93--dc21

97-31480 CIP

CHAPTER 1

NOD Mice and Related Strains: Origins, Husbandry and Biology Introduction Edward H. Leiter

N

onobese diabetic (NOD) is a recently generated inbred strain with a unique susceptibility to spontaneous development of autoimmune, insulin dependent diabetes mellitus (IDDM). Since their initial description in 1980 by Dr. Susumu Makino, NOD mice have been widely distributed and studied. These mice have provided important new immunogenetic and pathophysiologic insights into autoimmune disease and its prevention. The NOD genome is currently one of the most extensively characterized genomes of extant inbred strains. The strain is widely utilized in the analysis of how polygenically controlled immune defects confer susceptibility to IDDM. Multiple genes within its unique H2g7 major histocompatibility (MHC) haplotype confer the major component of IDDM susceptibility, consistent with the major contributions of MHC to autoimmune diseases, including IDDM, in humans. NOD mice have proven valuable in immunogenetic analysis to elucidate the pathogenic interactions between MHC and non-MHC diabetogenic loci, termed Idd loci (for Insulin Dependent Diabetes). However, interest in this strain is not limited to diabetes researchers. NOD mice develop a wide variety of organ-specific leukocytic infiltrates in addition to insulitis, and are susceptible to a broad spectrum of neoplasms. As detailed in the chapter by Greiner and Shultz in this volume, stocks of NOD mice congenic for the severe combined immunodeficiency (scid) mutation are proving exceptionally useful for NOD Mice and Related Strains: Research Applications in Diabetes, AIDS, Cancer and Other Diseases, edited by Edward Leiter and Mark Atkinson. © 1998 R.G. Landes Company.

2

NOD Mice and Related Strains: Research Applications in Diseases

growing both normal and neoplastic human tissues. The origin of the NOD mouse strain is fortuitous, the lucky combination of a prepared mind and serendipity. The fact that NOD mice were initially not released to the international research community willingly, and that certain NOD-related strains of obvious scientific value are still embargoed, highlights how corporate involvement in the development of an animal model of biomedical significance can adversely affect scientific progress. The history of this strain’s development in Japan, as well as the development of related strains, will be recounted briefly below.

DEVELOPMENT OF THE NOD AND RELATED STRAINS Although the NOD strain was developed in Japan, its heritage is international. NOD and related strains are derived from “Swiss” mice. “Swiss” mice were outbred, originating in Paris, but called “Swiss” because of the importation to the Rockefeller Institute for Medical Research in New York by Dr. Clara Lynch in 1926 of 2 albino males and 7 females from a colony maintained in Lausanne, Switzerland. The Rockefeller colony served as a source of “Swiss” mice to both research institutions and commercial breeders in the United States. SWR/ J and SJL/J represent two standardized strains derived from inbreeding progeny of Lynch’s “Swiss” mice. In 1947, Dr. Theodore S. Hauschka at the Institute for Cancer Research in Philadelphia received randombred “Swiss” mice from a commercial dealer and established the Ha/ ICR outbred stock. The stock was purposely kept outbred for cancer research purposes to reflect the genetically heterogeneous human population. In 1957, a commericial supplier, Charles River Laboratories, took Ha/ICR breeding stock and today markets their progeny as CD-1®. Both ICR and CD-1® mice have been widely used in research in Japan since the Second World War and are distributed there as Jcl:ICR by CLEA, Japan, and CD-1®:Crj (Charles River, Japan), respectively. In some publications, the ICR strain descriptor is occasionally (and erroneously) identified as “Imperial Cancer Research” instead of Institute for Cancer Research.2 The process of inbreeding ICR mice to produce new inbred strains in Japan continues to the present. Two of the more recently produced strains are ALS and ALR, selected respectively for susceptibility and resistance to the diabetogenic agent, alloxan.3

NOD Mice and Related Strains: Origins, Husbandry and Biology

3

The NOD and certain of the NOD-related inbred strains were developed from Jcl:ICR progenitors at the Shionogi Research Laboratories in Aburahi, Japan by Dr. Makino.1,4,5 Selection for cataract development was the initial goal of the breeding program. Inbreeding was begun in 1966 by Ohotori6 with progeny from an outbred Jcl:ICR female mouse exhibiting cataracts and microphthalmia. This selective breeding program led to the development of a strain in which all mice develop cataracts (now designated CTS, for Cataract Shionogi), as well as a cataract-free control strain separated at the 4th generation of inbreeding (designated NCT, both now beyond 100 generations of brother x sister matings). At the 6th generation of inbreeding (F6), two additional sublines, both free of cataracts, were initiated with the intent of developing a model for spontaneous diabetes development. At F6, Makino noticed some individuals exhibited high fasting blood glucose levels. By selective breeding of F6 sibs and their progeny that exhibited this phenotype, Makino hoped to develop a new inbred strain exhibiting spontaneous diabetes. At the same time, he recognized the need for a euglycemic control strain, and simultaneously inbred F6 sibs and their progeny exhibiting normal fasting blood glucose. In 1974, at the 20th generation of inbreeding (F20), a female spontaneously developed overt IDDM associated with heavy leukocytic infiltrations within the pancreatic islets (termed insulitis). Paradoxically, this female was not found in the line being selected for fasting hyperglycemia. Instead, the female with spontaneous IDDM was found in the line being inbred as a diabetes-free euglycemic control line. The progeny of this diabetic female were the founders of the Nonobese diabetic, or NOD strain. Since the line originally selected for high fasting glucose never progressed to overt diabetes at the time that IDDM appeared in the euglycemic control (now the NOD) line, the former strain was designated NON, for Nonobese normal,1 and more recently, redesignated as Nonobese nondiabetic.7 Obviously, it is quite serendipitous that Makino co-selected a euglycemic “control” line, and that his screening of this line was sufficiently rigorous so that the “exceptional” female was discovered and her progeny then selected to produce the NOD strain. Further adding to the luster of Makino’s achievement was that he maintained a husbandry environment sufficiently pathogen-free such that the diabetic phenotype could develop (see the section below on the critical effects of environment on the penetrance of the diabetic phenotype in NOD

4

NOD Mice and Related Strains: Research Applications in Diseases

mice). It should be noted that while subsequent inbreeding of ICR mice in Japan has led to other inbred strains, including the ILI8 (ICRL line-Ishibe), NSY (Nagoya-Shibata-Yasuda),9 ALS (alloxan-induced diabetes susceptible),3 and ALR (alloxan-induced diabetes resistant),3 none of the latter have developed spontaneous, autoimmune IDDM. Thus, Makino’s achievement is truly remarkable.

NOD DISTRIBUTION AND THE POLITICS OF INTERNATIONAL COMPETITION Makino’s first two publications describing the NOD strain1,10 appeared in Experimental Animals, a Japanese journal not widely seen by the diabetes and immunology research communities outside of Japan. A letter was sent by this author to Dr. Makino in December of 1980 asking whether it might be possible to import CTS and its related diabetic and non-diabetic substrains to The Jackson Laboratory for genetic analysis. A reply from Dr. Makino indicated that his employer, the Shionogi Company, a large Japanese pharmaceutical concern, had formed a “Committee on NOD Mice” and that the committee had decided to restrict distribution of these new strains within Japan. Investigations on diabetogenesis in NOD mice were carried out almost exclusively in Japan between 1980 and 1984 among scientists that constituted an “NOD Mouse Study Group” which met annually and published the proceedings of their annual meetings. During this period, the T cell basis for the autoimmune insulitis in NOD mice was firmly established. The culmination of this “All Nippon” approach to scientific progress was a volume on the NOD mouse, Insulitis and Type 1 Diabetes: Lessons from the NOD Mouse,11 published in 1986. This volume essentially summarized most of the work done in Japan on NOD mice during the four year “embargo” period. Two events opened the door to NOD’s worldwide immigration. The first was a set of letters in 1983 to the Shionogi Company from editors of the major Western diabetes-focused journals, Diabetes and Diabetologia, indicating that no further manuscripts on the NOD model would be accepted until the distribution barrier was dropped. The second event was the “unscheduled” migration of NOD mice to institutions in Australia (the NOD/Wehi strain progenitors, 1984) and the United States. A colony established at the University of California, Los Angeles by Dr. Yoko Mullen was the source of the current NOD/ Ym, NOD/Mrk, and NOD/MrkTac (Taconic Farms, Inc.) substrains. NOD and NON brought by Dr. M. Hattori from a colony in Kyoto to

NOD Mice and Related Strains: Origins, Husbandry and Biology

5

the Joslin Diabetes Center in Boston in 1984 (NOD/Jos and NON/ Jos) also provided two breeding pairs of NOD and three breeding pairs of NON mice to the author at The Jackson Laboratory (the nucleus of NOD/Lt and NON/Lt). Because of the exceptional breeding performance of NOD mice, it was soon possible to supply pedigreed breeding pairs of NOD/Lt mice to numerous investigators in the United States, Canada, Europe and Australia. By 1986, the Central Laboratory for Experimental Animals (CLEA, Japan) was receiving NOD/Shi breeding stock from the Shionogi Company’s source colony for international distribution. Research grants funded by the Juvenile Diabetes Foundation International provided many investigators with the financial support necessary to begin their investigations on the NOD mouse, such that by the time of the publication of the information obtained by the “NOD Study Group” in Japan,11 the international literature was burgeoning with publications on the immunopathogenesis of diabetes in the NOD mouse. In the five year period between 1986 and 1990, there were 192 entries on the NOD mouse in the bibliographic database of the National Library of Medicine (MEDLINE); in the ensuing five years between 1990-1995, there have been in excess of 669 entries in this database. A scientific meeting sponsored by the Juvenile Diabetes Foundation International in 1989 brought Japanese and non-Japanese NOD researchers together for the first time to share the benefits of international scientific cooperation. Currently, in addition to NOD/Shi distributed by CLEA, Japan, inbred NOD substrains are distributed in the United States by The Jackson Laboratory (NOD/LtJ and NOD/LtSz-scid) and Taconic Farms, Inc. (NOD/MrkTac). In Europe, NOD mice are distributed by Bomholtgard, Denmark. NON/LtJ and NOR/LtJ, an NOD-related strain (see below) are distributed by The Jackson Laboratory. At the time of this writing, open distribution of CTS/Shi mice is still forbidden by the Shionogi Company despite multiple publications by Shionogi investigators in the international literature.

NOD STRAIN CHARACTERISTICS Many of the observations provided in this section were obtained by observation of the author’s own substrain, NOD/Lt, at The Jackson Laboratory. Many details of the immunopathogenesis of IDDM will be omitted from this section, since they comprise the focus of later chapters in this volume.

6

NOD Mice and Related Strains: Research Applications in Diseases

SPONTANEOUS DEVELOPMENT OF INSULIN DEPENDENT DIABETES MELLITUS As would be inferred from their name, the most notable strain characteristic of NOD mice is their unique susceptibility to spontaneous development of IDDM in the absence of any cytopathic pancreatotropic viruses or other viruses. The frequency of diabetes achieved in any colony is critically dependent upon the specific pathogen-free (SPF) status of the colony.12-14 Whereas the Kilham rat virus is known to be a diabetogenic trigger in certain strains of inbred rats,15 exposure of NOD mice to a variety of murine pathogenic viruses prevents rather than triggers IDDM.12-14,16 During the initial period of NOD investigations in Japan, destruction of pancreatic b cells was firmly established as a T cell-mediated process (reviewed in ref. 5; also see other chapters in this volume). Insulitis initiates as leukocytic aggregates at the perimeters of islets (“peri-insulitis”). CD4+ T cells are the predominant leukocyte, followed by smaller numbers of CD8+ T cells, B cells and macrophages.17,18 As reviewed in chapter 3 both CD4+ and CD8+ T cell subsets are essential mediators of the spontaneously-developing disease. B cells and macrophages also are necessary contributors to pathogenesis. Autoantibodies, presumably maternal in origin, have been detected on b cells from 3-week-old NOD/Uf mice.19 As insulitic destruction of islets progresses, autoantibodies to a variety of b cell antigens, including insulin and glutamic acid decarboxylase (GAD) are generated. Although these autoantibodies are not central to the process of b cell destruction, the absence of IDDM in a congenic stock of B cell-deficient NOD/Lt mice establish that B cells are essential antigen presenting cells for both initiation and amplification of T cell responses to b cell autoantigens.20 The extent and severity of insulitis, as it progresses from a non-destructive peri-insulitis to a destructive intra-islet infiltration, is used as a semi-quantitative method to assess the efficacy of various antidiabetic therapies. A guide for scoring the various stages of insulitis is available.21 Although the incidence of IDDM in NOD/Shi is much higher in females (>80% ) than in males (<20%), widespread pancreatic lymphoaccumulation and insulitis are common to both sexes by nine weeks of age.22 An oft-repeated statement concerning NOD mice is that all develop insulitis. While this may be true for small experimental groups of NOD mice, it is probably not accurate for a larger population of animals. In the NOD/Lt colony, approximately

NOD Mice and Related Strains: Origins, Husbandry and Biology

7

10% of mice of both sexes sampled between 6-20 weeks were insulitisfree. This observation underscores the fact that development of IDDM in NOD mice is a threshold phenomenon, requiring the appropriate interactions between underlying susceptibility genes, the endocrine system the immune system, and the environment.23

DIFFERENCES IN CHARACTERISTICS OF NOD SUBSTRAINS Inbred colonies of NOD mice maintained separately for more than 10 generations from the Shionogi Aburahi source colony (NOD/Shi) are designated by the colony holder or institutional identification symbols. Accumulation of new mutations among these separated colonies can be expected to produce significant substrain divergence (and thus, potential alterations in strain characteristics) over time. Further, before the diabetogenic genes in the NOD/Shi strain had been fully fixed, some “lower incidence” lines were apparently distributed to members of the “NOD Mouse Study Group” in Japan, and some of these mice found their way out of Japan to form breeding nuclei at other institutions. NOD/Wehi is probably an example. This substrain, received from Dr. H. Asamoto in Kyoto and inbred since 1984 at the Walter and Eliza Hall Institute in Melbourne, Australia, exhibits high levels of insulitis in mice of both sexes, but a very low incidence in males (6% by 250 days of age), and a lower than expected incidence (47% by 250 days) in females.24 Based upon a recent report, the diabetes incidence in NOD/Shi mice beyond F20 in a specific pathogen-free (SPF) source colony at Aburahi Laboratories has remained relatively constant, with 70-80% incidence in females by 30 weeks of age versus a 20% incidence in males.5 This gender dimorphism is in part controlled by gonadal sex steroids. Gonadectomy at five weeks markedly increased diabetes development in NOD/Shi males while it depressed diabetes incidence in females.10 Similarly, pharmacologic levels of dihydrotestosterone implanted into young prediabetic NOD females prevents development of clinical symptoms, but not insulitis.25 While there is nothing unusual about the levels of gonadal sex steroids in the plasma of NOD/Lt mice, complex interactions between endocrine factors and environmental factors apparently interact to determine the frequency of IDDM observed in a given colony. A more detailed discussion of environmental parameters affecting diabetogenesis in NOD mice, especially exposure to pathogenic agents, may be found in chapter 6.

8

NOD Mice and Related Strains: Research Applications in Diseases

NOD males serve as sensitive indicators of the presence of environmental factors capable of modulating the penetrance of the inherited genetic susceptibility. An excellent illustration of this is the higher male IDDM frequency observed in the NOD/Lt substrain maintained by this author at The Jackson Laboratory. NOD/Lt males, inbred in a research colony at The Jackson Laboratory since 1984, are distinguished from NOD/Shi and NOD/Wehi males by a higher frequency (40-60%) of IDDM by 30-40 weeks of age. Unlike NOD/Shi mice, in which adolescent gonadectomy increases frequency of disease in males, the high incidence of IDDM in NOD/Lt males is not further increased by castration at five weeks of age. Although the difference in male incidence between the two substrains could represent differential exposure to specific pathogens, or genetic divergence, current evidence indicates that the difference resides in factors associated with husbandry, particularly dietary factors.26 A complex natural ingredient “chow” diet contains diabetogenic accelerators absent in semi-purified diets.27 Pregestimil™ is a defined formulation infant diet that delays onset and reduces frequency of IDDM development in NOD/Lt females, and completely blocks IDDM in NOD/Lt males. Protection mediated by this diet was associated with increases in total hepatic cytochrome p450 content as well as activity (this laboratory, unpublished study). Thus, the diabetogenic “accelerators” in chow diets may actually represent inhibitors of gender-dimorphic p450 activities that catabolize immunomodulatory steroids such as estrogens and glucocorticoids.

CLINICAL AND PATHOPHYSIOLOGIC FEATURES OF IDDM IN NOD MICE Onset of clinical diabetes (as measured by insulinopenia, hyperglycemia, glycosuria, polyuria and polydipsia) occurs after puberty, with the earliest cases generally appearing around 10 weeks of age. In a research colony of NOD/Lt mice at The Jackson Laboratory, 7090% of virgin, group-caged females are diabetic by 20 weeks of age, while onset is typically delayed in males, with a 40-50% incidence generally not achieved before 30 weeks of age. Female mice are stronger immune responders than are males, so the gender bias is consistent for an autoimmune disease in mice. Destruction of the pancreatic islets by leukocytic infiltrates occurs over a more protracted period in NOD mice as compared to the diabetes-prone BB rat (see ref. 26 for a comparative review of these two rodent models of IDDM). In T-

NOD Mice and Related Strains: Origins, Husbandry and Biology

9

lymphocytopenic and diabetes-prone BB rats, the islets are free of T cells until only 1-3 weeks prior to onset of clinical disease.28 Once T lymphocytic infiltrates have focused on the islets, transition from euglycemia to severe hyperglycemia is abrupt. In contrast, despite an earlier appearance (3-5 weeks in females, 5-7 weeks in males) of larger aggregations of leukocytes surrounding and/or penetrating the islets of young prediabetic NOD mice, precipitious declines in pancreatic insulin content (a reflection of destruction of b cell mass) is generally not observed before 12 weeks of age in NOD/Lt females, and later in NOD/Lt males.29 Interestingly, this is a time period in which many diabetes-prone BB rats are becoming acutely diabetic. Before the onset of non-fasting hyperglycemia or glycosuria, those prediabetic individuals approaching the diabetogenic threshold exhibit an impaired glucose tolerance test when challenged with 2.5 glucose/100 g body weight. When NOD mice are followed longitudinally (weekly testing intervals) to detect the first shift from euglycemia to hyperglycemia and/or glycosuria, the initial levels are followed by progressively higher values over a 3-4 week period, accompanied by only mild ketoacidosis and ketonuria, and only moderate weight loss during the first two weeks post-diagnosis of glycosuria. Glycosuria is not detected unless non-fasting plasma glucose levels reach 300 mg/dl or higher. The absence of severe ketoacidosis in diabetic NOD mice reflects a general ability of mice to metabolize ketone bodies to lactate in the liver.30 This in part may account for why NOD mice will live for 1-2 months after the detection of overt hyperglycemia (plasma glucose > 250 mg/dl) and glycosuria (>1/4%) without any administration of insulin. Plasma insulin levels of young normoglycemic NOD mice of both sexes usually range between 1-1.5 ng/ml (25-38 µU/ml). When chronic hyperglycemia is established, plasma insulin levels fall below the range of assay detection (<0.1 ng/ml). Plasma glucagon levels are elevated.31 Long-term management of diabetic NOD mice within normoglycemic levels by twice-daily insulin injections is difficult to achieve. Doses of 1 to 2 units of 1:1 mixtures of regular and slower acting porcine insulin preparations are injected i.p. as required by empirical determination of hyperglycemic status immediately before injection in the morning and evening. The fact that certain long-term secondary complications associated with chronic IDDM in humans such as retinopathies, neuropathies, and severe nephropathies are not common in diabetic NOD mice may in part be attributed to low survivorship after chronic diabetes is established.

10

NOD Mice and Related Strains: Research Applications in Diseases

As described in chapters 5 and 6, immunointervention therapies can reverse IDDM symptoms if initiated immediately after the onset of clinical symptoms, a time when residual b cell mass is still present. An interesting histopathologic feature of the NOD pancreas is that islets with insultic lesions often are larger than islets in standard inbred strains such as C57BL/6J or BALB/cByJ that are free of insulitic lesions. However, this distribution of larger-sized islets is independent of insulitis since comparably-sized islets are also observed in insulitis-free NOD-scid/scid mice.32 Insulitis appears during a time of islet growth; the finding of relatively constant pancreatic insulin content over the first 10 weeks post-partum29 despite development of a progressively more severe insulitis in many of the islets, implies that b cell hypertrophy and/or hyperplasia transiently keeps pace with any b cell loss due to cell-mediated or cytokine-elicited destruction. Not all peri-insular or intra-islet T cells are pathogenic effectors. CD4+ T cell clones that suppress IDDM have been isolated from the islets of young prediabetic NOD mice.33 A variety of cytokines appear to be elaborated by the various T cell clones that suppress diabetogenesis (see chapter 3). A strain-specific immunodeficiency characteristic of NOD/Lt mice is a weak ability of their antigen presenting cells to activate regulatory T cells in a syngeneic mixed lymphocyte reaction in vitro.34 Hence, cytopathic activation of a “destructive” insulitis may reflect an age- and sex-dependent shift in the balance of regulatory to effector cells systemically as well as in the insulitic infiltrates.35 NOD mice appear to be deficient both in the thymus and in the periphery of a relatively rare subset of TCR a/b+CD4–CD8– T cells, the so-called NK1+ T cell. These cells are reputedly high IL-4 producers and are discussed in detail in chapter 5 (JF Bach). They are difficult to study in NOD mice since they do not express the NK1.1 allotype. NOD mice congenic for the NK1.1 marker continue to show deficiencies in this subset; interestingly, the deficiency can be reversed by IL-7 administration.36

T-LYMPHOACCUMULATION AND THE POSSIBLE ORIGINS OF INSULITIS NOD mice are immunodeficient in a number of functions, one of which is the control of the number of T cells emigrating from the thymus into the periphery. This has been termed T-lymphoaccumulation. One of the most striking NOD strain-specific histopathologic features associated with T-lymphoaccumulation is the

NOD Mice and Related Strains: Origins, Husbandry and Biology

11

quantity of leukocytes filling the pancreatic lymphatic system and emanating from the pancreatic perivascular and periductular structures in an age-dependent fashion.37 This T-lymphoaccumulation is not reflected by a lymphoadenopathy as observed in mice with lymphoproliferative disease produced by mutations in the Fas or Fas ligand-encoding genes. Rather, lymphoid organs are normal in cellularity, but blood, spleen, and lymph nodes contain an elevated ratio of T to B cells compared to standard Swiss derived inbred strains such as SWR/J. The origins of T-lymphoaccumulation can be traced in part to defects in T cell maturation and/or selection in the thymus. Thymic and peripheral T cells from NOD mice display a proliferative unresponsiveness on stimulation through the T-cell receptor/CD3 complex, a defect associated with weak signalling through the p21ras pathway.38 Thymic involution in NOD/Lt mice is slow to develop compared to other inbred strains (and possibly other NOD substrains), and probably is associated with defects in apoptosis39-41 that allow increased numbers of mature T cells to emigrate from the thymus (see chapter5). Indeed, increased percentages of T cells (relative to total leukocytes) in both lymphoid organs, as well as anomalous accumulations of leukocytes in the parenchyma of various secretory glandular tissues, most notably pancreas, submandibular salivary gland, lacrimal glands, and Harderian glands, represents a strain characteristic.42 Analysis of thymectomized NOD/Lt mice in the author’s laboratory has provided illustrations of the remarkable T-lymphoproliferative potential of NOD/Lt mice. Whereas neonatal thymectomy was effective in producing long-lasting T-lymphocytopenia in other inbred strains in the author’s laboratory, this was not the case for NOD/Lt. In the latter, remaining thymic residua become fully repopulated as the mice aged so that T lymphopenia was only transient such that neither insulitis nor IDDM was prevented (unpublished observations). Similarly, when adolescent thymectomy was performed without concurrent irradiation, T cell numbers in the spleen declined rapidly over a 30-day period, but then rebounded to control levels. Pronounced development of thymic ectopies to the thyroid gland (as described by Many43 in non-thymectomized NOD mice) were observed in 80% of adolescent thymectomized mice (M. Cetkovic Cvrlje and E.H. Leiter, manuscript in preparation). A further reflection of this T-lymphoproliferative drive in NOD/Lt mice is the nearly 100% frequency of thymic lymphomas in NOD/LtSz-scid/scid mice.44 It is worthwhile

12

NOD Mice and Related Strains: Research Applications in Diseases

noting why the initial report from Japan erroneously concluded that NOD/Shi mice were T-lymphopenic.45 It is possible that the mice used in this early study were still segregating genes for the T-lymphopenia gene preventing emigration of mature thymocytes from the thymus that was fixed in the CTS strain (See refs. 46-48 and below). Also, this study utilized non-inbred ICR mice with larger lymphoid organs than found in Swiss-derived inbred strains as the basis for comparison. The importance of appropriate control strain selection for immunologic and pathophysiologic studies in NOD mice will be discussed later.

OTHER NOD STRAIN-CHARACTERISTIC IMMUNODEFICIENCIES ANTIGEN PRESENTING CELLS Since the immunodeficiencies associated with NOD antigen presenting functions are enumerated in detail in chapter 4, they will only be briefly mentioned here. NOD mice reportedly do not tolerize well to self-antigens,49 a defect that perhaps accounts for failure to negatively select autoreactive T cells in the thymus and/or to suppress their function in the periphery.34 Analysis in vitro of NOD bone marrowderived macrophage responses to myelopoietic growth factors (CSF-1) and differentiation factors (interferon gamma) show that they fail to mature completely.50-52 This is further indicated by subnormal lipopolys accharide (LPS) stimulated IL-1 secretion,50,51 the biologic potency of the IL-1 secreted is attenuated by a higher than normal production of IL-1 receptor antagonist (Serreze and Leiter, unpublished observations). This defect appears to be post-transcriptional, since IL-1 mRNA levels appear normal.53,54 Several lines of evidence argue that this decreased IL-1 secretory capacity may be of pathogenic significance. IL-1 supplementation in vitro restores the ability of NOD antigen presenting cells to activate immunoregulatory T cells in a syngeneic mixed lymphocyte reaction.53 Further, IDDM is blocked in NOD mice treated with recombinant IL-1 in vivo.55 Another aberrant function observed in NOD macrophages is that they lack normal regulation of the inducible prostaglandin E2 synthase gene (Pgst2 on Chromosome 1) and hence secrete higher than normal levels of prostaglandin E2.56,57 Finally, a congenic stock of NOD mice carrying C57BL/6J-derived diabetes resistance alleles on the distal part of Chro-

NOD Mice and Related Strains: Origins, Husbandry and Biology

13

mosome 3 in the Idd10 region exhibits a reduced frequency of IDDM.58 Marrow-derived macrophages from this congenic stock exhibits a more balanced IL-1 and IL-receptor antagonist secretory response to LPS (Serreze and Leiter, unpublished). Defects in both the high affinity Fcg1 receptor59 and Fcg2 recep60 tor have been reported. Signaling through the Fcg receptors modulates the balance of IL-1/IL-1 receptor antagonist secreted by human monocytes,61 such that these genetic defects may be associated with aberrant secretory responses exhibited by NOD/Lt macrophages in response to LPS stimulation. Both defects would also be expected to affect the ability of macrophages to phagocytose monomeric IgG2a and IgG2b antibodies, respectively. A possible example of substrain divergence is the finding that in NOD/Lt mice, IgG2b autoantibodies to insulin62 and GAD63 predominate, whereas a predominance of IgG2a autoantibodies to GAD have been reported in other NOD substrains.64 The mutation in the Fcγr1 gene, eliminating 300 amino acids of the cytoplasmic tail of the molecule, was originally proposed as a candidate gene for Idd10, but recombination analysis has eliminated this possibility (Dr. Linda Wicker, personal communication). However, a more balanced response to LPS-stimulated secretion of bioreactive IL-1 was observed when several of these recombinants were tested (D.V. Serreze and E.H. Leiter, unpublished). Interestingly, the frequency of IDDM was increased rather than decreased in a NOD congenic stock in which the defective Fcgr2 locus on distal Chromosome 1 was replaced by a wild-type allele from C57B6/J mice.60 T CELL FUNCTIONS As will be detailed in other chapters in this volume, NOD CD4+ T-helper cells tend to respond to antigenic stimulation with a T-helper 1-predominant cytokine profile. This may account, in part, for why NOD mice are strongly resistant to a variety of murine pathogenic agents. NOD T cells are good producers of interferon gamma,35,65 whereas NOD CD4+ T cells are especially low producers of IL-4,66 and tend towards low production of IL-10.67 Whereas NOD bone marrow stem cells do not respond normally to IL-3,52 NOD/Lt T cells are high producers of IL-3. As discussed in the next chapter, one of the stronger of the non-MHC diabetes susceptibility loci, Idd3 on Chromosome 3, is localized to a region of less than 0.3 cM containing the IL-2 structural gene.68 Although splenic T cells from NOD/Lt mice secrete levels of IL-2 in vitro that are comparable to SWR/Bm con-

14

NOD Mice and Related Strains: Research Applications in Diseases

trols when activated by Concanavalin A, purified NOD T cells are low IL-2 secretors in a syngeneic mixed lymphocyte reaction compared to purified SWR T cells.53 Furthermore, NOD spleens are enriched for T cells compared to SWR mice such that, on a per cell basis, NOD T cells probably secrete less IL-2 after mitogenic or allogeneic stimulation. Hence, although the functional activity of the IL-2 produced by NOD T cells appears to be normal,69 defects in the regulation of this locus are indicated. Combined with the exceedingly low IL-4 production by NOD thymocytes, low IL-2 production in response to selfantigen presentation in the thymus may underlie some of the apoptotic defects that in turn produce peripheral T-lymphoaccumulation. Nevertheless, even after activation in vitro by IL-2, NOD T cells are still more resistant to apoptosis induction than are lymphocytes from other strains.39 NATURAL KILLER (NK) CELLS NOD/Lt mice are deficient in functional NK cells as measured by inability to kill YAC-1 cell targets.45,54,70 This deficiency was not corrected by treatment in vivo with IL-2 or poly I:C.54 B CELLS Because autoantibodies appear to play a secondary role in IDDM pathogenesis in NOD mice,71 and because NOD B cells have only recently been established as an essential antigen presenting cell population for the initiation of disease,20 NOD strain-specific defects in Bc ell functions have not been as extensively studied as have T cell functions. Some of the NOD strain-specific defects reflected by poor activation of regulatory T cells in a syngeneic mixed lymphocyte reaction likely represent deficiencies in the B cell as well as the macrophage/dendritic cell population. These may include the report of unstable MHC class II (I-Ag7)/peptide complexes on NOD B cells.72 An abnormal development of the B cell repertoire has been indicated by continued usage of D-proximal variable heavy chain (VH) immunoglobulin genes characteristic of neonatal mice.73 NOD B cells, like Tc ells, exhibit extended survival in vitro.73 In this regard, we have observed high Bcl2 gene expression, and reciprocally lower Fas gene expression in the spleens of 8-week-old NOD mice compared to levels in spleens of three other inbred strains (M Cetkovic-Cvrlje and EH Leiter, unpublished observations). These molecular shifts would be expected to retard the process of programmed cell death.

NOD Mice and Related Strains: Origins, Husbandry and Biology

15

LEUKOCYTIC INFILTRATES IN NON-PANCREATIC GLANDULAR TISSUES SALIVARY AND LACRIMAL GLANDS The original description of the NOD mouse by Makino and colleagues noted the presence of heavy perivascular/periductular leukocytic infiltrates in the submandibular salivary glands at a high frequency.1 Since NOD mice also have leukocytic infiltrates in their lacrimal glands, it was proposed that NOD mice may serve as a model for Sjögren’s syndrome in humans. The histopathologic NOD salivary lesions, termed sialoadenitis or sialitis, developed after puberty and did not reflect a loss of glandular acinar cells as extensive as the progressive loss of b cells in the pancreas. These leukocytic infiltrates nevertheless impair function of the submandibular gland as evidenced by a progressive decline in both salivary flow rates and protein content of saliva.74 Epidermal growth factor, the product of submandibular gland duct cells, was lower in NOD/Uf (University of Florida) males than in BALB/c males, and was drastically reduced in diabetic NOD/Uf mice of both sexes.74 Autoantibodies to salivary gland and anti-nuclear proteins have also been reported.75,76 Thus, although loss of salivary gland structural integrity is not as extensive as in Sjögren’s syndrome in humans, xerostomia, or “dry mouth,” a clinical condition often associated with both Sjögren’s syndrome and IDDM in humans, is reflected in NOD mice. Sialoadentitis is efficiently transferred by splenic leukocytes or purified T cells from diabetic NOD/Lt donors into NOD/ LtSz-scid/scid recipients over the same time course as insulitis/IDDM transfer.77 Levels of insulin and insulin-like growth factors secreted into saliva are higher in NOD/Uf than in BALB/c mice.78 Since salivary cells may produce additional novel peptides in common with pancreatic b cells,79 it is tempting to speculate that a common autoantigen drives sialoadenitis and insulitis. Several lines of evidence argue against this. An aging associated sialoadenitis is commonly found in other strains not prone to autoimmune insulitis, including NON/ Lt, NZB, and C57BL/6J. The H2g7 MHC haplotype integral to insulits and diabetes development in NOD mice is not required for sialoadenitis development since the latter still develops in certain MHC congenic NOD mice resistant to insulitis and IDDM.80, 81 NOD mice do not express MHC class II I-E molecules on the cell surfaces of antigen presenting cells. It has been inferred that lack of I-E expression is the basis for sialoadenitis development in mice.82 This is a rather

16

NOD Mice and Related Strains: Research Applications in Diseases

astounding hypothesis since both NON/Lt and New Zealand Black (NZB) mice, both prone to aging-associated sialoadenitis development, express I-E molecules, and NOD/Lt mice expressing I-Ead transgenes still develop sialoadenitis.63 In the case of NOD/LtSz-scid/ scid mice, there appears to be a highly unusual loss of parotid and submandibular gland protein composition in the absence of sialoadenitis.83 This comparison was made to BALB/c instead of BALB/cscid/scid, so the effect of the mutation (and absence of normal levels of lymphokines) was not controlled. However, both NOD and NODscid/scid mice showed aberrant processing/secretion of two proteins, parotid secretory protein and proline rich protein compared to BALB/c.83 The infiltrates in NOD lacrimal glands are less extensive than in the submandibular glands, and more commonly, the Harderian rather than the lacrimal glands contain leukocytic infiltrates. Interestingly, the lacrimal glands also produce parotid secretory protein, and the NOD allele may be unique and transcriptionally regulated in an aberrant fashion compared to other inbred strains.84 THYROID A high incidence of thyroiditis and anti-thyroid autoantibodies have been reported in NOD mice.85 Splenocytes from diabetic NOD/Lt donors adoptively transferred thyroiditis into NOD-scid/scid recipients at a lower efficiency than insulitis.77 Spontaneously-developing lesions are generally quite mild, consisting of only a few leukocytes within thyroid follicles. Parathyroiditis has also been reported.86 A confounding factor in the diagnosis of both thyroiditis or parathyroiditis is the high frequency of thymic ectopies to the thyroid gland found in NOD mice.43 Clinical lesions comparable to Hashimoto’s thyroiditis can be induced in NOD mice by administration of a high sodium iodide containing diet following induction of goiters using a low iodine diet in combination with phenylthiourea treatment.87 An MHC congenic (H2h4) stock of NOD/Mrk mice will develop thyroiditis following sodium iodide exposure alone (Dr. Linda Wicker, Merck Laboratories, personal communication; also see ref. 88). MUSCLE, NERVE, KIDNEY AND COLON In non-diabetic NOD/Lt mice at a year and older, focal leukocytic infiltrates in muscle, nervous tissue, kidney and large intestine are observed. These multi-organ lesions are a further reflection of the

NOD Mice and Related Strains: Origins, Husbandry and Biology

17

T-lymphoaccumulation peculiar to this strain. Rarely are these lesions extensive enough to have clinical manifestations, although several cases of hind limb paralysis/meningitis have been diagnosed in the NOD/ Lt colony at The Jackson Laboratory. Focal glomerular lesions89 and changes in gene expression90 are apparent in NOD mice of both sexes prior to onset of diabetes. Cell lines established from NOD mesangial cells constitutively secrete more IGF-1 in vitro than do lines established from prediabetic mice, indicating stable phenotypic changes in growth factor production had occurred. There is no evidence that kidney functions are seriously compromised. Mild lesions are found at low frequency in NOD/Lt mice, mostly in the cecum.91 Flow-sorted CD45RBhigh CD4+ T cells adoptively transfer mild colitis into NODscid/scid recipients.92 EXPERIMENTALLY INDUCED LESIONS NOD mice are susceptible to a variety of experimentally-induced disease syndromes, including drug-induced diabetes, lupus, colitis, encephalomyelitis and thyroiditis. Not surprisingly, young NOD/Lt males, like outbred CD-1 males, are susceptible to induction of IDDM by multiple low doses of the fungal antibiotic, streptozotocin.93 This sensitivity is independent of autoimmune T cell involvement since NOD/LtSz-scid/scid males are even more sensitive to low doses of this diabetogen.93 Injection of a single dose of 2.6 x 107 heat-killed Mycobacterium tuberculosis [bacillus Calmette-Guerin (BCG)] i.v. into 8-week-old NOD mice prevented diabetes but precipitated a syndrome similar to systemic lupus erythematosus.94 Exposure of NOD/Lt mice to 3.5% dextran sodium sulfate in the drinking water for five days elicits a severe colitis.95 Experimental allergic encephalomyelitis can be induced in NOD mice using a peptide fragment (residues 56-70) from proteolipid protein.96 Experimental induction of thyroiditis was discussed in the preceding section.

MISCELLANEOUS NOD STRAIN CHARACTERISTICS DEAFNESS NOD mice are severely hearing impaired, developing a progressive hearing loss as assessed by auditory evoked brain stem responses (Dr. Ken Johnson, The Jackson Laboratory, personal communication). In most NOD mice, hearing loss progresses to complete deafness by

18

NOD Mice and Related Strains: Research Applications in Diseases

three months. CBA/J mice have normal hearing. Three recombinant congenic stocks have been produced by outcross of NOD mice to CBA/LtJ, followed by a backcross to CBA and then inbreeding of first backcross progeny. 97 Auditory evoked brain stem response testing of these three recombinant congenic lines carrying mixtures of NOD/Lt and CBA/LtJ genes indicates that this trait is under polygenic control since one line hears as well as CBA, and two lines have moderately impaired hearing. Since all three lines share the H2g7 MHC haplotype with NOD, genes within this complex are excluded. HEMOLYTIC ANEMIA Hemolytic anemia accompanied by development of Coombs’ positive autoantibodies, reticulocytosis, splenomegaly, and jaundice, has been observed in aging (>200 days) mice of both the NOD/Wehi and NOD/Lt substrains.98 In the low diabetes incidence NOD/Wehi substrain, 14/17 non-diabetic mice were affected by 550 days. TUMORS Aging NOD mice have been described as a veritable “Pandora’s Box” of aging-associated pathologies, including tumors.99 The description below will be limited to the authors’ NOD/Lt colony. It is rare for an NOD/Lt mouse that has not developed diabetes to survive beyond 1.5 years. Consistent with the presence of multiple immunodeficiencies in NOD/Lt mice, including the lack of functional NK cells, a broad variety of neoplasias were observed.13,99 In a relatively small sampling population of 54 diabetes-free mice, of which only 39 were older than eight months of age, the following neoplasms (number of cases in parenthesis) were detected: lymphomas and lymphosarcomas (5), osteosarcomas and osteochondrosarcomas (3), myoepitheliocarcinoma (1), rhabdomyosarcoma (1), mammary carcinoma (1) and hepatoma (1). Only one type of tumor was detected in a single individual. As mentioned above and as discussed in chapter 7 by D.G reiner and L.D. Shultz, thymic lymphomas are the major cause of mortality in NOD/LtSz-scid/scid mice maintained under SPF conditions.

REPRODUCTIVE AND DEVELOPMENTAL BIOLOGY NOD mice are excellent breeders. They mate at a young age and, for an inbred strain, produce exceptionally large litters (i.e., 9-14 pups). Little pre-weaning mortality in litters is experienced as long as the dam remains IDDM-free. In high incidence colonies, females seldom

NOD Mice and Related Strains: Origins, Husbandry and Biology

19

deliver more than two litters before females develop IDDM. Offspring of diabetic dams are larger at birth, making NOD mice a model for diabetic pregnancy.100 Testicular alterations in diabetic males have also been reported, including germ cell degeneration, tubular fibrosis and calcification, and disruption of spermatogenesis.101 High incidences of developmental anomalies have also been reported in embryos of diabetic dams.102,103 Injection of a single dose of 50 µl of Complete Freund’s Adjuvant into one hind footpad of each breeder female and male prevents IDDM onset in NOD/Lt females until after at least two and usually more litters have been born and weaned.13 Although NOD females are good natural ovulators, they do not superovulate well in response to the standard doses of gonadotropins used to superovulate other inbred strains, and the embryos obtained following treatment with gonadotropins are fragile and do not survive microinjection/reimplantation. Thus, while transgenes can be inserted directly into the NOD genome instead of that of hybrid strain combinations, yields of founder transgenic mice are low. Further, NOD preimplantation embryos do not progress beyond the two-cell stage when cultured in vitro in conventional culture media, but will develop to blastocyst in more advanced media.21 The important features of these media are reduced sodium, increased potassium, decreased glucose and phosphate, and addition of EDTA (Dr. John Eppig, The Jackson Laboratory, personal communication).

CHARACTERISTICS OF NOD-RELATED STRAINS Because certain of the NOD-related strains described above are often used for genetic, immunologic or physiologic comparison to NOD, a brief description of their strain characteristics will be included here, while salient genetic features of each strain will be covered in the next chapter. NON (NONOBESE NONDIABETIC) The inbred strain developed from the ICR-derived line with the high fasting blood glucose levels has been designated NON (Nonobese nondiabetic).5 Because of its relatedness to the NOD strain, NON mice are extremely useful for genetic analyses since they apparently share with NOD, some, but not all of the diabetogenic loci predisposing to diabetes. 104,105 Aging NON/Lt mice develop small foci of perivascular/periductular leukocytic infiltrates into the pancreas that do not progress to destructive insulitis. Focal infiltrates in the

20

NOD Mice and Related Strains: Research Applications in Diseases

submanidbular salivary glands are also increasingly common in older NON/Lt mice. Although NON mice do not develop autoimmune diabetes, and never become hyperglycemic under non-fasting conditions, NON/Lt males at The Jackson Laboratory develop a marked obesity by 20 weeks of age.106 This may be caused by low levels of the hormone leptin in adipocytes and in serum.107 Moreover, NON mice of both sexes are intolerant to glucose loading, and develop severe glomerulosclerotic kidney lesions.108 As NON/Lt mice age, immunoregulatory anomalies become apparent. Although T cell numbers and function are normal in young NON/Lt mice, by 20 weeks of age, T-lymphocytopenia, particularly in the CD8+ subset, is demonstrable in the spleen.109 This loss of T cells is reflected by a decline in T cell mitogen responses, and probably is associated with an NON strainspecific defect in the Tap1 (transporter associated with antigen processing) gene.110 A 113 bp insertion approximately 1 kb upstream of the Insulin 2 (Ins2) locus on Chromosome 7 in NON/Shi has been reported.111 However, it was subsequently found that this represented a wild-type sequence.112 NON/Lt mice are not hearing impaired like NOD/Lt, but are blind due to the presence of the retinal degeneration (rd) mutation on proximal Chromosome 5 associated with integration of a xenotropic proviral genome (Xmv25). Thus, NON mice are not necessarily the best controls for establishing certain “normative” behavioral, physiologic or immunologic baselines. However, because the genome of the NON strain has been as extensively characterized for simple sequence repeat polymorphisms as has NOD, this strain is extremely useful for genetic outcross experiments with NOD.105,113 CTS (CATARACT SHIONOGI) AND NCT (NON CATARACT) This strain, initially selected for cataracts with microphthalmia, has fixed an autosomal recessive mutation producing T-lymphopenia and T-lymphocytopenia that expresses at the level of the thymus.4648 The mutation may well be the mouse homolog of the lymphopenia (Lyp) gene in diabetes-prone BB rats.114 If so, it should map to mouse Chromosome 6 in linkage with the neuropeptide Y locus based upon homology with the map position of rat linkage markers.115 Although thymus and thymocyte development appear normal, a defect in ability of mature thymocytes to emigrate to the periphery produces a profound peripheral T-lymphopenia and lymphocytopenia evident

NOD Mice and Related Strains: Origins, Husbandry and Biology

21

at weaning.46-48 CTS mice fail to reject NOD skin grafts. Peripheral B cell numbers are normal. Mature thymocytes express the Mel-14 antigen and exhibit normal homing to the lymph nodes when explanted thymocytes were injected intravenously.48 This peripheral T-lymphocytopenia, then, represents the mirror image of the T-lymphoaccumulation characteristic of NOD mice. CTS thymocytes exhibit strong mitogenic responses upon Concanavalin A or anti-CD3 activation48 whereas NOD thymocytes are anergic. The NON strain is intermediate between NOD and CTS in that peripheral T cell numbers are normal at weaning, but decline rapidly with aging, with the loss of the CD8+ subset being especially marked. The CTS strain expresses a rare H2 haplotype, H2ct, which shares with NOD the same unique MHC class II and one of the unique class III (Hsp70) alleles, but differs at class I genes, some of which are unique to CTS.116 NCT is a cataract-free control strain separated from CTS at the 4th generation of inbreeding. There is no mention in the Japanese publications describing the CTS strain as to whether NCT mice carrying the T-lymphocytopenia gene, nor have details of its MHC been published. Both NCT and CTS apparently develop perivascular/periductular infiltrates at low frequency in females as they age.8 ILI (ICR-L LINE-ISHIBE) This strain was independently derived from outbred Jcl:ICR mice. Serologic analysis shows that ILI mice share common MHC class I and class II alleles with NOD.8 ILI mice, like NON mice, become obese with aging (M. Hattori, Joslin Diabetes Center, personal communication). They are apparently not T-lymphocytopenic since unlike CTS mice, they rapidly reject NOD skin grafts.8 ILI mice develop perivascular/periductular infiltrates, but not insulitis nor IDDM. ILI and NOD lymphocytes failed to respond reciprocally in a mixed lymphocyte reaction.8 Splenocytes from diabetic NOD donors transfer insulitis, but not IDDM, into ILI-nu/nu recipients. ILI could potentially serve as an excellent non-diabetic control strain for many types of immunologic studies conducted in NOD mice. NOR/Lt NOR/Lt is a recombinant congenic stock produced following an outcross between NOD/Lt and C57BLKS/J, and inbreeding initiated at the second backcross to NOD.117,118 These mice are albino and

22

NOD Mice and Related Strains: Research Applications in Diseases

carry NOD alleles at approximately 88% of all polymorphic markers typed, including the H2g7 haplotype as well as NOD markers for Idd susceptibility genes on Chromosomes 2, 3, and elsewhere. The NOR/Lt stock exhibits the same degree of hearing loss as NOD/LtJ mice. However, NOR mice only very rarely develop IDDM spontaneously, although they exhibit T-lymphoaccumulation, sialoadenitis and perivascular/periductular leukocytic infiltrates in the pancreas that progress to insulitis in a few islets as the mice age. Analysis of congenic stocks has demonstrated that C57BL-derived genome on Chromosome 2 provides a major component of the diabetes resistance.118 Young NOR mice of both sexes are resistant to cyclophosphamide induced IDDM. However, as they age, 20-30% can be rendered diabetic following administration of two injections of 200 mg/kg cyclophosphamide injected two weeks apart (this laboratory, unpublished). NOR/Lt mice exhibit a more robust syngeneic mixed lymphocyte reaction than do NOD/Lt mice, and certain aberrant antigen presenting cell responses to interferon gamma observed in NOD/Lt mice are not exhibited by NOR/Lt.50 ALS, ALR (ALLOXAN SUSCEPTIBLE, ALLOXAN RESISTANT) These two inbred strains have been recently derived by inbreeding of outbred CD-1 mice with selection for susceptibility (ALS) or resistance (ALR) to diabetes induced within a week after administration of 45-47 mg/kg alloxan.3 Other strain characteristics have not been described, although a preliminary genetic comparison has been reported.119 Six-month old ALR, but not ALS mice, are completely deaf (Dr. Ken Johnson, The Jackson Laboratory, personal communication). IQI/JIC This strain was produced by inbreeding ICR mice in Japan. Treatment of these mice with mercuric chloride induced antinucleolar autoantibodies.120 Similar to a report for the NOD strain,121 the presence of large numbers of thymic B cells has recently been reported in IQI/Jic mice.122 At the time of this writing, there has been no genetic or immunogenetic characterization of these mice other than the anecdotal report of a slowly progressive sialitis in females.122

Closely related to NOD, same MHC class II genes, different class I genes

closely related to NOD, MHC-identical

progenitor stock for NOD and related strains

Swiss-derived like ICR and NOD, genetically very different from but inbred and without NOD, including MHC immunodeficiencies; available

NOD-derived recombinant exhibits some but not all of NOD’s congenic stock; same MHC, immune dysfunctions. differs at relatively few non-MHC loci.

CTS/Shi

ILI/Jic

ICR (usually available as CD-1)

SWR/J

NOR/Lt

randomly bred

not commonly available

early developing T-lymphocytopenia; unavailable

Closely related to NOD; diabetes Develops obesity, impaired glucose resistant MHC tolerance, immunodeficiencies, difficult to breed.

NON/Lt

Disadvantage

Advantage

Strain

Table 1.1. Diabetes-free or resistant control strains for NOD mice References

53

117-118

Analysis to establish which Idd genes control aberrant immunopheno-types essential to pathogenesis.

116

8

46-48

control for immune functions that are aberrant in NOD

analysis of population frequency of rare genetic polymorphisms present NOD

Genetic analysis of non-MHC -linked Idd genes

Genetic analysis of the contributory role of MHC class II as well as other H2g7 alleles in the Idd1 complex

Resistant MHC 7, 13 Genetic analysis of Idd genes, potential model for type II diabetes

Best Use

NOD Mice and Related Strains: Origins, Husbandry, and Biology 23

no endogenous T- or B-lymphocyte functions

NOD. CB17-scid and NOD.CB.17 -scid Emv30null

NOD.B10H2b

diabetes-resistant MHC from SWR/J, available diabetes-resistant MHC from C57BL/10J, available

NOD-SWR-H2q

NON.NOD-H2g7 diabetogenic MHC from NOD/Lt; available

develops high incidence of thymoma with age; slower onset in Emv30null stock

delineation of the role of T cell 7,125-126 subsets; source of NOD islets free of insulitis; growth of human tissues

63

References

exhibits some but not all of NOD’s all MHC congenic stocks are 50-51 immune dysfunctions extremely useful in dissecting the role of MHC exhibits some but not all of NOD’s versus non-MHC genes 50-51 immune dysfunctions in pathogenesis and in identifying aberrant immunoexhibits some but not all of NOD’s phenotypes under MHC control 123 immune dysfunctions exhibits some but not all of NOD’s 81, 124 immune dysfunctions

analysis of T-helper cell repertoire development; presentation of β cell antigens

Best Use

NOD.NON-H2nb1 diabetes-resistant MHC from NON/Lt; available

Disadvantage NOD mice not tolerant to I-E+ antigen presenting cells from these mice

Advantage

NOD-Tg(Ead)Lt one gene difference confers Line 5 transgenics diabetes-resistance; several independent lines available in Europe, U.S., and Japan

Strain

Table1.2. Diabetes-free or resistant controls for NOD: available transgenic and congenic stocks

24 NOD Mice and Related Strains: Research Applications in Diseases

NOD.NON-Thy1a useful T cell allotypic marker for adoptive transfer studies into NOD-scid recipients (which are Thy1b)

127

77

130

123,128-129

delineation of essential pathogenic role of B lymphocytes

delineation of role of MHC class I and CD8+ T cells in pathogenesis

same as scid mice

retarded onset of diabetes, requiring various types of adoptive that more mice be aged to provide transfer studies diabetic donors

not tolerant to Ig+ cells

NOD.Igµnull

B-lymphocyte deficient, available

B2mnull stock not tolerant to class I+ cells; scidB2mnulls stock is tolerant.

no endogenous T- or B-lymphocyte no thymic lymphomas functions; not yet widely available reported

NOD.B2mnull no MHC class I on cell surfaces in NOD.scidB2mnull the absence of b2-microglobulin, CD8+/-deficient, available

NOD.Rag2

NOD Mice and Related Strains: Origins, Husbandry, and Biology 25

26

NOD Mice and Related Strains: Research Applications in Diseases

APPROPRIATE CONTROLS FOR NOD MICE The question of the appropriate control to use for experimentation with NOD mice often arises. With the exception of the NON/Lt, SWR/J, and NOR/Lt stocks that are available from The Jackson Laboratory, most of the other NOD-related strains have not been distributed outside of Japan. Table 1.1 lists some of the potential strains and congenic stocks that have been used to establish experimental “baseline” parameters in the absence of the insulitis and diabetes characteristic of NOD mice. The most appropriate choice for a control strain depends upon the nature of the investigation being undertaken (endocrinologic, immunologic, physiologic, etc.). Table 1.1 and Table 1.2 summarize the more well-defined control strains or stocks. ACKNOWLEDGMENTS This writing has been supported by the National Institutes of Health grants DK 36175 and DK27722, and a grant from The Juvenile Diabetes Foundation International. The author thanks Drs. David Serreze, Len Shultz and Linda Wicker for critical reviews. REFERENCES 1. Makino S, Kunimoto K, Muraoka Y, Mizushima Y, Katagiri K, Tochino Y. Breeding of a non-obese, diabetic strain of mice. Exp Anim 1980; 29:1-8. 2. Ikegami H, Eisenbarth GS, Hattori M. Major histocompatibility complex-linked diabetogenic gene of the nonobese diabetic mouse. Analysis of genomic DNA amplified by the polymerase chain reaction. J Clin Invest 1990; 85:18-24. 3. Ino T, Kawamoto Y, Sato K, et al. Selection of mouse strains showing high and low incidences of alloxan-induced diabetes. Exp Anim 1991; 40:61-67. 4. Makino S, Hayashi Y, Muraoka Y, Tochino Y. Establishment of the nonobese-diabetic (NOD) mouse. In: Sakamoto N, Min HK, Baba S, ed. Current Topics in Clinical and Experimental Aspects of Diabetes Mellitus. Amsterdam: Elsevier 1985:25-32. 5. Kikutani H, Makino S. The murine autoimmune diabetes model: NOD and related strains. In: Dixon FJ, ed. Advances in Immunology NY: Academic Press, 1992; 50:285-322. 6. Ohotori H, Yoshida T, Inuta T. “Small eye” and “cataract”, a new dominant mutation in the mouse. Exp Anim 1968; 17:91-96. 7. Makino S, Yamashita H, Kunimoto K, et al. Breeding of the NON mouse and its genetic characteristics. In: Sakamoto N, Hotta N, Uchida K, ed. Current Concepts of a New Animal Model: The NON mouse. Tokyo: Elsevier Science Publishers BV, 1992:4-10.

NOD Mice and Related Strains: Origins, Husbandry and Biology

27

8. Hattori M, Fukuda M, Ichikawa T, et al. A single recessive non-MHC diabetogenic gene determines the development of insulitis in the presence of an MHC-linked diabetogenic gene in NOD mice. J Autoimmun 1990; 3:1-10. 9. Ueda H, Ikegami H, Yamato E, et al. The NSY mouse: A new animal model of spontaneous NIDDM with moderate obesity. Diabetologia 1995; 39:503-508. 10. Makino S, Kunimoto K, Muraoka Y, Katagiri K. Effect of castration on the appearance of diabetes in NOD mouse. Exp Anim 1981; 30:137-140. 11. Tarui S, Tochino Y, Nonaka K. Insulitis and Type 1 Diabetes. Lessons from the NOD Mouse.Tokyo: Academic Press, 1986:294. 12. Leiter EH. The role of environmental factors in modulating insulin dependent diabetes. In: deVries R, Cohen I, van Rood JJ, eds. Current Topics in Immunology and Microbiology. The Role of Microorganisms in Non-infectious Disease. Berlin: Springer Verlag, 1990: 39-55. 13. Leiter EH. The nonobese diabetic mouse: a model for analyzing the interplay between heredity and environment in development of autoimmune disease. ILAR News 1993; 35:4-14. 14. Ohsugi T, Kurosawa T. Increased incidence of diabetes mellitus in specific pathogen-eliminated offspring produced by embryo transfer in NOD mice with low incidence of the disease. Lab Anim Sci 1994; 44(4):386-388. 15. Ellerman KE, Richards CA, Guberski DL, et al. Kilham rat virus triggers t-cell-dependent autoimmune diabetes in multiple strains of rat. Diabetes 1996; 45(5):557-562. 16. Bowman MA, Leiter EH, Atkinson MA. Autoimmune diabetes in NOD mice: a genetic programme interruptible by environmental manipulation. Immunol Today 1994; 15:115-120. 17. Miyazaki A, Hanafusa T, Yamada K, et al. Predominance of T lymphocytes in pancreatic islets and spleen of pre-diabetic non-obese diabetic (NOD) mice: a longitudinal study. Clin Exp Immunol 1985; 60:622-630. 18. Jarpe AJ, Hickman MR, Anderson JT, et al. Flow cytometric enumeration of mononuclear cell populations infiltrating the islets of Langerhans in prediabetic NOD mice: development of a model of autoimmune insulitis for type 1 diabetes. Regional Immunol 1991; 3:305-317. 19. Shieh D-C, Cornelius J, Winter W, Peck A. Insulin-dependent diabetes in the NOD mouse model. I. Detection and characterization of autoantibody bound to the surface of pancreatic beta cells prior to development of the insulitis lesion in prediabetic NOD mice. Autoimmunity 1993;15:123-135. 20. Serreze D, Chapman H, Varnum D, et al. B lymphocytes are essential for the initiation of T cell-mediated autoimmune diabetes: analysis of a new speed congenic stock of NOD.Igmnull mice. J Exp Med 1996; 184(5):2049-2053.

28

NOD Mice and Related Strains: Research Applications in Diseases

21. Leiter E. The NOD mouse: a model for insulin dependent diabetes mellitus. In: Shevach EM, Coico R, ed. In: Current Protocols in Immunology. New York: John Wiley & Sons, Inc., 1997: vol 3, in press. 22. Makino S, Muraoka Y, Kishimoto Y, Hayashi Y. Genetic analysis for insulitis in NOD mice. Exp Anim 1985; 34:425-432. 23. Leiter EH. Lessons from the animal models: the NOD mouse. In: Palmer JP, ed. Diabetes Prediction, Prevention, and Genetic Counselling. London: John Wiley & Sons, 1996:201-226. 24. Baxter AG, Adams MA, Mandel TE. Comparison of high- and lowdiabetes incidence NOD mouse strains. Diabetes 1989; 38:1296-1300. 25. Fox HS. Androgen treatment prevents diabetes in nonobese diabetic mice. J Exp Med 1992;175:1409-1412. 26. Serreze DV, Leiter EH. Genetic and pathogenic basis for autoimmune diabetes in NOD mice. Current Opin Immunol 1994; 6:900-906. 27. Coleman DL, Kuzava JE, Leiter EH. Effect of diet on the incidence of diabetes in non-obese diabetic (NOD) mice. Diabetes 1990; 39:432-436. 28. Logothetopoulos J, Valiquette N, Madura E, Cvet D. The onset and progression of pancreatic insulitis in the overt, spontaneously diabetic, young adult BB rat studied by pancreatic biopsy. Diabetes 1984; 33:33-36. 29. Gaskins HR, Prochazka M, Hamaguchi K, Serreze DV, Leiter EH. Beta cell expression of endogenous xenotropic retrovirus distinguishes diabetes susceptible NOD/Lt from resistant NON/Lt mice. J Clin Invest 1992; 90:2220-2227. 30. Coleman DL. Acetone metabolism in mice: increased activity in mice heterozygous for obesity genes. Proc Natl Acad Sci USA 1980; 77:290-293. 31. Ohneda A, Kobayashi T, Nihei J, et al. Insulin and glucagon in spontaneously diabetic non-obese mice. Diabetologia 1984; 27:460-463. 32. Rosmalen J, Jansen A, Homo-Delarche F, et al. Effect of prophylacticinsulin treatment on the number of antigen presenting cells and macrophages inthe pancreas of NOD mice. Is the prevention of diabetes based on b-cell rest? J Autoimmunity 1995; 21(1):28-29. 33. Akhtar I, Gold JP, Pan LY, et al. CD4(+) beta islet cell-reactive T cell clones that suppress autoimmune diabetes in nonobese diabetic mice. J Exp Med 1995; 182(1):87-97. 34. Serreze D. Autoimmune diabetes results from genetic defects manifest by antigen presenting cells. FASEB J 1993; 7:1092-1096. 35. Liblau RS, Singer SM, McDevitt HO. Th1 and Th2 CD4(+) T cells in the pathogenesis of organ-specific autoimmune diseases. Immunol Today 1995;16(1):34-38. 36. Gombert J, Tancredebohin E, Hameg A, et al. IL-7 reverses NK1(+) T cell-defective IL-4 production in the non-obese diabetic mouse. Int Immunol 1996;8(11):1751-1758.

NOD Mice and Related Strains: Origins, Husbandry and Biology

29

37. Fujino-Kurihara H, Fujita H, Hakura A, et al. Morphological aspects on pancreatic islets of non-obese diabetic (NOD) mice. Virchows Arch, Cell Pathol 1985; 49:107-120. 38. Jaramillo A, Gill BM, Delovitch TL. Minireview: insulin dependent diabetes mellitus in the non-obese diabetic mouse: a disease mediated by T cell anergy? Life Sci 1994; 55(15):1163-1177. 39. Garchon H-J, Luan J-J, Eloy L, Bédossa P, Bach J-F. Genetic analysis of immune dysfunction in non-obese diabetic (NOD) mice: mapping of a susceptibility locus close to the Bcl-2 gene correlates with increased resistance of NOD T cells to apoptosis induction. Eur J Immunol 1994; 24:380-384. 40. Colucci F, Cilio CM, Lejon K, et al. Programmed cell death in the pathogenesis of murine IDDM: resistance to apoptosis induced in lymphocytes by cyclophosphamide. J Autoimmun 1996; 9(2):271-276. 41. Penha-Goncalves C, Leijon K, Persson L, Holmberg D. Type 1 diabetes and the control of dexamethazone-induced apoptosis in mice maps to the same region on chromosome 6. Genomics 1995; 28(3):398-404. 42. Serreze D, Leiter E. Insulin Dependent Diabetes Mellitus (IDDM) in NOD Mice and BB Rats: Origins in Hematopoietic Stem Cell Defects and Implications for Therapy. In: Shafrir E, ed. Lessons from Animal Diabetes. V. London: Smith-Gordon, 1995: 59-73. 43. Many M-C, Drexhage HA, Denef J-F. High frequency of thymic ectopy in thyroids from autoimmune prone nonobese diabetic female mice. Lab Invest 1993; 69(3):364-367. 44. Prochazka M, Gaskins HR, Shultz LD, Leiter EH. The NOD-scid mouse: a model for spontaneous thymomagenesis associated with immunodeficiency. Proc Natl Acad Sci, USA 1992; 89:3290-3294. 45. Kataoka S, Satoh J, Fujiya H, et al. Immunologic aspects of the nonobese diabetic (NOD) mouse. Abnormalities of cellular immunity. Diabetes 1983; 32:247-253. 46. Yagi H, Suzuki S, Matsumoto M, Makino S, Harada M. Immune deficiency of the CTS mouse. I. Deficiency of in vitro T cell-mediated immune response. Immunol Invest 1990; 19:279-295. 47. Yagi H, Nagata M, Takeuchi M, et al. Immune deficiency of the CTS mouse. II. Impaired in vivo T cell-mediated immune response. Immunol Invest 1990;19:493-505. 48. Yagi H, Matsumoto M, Nakamura M, et al. Defect of thymocyte emigration in a T cell deficiency strain (CTS) of the mouse. J Immunol 1996; 157:3412-3419. 49. Shimada A, Charlton B, Rohane P, et al. Immune regulation in type 1 diabetes. J Autoimmun 1996; 9(2):263-269. 50. Serreze DV, Gaskins HR, Leiter EH. Defects in the differentiation and function of antigen presenting cells in NOD/Lt mice. J Immunol 1993;150:2534-2543.

30

NOD Mice and Related Strains: Research Applications in Diseases

51. Serreze DV, Gaedeke JW, Leiter EH. Hematopoietic stem cell defects underlying abnormal macrophage development and maturation in NOD/Lt mice: defective regulation of cytokine receptors and protein kinase C. Proc Natl Acad Sci USA 1993; 90:9625-9629. 52. Langmuir P, Bridgett M, Bothwell A, Crispe I. Bone marrow abnormalities in the non-obese diabetic mouse. Internat Immunol 1993; 5:169-177. 53. Serreze DV, Leiter EH. Defective activation of T suppressor cell function in Nonobese Diabetic mice. Potential relation to cytokine deficiencies. J Immunol 1988; 140:3801-3807. 54. Serreze DV, Hamaguchi K, Leiter EH. Immunostimulation circumvents diabetes in NOD/Lt mice. J Autoimmunity 1990; 2:759-776. 55. Jacob CO, Aiso S, Michie SA, et al. Prevention of diabetes in nonobese diabetic mice by tumor necrosis factor (TNF); similarities between TNF-a and interleukin 1. Proc Natl Acad Sci USA 1990; 87:968-972. 56. Xie T, Hofig A, Yui M, et al. Spontaneous prostaglandin synthase-2 (Pgs2) gene expression in macrophages of NOD and congenic mice. Autoimmunity 1995; 21(1):17A. 57. Xie T, Reddy S, Hofig A, et al. Regulation of prostaglandin synthase2 (Pgs-2) in NOD macrophages. Autoimmunity 1996; 24(1):23A. 58. Wicker LS, Todd JA, Prins J-B, et al. Resistance alleles in two nonMHC-linked insulin dependent diabetes loci on chromosome 3, Idd3 and Idd10, protect NOD mice from diabetes. J Exp Med 1994; 180:1705-1713. 59. Prins J-B, Todd J, Rodriques N, et al. Linkage on chromosome 3 of autoimmune diabetes and defective Fc receptor for IgG in NOD mice. Science 1993; 260:695-698. 60. Luan J, Monteiro R, Sautes C, et al. Defective Fc gamma RII gene expression in macrophages of NOD mice—genetic linkage with upregulation of IgG1 and IgG2b in serum. J Immunol 1996; 157(10): 4707-4716. 61. Marsh CB, Pope HA, Wewers MD. FCg recptor cross-linking downregulates IL-1 receptor antagonist and induces IL-1b in mononuclear phagocytes stimulated with endotoxin or Staphylococcus aureus. J Immunol 1994; 152:4604-4611. 62. Serreze DV, Leiter EH, Kuff EL, et al. Molecular mimicry between insulin and retroviral antigen p73. Development of cross-reactive autoantibodies in sera of NOD and C57BL/KsJ-db/db mice. Diabetes 1988; 37:351-358. 63. Hanson MS, Cetkovic-Cvrlje M, Ramiya V, et al. Quantitative thresholds of MHC Class II I-E expression on hematopoietically derived APC in transgenic NOD/Lt Mice determine level of diabetes resistance and indicate mechanism of protection. J Immunol 1996; 157:1279-1287. 64. Lenschow D, Herold K, Rhee L, et al. CD28/B7 regulation of Th1 and Th2 subsets in the development of autoimmune diabetes. Immunity 1996; 5(3):285-293.

NOD Mice and Related Strains: Origins, Husbandry and Biology

31

65. Tsumura H, Komada H, Ito Y, Shimura K. In vitro and in vivo interferon production in NOD mice. Lab Anim Sci 1989; 39:575-578. 66. Rapoport M, Zipris D, Lazarus A, et al. IL-4 reverses thymic T cell anergy and prevents the onset of diabetes in NOD mice. J Exp Med 1993; 178:87-99. 67. Rabinovitch A. Immunoregulatory and cytokine imbalances in the pathogenesis of IDDM. Diabetes 1994; 43(5):613-621. 68. Lord CJ, Bohlander SK, Hopes EA, et al. Mapping the diabetes polygene Idd3 on mouse chromosome 3 by use of novel congenic strains. Mamm Genome 1995; 6(9):563-570. 69. Chesnut K, Shie J-X, Cheng I, Muralidharan K, Wakeland EK. Characterization of candidate genes for IDD susceptibility from the diabetes-prone NOD mouse strain. Mamm Genome 1993; 4:549-554. 70. Suzuki T, Yamada T, Takao T, et al. Diabetogenic effects of lymphocyte transfusion on the NOD or NOD nude mouse. In: J. Rygaard NB, N. Graem, M. Sprang-Thomsen, ed. Immune-Deficient Animals in Biomedical Research. Basel: Karger, 1987:112-116. 71. Bendelac A, Boitard C, Bedossa P, et al. Adoptive T cell transfer of autoimmune nonobese diabetic mouse diabetes does not require recruitment of host B lymphocytes. J Immunol 1988; 141:2625-2628. 72. Carrasco-Marins E, Shimizu J, Kanagawa O, Unanue E. The class II MHC I-Ag7 molecules from Nonobese Diabetic mice are poor peptide binders. J Immunol 1996; 156:450-458. 73. Leijon K, Hammarström B, Holmberg D. Non-obese diabetic (NOD) mice display enhanced immune responses and prolonged survival of lymphoid cells. Int Immunol 1994; 6:339-345. 74. Hu Y, Nakagawa Y, Purushotham KR, Humphreys-Beher MG. Functional changes in salivary glands of autoimmune disease-prone NOD mice. Am J Physiol 1992; 263:E607-E614. 75. Humphreys-Beher MG, Brinkley L, Purushotham KR, et al. Characterization of antinuclear autoantibodies present in the serum from nonobese diabetic (NOD) mice. Clin Immun Immunopath 1993; 68(3):350-356. 76. Skarstein K, Wahren M, Zaura E, Hattori M, Jonsson R. Characterization of T-cell receptor repertoire and anti-ro/ssa autoantibodies in relation to sialadenitis of NOD mice. Autoimmun 1996; 22(2):9-16. 77. Christianson SW, Shultz LD, Leiter EH. Adoptive transfer of diabetes into immunodeficient NOD-scid/scid mice: relative contributions of CD4+ and CD8+ T lymphocytes from diabetic versus prediabetic NOD.NON-Thy-1a donors. Diabetes 1993; 42:44-55. 78. Kerr M, Lee A, Wang P, et al. Detection of insulin and insulin-like growth factors I and II in saliva and potential synthesis in the salivary glands of mice. Biochemical Pharmacology 1995; 49(10): 1521-1531. 79. Ito S, Suzuki T, Isemura S, et al. ‘Salivary peptide P-C’ of human pancreatic B-cells shares only partly immunoreactivity with salivary

32

80.

81.

82.

83.

84.

85. 86.

87.

88. 89.

90.

91.

92.

NOD Mice and Related Strains: Research Applications in Diseases

peptide P-C indicating a new B-cell protein which is different from insulin. Acta Endocrinol Copenh 1989; 120(1):62-8. Prochazka M, Serreze DV, Worthen SM, Leiter EH. Genetic control of diabetogenesis in NOD/Lt mice: development and analysis of congenic stocks. Diabetes 1989; 38:1446-1455. Wicker LS, DeLarto NH, Pressey A, Peterson LB. Genetic control of diabetes and insulitis in the nonobese diabetic mouse: analysis of the NOD.H-2b and B10.H-2nod strains. In: Alt F, Vogel H, ed. Molecular Mechanisms of Immunological Self-Recognition. New York: Academic Press, 1993:173-181. Li X, Golden J, Faustman DL. Faulty major histocompatability complex class II I-E expression is associated with autoimmunity in diverse strains of mice. Diabetes 1993; 42:1166-1172. Robinson CP, Yamamoto H, Peck AB, Humphreysbeher MG. Genetically programmed development of salivary gland abnormalities in the NOD (nonobese diabetic)-scid mouse in the absence of detectable lymphocytic infiltration: a potential trigger for sialoadenitis of NOD mice. Clin Immunol Immunopathol 1996; 79(1):50-59. Robinson CP, et al. Expression of parotid secretory protein (PSP) in murine lacrimal glands and its possible function as a bacterial binding protein in exocrine secretions. Am J Physiol 1997; 275(Gastrointestinal Liver Physiol 35):G863-G871. Bernard NF, Ertug F, Margolese H. High incidence of throiditis and anti-throid autoantibodies in NOD mice. Diabetes 1991; 41:40-46. Krug J, Williams AJK, Beales PE, Doniach I, Gale EAM, Pozzilli P. Parathyroiditis in the non-obese diabetic mouse-a new finding. J Endocrinol 1991; 131:193-196. Many M, Maniratunga S, Varis I, et al. Two-step development of Hashimoto-like thyroiditis in genetically autoimmune prone nonobese diabetic mice: effects of iodine-induced cell necrosis. Journal of Endocrinology 1995; 147:311-320. Wicker LS, Todd JA, Peterson LB. Genetic control of autoimmune diabetes in the NOD mouse. Ann Rev Immunol 1995; 13:179-200. Doi T, Hattori M, Agodoa LYC, et al. Glomerular lesions in nonobese diabetic mouse: before and after the onset of hyperglycemia. Lab Invest 1990; 63:204-212. Yang CW, Hattori M, Vlassara H, et al. Overexpression of transforming growth-factor-beta-1 messenger-RNA is associated with upregulation of glomerular tenascin and laminin gene-expression in nonobese diabetic mice. J Am S Neph 1995; 5:1610-1617. Leiter EH. Multifactorial control of autoimmune insulin dependent diabetes in NOD mice: Lessons for inflammatory bowel disease. In: Sutherland LR, Collins SM, Martin P, ed. Inflammatory Bowel Diseases: Basic research, clinical implications, and trends in therapy. London: Kluwer Academic Publishers, 1994:24-34. Shimada A, Rohane P, Fathman CG, Charlton B. Pathogenic and protective roles of CD45RB(low) CD4(+) cells correlate with cytokine

NOD Mice and Related Strains: Origins, Husbandry and Biology

93.

94.

95.

96.

97.

98. 99.

100.

101. 102.

103.

104. 105.

106.

33

profiles in the spontaneously autoimmune diabetic mouse. Diabetes 1996; 45(1):71-78. Gerling IC, Freidman H, Greiner DL, Shultz LD, Leiter EH. Multiple low dose streptozotocin-induced diabetes in NOD-scid /scid mice in the absence of functional lymphocytes. Diabetes 1994; 43:433-440. Baxter AG, Horsfall AC, Healey D, et al. Mycobacteria precipitate an SLE-like syndrome in diabetes-prone NOD mice. Immunology 1994; 83(2):227-231. Mahler M, Leiter E, Birkenmeier E, Bristol I, Elson C, Sundberg J. Differential susceptibility of inbred mouse strains to dextran sulfate sodium-induced colitis. Am J Physiol 1997; submitted. Amor S, Baker D, Groome N, Turk JL. Identification of major encephalitogenic epitope of proteolipid protein (residues 56-70) for the induction of experimental allergic encephalomyelitis in Biozzi AB/H and nonobese diabetic mice. J Immunol 1993; 150:5666-5672. Reifsnyder P, Liu O, Anderson N, et al. Genomic characterization of new recombinant congenic stocks between CBA/LsLt and NOD/Lt. Mamm Genome; manuscript in preparation. Baxter AG, Mandel TE. Hemolytic anemia in non-obese diabetic mice. Eur J Immunol 1991; 21:2051-2055. Leiter EH. The NOD mouse meets the “Nerup Hypothesis”. Is diabetogenesis the result of a collection of common alleles present in unfavorable combinations? In: Vardi P, Shafrir E, ed. Frontiers in Diabetes Research: Lessons from Animal Diabetes III. London: SmithGordon, 1990:54-58. Formby B, Schmid-Formby F, Jovanovic L, Peterson CM. The offspring of the female diabetic “Nonobese Diabetic” (NOD) mouse are large for gestational age and have elevated pancreatic insulin content: a new animal model for human diabetic pregnancy. Proc Soc Exp Biol Med 1987; 184:291-294. Tarleton G, Gondos B, Formby B. Testicular alterations in the nonobese diabetic mouse. Endocr Pathol 1990; 1:85-93. Tatewaki R, Otani H, Ando S, et al. Chromosome analysis of postimplantation stage embryos for studying possible causes of developmental abnormalities in nonobese diabetic mice. Biol Neonate 1991; 60:395-402. Otani H, Tanaka O, Tatewaki R, et al. Diabetic environment and genetic predisposition as causes of congenital malformations in NOD mouse embryos. Diabetes 1991; 40:1245-1250. Leiter EH. The genetics of diabetes susceptibility in mice. FASEB J 1989; 3:2231-2241. McAleer MA, Reifsnyder P, Palmer SM, et al. Crosses of NOD mice with the related NON strain: a polygenic model for type I diabetes. Diabetes 1995; 44:1186-1195. Committee on Immunologically Compromised Rodents. In: Immunodeficient Rodents. A Guide to their Immunobiology, Husbandry,

34

107.

108.

109.

110. 111.

112.

113.

114. 115. 116.

117.

118.

119.

120.

NOD Mice and Related Strains: Research Applications in Diseases

and Use. Washington: National Research Council, National Academy Press 1989:103-104. Igel M, Becker W, Herberg L, Joost H-G. Evidence that reduced leptin levels, but not an aberrant sequence of leptin or its receptor, contributre to the obesity syndrome in NON mice. Horm Metab Res 1996; 28:669-673. Tochino Y, Kanaya T, Makino S. Microangiopathy in the spontaneously diabetic nonobese mouse (NOD mouse) with insulitis. In: Abe H, Mitsuru H, ed. Diabetic Microangiopathy. Tokyo: Univ of Tokyo Press, 1983:423-432. Leiter E, Prochazka M, Coleman DL, et al. Genetic factors predisposing to diabetes susceptibility in mice. In: Jaworsk M, Molnar G, Rajotte R, Singh B, ed. The Immunology of Diabetes Mellitus. Amsterdam: Elsevier, 1986:29-36. Pearce RB, Trigler L, Svaasand EK, Peterson CM. Polymorphism in the mouse Tap-1 gene. J Immunol 1993; 151:5338-5347. Sawa T, Ohgaku S, Morioka H, Yano S. Molecular cloning and DNA sequence analysis of preproinsulin genes in the NON mouse, an animal model of human non-obese, non-insulin dependent diabetes mellitus. J Mol Endocrinol 1990; 5:61-67. Pearce R. Clarification of the Ins2 gene sequence: relevance to glucose intolerance in NON/Lt mice. Mammalian Genome 1996; 7:143-144. Prochazka M, Leiter EH, Serreze DV, Coleman DL. Three recessive loci required for insulin-dependent diabetes in NOD mice. Science 1987; 237:286-289. Awata T, Kanazawa Y. Genetic-markers for insulin-dependent diabetes-mellitus in Japanese. Diabet Res Clin Prac 1994; 24(S):S83-S87. Jacob HJ, Pettersson A, Wilson D, et al. Genetic dissection of autoimmune type 1 diabetes in the BB rat. Nature Genet 1992; 2:56-60. Ikegami H, Makino S, Yamato Y, et al. Identification of a new susceptibility locus for insulin dependent diabetes mellitus by ancestral haplotype congenic mapping. J Clin Invest 1995; 96:1936-1942. Prochazka M, Serreze DV, Frankel WN, Leiter EH. NOR/Lt; MHCmatched diabetes-resistant control strain for NOD mice. Diabetes 1992; 41:98-106. Serreze DV, Prochazka M, Reifsnyder PC, et al. Use of recombinant congenic and congenic strains of NOD mice to identify a new insulin dependent diabetes resistance gene. J Exp Med 1994; 180: 1553-1558. Sekiguchi F, Ishibashi K, Katoh H, et al. Genetic profile of alloxaninduced diabetes-susceptible mice (ALS) and resistant mice (ALR). Exp Anim 1990; 39:269-272. Saegusa J, Kiuchi Y, Itoh T. Antinucleolar autoantibody induced in mice by mercuric chloride. Strain differences in susceptibility. Exp Anim 1990; 39:597-599.

NOD Mice and Related Strains: Origins, Husbandry and Biology

35

121. Savino W, Boitard C, Bach J-F, Dardenne M. Studies on the thymus in Nonobese Diabetic Mouse I. Changes in the microenvironmental compartments. Lab Invest 1991; 64:405-417. 122. Saegusa J, Yasuda A, Kubota H. IQI/Jic mice have thymic B cells. Exp Anim 1996; 45:353-360. 123. Serreze D, Gallischan W, Snider D, et al. MHC Class I-mediated antigen presentation and induction of CD8+ cytotoxic T-cell responses in autoimmune diabetes-prone NOD mice. Diabetes 1996; 45:902-908. 124. Wicker LS, Todd JA, Peterson L. Genetic control of autoimmune diabetes in the NOD mouse. Ann Rev Immunol 1995; 13:179-200. 125. Shultz LD, Schweitzer PA, Christianson SW, et al. Multiple defects in innate and adaptive immunological function in NOD/LtSZ-scid mice. J Immunol 1995; 154:180-191. 126. Serreze DV, Leiter EH, Hanson MS, et al. Emv30null NOD-scid mice: an improved host for adoptive transfer of autoimmune diabetes and growth of human lymphohematopoietic cells. Diabetes 1995; 44:1392-1398. 127. Soderstrom I, Bergman ML, Colucci F, Lejon K, Bergqvist I, Holmberg D. Establishment and characterization of RAG-2 deficient non-obese diabetic mice. Scand J Immunol 1996; 43:525-30. 128. Serreze DV, Chapman HD, Gerling IC, et al. Initiation of autoimmune diabetes in NOD/Lt mice is MHC class I-dependent. J Immunol 1997; 158:3978-3986. 129. Wicker LS, Leiter EH, Todd JA, et al. b2 microglobulin-deficient NOD mice do not develop insulitis or diabetes. Diabetes 1994; 43:500-504. 130. Serreze DV, Chapman HD, Varnum DS, et al. B lymphocytes are essential for the initiation of T cell-mediated autoimmune diabetes: analysis of a new ‘‘speed congenic’’ stock of NOD.Igmnull mice. J Exp Med 1996; 184(5):2049-2053.

CHAPTER 2

Genetics and Immunogenetics of NOD Mice and Related Strains Edward H. Leiter

INTRODUCTON

S

ince the NOD mouse is distinguished from mice of other inbred strains in terms of its unique predisposition to develop autoimmune, insulin dependent diabetes mellitus (IDDM), genetic analysis has tended to focus upon genes controlling development of aberrant autoimmune responses. Indeed, most of the chapters in this volume attest to the extensive knowledge gained about the immune system of NOD mice. Yet the NOD mouse is of considerable value to mouse and human geneticists interested in mapping complex traits that may be totally unrelated to autoimmune disease in general and IDDM in particular. In most cases, a newly-developed inbred strain is of less utility to a geneticist than long-standing inbred strains because of the lack of genetic characterization of the former. However, because of its importance to autoimmunity research, the NOD strain, or more precisely DNA from NOD mice, represents one of the “select few” 11 strains included in the Massachusetts Institute of Technology (MIT) Genome Center’s survey of over 6,000 PCR-typed and geneticallymapped simple sequence repeat (SSR) polymorphisms. Hence, NOD and its related NON strain, despite their recent origins, represent two of the most genetically well-characterized inbred strains. The SSR

NOD Mice and Related Strains: Research Applications in Diabetes, AIDS, Cancer and Other Diseases, edited by Edward Leiter and Mark Atkinson. © 1998 R.G. Landes Company.

38

NOD Mice and Related Strains: Research Applications in Diseases

markers, distributed throughout the genome, cover most of the 20 mouse chromosomes with the exception of the Y chromosome at a density of ~3cM. Commerically-available PCR primer sets are available for each of these marker loci. Strain-specific polymorphic allele sizes are available on-line (www.mit.genome.edu). SSR chromosomal map positions are integrated into the genetic map of the mouse. Information for a given SSR-based polymorphic locus is available online (www.informatics.jax.org or www.resgen.org), and its map position is constantly being updated as new linkage data become available for identified structural genes mapped to the same chromosomal region. Commercially-available yeast artificial chromosome (YAC) contig libraries based upon these SSR markers permit physical ordering of DNA sequences found within polymorphic regions. NOD mice differ at approximately 50% of the >6,000 polymorphic SSR-containing loci when compared to other commonly-used, non-Swiss derived inbred mouse strains such as C57BL/6J (B6), C3H/HeJ (C3H) and CBA/J. The simple-to-perform PCR technology allows the investigator to select a defined NOD set of polymorphisms distributed over the 20 chromosomes, and follow their segregation through multiple outcross/intercross or outcross/backcross cycles (by PCR-typing of DNA that can be easily prepared from peripheral blood). This technique is known as a genome wide scan. DNAs from progeny are scanned for distortions in the frequency of polymorphic parental alleles between affected and unaffected progeny in the case of a discontinuous trait such as IDDM, or for linkage to a high or a low response in a continously varying quantitative trait. As discussed in the previous chapter, NOD mice breed relatively early; dams produce uncommonly large litters for an inbred strain, and are excellent mothers, nursing nearly all of their pups to weaning. The developmental biologist should note that the excellent maternal instincts of NOD females makes the pseudopregnant NOD female an excellent host for blastocyst transfer when chimeric mice are being produced (Dr. Robyn Slattery, John Curtin School of Medicine, Canberra, Australia, personal communication). This excellent breeding performance, combined with the increasing numbers of congenic and transgenic stocks, as well as of defined gene “knock-outs” being developed on the NOD inbred strain background, is proving especially useful for establishing genetic linkages for multigenically controlled traits. If the mouse geneticist is trying to map the gene con-

Genetics and Immunogenetics of NOD Mice and Related Strains

39

trolling a monogenic trait, he immediately thinks of initiating an interspecific outcross to maximize genetic polymorphisms between the strain susceptible to the disorder and a resistant strain. However, as will be described below, for complex multifactorial diseases like IDDM or inflammatory bowel disease, this strategem is often counterproductive because the trait is not detectable or detectable only at very low frequency in either an F2 or first or second backcross (BC1, BC2) generation. If intraspecific crosses are performed instead, the frequency of trait expression can be expected to be higher. To illustrate, in the previous chapter, it was mentioned that NOD mice develop a progressive hearing loss that is multigenically controlled whereas CBA/J mice do not. Following an interspecific outcross between NOD and a wild-derived inbred strain, such as SPRET/Ei or CAST/Ei (derived from Mus spretus and Mus castaneus, respectively), there is a high probability that the frequency of the deafness trait may be so low that very large numbers of F2 mice would have to be produced, or multiple backcrosses would have to be performed before the penetrance of the deafness phenotype made the genetic analysis feasible. In contrast, if NOD were outcrossed to CBA, a higher penetrance of the disease might be anticipated. Further, recombinant congenic stocks in which defined NOD chromosomal regions are fixed on a predominantly CBA genetic background exist. Although a geneticist might eschew the use of the NOD genome for mapping purposes based upon fear that mice might be lost from a study through development of IDDM, this is an easily obviated problem. As detailed in the preceding chapter, MHC congenic stocks of NOD mice can be used that do not develop IDDM. Thus, the remarkable breeding performance of this inbred strain, coupled with its well-characterized genome, should be kept in mind for use in the genetic dissection of complex traits such as obesity, cardiovascular disease, and inflammatory bowel disease.

NOD AND THE GENETICS OF AUTOIMMUNITY THE GENETIC BASIS OF IDDM The fact that NOD mice and BB/Wor-DP rats develop spontaneous IDDM as a result of selective breeding confirms the primary role of genetics in establishing disease susceptibility in these rodent models. However, not all NOD mice or BB-DP rats will develop IDDM if the intrinsic and extrinsic environment is not optimum. Because of

40

NOD Mice and Related Strains: Research Applications in Diseases

this complex, multifactorial control, it is most accurate to refer to NOD mice simply by the abbreviation “NOD”, and not as “nonobese diabetic”, since, in most vivaria, not all NOD mice develop clinical features of diabetes. In humans, comparison of disease concordance frequency between monozygotic versus dizygotic twins is a classic way to establish the relative contributions of heredity versus environment. In the case of IDDM in humans, the higher concordance rates for monozygotic twins (~30-50%) versus dizygotic twins (~5%) clearly establishes a primary etiopathologic contribution for disease-predisposing genetic factors. However, since concordance rates for IDDM in monozygotic twins rarely exceeds 50%, either random somatic genetic changes are important in development of a fully susceptible individual (e.g., in the constitution of T cell receptor and immunoglobulin repertoires), or the penetrance of the diabetogenic genes is strongly influenced by environment, or both. Further complicating elucidation of the genetic contributions to development of a complex, multifactorial disease represented by IDDM in humans is the fact that the disease is genetically quite heterogeneous. Only about 20% of new cases of IDDM occur in families in which another family member is affected; the remainder are “sporadic”, suggesting that new combinations of potentially diabetogenic collections of genes are constantly being generated in a randomly breeding human population. Disease heterogeneity is reflected by the finding that not all cases of IDDM have an autoimmune etiology. Even within the demonstrated cases of autoimmune IDDM, where genes within the major histocompatibility (MHC) human leukocyte antigen (HLA) complex on Chromosome 6p21 are clearly major determinants of susceptibility, heterogeneity in terms of the predisposing HLA alleles, as well as in the non-HLA genetic linkages is encountered within and between racial and ethnic groups.1 Given this background of genetic complexity in humans, the NOD mouse must be viewed as a single case study because an inbred mouse essentially represents a single individual reproduced in multiple copies. Same sex individuals can exchange organ and tissue grafts without rejection. However, within the strain, males and females will be distinguished by gender-dimorphic patterns of gene expression, including maternal versus paternal imprinting of genes, as well as by sex chromosome-controlled differences. One of the latter is the socalled H-Y antigen, which precludes the transfer of NOD male tissues

Genetics and Immunogenetics of NOD Mice and Related Strains

41

into female recipients.2 Gender-independent variations between individuals within the strain will also arise through stochastic somatic processes associated with immune repertoire development, including rearrangement of immunoglobulin and T cell receptor genes, as well as differential amplification of the adaptive immune system in response to variable exposure to environmental antigens. Finally, as discussed in the previous chapter, random mutation in an individual may become fixed in a breeding colony if it improves fitness, reproductive performance, etc., and such genetic shifts will eventually lead to substrain divergence.3 WHAT MAKES THE NOD GENOME UNIQUELY PERMISSIVE FOR SPONTANEOUS IDDM DEVELOPMENT? The reader of this book should have no difficulty finding current reviews describing the latest advances in knowledge regarding the immunogenetic basis of IDDM susceptibility in NOD mice. The following reviews are recommended at the time of this writing.4-7 All workers in the field agree that the unique H2g7 MHC haplotype of NOD mice is the predominant contributor of IDDM susceptibility in NOD mice. Depending upon the strain paired with NOD in genetic segregation analysis, this complex locus is estimated to contribute upwards of 40% of the relative risk for diabetes development.8 This is fully consistent with what is known about the genetics of autoimmune IDDM in humans, wherein diabetes-predisposing HLA alleles also are the strongest known determinants of genetic susceptibility to IDDM.9 Before the specific details of the diabetogenic MHC and nonMHC loci are summarized, we will consider the more general question of why the NOD strain, among all the extant mouse strains, is the only one to develop IDDM spontaneously. As described in the preceding chapter, the ICR-derived ILI strain shares the same diabetogenic H2g7 haplotype as NOD, and the CTS strain shares class II MHC alleles, yet neither ILI nor CTS mice nor other strains of mice congenic for H2g7 develop IDDM.10 Two possibilities have been debated. The first is that the NOD strain has accumulated a set of rare mutations in both MHC and nonMHC genes that, collectively, set this strain apart from virtually all other extant strains. A listing of some of the non-MHC genetic loci reported to be rare or defective in NOD is provided in Table 2.1. The second possibility, termed the “Nerup hypothesis”11 after Dr. Jørn

42

NOD Mice and Related Strains: Research Applications in Diseases

Table 2.1. Important Genetic Polymorphisms Distinguishing NOD Mice Gene (chromosome)

Phenotype

Fcγr2 (low affinity Fc receptor gamma II, Chr 1)

multiple mutations in promoter regions led to low expression in macrophage, impaired binding of IgG1 and IgG2b.

Ref. 98

null mutation (Hc0 allele) in the gene so that NOD 99 lacks the C5a component such that NOD serum does not support complement dependent lysis. Fcγr1 (high affinity multiple mutations in both cytoplasmic and 80,10 Fc receptor gamma 1, extracellular domains affect both surface expression Chr 3) and ligand binding, as well as impaired clearance of IgG2a. Il2 (interleukin-2, NOD expresses the Il2b allele, the product of which 55, Chr 3) may be less potent (due to glycosylation) than the 101 Il2a allele product. Slc1a1 (formerly Nhe1 ubiquitously expressed membrane ion exchanger 56 solute carrier family 9, maintaining pH homeostasis; NOD has low activity sodium/hydrogen allelle. exchanger, isoform 1, Chr 4) Nk1 (natural killer cell NOD mice fail to develop functional NK1+ T cells 102, antigen-1, Chr 6) and both NOD and NOD-scid mice lack functional 103 NK killing activity. NOD mice exhibit rare poly– morphisms in the NK1-linked mouse homolog to the human gene, NKR-P1, encoding an NK signal transduction molecule (Leiter lab., unpublished) Art2a, 2b (ADP-riboNOD has rare restriction fragment length 104, syltransferase, polymorphism, somewhat reduced expression 105 formerly Rt6, Chr 7) levels in spleens of young but not old mice. Mrv6 (Mouse MAIDS defective endogenous proviral genome; no virus-related proviral evidence for expression in NOD thymus or islets. 76 gene), Chr 14 Ly6c (lymphocyte impaired expression on bone marrow, splenocytes, 106 antigen 6C, Chr 15) and lymph node cells correlates with mutations in the 5” flanking region. H2g7 complex, Chr 17 H2-K common allele, but acquires diabetogenic function 31,33 with rest of haplotype Lmp2 same allele as H2r,q, 109 Tap1 unusual RFLV in intron; tryptophan replaces 23,110 cysteine at residue 174 of cDNA Lmp7 same allele as H2r,cas4, 109 g7 H2-A rare amino acid substitutions in Ab chain 16 H2-E nonsense mutation in Ea chain, no surface expression 14

Hc (hemolytic complement, Chr 2)

Genetics and Immunogenetics of NOD Mice and Related Strains

43

Gene (chromosome)

Phenotype

Hsp70 Tnfa H2-D

unusual RFLV suggest novel allele 22 reduced transcription, secretion in some studies 107 common allele, but acquires diabetogenic function 31-33 with rest of haplotype

Qa region Xmv66 Ptgs2 (prostaglandin synthase 2, Chr 1) Reg (rat regenerating, islet derived, mouse homolog 1, 2, Chr 3 & 12)

Ref.

deletions of some Qa family genes indicated novel 108 xenotropic proviral genome distal to H2-K 8 cytokine-inducible cyclooxygenase with anomalous 65,66 high constitutive expression in NOD macrophage. high levels of expression in prediabetic and diabetic 112 NOD pancreas.

Nerup of Denmark, who first raised the issue for debate,12 is that the NOD genome does not contain diabetes-specific genes, but rather, by chance, has accumulated unfavorable combinations of relatively common alleles. Accumulated evidence at this point suggests that the NOD strain is both a collection of some relatively rare diabetogenic alleles working disharmoniously with a larger collection of common alleles.

THE H2g7 HAPLOTYPE: A SCAFFOLDING FOR INITIATION OF INSULITIS AND DIABETES MHC class I and class II molecules expressed on antigen presenting cells (APC) select the repertoire of T cells intrathymically, and regulate their functions in the periphery such that, in normal mice, Tc ell autoreactivity against cells in glandular tissues is prevented.13 Genetic segregation analysis initially identified the H2 region as containing a gene or genes controlling susceptibility to IDDM in NOD mice.14,15 A provisional nomenclature was proposed wherein chromosomal regions containing such genes were sequentially identified as Idd loci.15 The region linked to H2 on Chromosome 17 was provisionally designated Idd1. A similar provisional nomenclature has since been adopted for human IDDM susceptibility conferring loci, with the homologous HLA region designated as IDDM1.1 MHC haplotype designations in mice are based in part upon allele typing at the H2K and H2D markers at the proximal and distal ends of the MHC complex. H2g denotes a recombinant haplotype containing a H2Kd and a H2Db allele. The H2g7 haplotype assigned to NOD mice encompasses additional unique and/or rare alleles within the class II and class III region of the complex. The notation I-Ag7 denotes a class II

44

NOD Mice and Related Strains: Research Applications in Diseases

molecule encoded by a common Aad allele and a very rare Abg7 allele in which a nested set of five nucleotide substitutions between position 248-252 converts a conserved proline residue at amino acid position 56 to histidine and a common aspartic acid residue at position 57 to serine.16 I-Ag7 is present in the related CTS and ILI strains, and has been found in randomly screened outbred ICR mice17 as well as in the Biozzi high antibody responder strain.18 The Abg7 of NOD is therefore homologous to “diabetogenic” HLA-DQβ non-aspartic acid57 containing alleles in humans.19 MHC class II molecules (generally only expressed on APC) are essential for the initiation and maintenance of both cell-mediated and humoral immune responses. The Abg7 amino acid substitutions are located on parts of the molecule that affect the binding of peptides to the antigen-binding cleft, with the consequence that the I-Ag7 peptide binding cleft is able to selectively bind and present peptides with acidic residues at their C-terminus.20 A second “diabetogenic” mutation is represented by the Eab allele in the H2g7 complex (homologous to the human HLA-DR region). This allele is relatively common in inbred strains such as B6 that fail to express cell surface IE molecules due to a deletion within the first exon of the Ea gene that prevents transcription.14 The Eb allele in H2g7 is Ebd 21 and is expressed normally, but in the absence of Eα chains, no I-E heterodimers form at the cell surfaces of APC. The class III region distal to the Ebd locus contains a unique heat shock protein allele (Hsp70).22 Proximal to the Ab locus, the H2g7 haplotype also contains rare alleles at Tap1 and Tap2 (for transporters associated with antigen processing, and previously designated Ham1 and Ham2).23 Confirmation of the diabetogenic contributions of the unique I-Ag7 molecule in the absence of cell surface I-E molecules was provided by transgenic experiments (reviewed by Slattery and Miller24 and by Cooke et al25). In brief, these studies confirmed that both the proline to histidine substitution at residue 56, and the aspartic acid to serine substitution at position 57 of the Aβ chain contribute independently to IDDM susceptibility. Similarly, insertion of Ea d transgenes leading to I-E cell surface expression also protected against insulitis and IDDM development. Protection was apparently not mediated by negative selection within the thymus of diabetogenic effectors.26,27 Rather, alterations in peptide presentation by APC are indicated by the finding that protection is associated with a shift in the cytokine profile toward reduced T cell production of interferon gamma

Genetics and Immunogenetics of NOD Mice and Related Strains

45

and increased production of cytokines associated with T-helper 2 functions.26,28 This deviation in T cell functional activities apparently leads to peripheral suppression of autoimmune effectors. The I-A g7 heterodimer is reportedly unstable on cell surfaces of NOD APC; transgenic expression of I-E molecules does not correct this anomaly.29 High levels of I-E expression on NOD APC result in lower levels of T cell proliferative responses in a recall assay in vitro following priming in vivo with putative β cell autoantigens such as glutamic acid decarboxylase.26 These I-E effects do not seem to reflect competition with I-Ag7 for binding of diabetogenic peptides, but rather, may reflect further quantitative reductions in the absolute numbers of I-Ag7 molecules expressed on APC.26 The respective diabetogenic contributions of the H2g7 class II region genes partially answer the question posed at the beginning of this section. The two diabetogenic mutations in Abg7 are uncommon whereas the “diabetogenic” mutation in the Eab gene is relatively common. Hence, genetic predisposition to IDDM in NOD mice indeed is comprised in one instance by the presence of a common null mutation (Eab) that is not inherently diabetogenic. However, when an expressed allele is inserted by transgenesis, the I-E region is typed as an Idd locus because the I-E molecules epistatically suppress the diabetogenic contributions of the I-Ag7 locus. Epistasis is a form of gene interaction wherein one gene interferes with or otherwise modifies the phenotypic influence of another nonalleleic gene in the same individual. THE USEFULNESS OF CONGENIC STOCKS The classic technique in mouse genetics for demonstrating the role of a specific genetic locus in the control of a given phenotype was pioneered by Nobel laureate Dr. George Snell at The Jackson Laboratory in the 1950s. Snell investigated the role of H2 (then thought to be a single gene called Histocompatibility-2!) from a donor strain onto a recipient strain. After eight backcrosses, approximately 99.2% of the total genome (comprising ~100,000 expressed genes) is recipient-derived, with the major concentration of donor genes being linked together on the specifically introgressed chromosomal segment. This classic technique provided further confirmation of the primary role of MHC in controlling beta cell autoimmunity in NOD mice. Replacement of the diabetogenic H2g7 haplotype with a protective MHC haplotype (H2b, H2q or H2nb1) completely eliminated islet-invasive

46

NOD Mice and Related Strains: Research Applications in Diseases

insulitis and spontaneous IDDM development in NOD mice.4,5 This fact, coupled with evidence (described below) that no single non-MHC locus has a comparable effect, firmly establishes this complex locus as the major component of susceptibility. Although immunogenetic analysis has concentrated on the diabetogenic contributions of the MHC class II region of the H2g7 haplotype, current evidence suggests that the haplotype as a whole must be considered as contributing to susceptibility. The most compelling evidence comes from the congenic transfer of the unique MHC haplotype of the related CTS/Shi strain onto the NOD/Shi genetic background. The MHC of CTS (H2ct) mice apparently contains the same class II alleles as NOD, but distinct class I loci, indicating that loci between these markers may differ as well. When this CTS haplotype was transferred onto the NOD inbred background and compared in homozygous state to segregants homozygous for the H2g7 haplotype, a lower incidence of diabetes and insulitis was observed in the H2ct homozygous mice than in segregants homozygous for H2g7.30 The reduced diabetogenic potency of the H2ct thus provides strong support for the concept that, while the class II region is clearly important to disease development, other loci within the extended H2g7 haplotype also contribute. One of these has been dubbed Idd16, although the position of the locus in the complex was not established.31 Even though the MHC class I alleles of NOD mice (H2Kd, H2Db) are common in non-autoimmune prone strains, they acquire in NOD mice a diabetogenic function—the selection and targeting of CD8+ T cells essential for initiation of the diabetogenic process.32 Thus, the class I alleles also contribute to the overall susceptibility represented by the extended H2g7 haplotype. The NOD allele at the β2-microglobulin locus (B2ma) is an excellent illustration of the complexity of diabetes genetics in this mouse. It is not uncommon, nor is it defective in NOD mice. However, current unpublished evidence indicates that it may be one of several genes within the Idd13 locus on Chromosome 2 that promotes diabetogenesis by affecting the conformation of MHC class I molecules.33 If the role of B2ma as a diabetogenic contributor is confirmed by transgenic insertion of B2ma and B2mb alleles separately into a congenic stock of NOD/Lt-B2mnull mice, then this will be a confirmation of the “Nerup hypothesis” that common alleles can acquire diabetogenic functions in certain unfortunate combination with other genes.

Genetics and Immunogenetics of NOD Mice and Related Strains

47

RARE ALLELES DO NOT NECESSARILY EQUATE WITH DIABETOGENIC ALLELES: THE TAP GENE IMBROGLIO If common alleles such as H2Kd, H2Db, Eab and B2ma can subserve a diabetogenic function when combined with rare diabetogenic genes such as Abg7, the inverse also holds true—rare alleles do not necessarily equate with diabetogenic genes. The intra-MHC Tap genes in H2g7 provide an object lesson in this regard. It was noted previously that both the Tap1 and Tap2 alleles in H2g7 were uncommon based upon restriction fragment length polymorphisms.23 The products of these loci are members of a superfamily of ATP-dependent transport proteins. They form heterodimers that transport processed antigenic peptide fragments generated in endosomal compartments into the lumen of the endoplasmic reticulum. These processed peptides are then complexed with MHC class I molecules and are subsequently translocated to the cell surface for presentation to CD8+ T cells.34 Many mutant mouse and human cell lines lacking the ability to form stable MHC class I-peptide complexes carry mutations which map to regions encoding Tap1 or homologous genes.34 Accordingly, if comparable mutations existed in the NOD’s Tap loci that impaired antigen presentation, they too could represent Idd loci. Both the author’s laboratory23 and another laboratory35 observed and reported the same unique polymorphism in a non-coding region of the NOD Tap1 gene. However, subsequent analysis of NOD’s Tap1 and Tap2 coding regions showed completely normal sequences.36 Several groups have shown normal levels of Tap mRNA expression in NOD splenic leukocytes.23,36,37 The Tap gene products of the H2g7 haplotype did not differ from those encoded within four other MHC haplotypes in affinity for ATP, kinetics of peptide uptake, and substrate specificity.38 These facts notwithstanding, Faustman et al35,39 have asserted that presumed Tap1 gene defects likely explain the diabetogenic contribution of MHC in mice and humans. These workers had difficulty in observing Tap1 transcripts from NOD spleen by Northern analysis,35,39 and further reported that splenocytes from the NOD mice in their colony were expressing diminished constitutive levels of cell surface MHC class I molecules. These diminished class I levels were inferred to be the consequence of a null mutation in the Tap1 gene. It was further proposed that the low class I expression was the basis for failure of NOD mice to develop immune tolerance to endogenous antigens, and thus represented the molecular basis for the MHC-associated susceptibility.

48

NOD Mice and Related Strains: Research Applications in Diseases

This latter premise was buttressed by the observation of glucose intolerance in association with low grade islet-associated leukocytic infiltrates in older 129↔C57BL/6 chimeric mice rendered MHC class I deficient by a “knockout” of the β2-microglobulin (B2m ) gene.35 In a subsequent study, the Tap gene function and MHC class I mediated antigen presentation was reported by this group to be impaired only in diabetes-prone NOD females, but not in more resistant males.40 This claim of low class I expression has been widely refuted by numerous laboratories.20,36,41,42 Since the T-lymphoaccumulation described in the preceding chapter produces a reduction in the ratio of splenic B cells to T cells, the perceived underexpression of class I molecules in fact represented this strain-characteristic reduction in the numbers of splenic B cells (with larger cell volumes). Reduction in Bc ell numbers is not inherently diabetogenic; on the contrary, NOD mice with a targeted mutation in the immunoglobulin heavy chain (Igh6tm1Cgm) gene do not develop B cells, and are insulitis and diabetes-resistant as a consequence of the loss of this population of APC.43 Further, four different laboratories44-47 have analyzed independentlyproduced stocks of NOD mice demonstrating MHC class I- and CD8+ T cell-deficiency following congenic introduction of a disrupted B2m gene. These congenic NOD.B2mnull stocks were all insulitis- and IDDM-resistant, firmly refuting the contention by Faustman et al35 that T cells become inherently diabetogenic when maturing in a MHC class I-deficient environment. A defect in IFNγ-upregulation of MHC class I has been reported in NOD/Lt peritoneal macrophages.42 However, this is a Tap gene-independent trans-effect entailing defective signaling via transcriptional activators, as demonstrated by the finding that the H2g7-identical NOR/Lt strain, with the same Tap1 sequence as NOD, shows normal IFNγ regulation of MHC class I.42 Finally, another assertion made by the Faustman group,48 that constitutive levels of MHC class I expression are diminished on peripheral blood leukocytes (PBL) from human IDDM patients, has also failed to be confirmed.49 This discussion should convince the reader that caution should be exercised to avoid over-interpreting the clinical significance of rare genomic polymorphisms. Tap gene polymorphisms are certainly useful to human geneticists. Although primary associations between human TAP2 (and not TAP1) allelic variants and IDDM have been suggested,50 these appear to result from linkage disequilibrium with diabetogenic class II alleles.51,52 Availability of Tap polymorphisms, however, helps to better identify genetic subtypes of IDDM in humans.53

Genetics and Immunogenetics of NOD Mice and Related Strains

49

GENETIC ANALYSIS OF INSULITIS: IDDM SUSCEPTIBILITY INHERITED AS A THRESHOLD LIABILITY As described above, analysis of transgenic and congenic stocks of NOD mice in which the H2g7 haplotype has been altered or replaced has shown that multiple Ioci within or tightly-linked to this MHC are essential for insulitis development. NOR/Lt, a closely-related recombinant congenic stock sharing approximately 88% of its genome with NOD (including H2g7) perfectly illustrates the requirement for an interaction between MHC and non-MHC ldd loci. NOR/Lt mice are resistant to development of destructive insulitis, even though T-lymphoaccumulation in the pancreas still occurs.54 Genetic modeling based upon the large numbers of Idd loci identified by segregation analysis8,54-56 indicates that a polygenic threshold liability (or multiplicative) model best fits the experimental results.8,57 This is also the case for polygenic, spontaneously developing lupus-like syndromes in mice.58 The alternative model would be a genetic heterogeneity model in which widely variable combinations of the Idd loci would be capable of triggering insulitis sufficiently aggressive to mediate clinical symptoms. Penetrance of the phenotype of clinical IDDM requires destruction of approximately 90% of the β cell mass in the pancreas. Most NOD mice of both sexes will develop an islet-invasive insulitis as they age, yet not all develop overt clinical symptoms of diabetes. The NOD/Wehi substrain is an excellent example of this phenomenon.59 A genetic program for development of autoimmune IDDM is contained in hematopoietic stem cells of NOD mice.60,61 As discussed in chapter 6 by M.A. Atkinson, environmental factors strongly influence the extent to which this program is expressed. Early genetic analysis in Japan and the United States (reviewed in Kikutani and Makino10) showed a much higher frequency of insulitis than of clinical IDDM in backcross progeny following outcross of NOD to other inbred strains, suggesting that only a few of the diabetogenic NOD genes controlled development of islet-destructive insulitis. Insulitis is properly assessed not as a discontinuous trait (+ or -), but rather as a continuously varying trait with phenotypes scored between a range of “benign” peri-insulitis to fully-destructive intra-islet insulitis. Perception of the number of “insulitis” genes segregating in a given cross is a function of the partner strain used in an intraspecific outcross.61 When interspecific outcross with Mus spretus is employed, the genetic control of insulitis is seen to be considerably more complex than when

50

NOD Mice and Related Strains: Research Applications in Diseases

intraspecific crosses are utilized.62 This serves to emphasize that an inbred partner strain, especially if it is Swiss-derived, may share with NOD a number of susceptibility genes not shared by the wild-derived strain. As will be discussed further below, the partner strain may also contribute diabetes acceleration genes in addition to resistance genes. The susceptibility genes, regardless of whether they originate from NOD or from the partner strain used in outcross, are assumed to express at the level of the immune system rather than in pancreatic β cells. For example, NOD alleles of genes controlling the rate and level of T cell apoptosis have been implicated,63 as have genes controlling the biochemical functions of peri-insular leukocytes.64-66 If certain of these were estrous cycle dependent, as is apparently the case for one candidate gene, prostaglandin E2 synthase-2 (Ptgs2) on Chromosome 1 (see chapter 4 by M. Clare-Salzler), a mechanism for the gender-dimorphic destructiveness of the insulitic process in females versus males would be provided.

GENETIC SEGREGATION ANALYSIS AS A FIRST STEP IN IDENTIFICATION OF NON-MHC IDD LOCI Outcross of NOD to mice of various other inbred strains have identified over 16 loci on 13 different chromosomes that influence the diabetogenic process in hybrids.67 By definition, any locus which affects susceptibility/resistance to IDDM development in NOD mice is an Idd locus. By this broad definition, any mutant locus (spontaneous or targeted disruption by homologous recombination) introduced into the NOD genome that prevents or accelerates IDDM development is an Idd locus. By this definition, there are well over 100 loci that require the presence of the undisturbed NOD allele to promote IDDM development. These would include spontaneous or targeted mutations at the scid, Rag, nu, Igh, and B2m loci since mutations affecting function of all of these loci are capable of preventing IDDM in NOD mice. H2g7 exerts a codominant permissive effect for insulitis development against which the non-MHC loci are identified.60,68 Widespread, destructive patterns of insulitis do not develop in F2 segregants which have inherited zero copies of H2g7, while it is prevalent in both H2g7 heterozygous and homozygous segregants. Similarly, with few exceptions (reviewed by Wicker et al,4 e.g., NOD.H2h2, and NOD.H2h4 as well as NOD.H2ct 30) replacement of H2g7 by other MHC haplotypes

Genetics and Immunogenetics of NOD Mice and Related Strains

51

suppresses development of destructive insulitis. Given the predominant role of H2g7 in setting the basal “threshold” liability, the most informative segregation analyses of non-MHC Idd loci have employed outcross of NOD with H2g7-congenic stocks of B10, B6 or NON.8,55,69 Segregation analysis has primarily concentrated on genome wide scan of diabetic versus non-diabetic progeny in either a first backcross (BC1) or F2 generation. A contingency Chi-square analysis is performed to identify significant deviations between expected and observed frequencies of diabetic versus non-diabetic segregants at all typed genetic markers. A first backcross generation offers the advantage that, if many genes are segregating, a sufficient number of NOD susceptibility genes will be present in homozygous state to permit a sufficiently high spontaneous IDDM incidence necessary for statistical comparison to non-diabetic segregants. However, NOD loci whose contributions to susceptibility are strongly dominant cannot be assessed in a BC1 generation since no significant difference in frequency would be expected between homozygous versus heterozygous BC1 segregants. To assess dominant contributions of a diabetogenic Idd locus from either parental strain, an F2 analysis is required since comparisons between heterozygotes versus both parental homozygotes can be made. As will be noted by examining the listing of currently-identified chromosomal regions carrying Idd loci, both BC1 and F2 analysis has uncovered the fact that the diabetes-resistant parental genome also contributes susceptibility loci to hybrid progeny (Table 2.2). Genes effecting selection of the T cell repertoire would be excellent candidates for Idd genes. Notable in this regard are the products produced by endogenous mammary tumor virus (Mtv) proviral loci. These retroviral loci encode proteins that mimic superantigens by interacting with the Vβ portion of the T cell receptor, effecting deletions of specific Vβ clonotypes.70 The NOD/Lt genome carries Mtv3 on Chromosome 11,71 Mtv17 on Chromosome 471 and Mtv31 on the Y chromosome.72 The NOD/Crc substrain is segregating for a new Mtv, Mtv45, also on proximal Chromosome 11.73 However, genetic segregation analysis using either NON/Lt or B6 or B10 mice as outcross partner strains has not yet implicated any of these Mtv loci. The effect of a superantigen-like factor controlling transient shifts in the Vβ8.3 CD8 T cell subset has been proposed as the basis for the differential female:male susceptibility.74 However, as described in chapter 3 by D.V. Serreze, this subset is not required for diabetogenesis. The

52

NOD Mice and Related Strains: Research Applications in Diseases

NOD genome contains a wide variety of other classes of proviral elements (summarized by Gaskins et al75). A MAIDS (mouse AIDS)-like proviral genome (Mrv6) has been mapped to Chromosome 14, but not in the region associated with the Idd8 locus.76 NOD mice express the Fv1n (N-tropic) allele on Chromosome 4 controlling susceptibility to infection by naturally occuring ecotropic leukemia viruses.

SPEED CONGENICS Another form of segregation analysis is accomplished through construction of a “speed congenic” stock. This entails outcrossing NOD with the strain thought to carry a gene controlling IDDM development (potentially a congenic chromosomal segment from a diabetes-resistant strain or stock, or a genetically disrupted gene such as B2m or the immunoglobulin heavy chain gene (Igµ). The F1 hybrid is backcrossed to NOD, but before the second backcross is initiated, BC1 individuals are typed for the known NOD-derived chromosomal segments presumed to carry Idd loci (listed in Table 2.2). Individuals are then selected for the next backcross that are homozygous for the greatest number of NOD-derived Idd loci, while heterozygous for allogeneic markers demarcating the introgressing congenic segment or for a marker homologously recombined into a gene as part of a gene targeting strategy (e.g., a neomycin resistance gene). In this situation, the assumption is that the introgressed resistance allele carried through mutliple backcrosses in heterozygous state will be protective when intercrossed to generate segregants homozygous for the locus. Generally, at least nine cycles of backcrossing following outcross (N10) are required to eliminate residual heterozygosity at markers on nonselected chromosomes, and 20 generations of backcrossing are desirable. However, by employing the speed congenic method, a high incidence of IDDM can be demonstrated after only six cycles of backcrossing (in N7F1 segregants). The intercross generates segregants homozygous or heterozygous for the resistance (or disrupted) allele, as well as homozygous for the NOD susceptibility (or non-disrupted) allele. As an example, N7 segregants homozygous or heterozygous for the NOD allele at the Igµ locus exhibit high incidences of IDDM, confirming that sufficient numbers of NOD diabetogenic genes have been accumulated in the stock to reconstitute the susceptible genotype. This is the necessary finding if the effect of the “knock-out” is to be assessed. In the example cited, the complete absence of B cells ef-

Genetics and Immunogenetics of NOD Mice and Related Strains

53

fected by the Igµ “knock-out” completely suppressed diabetogenesis.43 Particular caution needs to be taken when introgressing a disrupted alllele (usually derived from the 129/Sv or 129/Ola genome providing the embryonal stem (ES) cells used in the targeting procedure). This introgression process not only brings in the targeted gene mutation, but also a considerable number of strain 129 alleles flanking the “knock-out” locus. These linked genes could modify IDDM susceptibility independently of any effect of the disrupted gene to which they are linked. A striking example of the latter was the erroneous conclusion that a disrupted T cell receptor (TCR)alpha gene introgressed into NOD mice blocked development of autoimmunity, when in fact, protection was derived from another strain 129-derived resistance allele on Chr 14, possibly at the Idd 8 or Idd12 locus.77 The only way to control for this is to introgress in parallel with the “knock-out” gene a congenic segment from the 129 genome carrying the wildtype allele.

CONGENIC AND SUBCONGENIC STOCKS INDICATE THE PRESENCE OF MULTIPLE NON-MHC SUSCEPTIBILITY LOCI ON NUMEROUS CHROMOSOMES The congenic or speed congenic method represents the most utilized method to date not only for confirming the existence and strength of an Idd locus or a candidate gene, but for defining candidate gene or locus function in the control of immunophenotypes aberrantly regulated in standard NOD mice. Introgression into the NOD genome of a congenic segment conferring resistance is the more common approach, but the reciprocal may also be done (introgressing the NOD allele onto a resistant background). When the latter approach is taken, a bicongenic stock containing either H2g7 plus a single nonMHC Idd region, or else mutliple Idd genes on a single non-MHC region usually needs to be produced to see a shift in insulitis level.78 Congenic mapping allows more precise positioning of an Idd locus, since with each backcross cycle, individuals can be selected in which the length of the original congenic segment has been reduced by recombination. This results in a series of interval specific congenic stocks; if diabetes resistance or some other donor-strain specific subphenotype is retained within a specific interval, then the Idd locus (or loci) are assumed to be within the interval. The Idd3 locus on proximal Chromosome 3 has been fine-mapped to within less than a

54

NOD Mice and Related Strains: Research Applications in Diseases

one centimorgan segment containing the Il2 candidate gene by this technique.79 With the exception of the MHC class II genes, the identities of the non-MHC loci are unknown, although, as indicated above, all non-MHC Idd regions contain interesting candidate genes to tantalize the interested investigator. Idd loci on 5 NOD chromosomes have been reported to control the insulitic process directly (see Wicker et al for a review4). Although only single Idd loci were originally detected by exclusion mapping on these NOD chromosomes, further analysis of interval specific subcongenic stocks has led to the realization that most chromosomes carrying what was originally mapped as a single Idd locus, in fact, carries more. Because of the strong contribution to insulitis development by H2g7, inability to demonstrate that a specific chromosomal region does not affect the extent of insulitis in a NOD congenic stock is not always reason to exclude the region as containing insulitis determinants. H2g7 expressing bi- or polycongenic stocks would be more informative in this regard, particularly when the role of an NOD-derived “insulitis” locus is being examined on a diabetes resistant background such as B6.78 Analysis of B6 congenic stocks carrying NOD-derived susceptibility loci have led to the suggestion that Chromosomes 1 and 17 regulate mononuclear cell trafficking into the islets, while other loci are involved in regulation of tolerance to islet cell autoantigens.78 To illustrate the complex interactions between Idd genes, only one locus on Chromosome 3 (Idd3) was initially detected.69 However, the complexity of the Chromosome 3 contribution to diabetogenesis was appreciated when Chromosome 3 subcongenic stocks were produced. At least two widely separated loci (Idd3 and Idd10 ) on Chromosome 3 are now known to interact epistatically in bicongenic stocks to modulate the frequency and destructiveness of insulitis.4 As discussed above, the Idd3 region has been narrowed to a 0.3 cM region containing the Il2 locus.79 Although the defective NOD Fc gamma receptor-1 (Fcgr1) allele was originally proposed as an Idd10 candidate gene,80 it has recently been excluded by fine mapping (Dr. Linda Wicker, personal communication). Even more Idd genes are likely to be identified on this chromosome (Dr. Linda Wicker, personal communication). Chromosome 1 is one of the longer mouse chromosomes. Although separate loci controlling insulitis/diabetes [Idd5, near the IL-1 receptor (Il1r) and the cytotoxic T lymphocyte associated protein-4 (CTLA4) gene] and peri-insulitis (“Nod1”, near Bcl2, a protooncogene

Genetics and Immunogenetics of NOD Mice and Related Strains

55

Table 2.2. Currently-identified Idd loci controlling IDDM in NOD mice Locus(Chromosome, Contribution of Comments linkage marker) NOD allele to IDDM

Idd1 = H2g7 (17)

susceptibility

Idd2 (9, Thy1)

susceptibility

Idd3 (3, Il2))

susceptibility

Idd4 (11, Acrb)

susceptibility

Idd5 (1, Il1r, Ctla4) susceptibility, insulitis (1, Bcl2) susceptibility, peri-insulitis Idd6 (6, D6Nds1)

susceptibility

Idd7 (7, Ckmm) Idd8 (14, Plau)

resistance resistance resistance

Idd9 (4, Nppa)

susceptibility

Idd10 (3, Tshb)

susceptibility

Idd11 (4, Nhe1)

susceptibility

Idd12 (14, Plau)

susceptibility

Idd13 (2, Il1)

susceptibility

Idd14 (13, D13Mit61) Idd15 (5, Xmv65)

susceptibility susceptibility

Idd16 (17, MHC)

susceptibility

MHC class I and II loci contribute; unique Tap and Hsp70 alleles affects timing of IDDM onset; effect more pronounced in outcross with NON than with B10 or B6 controls frequency and severity of insulitis; may be the Il2 gene itself affects timing of IDDM onset in B10, but not NON outcross analysis Bcl2 locus may control reduced susceptibility of NOD T cells to apoptotic cell death. Segregates in outcross with C57 strains, but not NON (which also develops peri-insulitis) NON contributes a gene in this region that is more diabetogenic than the NOD allele at Idd6. B10 allele is protective Both NON and C57BL strains contribute a diabetogenic allele Diabetogenic allele contributed by B10, but not NON diabetogenic effect clearly demonstrable in F2 Protective effect in B10, but negligible in NON outcross Observed in outcross/backcross with B6 and SJL Observed in outcross/backcross with B6 and SJL; possibly the same locus as Idd8, but B6 and SJL alleles are not diabetogenic Protective allele observed in outcross with related NOR/Lt stock discovered in F2 cross with NON discovered in F2 cross with NON; Xmv65 a xenotropic proviral locus possibly encoding a defective retrovirus expressed in NOD beta cells based upon observation that NOD-H2ct congenic stock exhibits lower IDDM incidence than H2g7; position proximal or distal to class II loci unclear

56

NOD Mice and Related Strains: Research Applications in Diseases

associated with resistance to programmed cell death) have been identified, this chromosome is laden with many more loci implicated in autoimmune disease. Currently, CTLA4 is receiving attention because it lies in a region homologous to a human IDDM locus. Table 2.3 indicates the known human IDDM linkages and potential homologs in the mouse. It is certain that many more insulitis/diabetes controlling genes will eventually be identified in humans and in mice. The ability to identify an Idd locus in mice is wholly contigent upon the genome of the partner strain used in the initial outcross. Other than H2g7, the only non-MHC locus that consistently appears in multiple NOD outcrosses is the Idd3 locus.

CONNECTING Idd LOCI WITH IMMUNOPHENOTYPES Insulitis is too broad a phenotype to provide critical insight as to how a given Idd allele may be functioning. Hence, more specific immune anomalies must be identified in the NOD genome that are corrected or ameliorated when a given NOD allele is replaced by a resistance allele in an interval-specific congenic stock. Certain of the defects in functional maturation of marrow derived NOD macrophages (discussed in detail in chapter 1) appear to be corrected in NOD mice congenic for B6 resistance alleles at both Idd3 and Idd10. Marrow-derived macrophages from this diabetes-resistant congenic stock exhibit normalized growth and functional responses to myeloid growth factors previously reported to be an NOD strain-specific defect.42,81 Further, marrow from these Idd3/Idd10 congenic mice does not transfer IDDM into lethally irradiated NOD/Lt mice, unlike standard NOD/Lt marrow. Associated with the developmental defects of these marrow-derived macrophages is an aberrant release of bioactive IL-1 following lipopolysaccharide (LPS) stimulation. In comparison to B6 marrow-derived macrophages, this is manifest as a lower release of IL-1β coupled with very high levels of IL-1 receptor gene transcription coupled with release of an IL-1 antagonist (D.V. Serreze and E.H. Leiter, unpublished observations). This NOD strain-specific phenotype is ameliorated in Chromosome 3 congenic mice in which the defective NOD Fcrg1 allele, previously suggested as a possible Idd10 candidate gene,80 was replaced by a wild-type B6 allele. However, this congenic stock, maintained by Drs. Linda Wicker and Larry Peterson (Merck Research Laboratories, Rahway, NJ) carries diabetogenic NOD alleles proximal and distal to the small region of B6 genome around Fcgr1, and is not protected from IDDM. Thus, correction of this de-

Genetics and Immunogenetics of NOD Mice and Related Strains

57

Table 2.3. Comparative mapping of human and mouse Idd genes Human Locus (Chr) Marker/ Candidate

Potential NOD Idd homolog (Chr)

Marker

IDDM1 (6p21) IDDM2 (11p15.5) IDDM3 (15q26) IDDM4 (11q13) IDDM5 (6q25) IDDM6 (18q) IDDM7 (2q31-33) IDDM8 (6q25-27)

HLA-DQB1 INS/VNTR D15S107 FGF3 ESR D18S64 D2S326 D6S264

H2g7

IDDM9 (3q21-q25) IDDM10 (10p11.2-q11.2) IDDM11 (14q24.3-q31) IDDM12 (2q31-33) IDDM13 (2q34) GCK (7p) IDDM15 (6q21)

D3S1303 GAD2

Idd1 (17) none yet identified (distal 7) ?Idd2 (middle 9) none yet identified (distal 7) none yet identified (proximal 10) ? “Idd5 “(1) ? “Idd5” (1) none yet identified (proximal 10 or 17) none yet identified (middle 6) none yet identified (proximal 2)

D14S67

none yet identified (middle 12)

IGFBP-2,5 GCK D6S283

?”Idd5 “ (1) ?”Idd5 “ (1) none yet identified (proximal 11) ?Idd14 (13)

?Cyp19

Bcl2 Ctla4, Il1r

Bcl2 Igfbp-5 D13Mit61

Idd5 placed in quotation marks because there appear to be multiple Idd loci on mouse Chr 1

viant immunophenotype is apparently dissociated from other B6 alleles on distal Chromosome 3 conferring IDDM protection. Genetic segregation analysis has also been used to identify immunophenotypes controlled by Idd loci. Currently lumped together as “Idd5”, two loci on Chromosome 1 (one or more proximal, near the IL-1 receptor (Il1r) and CTLA loci controlling insulitis and diabetes, and one or more distal, near the Bcl2 proto-oconcogene controlling insulitis and peri-insulitis respectively have been identified55,82 Idd genes on Chr 9 (Idd2) and Chr 11 (Idd4) appear to act as “timing” genes,55 determining the rate of activation of cytopathic effectors in the insulitic infiltrates. A gene in the Idd4 region has been linked to thymocyte proliferative unresponsiveness in NOD mice.64 A gene on Chromosome 6 presumably near and possibly the same as Idd6, controls dexamethasone-induced apoptosis in NOD thymocytes.83 A locus controlling the sex-dependent elevation of CD4+ T cells (Tlf, for T lymphocyte fraction) circulating in peripheral blood segregated with IDDM in an NOD x NON outcross.84 This locus is near, and possibly the same as Idd2 on Chromosome 9.84 The Tlf locus is androgen-sen-

58

NOD Mice and Related Strains: Research Applications in Diseases

sitive, correlating in backcross analysis with the stronger protective action of the NON allele at Idd2 in males than in females.84 Mice congenic for the NON-derived Idd2 containing segment of Chromosome 9 are more resistant to cyclophosphamide-induced IDDM compared to standard NOD/Lt sex-matched controls (this laboratory, unpublished). The interferon gamma inducing factor (i.e., IL-18) gene has been mapped to the Idd2 region and is abnormally upregulated in NOD but not BALB/c macrophages following cyclophosphamide administration.85

DO NOD ISLETS EXPRESS STRAIN-SPECIFIC Idd GENES? It remains an open question as to whether any of the Idd genes described above exhibit a β cell restricted pattern of expression. Intracisternal type A retroviral genomes are expressed in NOD β cells, but it is expression of xenotropic forms that distinguish NOD from related strains.86 A gene whose product is expressed in NOD islets and typed by proliferation of a cytopathic islet-reactive CD4+ T cell line from NOD spleen has been mapped to Chromosome 6 near the Idd6 region.87 β cell expression of a variety of proviral genomes present in NOD mice could represent targets of β-cytotoxic T cells. Humoral responses to endogenous retroviral gene products are typically found in autoimmune-prone strains of mice, including NOD.75,88 Initiation of insulitis in NOD islets may represent a response to β cell expression of xenotropic retroviral antigens encoded by two endogenous proviral loci, Xmv65 and Xmv66.75 The Xmv65 locus on proximal Chr 58 segregates with IDDM in an (NOD x NON.H2g7)F2 cross (Idd15 marker). Xmv66 has been mapped just distal to H2g7 (this laboratory, unpublished) and thus would be a potential candidate for Idd16.31 Finally, certain genes not normally expressed by NOD β cells can be induced either by conditions of neoplastic transformation or by cytokines, especially interferon gamma. One of these is Emv30, an endogenous ecotropic proviral genome whose expression is induced in transformed NOD β cells.89 A peculiar antigenic specificity induced in NOD macrophages and islet cells by gamma interferon is typed by anomalous binding of 28-13-3, an “irrelevant” H2Kb monoclonal antibody (NOD mice express H2Kb). The antigen, termed the “occult” antigen, is MHC-linked, with complex trans-regulation implicated.90 Although it is unknown whether the new serologic specificity represents induction of a “spare” class I gene, or simply reflects altered conformation of the standard NOD class I genes due to binding of new

Genetics and Immunogenetics of NOD Mice and Related Strains

59

peptides, the phenomenon itself further illustrates the possibility that exposure of NOD β cells to high concentrations of cytokines may elicit altered forms of “self ”. NOD β cells have also been reported to express low voltage activated calcium channels.91 This expression is reportedly abnormal, leading to higher levels of intracellular calcium (and possibly enhanced excitability at higher glucose concentrations) than found in β cells of a control strain.91

LESSONS FROM GENETICS OF IDDM IN NOD MICE FOR THE GENETIC PREDICTION OF IDDM IN HUMANS Currently, pre-type I autoimmune diabetes in humans is detected by testing for autoantibody development to a variety of islet cell autoantigens, coupled with the demonstration of impaired insulin secretory capacity. In combination with evidence for impaired β cell function, detection of diagnostic high levels of 2-3 of these antibodies (e.g., with specifities to glutamic acid decarboxylase, insulin, and other candidate autoantigens such as IA-2)92 essentially means that the autoimmune destruction of β cells has already been initiated. Identification of genotypes conferring high risk for IDDM in infants at birth would greatly enhance the likelihood that intervention therapy would be successful if initiated prior to the onset of β cell damage. Since “high risk” HLA class II genes are so prevalent in Caucasian populations, neonatal screening for these genes at a population level are not predictive per se of future IDDM development. However, if there were a relatively limited set of high risk non-MHC susceptibility loci that were commonly associated with high risk HLA haplotypes, genetic risk assessment of prediabetes would be greatly enhanced. The original hope was that elucidation of the locations of the major nonMHC Idd genes in NOD mice would directly lead to identification of homologous counterparts in the human genome (designated IDDM loci). This direct extrapolation has not been realized.67 Recent analysis in human multiplex families reported evidence for IDDM linkage at 18 chromosomal regions.93 The most frequently detected IDDM linkage outside of HLA (IDDM1) in humans is IDDM2, associated with different alleles of a variable number tandem repeat (VNTR) located upstream of the insulin (INS) locus.94 Evidence exists that shorter alleles express at lower levels in fetal thymus, such that less tolerance to (pro)insulin may be induced while the immune repertoire is being generated.95,96 An Idd homolog near the Ins2 gene on mouse Chromosome 7 has not yet been identified in any outcross

60

NOD Mice and Related Strains: Research Applications in Diseases

between NOD and diabetes-resistant strains, perhaps because mice have two insulin genes so that (pro)insulin levels in fetal thymus are not contingent on one locus only. Combining linkage analysis in NOD mice with identification of the immunophenotypes controlled by the Idd loci may allow identifcation of similar immune defects essential to IDDM development in humans, even though the genetic events underlying to the defect may be different. An important insight gained from the genetic analysis of IDDM susceptibility in NOD mice (and illustrated in Fig. 2.1) is that the NOD genome contains but one subset of a much larger set of potential Idd genes predisposing to autoimmune disease. This is illustrated in Fig. 2.1, showing that the risk of IDDM development increases as H2g7 is combined with increasing numbers of non-MHC Idd loci. When the NOD-specific diabetogenic interactions are disrupted by outcross, and then reassorted into diabetogenic combinations through intercross or backcross, the same fixed set of IDDM susceptibility modifiers defining the NOD genome need not be fully reconstituted to elicit IDDM. One of the most interesting discoveries (see refs. 8, 55), made by outcross of NOD with IDDM-resistant strains such as B10 and NON, is that the “normal” parental strain contributes susceptibility as well as resistance alleles (e.g., B10 derived susceptibility alleles at Idd7 and Idd8, and NON derived susceptibility alleles at Idd6 and Idd7, but not at Idd8). The NON strain was, in fact, originally selected for impaired glucose homeostasis. Indeed, NON/Lt mice become obese with age and exhibit impaired glucose tolerance. Accordingly, NON genes contributing to glucose intolerance apparently synergize deleteriously with the collection of NOD-derived Idd susceptibility genes to increase the penetrance of the disease phenotype. Similarly, following outcross of NOD to an inbred Mus spretus mouse, undefined M. spretus alleles precipitated a non-insulin dependent form of diabetes in males,97 emphasizing not only the heterogeneity in the genetics of diabetes, but also in its pathophysiology. In addition, the activities of certain immune mediators such as Natural Killer (NK) cells and lytic complement which could possibly participate in autoimmune destruction of pancreatic β cells, are deficient in NOD, but not B10 or NON mice. Idd susceptibility loci from these “resistant” strains might include genes promoting more normal activation of immune responses defective in NOD, expanding the repertoire by which β cells may be destroyed in hybrid mice. A possible illustration

Genetics and Immunogenetics of NOD Mice and Related Strains

61

Fig. 2.1. Stepwise logistic regression analysis shows that the probability of developing diabetes increases as a function of the number of deleterious nonMHC genotypes contributed by each parental genome. Data are from genotyping 245 diabetic and nondiabetic first backcross progeny (to NOD) following intercross between NOD/Lt x NON/Lt.H2g7. The threshold susceptibility conferred by H2g7 homozygosity is present in individuals typed as “zero” for the non-MHC loci entered into this analysis. A deleterious gene was considered present when a backcross individual was homozygous for any of the NOD-derived susceptibility markers or hetero-zygous for the NON-derived Idd7 susceptibility locus marker, Ckmm. Reprinted in modified form by permission from McAleer et al (1995). Crosses of NOD mice with the related NON strain: a polygenic model for type I diabetes. Diabetes 44:1186-1195.

of this is that congenic replacement of a defective NOD Fcgr2 allele on distal Chromosome 1 accelerates rather than retards diabetogen– esis.98 The finding that the genome of the NOD mouse represents but one subset of a larger spectrum of potentially pathogenic genes combinations certainly indicates that the genetics of IDDM in randomlyreproducing human populations will be no less complicated.

62

NOD Mice and Related Strains: Research Applications in Diseases

ACKNOWLEDGMENTS This writing has been supported by NIH grants DK 36175 and DK27722, and a grant from The Juvenile Diabetes Foundation International. The author thanks Drs. David Serreze, Len Shultz, and Linda Wicker for critical reviews. REFERENCES 1. Merriman T, Todd J. Genetics of insulin-dependent diabetes; nonmajor histocompatibility genes. Horm Metab Res 1996; 28(6):289-293. 2. Chandler P, Fairchild S, Simpson E. H-Y responses of non-obese diabetic (NOD) mice. J Immunogenet 1988; 15:321-330. 3. Bailey DW. How pure are inbred strains of mice? Immunol Today 1982; 3:210-214. 4. Wicker LS, Todd JA, Peterson LB. Genetic control of autoimmune diabetes in the NOD mouse. Ann Rev Immunol 1995; 13:179-200. 5. Serreze DV, Leiter EH. Insulin Dependent Diabetes Mellitus (IDDM) in NOD Mice and BB Rats: Origins in Hematopoietic Stem Cell Defects and Implications for Therapy. In: Shafrir E, ed. Lessons from Animal Diabetes. V London: Smith-Gordon, 1995:59-73. 6. Slattery R. Transgenic approaches to understanding the role of MHC genes in insulin dependent diabetes mellitus. II. The non-obese diabetic (NOD) mouse. Baillieres-Clin-Endocrinol-Metab 1991; 5(3): 449-54. 7. Vyse TJ, Todd JA. Genetic analysis of autoimmune disease. Cell 1996; 85(3):311-318. 8. McAleer MA, Reifsnyder P, Palmer SM et al. Crosses of NOD mice with the related NON strain: a polygenic model for type I diabetes. Diabetes 1995; 44:1186-1195. 9. Nepom GT. Class II antigens and disease susceptibility. Annu Rev Med 1995; 46:17-25. 10. Kikutani H, Makino S. The murine autoimmune diabetes model: NOD and related strains. In: Dixon FJ, ed. Adv Immunol NY: Academic Press, 1992; 51:285-322. 11. Leiter EH. The NOD mouse meets the “Nerup Hypothesis”. Is diabetogenesis the result of a collection of common alleles present in unfavorable combinations? In: Vardi P, Shafrir E, ed. Frontiers in Diabetes Research: Lessons from Animal Diabetes III. London: SmithGordon, 1990:54-58. 12. Nerup J, Mandruppoulsen T, Helqvist S et al. On the pathogenesis of IDDM. Diabetologia 1994; 37(Suppl 2):S82-S89. 13. Serreze D. Autoimmune diabetes results from genetic defects manifest by antigen presenting cells. FASEB J 1993; 7:1092-1096. 14. Hattori M, Buse JB, Jackson RA et al. The NOD mouse: recessive diabetogenic gene in the major histocompatability complex. Science 1986; 231:733-735.

Genetics and Immunogenetics of NOD Mice and Related Strains

63

15. Prochazka M, Leiter EH, Serreze DV, Coleman DL. Three recessive loci required for insulin-dependent diabetes in NOD mice. Science 1987; 237:286-289. 16. Acha-Orbea H, McDevitt HO. The first external domain of the Nonobese diabetic mouse class II I-Aβ chain is unique. Proc Natl Acad Sci USA 1987; 84:2435-2439. 17. Ikegami H, Ogihara T. Genetics of insulin-dependent diabetes-mellitus (Review). Endocr J 1996; 43(6):605-613. 18. Liu G, Baker D, Fairchild S et al. Complete characterization of the expressed immune response genes in Biozzi AB/H mice: structural and functional identity between AB/H and NOD A region molecules. Immunogenetics 1993; 37:296-300. 19. Todd JA, Acha-Orbea H, Bell JI et al. A molecular basis for MHC class II-associated autoimmunity. Science 1988; 240:1003-1009. 20. Reich E-P, von Grafenstein H, Barlow A, Swenson KE, Williams K, Janeway CA. Self peptides isolated from MHC glycoproteins of Nonobese diabetic mice. J Immunol 1994; 152:2279-2288. 21. Acha-Orbea H, Scarpellino L. Nonobese diabetic and Nonobese nondiabetic mice have unique MHC class II haplotypes. Immunogenetics 1991; 34:57-59. 22. Gaskins HR, Prochazka MP, Nadeau JH et al. Localization of a mouse heat shock protein Hsp70 gene within the H-2 complex. Immunogenetics 1990; 32:286-289. 23. Gaskins HR, Monaco JJ, Leiter EH. Intra-MHC transporter (Ham) genes in diabetes susceptible NOD/Lt mice. Science 1992; 256: 1826-1828. 24. Slattery R, Miller J. Influence of T-lymphocytes and major histocompatibility complex class-II genes on diabetes susceptibility in the NOD mouse. In: Chisari F, Oldstone M, ed. Curr Top Microbiol Immunol vol 206. Transgenic Models of Human Viral and Immunological Disease); Berlin: Springer-Verlag, 1996:51-66. 25. Cooke A, O’Reilly LA, Baxter AG et al. Effect of MHC class II encoding transgenes on autoimmunity in Nonobese Diabetic mice. In: Bluethmann H, Ohashi PS, ed. Transgenesis and Targeted Mutagenesis in Immunology. San Diego: Academic Press, 1994:183-190. 26. Hanson MS, Cetkovic-Cvrlje M, Ramiya V et al. Quantitative thresholds of MHC Class II I-E expression on hematopoietically derived APC in transgenic NOD/Lt Mice determine level of diabetes resistance and indicate mechanism of protection. J Immunol 1996; 157:1279-1287. 27. Parish N, Acha-Orbea H, Simpson E et al. A comparative study of T-cell receptor Vβ usage in non-obese diabetic (NOD) and I-E transgenic NOD mice. Immunology 1993; 78:606-610. 28. McDevitt H, Singer S, Tisch R. The role of MHC class II genes in susceptibility and resistance to type i diabetes mellitus in the NOD mouse. Horm Metab Res 1996; 28(6):287-288.

64

NOD Mice and Related Strains: Research Applications in Diseases

29. Carrasco-Marins E, Shimizu J, Kanagawa O, Unanue E. The class II MHC I-Ag7 molecules from Nonobese Diabetic mice are poor peptide binders. J Immunol 1996; 156:450-458. 30. Ikegami H, Makino S. Genetic susceptibility to insulin-dependent diabetes mellitus: from NOD mice to humans. In: Shafrir E, ed. Lesson from Animal Diabetes IV. London: Smith Gordon, 1993; IV:39-50. 31. Ikegami H, Makino S, Yamato Y et al. Identification of a new susceptibility locus for insulin dependent diabetes mellitus by ancestral haplotype congenic mapping. J Clin Invest 1995; 96:1936-1942. 32. Serreze DV, Chapman HC, Gerling I et al. Initiation of autoimmune diabetes in NOD/Lt mice is MHC class I-dependent. J Immunol 1996; 158:3978-3986. 33. Serreze D, Chen E, Bridgett M et al. Subcongenic analysis of the Idd13 locus in NOD/Lt mice: evidence for several susceptibility genes including a possible diabetogenic role for B2-microglobulin. J Immunol 199-; manuscript submitted. 34. Monaco JJ. A molecular model of MHC class I-restricted antigen processing. Immunol Today 1992; 13:173-179. 35. Faustman D, Li X, Lin HY et al. Linkage of faulty major histocompatibility complex class I to autoimmune diabetes. Science 1991; 254:1756-1761. 36. Pearce RB, Trigler L, Svaasand EK, Chen HM, Peterson CM. Levels of Tap-1 and Tap-2 mRNA and expression of Kd and Db on splenic lymphocytes are normal in NOD mice. Diabetes 1995; 44(5):572-579. 37. Pearce RB, Trigler L, Svaasand EK, Peterson CM. Polymorphism in the mouse Tap-1 gene. J Immunol 1993; 151(10):5338-5347. 38. Schumacher TNM, Kantesaria DV, Serreze DV, Roopenian DC, Ploegh HL. Transporters from H-2b, H-2d, H-2s, H-2k, and H-2g7 (NOD/Lt) haplotype translocate similar sets of peptides. Proc Natl Acad Sci USA 1994; 91:13004-13008. 39. Faustman D. Mechanisms of autoimmunity in Type 1 diabetes. J Clin Immunol 1993; 13:1-7. 40. Li FQ, Guo J, Fu YN et al. Abnormal class I assembly and peptide presentation in the nonobese diabetic mouse. Proc Natl Acad Sci USA 1994; 91(23):11128-11132. 41. Wicker L, Podolin P, Fischer P et al. Expression of intra-MHC transporter (Ham) genes and class I antigens in diabetes-susceptible NOD mice. Science 1992; 256:1828-1830. 42. Serreze DV, Gaskins HR, Leiter EH. Defects in the differentiation and function of antigen presenting cells in NOD/Lt mice. J Immunol 1993; 150:2534-2543. 43. Serreze D, Chapman H, Varnum D et al. B lymphocytes are essential for the initiation of T cell-mediated autoimmune diabetes: analysis of a new ‘‘speed congenic’’ stock of NOD.Igµnull mice. J Exp Med 1996; 184(5):2049-2053. 44. Wicker LS, Leiter EH, Todd JA et al. β2 microglobulin-deficient NOD mice do not develop insulitis or diabetes. Diabetes 1994; 43:500-504.

Genetics and Immunogenetics of NOD Mice and Related Strains

65

45. Serreze DV, Leiter EH, Christianson GJ, et al. MHC class I deficient NOD-B2mnull mice are diabetes and insulitis resistant. Diabetes 1994; 43:505-509. 46. Sumida T, Furukawa M, Sakamoto A et al. Prevention of insulitis and diabetes in beta(2)-microglobulin-deficient non-obese diabetic mice. Int Immunol 1994; 6(9):1445-1449. 47. Katz J, Benoist C, Mathis D. Major histocompatibility complex class I molecules are required for the development of insulitis in non-obese diabetic mice. Eur J Immunol 1993; 23:3358-3360. 48. Fu Y, Nathan DM, Li F, Li X, Faustman DL. Defective major histocompatibility complex Class I expression on lymphoid cells in autoimmunity. J Clin Invest 1993; 91:2301-2307. 49. Hao W, Gladstone P, Engardt S et al. Major histocompatibility complex class I molecule expression is normal on peripheral blood lymphocytes from patients with insulin-dependent diabetes mellitus. J Clin Invest 1996; 98(7):1613-1618. 50. Caillat-Zucman S, Bertin E, Timsit J et al. TAP1 and TAP2 transporter genes and predisposition to insulin dependent diabetes mellitus. CR Acad Sci, Paris 1992; 315:535-539. 51. Ronningen KS, Undlien DE, Ploski R et al. Linkage disequilibrium between TAP2 variants and HLA class II alleles; no primary association between TAP2 variants and insulin dependent diabetes mellitus. Eur J Immunol 1993; 23:1050-1056. 52. Kawaguchi Y, Ikegami H, Fukuda M et al. Absence of association of TAP and LMP genes with type 1 (insulin dependent) diabetes mellitus. Life Sci, 1994; 54:2049-2053. 53. Maugendre D, Alizadeh M, Gauthier A et al. Genetic-heterogeneity between type 1A and type 1B insulin-dependent diabetes-mellitus— HLA class-II and TAP gene analysis. Tissue Antigens 1996; 48(5):540-548. 54. Serreze DV, Prochazka M, Reifsnyder PC et al. Use of recombinant congenic and congenic strains of NOD mice to identify a new insulin dependent diabetes resistance gene. J Exp Med 1994; 180: 1553-1558. 55. Ghosh S, Palmer SM, Rodrigues NR et al. Polygenic control of autoimmune diabetes in nonobese diabetic mice. Nature Genet 1993; 4:404-409. 56. Morahan G, McClive P, Huang D, Little P, Baxter A. Genetic and physiological association of diabetes susceptibility with raised Na+/H+ exchange activity. Proc Natl Acad Sci, USA 1994; 91:5898-5902. 57. Risch N, Ghosh S, Todd JA. Statistical evaluation of multiple-locus linkage data in experimental species and its relevance to human studies: application to Nonobese Diabetic (NOD) mouse and human insulin-dependent diabetes mellitus (IDDM). Am J Human Genet 1993; 53:702-714.

66

NOD Mice and Related Strains: Research Applications in Diseases

58. Morel L, Rudofsky U, Longmate J et al. Polygenic control of susceptibility to murine systemic lupus erythematosus. Immunity 1994; 1:219-229. 59. Baxter AG, Koulamanda M, Mandel TE. High and low diabetes incidence nonobese diabetic (NOD) mice. Origins and characterization. Autoimmunity 1991; 9:61-67. 60. Leiter EH, Serreze DV. Antigen presenting cells and the immunogenetics of autoimmune diabetes in NOD mice. Regional Immunol 1992; 4:263-273. 61. Leiter EH. The genetics of diabetes susceptibility in mice. FASEB J 1989; 3:2231-2241. 62. De Gouyon B, Melanitou E, Richard MF et al. Genetic analysis of diabetes and insulitis in an interspecific cross of the nonobese diabetic mouse with Mus spretus. Proc Natl Acad Sci, USA 1993; 90:1877-1881. 63. Garchon HJ, Bedossa P, Eloy L, Bach JF. Identification and mapping to chromosome-1 of a susceptibility locus for peri-insulitis in nonobese diabetic mice. Nature 1991; 353:260-262. 64. Gill BM, Jaramillo A, Ma LL et al. Genetic linkage of thymic T-cell proliferative unresponsiveness to mouse chromosome 11 in NOD mice: a possible role for chemokine genes. Diabetes 1995; 44(6): 614-619. 65. Xie T, Reddy S, Hofig A et al. Regulation of prostaglandin synthase2 (Pgs-2) in NOD macrophages. Autoimmunity 1996; 24 (suppl 1): 23A. 66. Xie T, Hofig A, Yui M et al. Spontaneous prostaglandin synthase-2 (Pgs2) gene expression in macrophages of NOD and congenic mice. Autoimmunity 1995; 21(1):17A. 67. Leiter EH. Lessons from the animal models: the NOD mouse. In: Palmer JP, ed. Diabetes Prediction, Prevention, and Genetic Counselling. London: John Wiley & Sons, 1996: 201-226. 68. Wicker LS, Miller BJ, Fischer PA, Pressey A, Peterson LB. Genetic control of diabetes and insulitis in the nonobese diabetic mouse. Pedigree analysis of a diabetic H-2nod/b heterozygote. J Immunol 1989; 142:781-784. 69. Todd JA, Aitman TJ, Cornall RJ et al. Genetic analysis of autoimmune type 1 diabetes mellitus in mice. Nature 1991; 351:542-547. 70. Frankel WN, Rudy C, Coffin JM, Huber BT. Linkage of Mls genes to endogenous mammary tumor viruses of inbred mice. Nature 1991; 349:526-528. 71. Knight AM, Dyson PJ. Detection of DNA polymorphisms between two inbred mouse strains-limitations of restriction fragment length polymorphisms (RFLPs). Mol Cell Probes 1990; 4:497-504. 72. Prochazka M, Leiter EH. Identification of a novel mouse mammary tumor proviral locus (Mtv-31) on chromosome Y. Mouse Genome 1990; 87:111.

Genetics and Immunogenetics of NOD Mice and Related Strains

67

73. Fairchild S, Rosenwasser O, Dyson P, Tomonari K. Tcrb-V3+ T cell deletion and a new mouse mammary tumor provirus, Mtv-44. Immunogenetics 1992; 36:189-194. 74. McDuffie M, Ostrowska A. Superantigen-like effects and incidence of diabetes in NOD mice. Diab 1993; 42:1094-1098. 75. Gaskins HR, Prochazka M, Hamaguchi K, Serreze DV, Leiter EH. Beta cell expression of endogenous xenotropic retrovirus distinguishes diabetes susceptible NOD/Lt from resistant NON/Lt mice. J Clin Invest 1992; 90:2220-2227. 76. Tsumura H, Reifsnyder P, Leiter E. Mapping of a murine AIDS virus-related proviral gene (Mrv6) in NOD/Lt mice to chromosome 14. Mamm Genome 1996. 77. Elliott JI, Altmann DM. Non-obese diabetic mice hemizygous at the T cell receptor alpha locus are susceptible to diabetes and sialitis. Eur J Immunol 1996; 26(4):953-956. 78. Yui M, Muralidharan K, Moreno-Altamirano B, Perrin G, Chestnut K, Wakeland E. Production of congenic mouse strains carrying NODderived diabetogenic genetic intervals: an approach for the genetic dissection of complex traits. Mamm Genome 1996; 7:331-334. 79. Lord CJ, Bohlander SK, Hopes EA et al. Mapping the diabetes polygene Idd3 on mouse chromosome 3 by use of novel congenic strains. Mamm Genome 1995; 6(9):563-570. 80. Prins J-B, Todd J, Rodriques N et al. Linkage on chromosome 3 of autoimmune diabetes and defective Fc receptor for IgG in NOD mice. Science 1993; 260:695-698. 81. Serreze DV, Gaedeke JW, Leiter EH. Hematopoietic stem cell defects underlying abnormal macrophage development and maturation in NOD/Lt mice: defective regulation of cytokine receptors and protein kinase C. Proc Natl Acad Sci USA 1993; 90:9625-9629. 82. Garchon H-J, Luan J-J, Eloy L, Bédossa P, Bach J-F. Genetic analysis of immune dysfunction in non-obese diabetic (NOD) mice: mapping of a susceptibility locus close to the Bcl-2 gene correlates with increased resistance of NOD T cells to apoptosis induction. Eur J Immunol 1994; 24:380-384. 83. Penha-Goncalves C, Leijon K, Persson L, Holmberg D. Type 1 diabetes and the control of dexamethazone-induced apoptosis in mice maps to the same region on chromosome 6. Genomics 1995; 28(3):398-404. 84. Pearce RB, Formby B, Healy K, Peterson CM. Association of an androgen-responsive T cell phenotype with murine diabetes and Idd2. Autoimmunity 1995; 20(4):247-258. 85. Rothe H, Jenkins N, Copeland N, Kolb H. Active stage of autoimmune diabetes is associated with the expression of a novel cytokine, IGIF, which is located near Idd2. J Clin Invest 1997; 99:469-474. 86. Leiter EH, Hamaguchi K. Viruses and diabetes: diabetogenic role for endogenous retroviruses in NOD mice? J Autoimmunity 1990; (3 suppl):31-40.

68

NOD Mice and Related Strains: Research Applications in Diseases

87. Dallas-Pedretti A, McDuffie M, Haskins K. A diabetes-associated T-cell autoantigen maps to a telomeric locus on mouse chromosome 6. Proc Natl Acad Sci, USA 1995; 92:1386-1390. 88. Serreze DV, Leiter EH, Kuff EL, et al. Molecular mimicry between insulin and retroviral antigen p73. Development of cross-reactive autoantibodies in sera of NOD and C57BL/KsJ-db/db mice. Diabetes 1988; 37:351-358. 89. Hamaguchi K, Gaskins HR, Leiter EH. NIT-1, a pancreatic β cell line established from a transgenic NOD/Lt mouse. Diabetes 1991; 40:842-849. 90. Leiter E, Christianson G, Serreze D, Ting A, Worthen S. MHC antigen induction by interferon γ on cultured mouse pancreatic β cells and macrophages. Genetic analysis of strain differences and discovery of an “occult” class I-like antigen in NOD/Lt mice. J Exp Med 1989; 170:1243-1262. 91. Wang L, Bhattacharjee A, Fu J, Li M. Abnormally expressed lowvoltage-activated calcium channels in beta-cells from NOD mice and a related clonal cell line. Diabetes 1996; 45(12):1678-1683. 92. Gottlieb PA, Eisenbarth GS. Mouse and man: multiple genes and multiple autoantigens in the aetiology of type i DM and related autoimmune disorders. J Autoimmun 1996; 9(2):277-281. 93. Davies JL, Kawaguchi Y, Bennett ST, et al. A genome-wide search for human type 1 diabetes susceptibility genes. Nature 1994; 371:130-136. 94. Owerbach D, Gabbay K. The search for IDDM susceptibility genes: The next generation. Diabetes 1996; 45(5):544-551. 95. Pugliese A, Zeller M, Fernandez Jr A, et al. The insulin gene is transcribed in the human thymus and transcription levels correlate with allelic variation at the INS VNTR-IDDM2 susceptibility locus for type 1 diabetes. Nature Genetics 1997; 15:293-297. 96. Vafiadis P, Bennett S, Todd J, et al. Insulin expression in human thymus is modulated by INS VNTR alleles at the IDDM2 locus. Nature Genetics 1997; 15:289-292. 97. Hattori M, Yamato E, Matsumoto E, et al. Occurrence of pretype I diabetes (pre-IDDM) and type II diabetes (NIDDM) in BC1 [(NOD x Mus spretus)F1 x NOD] mice. In: Shafrir E, ed. Lessons from Animal Diabetes VI. Boston: Birkhaüser, 1996; VI:83-95. 98. Luan J, Monteiro R, Sautes C, et al. Defective Fc gamma RII gene expression in macrophages of NOD mice—genetic linkage with upregulation of IgG1 and IgG2b in serum. J Immunol 1996; 157(10): 4707-4716. 99. Baxter AG, Cooke A. Complement lytic activity has no role in the pathogenesis of autoimmune diabetes in NOD mice. Diabetes 1993; 42:1574-1578.

Genetics and Immunogenetics of NOD Mice and Related Strains

69

100. Gavin A, Hamilton J, Hogarth P. Extracellular mutations of Nonobese Diabetic mouse FcγRI modifiy surface expression and ligand binding. J Biol Chem 1996; 271:17091-17098. 101. Chesnut K, Shie J-X, Cheng I, et al. Characterization of candidate genes for IDD susceptibility from the diabetes-prone NOD mouse strain. Mamm Genome 1993; 4:549-554. 102. Shultz LD, Schweitzer PA, Christianson SW, et al. Multiple defects in innate and adaptive immunological function in NOD/LtSZ-scid mice. J Immunol 1995; 154:180-191. 103. Giorda R, Rudert WA, Vavassori C, et al. NKR-P1, a Signal Transduction Molecule on Natural Killer Cells. Science 1990; 249: 1298-1300. 104. Cetkovic-Cvrlje M, Leiter E. Mono-ADP ribosyltransferase genes and diabetes in NOD mice: is there a relationship? In: Haag F, KochNolte F, ed. ADP-Ribosylation IN Animal Tissues: Structure, Function and Biology of Mono(ADP-Ribosyl)Transferase and Related Enzymes. New York: Plenum Press, 1997:217-227. 105. Prochazka M, Gaskins HR, Leiter EH, et al. Chromosomal localization, DNA polymorphism, and expression of Rt-6, the mouse homologue of rat T cell lymphocyte differentiation marker RT6. Immunogenetics 1991; 33:152-156. 106. Philbrick WM, Maher SE, Bridgett MM, Bothwell ALM. A recombination event in the 5’ flanking region of the Ly-6C gene correlates with impaired expression in the NOD, NZB and ST strains of mice. EMBO 1990; 9:2485-2492. 107. Bazzoni F, Beutler B. Comparative expression of TNF-α alleles from normal and autoimmune-prone MHC haplotypes. J Inflamm 1995; 45:106-114. 108. Lund T, Shaikh S, Kendall E, et al. RFLP analysis of the MHC class III region defines unique haplotypes for the non-obese diabetic, cataract Shionogi and the non-obese non-diabetic mouse strains. Diabetologia 1993; 36:727-733. 110. Nandi D, Iyer MN, Monaco JJ. Molecular and serological analysis of polymorphisms in the murine major histocompatibility complex-encoded proteasome subunits, LMP-2 and LMP-7. Exp Clin Immunogenet 1996; 13(1):20-9. 111. Marusina K, Iyer M, Monaco J. Allelic variation in the mouse Tap-1 and Tap-2 transporter genes. J Immunol 1997; 158:5251-5256. 112. Baeza NJ, Moriscot CI, Renaud WP, Okamoto H, Figarella CG, Vialettes BH. Pancreatic regenerating gene overexpression in the nonobese diabetic mouse during active diabetogenesis. Diabetes 1996; 45(1):67-70.

CHAPTER 3

The Identity and Ontogenic Origins of Autoreactive TL ymphocytes in NOD Mice David V. Serreze

INTRODUCTION

I

nsulin dependent diabetes mellitus (IDDM) in both humans and the NOD mouse model is caused by autoimmune destruction of pancreatic β cells by T lymphocytes.1-4 The identity of T cells mediating autoimmune β cell destruction in IDDM, and the mechanisms that allow these effectors to be generated, has been a matter of intense investigation and controversy. This chapter will focus on studies in the NOD mouse which have provided insight into these questions.

OVERVIEW OF T CELL ONTOGENY AND SELECTION T lymphocytes are derived from presursor cells of bone marrow origin that subsequently differentiate within the thymus. Mature Tl ymphocytes that emigrate from the thymus and seed the blood and peripheral lymphoid organs can be divided into two major subsets based on cell surface expression of the CD4 or CD8 markers.5,6 The ability of an individual T cell within either subset to recognize a specific antigen is imparted by clonally distributed T cell receptor (TCR) molecules that are expressed as α/β chain heterodimers on the NOD Mice and Related Strains: Research Applications in Diabetes, AIDS, Cancer and Other Diseases, edited by Edward Leiter and Mark Atkinson. © 1998 R.G. Landes Company.

72

NOD Mice and Related Strains: Research Applications in Diseases

cell surface.7,8 While the total mouse genome contains on the order of 105 genes, at least 109 different TCR molecules can be generated.7 This broad diversity is generated as T cells differentiate in the thymus by splicing and rejoining of germline DNA sequences that encode various components of the TCR α and β chains followed by pairing of the resulting gene products.9 The germline sequence for the murine TCR β chain resides on chromosome 6 and consists of approximately 30 variable (V) region gene segments divided in 20 subfamilies of 1-3 members each, and two consecutive downstream clusters each consisting of 1 diversity (D), 6 joining (J) and 1 constant (C) region gene segment.10 The germline sequence for the TCR α chain on mouse chromosome 14 lacks D region segments, but contains approximately 100 V region gene segments, 50 J region segments, and 1 C region segment.10 The particular germline segments which are incorporated into any given VDJ (β chain) or VJ (α chain) gene rearrangement generates structural variability at the N-terminus of both resulting molecules, which upon pairing contributes to the antigenic specificity of the TCR.10 An additional factor contributing to the heterogeneity of rearranged TCR α and β chains is that any given V and J coding region can be linked by differing splice sequences.7 These V to J linkage sequences (which incorporate a D segment in the TCR β chain) are termed CDR3 regions and are also major contributors to the antigenic specificity of the rearranged TCR molecule.11 The constant region domain at the C-terminus of rearranged TCR α/β molecules noncovalently associates with the CD3 complex at the cell surface. Antigen binding to the rearranged TCR α/β molecules induces the CD3 complex to transmit signals that activate intracellular protein kinase C (PKC) mediated second messanger pathways that trigger appropriate T cell effector functions.12 Once a TCR gene rearrangement has occurred in any differentiating T cell, subsequent rearrangment of TCR α and β sequences remaining in the germline configuration is suppressed through a mechanism termed allelic exclusion, although this process is less complete for α than β chain sequences.13,14 Mice also generate a small number of T cells (< 5% of the total population) that express rearranged TCR γ / δ chain heterodimers encoded by germline DNA sequences on chromosomes 10 and 14, respectively.9 These γ/δ T cells will not be discussed further in this review since they have not yet been implicated in the pathogenesis of autoimmune IDDM in NOD mice.

The Identity and Origins of Autoreactive T Lymphocytes in NOD Mice

73

The TCR α/β chain heterodimers generated by the recombinatorial processes described above, recognize antigens which consist of peptide fragments bound to molecules encoded within the major histocompatability complex (MHC).15 There are two primary types of MHC gene products. MHC class I gene products (H2K and H2D in mice) are expressed on virtually all cell types and present peptides of intracellular origin, such as those derived from replicating viruses, to T cells that express TCR α/β chain heterodimers complexed with CD8 molecules. Such MHC class I restricted CD8+ T cells usually exert a cytotoxic function. In contrast, MHC class II gene products (I-A and I-E in mice) which exist as α/β chain heterodimers, are generally only expressed on thymic epithelial cells and bone marrow derived antigen presenting cells (APC) such as B-lymphocytes, macrophages, and dendritic cells. APC take up extracellular proteins and degrade them into peptide fragments which are then bound to MHC class II molecules for presentation to T cells that express TCR α/β chain heterodimers complexed with CD4 molecules. Following antigenic activation, MHC class II restricted CD4+ T cells usually provide helper functions that amplify other components of the immune response including cytotoxic CD8 + T cells and antibody production by B-lymphocytes. Neither MHC class I or class II molecules have an inherent capacity to discriminate between peptides derived from foreign pathogens or normal endogenous proteins.8 Thus, to prevent the development of deleterious autoimmune responses, it is necessary to destroy or inactivate any T cells that express a rearranged TCR that recognizes endogenous peptides bound to self MHC molecules. This normally occurs through several different mechanisms which select from the theoretical total pool of ~109 TCR clonotypes, a subset of effectors that can respond to foreign, but not endogenous peptides bound to self MHC molecules. One such tolerogenic mechanism occurs in the thymus.7,8 The TCR molecules of T cell precursors differentiating within the thymus interact with peptides presented by MHC gene products expressed on both thymic epithelial cells and hematopoietically derived APC. These interactions result in the positive selection of T cells capable of recognizing foreign antigens presented by self-MHC gene products. Immature T cells whose TCR engages endogenous peptides bound to self-MHC molecules are normally negatively selected in the thymus through an activation driven cell death

74

NOD Mice and Related Strains: Research Applications in Diseases

process known as apoptosis. Several recent studies have demonstrated that the threshold of T cell activation required to induce negative selection in the thymus is quantitatively greater than that required for positive selection.16,17 However, not all autoreactive T cells are negatively selected in the thymus. Such autoreactive T cells can undergo apoptotic cell death in the periphery when stimulated to a sufficiently high activation threshold by APC presenting large quantaties of the appropriate antigen.18-22 The activities of other autoreactive T cells escaping intrathymic deletion are suppressed in the periphery by immunoregulatory T cells which also must be stimulated by a highly activated APC in order to become functional.23 Another mechanism by which APC can downregulate autoimmune responses is by inducing a change in the cytokine profile produced by CD4+ T cells reacting against self peptides.24-26 Autoimmune tissue destruction appears to be promoted when self peptide reactive CD4+ T cells produce a Th1 pattern of cytokines including interleukin-2 (IL-2) and gammainterferon (γIFN). These Th1 cytokines amplify cytotoxic CD8+ T cell functions, and also support macrophage activation and delayed type hypersensitivity (DTH) responses. In contrast, autoimmune tissue destruction appears to be blocked when self peptide reactive CD4+ T cells produce a Th2 pattern of cytokines including IL-4, IL-5, IL-6, IL-10 and IL-13, which provide help for the activation of B lymphocyte mediated humoral immunity. Of these cytokines, IL-4 appears to be most important in switching CD4+ T cells from a Th1 to Th2 response profile. Generally CD4+ T cells switch from a Th1 to a Th2 profile as a function of increasing antigen dose presented by APC. The cytokine profile produced by CD4+ T cells can also vary depending upon the type of APC providing antigenic stimulation, with Blymphocytes tending to promote the activation of Th2 responses. The particular MHC class II gene product that an APC uses to present a given antigen can also determine whether a Th1 or Th2 CD4+ T cell response is elicited.27 These T cell response patterns can also be effected by APC produced cytokines, with macrophage production of IL-12 or IL-1 promoting Th1 and Th2 activation respectively.28-31 A constant for all of these immunoregulatory mechanisms, is that high levels of T cell stimulation tend to promote tolerance, while lower levels tend to promote immunological effector responses. Thus, any genetic defects that compromise the stimulatory capacity of APC and/ or impairs T cell responsiveness could preferentially diminish any or all of these tolerogenic mechanisms without fully abrogating immu-

The Identity and Origins of Autoreactive T Lymphocytes in NOD Mice

75

nological effector responses. As discussed later in this chapter, such defects are present in both APC and T cells from NOD mice and may be central to the development of autoimmune IDDM.

THE DIABETOGENIC ROLE OF CD4+ VERSUS CD8+ T CELLS IN NOD MICE In NOD mice, T cell infiltration of the pancreatic islets (insulitis) initiates at approximately 5 weeks of age, with the first appearance of overt diabetes in females generally occuring at 12-14 weeks of age.2,32 Among the major controversies in the etiopathogenesis of IDDM in NOD mice is the phenotypical and functional relationship of the autoreactive T cells that initiate β cell destruction, to those present at the onset of overt disease which can mediate rejection of subsequently implanted islet grafts. The class II region genes of the NOD H2g7 MHC haplotype encode the rare I-Ag7 gene product, but no expressible I-E molecules.33-36 Since the genes that encode this unusual MHC class II region provide a primary component of IDDM susceptibility in NOD mice (see chapter 2 by E. Leiter), it is not surprising that CD4+ T cells are major contributors to autoimmune β cell destruction. However, there has been much debate as to whether I-Ag7 positive APC activate a CD4+ T cell response that mediates autoimmune IDDM in an independent manner, or whether this process also requires contributions from CD8+ T cells that recognize antigens presented on the surface of pancreatic β cells by the relatively common class I gene products (e.g., Kd and Db) of the H2g7 haplotype. Supporting this latter concept was a report that both CD4+ and CD8+ T cells must be transferred from adult NOD donors to accelerate IDDM onset in neonatal or young sub-lethally irradiated syngeneic recipients.37,38 Similarly, using a series of β cell autoreactive CD4+ and CD8+ T cell clones isolated from the insulitic lesion of adult NOD mice, Reich et al39 found that both T cell subsets must be co-transferred to accelerate IDDM onset in young recipients. In contrast, other investigators have proposed that APC from NOD mice process soluble antigens common to all β cells, and bind these to their unusual I-Ag7 MHC class II molecules for presentation to CD4+ T cells which then mediate autoimmune β cell destruction in a manner analogous to a DTH response. This hypothesis was based on the finding that diabetic NOD mice rapidly reject leukocyte depleted islet, but not pituitary grafts, from allogeneic BALB/c donors.40 The same laboratory subsequently reported that diabetic NOD mice retain leukocyte depleted BALB/c islet grafts if the

76

NOD Mice and Related Strains: Research Applications in Diseases

recipients are first depleted of CD4+ T cells by monoclonal antibody treatment.41,42 This was taken as evidence that diabetogenic T cells in NOD mice are not restricted by MHC class I molecules expressed on the target β cell. Indeed, cloned lines of CD4+ T cells isolated from spleens of overtly diabetic NOD mice can also passively transfer IDDM and mediate islet graft rejection.43-46 However, it is important to note that the CD4+ T cells which independently transfer IDDM or mediate islet graft destruction were originally derived from overtly diabetic donors. This is significant since the repertoire of diabetogenic T cells present at the onset IDDM in NOD mice may not be reflective of the autoreactive effectors that actually initiate autoimmune β cell destruction. Major insights into the effector mechanisms responsible for the initiation of autoimmune β cell destruction have been provided by studies in which various T cell populations have been passively transferred into a stock of NOD mice made T and B-lymphocyte deficient by congenic transfer of the severe combined immunodeficiency (scid) mutation.47-49 Since these NOD-scid mice lack functional T-lymphocytes, they remain diabetes free. An obvious advantage of using T cell deficient NOD-scid mice as recipients for passive transfer studies, is that the transferred T cell population cannot activate effectors endogenous to the host. Furthermore, processed β cell autoantigens are present on the I-Ag7 MHC class II molecules of intra-islet APC from NOD-scid mice.50 This indicates that NOD islets contain APC with I-A g7 MHC class II molecules that are preloaded with β cell autoantigens prior to the infiltration of diabetogenic CD4+ T cells. When CD4+ T cells isolated from the spleens of overtly diabetic NOD donors are transferred into NOD-scid recipients, both insulitis and overt IDDM develop within 3-4 weeks.48 Similarly, some cloned lines of islet-reactive CD4+ T cells isolated from the spleens of overtly diabetic NOD mice can also transfer IDDM to NOD-scid recipients.51 IDDM also develops at a high frequency in a stock of NOD mice in which allelic exclusion has resulted in > 95% of the T cells expressing rearranged TCR α and β chain transgenes derived from one of these CD4+ β cell autoreactive T cell clones.52 Collectively, these results would seem to support the hypothesis that β cell autoantigens bound to IAg7 MHC class II molecules on intra-islet APC of NOD mice elicit a DTH, like response by CD4+ T cells that is sufficient to induce IDDM. However, passive transfer studies using NOD-scid recipients have provided solid evidence that while populations of CD4+ T cells present

The Identity and Origins of Autoreactive T Lymphocytes in NOD Mice

77

Fig. 3.1. Initiation and amplification of T cell mediated autoimmune β cell destruction in NOD mice.

in diabetic NOD mice are sufficient to transfer IDDM, as well as to mediate islet graft rejection,43,45,46 these effectors are only generated after β cell necrosis has been initiated by a process dependent upon MHC class I restricted CD8+ T cells. This was demonstrated by the finding that while CD4+ T cells isolated from young prediabetic NOD donors can “home” to NOD-scid islets, they cannot initiate IDDM in the absence of CD8+ T cells.48 Furthermore, it has been recently demonstrated that following in vitro activation on islet cells that express transgene encoded B7-1 co-stimulatory molecules, β cell autoreactive CD8+ Tc ell clones isolated from standard NOD mice can rapidly transfer IDDM to NOD-scid recipients in the absence of CD4+ T cells.53 Several laboratories have unequivocally demonstrated the essential role of MHC class I restricted CD8+ T cells for initiating autoimmune β cell destruction, by developing stocks of NOD mice which

78

NOD Mice and Related Strains: Research Applications in Diseases

fail to develop this T cell subset. This was done by congenically transferring a β2-microglobulin (B2m) gene that was functionally inactivated by homologous recombination onto the NOD inbred background. Intact B2m is required to transport MHC class I molecules to cell surfaces.54,55 Thus, NOD.B2mnull mice fail to express cell surface MHC class I molecules, and hence do not positively select CD8+ T cells.56–58 These NOD.B2mnull mice fail to develop either insulitis or IDDM. This conclusively demonstrates that the relatively common MHC class I gene products encoded by the H2g7 haplotype of NOD mice exert an autoimmune function that is essential for initiating autoimmune β cell destruction. This function most likely entails the selection of autoreactive CD8+ T cells and their targeting to pancreatic β cells. A recent study has demonstrated such MHC class I restricted autoreactive CD8+ T cells are likely to initiate pancreatic β cell destruction through a Fas/Fas-ligand, rather than a perforin mediated mechanism.59 The initiation of β cell necrosis in NOD mice by a process requiring MHC class I restricted CD8+ T cells, apparently results in the release of previously sequestered antigens which subsequently activate and amplify many additional effector T cell populations (Figure 3.1). Thus, by the time overt diabetes has developed, NOD mice have accumulated a greatly expanded repertoire of β cell autoreactive T cells. Some of the CD4+ T cells that accumulate in overtly diabetic NOD mice, or are present in stocks that express rearranged TCR α and β chain transgenes from these effectors, can clearly mediate autoimmune β cell destruction, but their mechanisms of action do not accurately reflect those of the T cells which actually initiate pathogenesis in young prediabetic NOD mice. A recent study has found that the cascade of autoreactive T cell responses required for the progression to overt IDDM in NOD mice is dependent upon the presence of B lymphocytes which may serve as a critical APC population at some point(s) in this process.60

TCR GENE REARRANGEMENTS ASSOCIATED WITH β CELL AUTOREACTIVITY IN NOD MICE As described above, the repertoire of T cells that initiate autoimmune β cell destruction in NOD mice is less diverse than that present at the onset of overt hyperglycemia. Many investigators have analyzed whether the T cells proposed to initiate autoimmune β cell destruction in NOD mice express a fixed and finite set of rearranged TCR α and β chain genes. Many of these efforts have been pursued with the

The Identity and Origins of Autoreactive T Lymphocytes in NOD Mice

79

hope that autoimmune IDDM is initiated by a T cell that expresses a single rearranged TCR clonotype that recognizes a single pancreatic β cell peptide. This has been reported to occur in other autoimmune syndromes such as experimentally induced allergic encephalitis.61 If autoimmune IDDM is also initiated by a single T cell clonotype, it has been proposed that it may be possible to inhibit the disease process in high risk individuals by pharmacological treatment with a peptide that specifically blocks the TCR of this autoreactive effector. As an alternative approach, it has also been proposed that it may be possible to prevent IDDM development through various protocols that induce tolerance to the autoantigenic peptide recognized by the initiating T cell clonotype. However, the development of such therapies will become more complicated with each increase in the number of Tc ell clonotypes found to initiate autoimmune β cell destruction in IDDM. Indeed, the studies described below indicate that while it is much less diverse than that present at the onset of overt hyperglycemia, the spectrum of T clonotypes contributing to the initiation of auto- immune β cell destruction in NOD mice is not monoclonal. The germline TCR α and β chain sequences of NOD mice are identical to those most commonly found in other inbred laboratory mouse strains.62,63 Thus, the development of T cells that initiate autoimmune β cell destruction in NOD mice cannot be ascribed to mutations in the TCR α or β chain germline sequences. This conclusion is further supported by the finding that IDDM still develops in a stock of NOD mice congenic for the germline TCR β chain sequences characterizing the SWR or C57L/J inbred strains.64,65 Significantly, in the SWR and C57L/J strains approximately one-half of the normal TCR Vβ germline elements are deleted including Vβ8, which had previously been implicated in an antibody depletion study to be a component of T cell clonotypes initiating autoimmune β cell destruction in NOD mice.66 This indicates either that the TCR β chain elements absent in the SWR and C57L/J strains, but present in NOD, are not required to generate β cell autoreactive T cells, or that there is great plasticity in the repertoire of T cell clonotypes capable of initiating autoimmune IDDM. Numerous lines of evidence support the latter possibility. The first line of evidence indicating that autoimmune β cell destruction is not initiated by a fixed set of T cell clonotypes was the finding that IDDM develops at a high frequency in a stock of NOD mice expressing rearranged TCR α and β chain transgenes specific for

80

NOD Mice and Related Strains: Research Applications in Diseases

a non-relevant antigen.67 Due to allelic exclusion >98% of the peripheral T cells in this stock express the rearranged TCR β transgene. However, as described earlier, allelic exclusion is less complete for TCR α than β chain germline sequences. Because of this, ~17% of the rearranged TCR α chain mRNA transcripts in this stock are not encoded by the transgene, but rather by endogenous α chain germline sequences. Even though they are greatly limited in scope compared to standard NOD mice, some portion of the rearranged TCR α chains derived from endogenous germline sequences are apparently capable of pairing with the single transgene derived TCR β chain to generate diabetogenic T cell clonotypes. Recent evidence indicates that the limited TCR recombinatorial processes available to this transgenic stock of NOD mice still enables them to generate a relatively broad spectrum of diabetogenic T cells that recognize multiple β cell autoantigens.68 The mosaic of TCR clonotypes that initiate and then amplify autoimmune β cell destruction in NOD mice has also been assessed using the technique of reverse-transcription polymerase chain reaction (RT-PCR) analysis. Such analyses have indicated that even at 3 weeks of age, islet infiltrating T cells of NOD mice utilize TCR genes in a polyclonal fashion.69,70 As an alternative approach, other investigators have isolated β cell autoreactive T cell clones from NOD mice and then correlated their patterns of TCR gene utilization with their ability to passively transfer IDDM. As shown in Table 3.1, both CD4+ and CD8+ T cell clones with the capacity to passively transfer IDDM utilize a wide spectrum of TCR genes. Interestingly, this approach has also identified T cell clones which block IDDM development. However, as observed with autoreactive effector populations, these protective clones do not appear to be characterized by a fixed pattern of TCR gene utilization. The nature of these IDDM protective T cell clones will be discussed later in this chapter. The studies described above clearly indicate that the T cells which initiate autoimmune IDDM in NOD mice do not incorporate a fixed set of V, D and J elements into their rearranged TCR molecules. On the other hand, there have been reports that a relatively limited set of CDR3 sequences and V, D and J elements are incorporated into the diabetogenic TCR molecules of NOD mice.76,79 However, the TCR sequences identified in these studies were not totally monoclonal in nature and differed from each other (Table 3.2). One factor that may influence the array of TCR gene rearrangements that are detected among diabetogenic T cells in NOD mice is whether the sequenced

The Identity and Origins of Autoreactive T Lymphocytes in NOD Mice

81

Table 3.1. β cell autoreactive T cell clones isolated from NOD mice Ref.

Source of T cell clones

CD4+ clone name

(43, 71) diabetic NOD BDC 2.4 spleen BDC 4.6 BDC 4.12 BDC 5.2 BDC 5.10 BDC 6.3 BDC 6.9 (39) diabetic NOD unnamed islets (72) diabetic NOD PR-3 islets (73) 8 wk old NOD IS-37SD islets (74) diabetic NOD 9.2.2 spleen 9.2.6 9.2.8 (75, 76) diabetic and None prediabetic NOD islets (77) prediabetic 1.19 NOD islets 2.20 2.35 2.40 4.1 4.4 (78) 4 wk old NOD-5 NOD islets NOD-14 NOD-21

Vβ/Vα TCR gene usage

Effect on CD8+ IDDM in clone passive name transfer

Vβ/Vα TCR gene usage

Effect IDDM in passive transfer

ND BDC 2.5 ND 19/? 6/12 ND ND 4/13.1 ND*

ND None 4/? promotes ND ND promotes ND ND promotes promotes unnamed

—

—

ND*

promote

2/ND

blocks

—

—

ND

promotes IS-2.15

11/ND

blocks

11/ND 11/ND 11/ND —

promotes 9.1.1.17 promotes 9.1.1.33 promotes — NY 2.3 NY 8.3

6/ND 6/ND

neutral neutral

6/ND ? 8.3/ND 8.3/ND 2/ND ? 14/10 8.1/4 8.1/1

promotes None promotes promotes promotes promotes promotes blocks None blocks blocks

None

11/4 promote 8.1/n1.1 promote —

—

—

—

*Originally reported as Vβ5 but subsequently withdrawn, ND-not determined, ?-Unable to equate with any previously identified TCR segment.

TCR genes are isolated from single or pooled islets. This was demonstrated by the finding that while T cells infiltrating an individual NOD islet only utilize one to six independent CDR3 sequences, a much broader diversity of such sequences are detected when comparing multiple T cell infiltrated islets from the same mouse.80 Thus, in NOD mice the diabetogenic T cell infiltrates within each individual islet

82

NOD Mice and Related Strains: Research Applications in Diseases

Table 3.2. TCR gene sequencing results from reports of limited heterogeneity among β cell autoreactive T cells in NOD mice Ref. Source of TCR genes analyzed

T cell subset

Vα/Jα # Different Vβ/Jβ TCR # Different TCR gene CDR3 splice gene usage CDR3 splice usage sequences sequences linking Vα to Jα linking Vβ to Jβ

(76) T cell clones CD8+ from prediabetic and diabetic NOD islets

(76) Islets from 30- to 40-day old NOD mice

(82) Islets from 14day-old NOD mice

1/20n n1.1/34 n1.2/30n 2/33 2/26n 2/25n 2/23n n2/25n 3/47n 3/20n 4/15 4/10n 4/33 4/47n 4/26n 5/34 5/7 8/8 8/32 8/34 10/20n 10/42n CD4+ ND plus CD8+

1 1 1 2 1 1 1 1 1 1 2 1 1 1 1 1 1 1 1 1 1 1 ND

ND

ND

ND

5.2/1.5 8.1/2.7 8.3/1.1 8.3/2.1 9/2.4 9/2.3 10/2.7 10/1.1 14/1.4 14/1.3 16/2.7 x/1.3

1 1 1 1 1 1 1 1 1 1 1 1

1/ND 2/ND 3/2.1 3/2.2 3/2.4 3/2.5 5/ND 6/ND 7/2.1 7/2.4 7/2.5 8.1/2.1 8.2/1.1 8.2/2.5 8.2/2.6 13/2.1 13/ND 13/2.5

— — 7 3 4 8 — — 9 4 6 1 1 1 1 3 1 3

The Identity and Origins of Autoreactive T Lymphocytes in NOD Mice

83

appear to develop autonomously from a limited but differing set of initiating clonotypes. The great plasticity in the repertoire of T cells capable of initiating autoimmune β cell destruction in NOD mice also suggests that this strain is characterized by broad based rather than antigen-specific immunotolerogenic defects. This conclusion is further supported by the fact that NOD mice also develop autoimmune pathologies directed against a number of tissues other than pancreatic β cells.81

PANCREATIC β CELL AUTOANTIGENS TARGETED BY DIABETOGENIC T LYMPHOCYTES IN NOD MICE A single peptide-MHC antigenic complex can be recognized by T cells expressing multiple TCR gene rearrangements.83-86 Thus, while the T cells which initiate autoimmune IDDM in NOD appear to utilize a relatively broad spectrum of TCR gene rearrangements, this does not automatically mean that these effectors recognize an equally diverse set of β cell autoantigens. The breadth of β cell antigens targeted by autoreactive T cells in NOD mice has been the subject of many investigations (Table 3.3). Proposed to be among the initial antigens targeted are the 65 kD and 67 kD isoforms of glutamic acid decarboxylase (GAD). Their potential role as primary β cell autoantigens is supported by two reports that CD4+ T cells in spleens of very young (3-4-week-old) prediabetic NOD mice demonstrate a spontaneous Th1 like response directed against two epitopes of GAD65 spanning amino acid positions 509-528 (p34) and 524-543 (p35).87,88 In older NOD mice with more pronounced insulitic lesions, CD4+ T cell reactivity was found to have spread to other epitopes within the GAD molecules, as well as to additional β cell autoantigens including insulin, heat shock proteins, peripherin and carboxypeptidase-H. In addition to insulin and carboxypeptidase-H, other unidentified β cell secretory granule proteins are also targeted by CD4+ T cell clonotypes activated in the latter stages of the NOD autoimmune cascade.89,90 Interestingly, one of these unknown β cell granule antigens is encoded by a gene on the region of NOD chromosome 6 that contains the IDDM susceptibility locus Idd6 (see chapter 2 by E. Leiter). The failure of NOD mice to establish T cell tolerance to the early β cell autoantigen GAD, can be overridden by injecting large quantities of this protein intrathymically88 or intravenously87 into 3-4-week-old recipients. These treatments not only blocked the development of primary NOD T cell responses to GAD, but also the spread of T cell

84

NOD Mice and Related Strains: Research Applications in Diseases

reactivity to secondary β cell autoantigens, and most importantly, inhibited the development of IDDM. Thus, the appearance of CD4+ T cell reactivity to GAD appears marks a key turning point in the development of autoimmune IDDM in NOD mice. While the appearance of CD4+ T cell reactivity to GAD may represent a key milestone in the pathogenesis of IDDM, it is unlikely to represent the earliest β cell autoreactive T cell response in NOD mice. This is supported by the finding that both GAD isoforms are intracellular proteins,94 and thus should be immunologically presented by

Table 3.3. Survey of β cell autoantigens recognized by NOD T cells Ref.

β cell protein or fraction recognized by NOD autoreactive T cells

(87)

GAD65

(88)

(91)

(89) (90)

(77, 92)

(93)

Epitope within T cell subtype β cell protein responding to recognized by indicated β cell autoreactive autoantigen T cells

p34 (AA 509-528) p35 (AA 524-543) ND ND ND ND ND ND ND ND ND

CD4+ CD4+ ND ND ND CD4+ CD4+ CD4+ CD4+ CD4+ ND

Hsp65 CPH Insulin GAD65 GAD67 Peripherin CPH Hsp60 30-60kD β cell cytosol fraction β cell insulin ND CD4+ BDC 2.5 granule (see Table 3.1) NOD Chr. 6 ND CD4+ BDC 6.9 encoded β cell (see Table 3.1) insulin granule Insulin B-chain recognize CD4+ by 22/40 islet (AA 9-23) reactive clones with diverse TCR gene utilization Proinsulin B-chain (AA 9-23) CD4+

Age of NOD mice exhibiting earliest T cell response to indicated β cell autoantigen

Anatomical source of T cells responding to indicated β cell autoantigen

3-4 weeks 3-4 weeks 6 weeks 8 weeks 12-15 weeks 4 weeks 4 weeks 6 weeks 6 weeks 6 weeks 8 days

Spleen Spleen Spleen Spleen Spleen Spleen Spleen Spleen Spleen Spleen Spleen

Overt diabetic

Spleen

Overt diabetic

Spleen

4 weeks

Pancreatic islets

7-10 weeks

Spleen

The Identity and Origins of Autoreactive T Lymphocytes in NOD Mice

85

MHC class I molecules that trigger CD8+ rather than CD4+ T cell responses. Therefore, it is likely that a CD4+ T cell response to GAD is only elicited after β cell lysis is initiated by T cells responding to other antigens. Indeed, splenic T cells from 8-day-old NOD mice reportedly respond to antigens contained within a 30-60 kD β cell cytosolic fraction but not to recombinant GAD.91 While it was not stated, the fact that the β cell extracts used in this study were soluble antigens indicates that they most likely stimulated CD4+ rather than CD8+ T cells from the 8-day-old NOD donors. As described earlier, the initiation of autoimmune β cell destruction in NOD mice clearly requires contributions from MHC class I restricted CD8+ T cells in addition to MHC class II restricted CD4+ T cells. To date there have been no reports identifying MHC class I restricted antigens recognized by CD8+ T cells contributing to any phase of autoimmune β cell destruction in NOD mice. The identification of such MHC class I restricted antigens would be of great significance, since they are likely to be recognized by T cell clonotypes that contribute to the earliest initiation phases of autoimmune β cell destruction in IDDM. An important, but often overlooked factor that may effect whether a given β cell antigen is defined as a primary or secondary target of the autoimmune attack in NOD mice, is the anatomical site from which the responding T cells are isolated. This is illustrated by an analysis of NOD T cell responses to insulin. As described above, splenic T cells demonstrating spontaneous reactivity to insulin were not detected in NOD mice younger than 12 weeks of age, a time point well after the establishment of autoimmune β cell destruction.87 Similarly, peripheral lymph node T cells from NOD mice only respond to exogenous insulin priming after significant levels of β cell necrosis has already occurred.95 However, other investigators have found insulin autoreactive CD4+ T cell clones can be isolated from pancreatic islets of NOD mice as young as four weeks of age.77,92 Greater than 50% (22/40) of the β cell autoreactive CD4+ T cell clones isolated by these investigators recognize an epitope within the insulin B-chain spanning amino acids 9-23. Despite recognizing this single antigenic epitope, these NOD insulin autoreactive T cell clones utilize a diverse array of TCR genes. This supports the concept that while NOD mice generate a very broad repertoire of β cell autoreactive T cell clonotypes, the array of antigens that they recognize may be relatively limited in scope. However, since NOD mice generate such a broad spectrum of

86

NOD Mice and Related Strains: Research Applications in Diseases

cell autoreactive T cell clonotypes, and the lesion within each islet develops autonomously, there is likely to be variability in the sequence that antigens are targeted. β

MECHANISTIC BASIS FOR THE DEVELOPMENT OF AUTOREACTIVE T CELLS IN NOD MICE Since knowledge gained from the NOD mouse indicates that the spectrum of T cell clonotypes which mediate autoimmune β cell destruction is broad and they recognize multiple antigens, the best hope for an IDDM prophylactic therapy is one that corrects the immunoregulatory dysfunctions that underlie the generation and activation of these effectors. As described earlier, autoreactive T cells are normally physically deleted either in the thymus or periphery, functionally suppressed by immunoregulatory T cells, or rendered nonpathogenic by a shift in their cytokine production pattern from a Th1 to Th2 profile. For each of these immunoregulatory mechanisms, the threshold of T cell activation required to induce tolerance is higher than that required to trigger an effector response. Thus, any defects that compromise the stimulatory capacity of APC and/or impair Tcell responsiveness could preferentially diminish tolerogenic mechanisms without fully abrogating immunological effector functions. Such dysfuctions are present in both APC and T cells from NOD mice, and may be major contributors to the development of autoimmune IDDM. The unusual H2g7 MHC haplotype of NOD mice contributes to several APC dysfunctions that may lead to the development of β cell autoreactive T cells. IDDM rarely develops (<3% incidence) in congenic stocks of NOD mice that heterozygously express MHC haplotypes from other strains.96-98 Thus, the immunotolerogenic defects which underlie the development of β cell autoreactive T cells in NOD mice occur most readily when H2g7 is homozygously expressed. One explanation for this could be that H2g7 MHC molecules are unable to present antigens in an efficient fashion to T cell clonotypes with potential β cell autoreactivity, and this results in the preferential activation of effector rather than immunotolerogenic functions. Indeed, it has been recently demonstrated that the α/β chain complexes which comprise I-Ag7 MHC class II molecules in NOD mice do not dimerize in a stable fashion, and this decreases the efficiency by which they bind and present antigen.99 One immunotolerogenic defect associated with homozygous expression of the H2g7 MHC haplotype is the reduced ability of NOD

The Identity and Origins of Autoreactive T Lymphocytes in NOD Mice

87

APC to activate immunoregulatory T cells in a syngeneic mixed leukocyte reaction (SMLR).100,101 Several lines of evidence indicate that the failure to activate immunoregulatory T cells in a SMLR contributes to the pathogenesis of autoimmune IDDM in NOD mice. First, the SMLR defect is overridden and IDDM inhibited in NOD mice treated with recombinant IL-2 in vivo.102 Secondly, IDDM is inhibited in NOD mice injected at young age with cloned lines of immunoregulatory T cells propagated from an SMLR supplemented with IL-2 in vitro.72,78,103 The reduced ability of NOD mice to produce immunoregulatory T cells that block the function of diabetogenic effectors appears to result from homozygous expression of the unusual I-Ag7 MHC class II gene product. This is supported by the fact that IDDM is inhibited in NOD mice that express transgenes encoding other I-A MHC class II gene products,104–106 and that T cells from such stocks can passively transfer disease resistance.107 IDDM development is also blocked in NOD mice that express transgenes which restore I-E MHC class II expression on APC.105,108,109 This indicates that in addition to those brought about by the presence of I-Ag7 molecules, some of the immunoregulatory dysfunctions which underlie the development of β cell autoreactive T cells in NOD mice are caused by their failure to express an I-E MHC class II gene product on APC. One study has found that the restoration of I-E expression on APC may block IDDM development in NOD mice by inducing a Th1 to Th2 cytokine shift by CD4+ T cells responding to the β cell autoantigen GAD.109 While transgenic expression of an I-A or I-E class II gene product derived from a diabetes resistant MHC haplotype restores the ability of NOD APC to activate other immunoregulatory mechanisms, they remain incapable of mediating clonal deletion of β cell autoreactive T cells.110,111 However, in a competitive bone marrow reconstitution system, it was found that the β cell autoreactive T cells which normally develop from NOD bone marrow are deleted when forced to mature in the presence of APC from a stock of NOD mice that congenically express an entire MHC haplotype associated with IDDM resistance.112 This may result from the protective APC population expressing class I as well as class II gene products from a diabetes resistant MHC haplotype. Collectively, these results indicate that as APC express increasing numbers of genes from a diabetes resistant MHC haplotype, they acquire the ability to activate a wider array of immunotolerogenic mechanisms that limit both the development and

88

NOD Mice and Related Strains: Research Applications in Diseases

function of β cell autoreactive T cells. Some non-MHC controlled defects also contribute to the reduced ability of NOD APC to activate immunoregulatory functions (see chapter 4 by M. Clare-Salzler). These non-MHC controlled defects appear to entail a reduced ability of NOD APC to provide T cell co-stimulatory signals at levels sufficient to induce tolerogenic rather than effector functions. One factor that contributes to the reduced T cell co-stimulatory capacity of APC from NOD mice is that their macrophages produce significantly less IL-1 than those from IDDM resistant strains.100,113 The reduced ability of NOD macrophages to produce IL-1 results from their failure to fully differentiate from precursor cells in bone marrow.114-116 This defect may be of pathogenic significance since IDDM is blocked in NOD mice treated with IL-1 in vivo.113 IDDM protection may result from the fact that IL-1 supplementation in vitro has been shown to restore the ability of NOD APC to activate immunoregulatory T cells in an SMLR.100 Another, not mutually exclusive possibility, is that IDDM protection results from the ability of IL-1 to preferentially activate a Th2 rather than Th1 response by CD4+ T cells.28,29,31 An additional factor contributing to the reduced T cell co-stimulatory capacity of APC from NOD mice, is that their macrophages produce unusually high levels of prostaglandin E-2 (PGE2).117 This could contribute to the development of β cell autoreactive T cells through several mechanisms. First was the finding that when purified, dendritic cells from NOD mice represent a highly efficient APC population for activating immunoregulatory T cells in an SMLR, but this function is normally suppressed by the high levels of PGE2 produced by macrophages.117 PGE2 can also downregulate macrophage function in an autocrine fashion by elevating intracellular cAMP levels which in turn antagonize the PKC mediated second messanger pathways required for IL-1 secretion.118-121 As described above, the resulting decrease in IL-1 mediated T cell co-stimulatory activity could then block the induction of various immunoregulatory functions. In a similar way, a PGE2 induced elevation of intracellular cAMP, can block the induction of immunoregulatory mechanisms directly at the T-cell level by antagonizing TCR coupled PKC second messanger activities. Indeed, the presence of PGE2 has been shown to inhibit the ability of T cells to be driven to an activation state sufficient to induce apoptotic cell death following TCR cross linking.122

The Identity and Origins of Autoreactive T Lymphocytes in NOD Mice

89

If diminution of APC functions brought about by synergistic interactions between MHC and non-MHC genes is truly central to the pathogenesis of autoimmune IDDM, this could provide an avenue for the development of prophylactic therapies. One possible approach could be the use of non-specific immunostimulatory agents to upregulate production of T cell co-stimulatory factors by APC. Indeed, this may provide an explanation for the seemingly paradoxical finding that IDDM is inhibited in NOD mice treated with a wide range of non-specific immunostimulatory agents.24,123,124 In addition, the reduced ability of H2 g7 MHC molecules to present β cell autoantigens in a manner efficient enough to induce tolerogenic rather than effector T cell responses, can be overridden by exposing NOD APC to large quantities of these antigens. This was demonstrated by the finding that the development of β cell autoreactive T cells and IDDM are both inhibited in NOD mice intrathymically injected at a young age with whole syngeneic islet cells.125 As described earlier, similar results were observed in NOD mice intrathymically injected with the β cell autoantigen GAD.88 Overriding inefficient presentation of β cell autoantigens by H2g7 MHC molecules would also account for the finding that IDDM development is inhibited in NOD mice injected with dendritic cells purified from pancreatic lymph nodes (presumably presenting high levels of β cell antigens), but not from cervical or axillary nodes.126 Also supporting the concept that H2g7 positive APC can be made tolerogenic when pulsed with high doses of β cell autoantigens, is that NOD T cells isolated from pancreatic lymph nodes are less efficient than those obtained from other anatomical sites in passively transferring IDDM.127 There is also evidence that some immunotolerogenic defects in NOD mice are intrinsic to T cells. Specifically, NOD thymocytes proliferate poorly in response to TCR cross-linking agents.128 It was subsequently demonstrated that this defect results from a reduced ability of NOD thymocytes to activate TCR coupled PKC second messenger pathways.129 This could contribute IDDM development by inhibiting the ability of immature T cells with potential β cell autoreactivity to be driven to an activation state sufficient to induce their deletion via apoptosis. Support for this possibility is provided by the finding that the NOD thymocyte response defect is at least partially controlled by a gene(s) that maps to the region of chromosome 11 previously shown to contain the diabetes susceptibility locus Idd4.130

90

NOD Mice and Related Strains: Research Applications in Diseases

CONCLUSIONS NOD mice generate a diverse array of both MHC class I and class II restricted T cell clonotypes capable of initiating and then amplifying autoimmune destruction of pancreatic β cells. While apparently more limited in scope than the T cell clonotypes that respond to them, a relatively broad spectrum of β cell antigens are targets of the autoimmune response in NOD mice. Thus, if a similar situation exists in humans it will be impossible to develop IDDM prophylactic therapies targeted to “the diabetogenic T cell clonotype” or “the β cell autoantigen”. However, studies in the NOD mouse indicate that it may be possible to prevent IDDM with therapies that correct defects which underlie the genesis of β cell autoreactive T cells. Paradoxically, it appears that these therapies would have to be immunostimulatory in nature, and thus enable T cell-APC interactions to occur at levels vigorous enough to preferentially activate tolerogenic rather than effector functions. ACKNOWLEDGMENTS Cited work from the author’s laboratory was supported by grants from The National Institutes of Health (DK46266 and DK51090), The Juvenile Diabetes Foundation International, and Cancer Center Support (CORE) (CA34196). REFERENCES 1. Castano L, Eisenbarth GS. Type 1 diabetes: a chronic autoimmune disease of human, mouse, and rat. Ann Rev Immunol 1990; 8:647-680. 2. Kikutani H, Makino S. The murine autoimmune diabetes model: NOD and related strains. Adv Immunol 1992; 51:285-322. 3. Serreze DV, Leiter EH. Genetic and pathogenic basis of autoimmune diabetes in NOD mice. Curr Opin Immunol 1994; 6:900-906. 4. Wicker LS, Todd JA, Peterson LB. Genetic control of autoimmune diabetes in the NOD mouse. Annual Rev of Immunol 1995; 13:179-200. 5. Ramsdell F, Fowlkes BJ. Clonal deletion versus clonal anergy: the role of the thymus in inducing self tolerance. Science 1990; 248:1342-1348. 6. von Boehmer H, Kisielow P. Lymphocyte lineage commitment: instruction versus selection. Cell 1993; 73:207-208. 7. Blackman M, Kappler J, Marrack P. The role of the T cell receptor in positive and negative selection of developing T cells. Science 1990;248:1335-1341. 8. von Boehmer H, Kisielow P. Self-nonself discrimination by T cells. Science 1990;248:1369-1373.

The Identity and Origins of Autoreactive T Lymphocytes in NOD Mice

91

9. Strominger JL. Developmental biology of T cell receptors. Science 1989;244:943-950. 10. Sprent J, Webb SR. Function and specificity of T cell subsets in the mouse. Adv Immunol 1987;41:39-133. 11. Davis MM, Bjorkman PJ. T cell antigen receptor and T cell recognition. Nature 1988;334:395-402. 12. Weiss A. T cell antigen receptor signal transduction: a tale of tails and cytoplasmic protein-tyrosine kinases. Cell 1993;73:209-212. 13. Fondell JD, Marolleau JP, Primi D, Marcu KB. On the mechanism of non-allelically excluded Vα-Jα T cell receptor secondary rearrangements in a murine T cell lymphoma. J Immunol 1990;144:1094-1103. 14. Hue I, Trucy J, McCoy C et al. A novel type of aberrant T cell receptor α-chain gene rearrangement. J Immunol 1990;144:4410-4419. 15. Serreze DV. Autoimmune diabetes results from genetic defects manifest by antigen presenting cells. FASEB 1993;7:1092-1096. 16. Ashton-Rickardt PG, Bandeira A, Delany JR et al. Evidence for a differential avidity model of T cell selection in the thymus. Cell 1994;76:651-663. 17. Sebzda E, Wallace VA, Mayer J, Yeung RSM, Mak T, Ohashi PS. Positive and negative thymocyte selection induced by different concentrations of a single peptide. Science 1994;263:1615-1618. 18. Rocha B, Von Boehmer H. Peripheral selection of the T cell repertoire. Science 1991;251:1225-1231. 19. Zhang L, Martin DR, Fung-Leung W-P, Teh H-S, Miller RG. Peripheral deletion of mature CD8+ antigen specific T cells after in vivo exposure to male antigen. J Immunol 1992;148:3740-3745. 20. Ucker DS, Meyers J, Obermiller PS. Activation-driven T cell death. II. Quantitative differences alone distinguish stimuli triggering nontransformed T cell proliferation or death. J Immunol 1992; 149:1583-1592. 21. Critchfield JM, Racke MK, Zuniga-Pflucker JC et al. T cell deletion in high antigen dose therapy of autoimmune encephalomyelitis. Science 1994;263:1139-1143. 22. Pelfry CM, Tranquill LR, Boehme SA, McFarland HF, Lenardo MJ. Two mechanisms of antigen-specific apoptosis of myelin basic protein (MBP)-specific T lymphocytes derived from multiple sclerosis patients and normal individuals. J Immunol 1995;154:6191-6202. 23. Ishikura H, Jayaraman S, Kuchroo V, Diamond B, Saito S, Dorf ME. Functional analysis of cloned macrophage hybridomas. VII. Modulation of suppressor T cell-inducing activity. J Immunol 1989; 143:414-419. 24. Rabinovitch A. Immunoregulatory and cytokine imbalances in the pathogenesis of IDDM: therapeutic intervention by immunostimulation? Diabetes 1994; (43):613-621. 25. Paul WE, Seder RS. Lymphocyte responses and cytokines. Cell 1994;76:241-251.

92

NOD Mice and Related Strains: Research Applications in Diseases

26. Liblau RS, Singer SM, McDevitt HO. Th1 and Th2 CD4+ T cells in the pathogenesis of organ specific autoimmune diseases. Immunol Today 1995; 16:34-38. 27. Murray JS, Madri J, Pasqualini T, Bottomly K. Functional CD4 T cell subset interplay in an intact immune system. J Immunol 1993; 150:4270-4276. 28. Greenbaum LA, Horowitz JB, Woods A, Pasqualini T, Reich E-P, Bottomly K. Autocrine growth of CD4+ T cells: differential effects of IL-1 on helper and inflammatory T cells. J Immunol 1988; 140: 1555-1560. 29. McAuthor JG, Raulet DH. CD28-induced costimulation of T helper type 2 cells mediated by induction of responsiveness to interleukin 4. J Exp Med 1993; 178:1645-1653. 30. Germann T, Gately MK, Schoehaut DS et al. Interleukin-12/T cell stimulating factor, a cytokine with multiple effects on T helper 1 (Th1) but not on Th2 cells. Eur J Immunol 1993; 23:1762-1770. 31. Taylor-Robinson AW, Phillips RS. Expression of IL-1 receptor discriminates Th2 from Th1 cloned CD4+ T cells specific for Plasmodium chabaudi. Immunology 1994; 81:216-221. 32. Serreze DV, Leiter EH. Insulin dependent diabetes mellitus (IDDM) in NOD mice and BB rats: origins in hematopoietic stem cell defects and implications for therapy. In: Shafrir E, ed. Lessons from Animal Diabetes V. London: Smith-Gordon, 1996: 59-73. 33. Acha-Orbea H, McDevitt HO. The first external domain of the nonobese diabetic mouse class II I-Aβ chain is unique. Proc Natl Acad Sci, USA 1987; 84:2435-2439. 34. Hattori M, Buse JB, Jackson RA et al. The NOD mouse: recessive diabetogenic gene in the major histocompatability complex. Science 1986; 231:733-735. 35. Lund T, Simpson E, Cooke A. Restriction fragment length polymorphism in the major histocompatibility complex of the Non-obese Diabetic mouse. J Autoimmunity 1990; 3:289-298. 36. Fujishima Y, Koide Y, Kaidoh T, Nishimura M, Yoshida TO. Restriction fragment length polymorphism analysis of major histocompatibility complex genes in the non-obese diabetic mouse strain and its non-diabetic sister strains. Diabetologia 1989; 32:118-125. 37. Miller BJ, Appel MC, O’Neil JJ, Wicker LS. Both the Lyt-2+ and L3T4+ T cell subsets are required for the transfer of diabetes in nonobese diabetic mice. J Immunol 1988; 140:52-58. 38. Bendelac A, Carnaud C, Boitard C, Bach JF. Syngeneic transfer of autoimmune diabetes from diabetic NOD mice to healthy neonates: requirement for both L3T4+ and Lyt-2+ T cells. J Exp Med 1987; 166:823-832. 39. Reich E-P, Sherwin RS, Kanagawa O, Janeway CA. An explanation for the protective effect of the MHC class II I-E molecule in murine diabetes. Nature 1989; 341:326-328.

The Identity and Origins of Autoreactive T Lymphocytes in NOD Mice

93

40. Nomikos IN, Prowse SJ, Carotenuto P, Lafferty KJ. Combined treatment with nicotinamide and desferrioxanine prevents islet allograft destruction in NOD mice. Diabetes 1987; 35:1302-1304. 41. Wang Y, Hao L, G. GR, Lafferty KJ. Autoimmune diabetes in NOD mouse is L3T4 T-lymphocyte dependent. Diabetes 1987; 36:535-538. 42. Wang Y, Pontesilli O, Gill RG, La Rosa FG, Lafferty KJ. The role of CD4+ and CD8+ T cells in the destruction of islet grafts by spontaneously diabetic mice. Proc Natl Acad Sci USA 1991; 88:527-531. 43. Haskins K, Portas M, Bergman B, Lafferty K, Bradley B. Pancreatic islet-specific T-cell clones from nonobese diabetic mice. Proc Natl Acad Sci USA 1989; 86:8000-8004. 44. Haskins K, McDuffie M. Acceleration of diabetes in young NOD mice with a CD4+ islet-specific T cell clone. Science 1990; 249:1433-1436. 45. Bradley BJ, Wang Y, Lafferty KJ, Haskins K. In vivo activity of an islet-reactive T-cell clone. J Autoimmun 1990; 3:449-456. 46. Bradley BJ, Haskins K, La Rosa FG, Lafferty KJ. CD8 T cells are not required for islet destruction induced by a CD4+ islet-specific T-cell clone. Diabetes 1992; 41:1603-1608. 47. Prochazka M, Gaskins HR, Shultz LD, Leiter EH. The nonobese diabetic scid mouse: model for spontaneous thymomagenesis associated with immunodeficiency. Proc Natl Acad Sci 1992; 89:3290-3294. 48. Christianson SW, Shultz LD, Leiter EH. Adoptive transfer of diabetes into immunodeficient NOD-scid/scid mice: Relative contributions of CD4+ and CD8+ T lymphocytes from diabetic versus prediabetic NOD.NON-Thy 1a donors. Diabetes 1993; 42:44-55. 49. Rohane PW, Shimada A, Kim DT et al. Islet infiltrating lymphocytes from prediabetic NOD mice rapidly transfer diabetes to NOD-scid/ scid mice. Diabetes 1995; 44:550-554. 50. Shimizu J, Carrasco-Marin E, Kanagawa O, Unanue ER. Relationship between β cell injury and antigen presentation in NOD mice. J Immunol 1995; 155:4095-4099. 51. Peterson JD, Haskins K. Transfer of diabetes in the NOD-scid mouse by CD4 T-cell clones: differential requirement for CD8 T-cells. Diabetes 1996; 45:328-336. 52. Katz JD, Wang B, Haskins K, Benoist C, Mathis D. Following a diabetogenic T cell from genesis through pathogenesis. Cell 1993; 74:1089-1100. 53. Wong FS, Visintin I, Wen L, Flavell RA, Janeway CA. CD8 T cell clones from young nonobese diabetic (NOD) islets can transfer rapid onset of diabetes in NOD mice in the absence of CD4 cells. J Exp Med 1996; 183:67-76. 54. Koller BH, Marrack P, Kappler JW, Smithes O. Normal development of mice deficient in β2m, MHC class I proteins, and CD8+ T cells. Science 1990; 248:1227-1230. 55. Zijlstra M, Bix M, Simister NE, Loring JM, Raulet DH, Jaenisch R. β2-microglobulin deficient mice lack CD4-8+ cytolytic T cells. Nature 1990; 344:742-746.

94

NOD Mice and Related Strains: Research Applications in Diseases

56. Serreze DV, Leiter EH, Christianson GJ, Greiner D, Roopenian DC. MHC class I deficient NOD-B2mnull mice are diabetes and insulitis resistant. Diabetes 1994; 43:505-509. 57. Wicker LS, Leiter EH, Todd JA et al. β2-microglobulin-deficient NOD mice do not develop insulitis or diabetes. Diabetes 1994; 43:500-504. 58. Sumida T, Furukawa M, Sakamoto A et al. Prevention of insulitis and diabetes in beta(2)-microglobulin-deficient non-obese diabetic mice. Int Immunol 1994; 6:1445-1449. 59. Chervonsky AV, Wang Y, Wong FS et al. The role of Fas in autoimmune diabetes. Cell 1997; 89:17-24. 60. Serreze DV, Chapman HD, Varnum DS et al. B lymphocytes are essential for the initiation of T cell mediated autoimmune diabetes: analysis of a new “speed congenic” stock of NOD.Igµnull mice. J Exp Med 1996; 184:2049-2053. 61. Wraith DC, McDevitt HO, Steinman L, Acha-Orbea H. T cell recognition as the target for immune intervention in autoimmune disease. Cell 1989; 57:709-715. 62. Livingstone A, Edwards CT, Shizuru JA, Fathman CG. Genetic analysis of diabetes in the nonobese diabetic mouse. I. MHC and T cell receptor β gene expression. J Immunol 1991; 146:529-534. 63. Lund T, Shaikh S, Hattori M, Makino S. Analysis of the T cell receptor (TcR) regions in the NOD, NON and CTS mouse strains define new TcR Vα haplotypes and new deletions in the TcR Vβ region. Eur J Immunol 1992; 22:871-874. 64. Shizuru JA, Taylor-Edwards C, Livingstone A, Fathman CG. Genetic dissection of T cell receptor Vβ gene requirements for spontaneous murine diabetes. J Exp Med 1991; 174:633-638. 65. McDuffie M. Diabetes in NOD mice does not require T lymphocytes expressing Vβ8 or Vβ5. Diabetes 1991; 40:1555-1559. 66. Bacelj A, Charlton B, Mandel TE. Prevention of cyclophosphamideinduced diabetes by anti-Vβ8 T-lymphocyte-receptor monoclonal antibody therapy in NOD/Wehi mice. Diabetes 1989; 38:1492-1495. 67. Lipes M, Rosenzweig A, Tan K et al. Progression to diabetes in nonobese diabetic (NOD) mice with transgenic T cell receptors. Science 1993; 259:1165-1168. 68. Gottlieb PA, Wegmann DR, Babu SR, Lipes MA. Characterization of infiltrating T cells from TCR transgenic NOD mice. Autoimmunity 1995; 21(1):56. 69. Maeda T, Sumida T, Kurasawa K et al. T-lymphocyte-receptor repertoire of infiltrating T lymphocytes into NOD mouse pancreas. Diabetes 1991; 40:1580-1585. 70. Waters SH, O’Neil JJ, Melican DT, Appel MC. Multiple TCR Vβ gene usage by infiltrates of young NOD mouse islets of Langerhans: a polymerase chain reaction analysis. Diabetes 1992; 41:308-312. 71. Candeias S, Katz J, Benoist C, Mathis D, Haskins K. Islet-specific T cell clones from nonobese diabetic mice express heterogeneous Tcell receptors. Proc Natl Acad Sci USA 1991; 88:6167-6170.

The Identity and Origins of Autoreactive T Lymphocytes in NOD Mice

95

72. Reich E-P, Scaringe D, Yagi J, Sherwin RS, Janeway CA. Prevention of diabetes in NOD mice by injection of autoreactive T lymphocytes. Diabetes 1989; 38:1647-1651. 73. Pankewycz O, Strom TB, Rubin-Kelly V. Islet-infiltrating T cell clones from non-obese diabetic mice that promote or prevent accelerated onset diabetes. Eur J Immunol 1991; 21:873-879. 74. Shimizu J, Kanagawa O, Unanue ER. Presentation of β cell antigens to CD4+ and CD8+ T cells of non-obese diabetic mice. J Immunol 1993; 151:1723-1730. 75. Nagata M, Santamaria P, Kawamura T, Utsugi T, Yoon J-W. Evidence for the role of CD8+ cytotoxic T cells in the destruction of pancreatic β-cells in nonobese diabetic mice. J Immunol 1994; 152:2042-2050. 76. Santamaria P, Utsugi T, Park B-J, Averill N, Kawazu S, Yoon J-W. Beta-cell-cytotoxic CD8+ T cells from nonobese diabetic mice use highly homologous T cell receptor α chain CDR3 sequences. J Immunol 1995; 154:2494-2503. 77. Daniel D, Gill RG, Schloot N, Wegmann D. Epitope specificity, cytokine production profile and diabetogenic activity of insulin-specific T cell clones isolated from NOD mice. Eur J Immunol 1995; 25:1056-1062. 78. Akhtar I, Gold JP, Pan L-Y, et al. CD4+ β islet cell-reactive T cell clones that suppress autoimmune diabetes in nonobese diabetic mice. J Exp Med 1995; 182:87-97. 79. Galley KA, Danska JS. Peri-islet infiltrates of young non-obese diabetic mice display restricted TCR β-chain diversity. J Immunol 1995; 154:2969-2982. 80. Sarukhan A, Bedossa P, Garchon H-J, Bach J-F, Carnaud C. Molecular analysis of TCR junctional variability in individual infiltrated islets of non-obese diabetic mice: evidence for the constitution of largely autonomous T cell foci within the same pancreas. Int Immunol 1995; 7:139-146. 81. Leiter EH. The NOD mouse meets the “Nerup Hypothesis”. Is diabetogenesis the result of a collection of common alleles present in unfavorable combinations? In: Vardi P, Shafrir E, ed. Frontiers in Diabetes Research: Lessons from Animal Diabetes III. London: SmithGordon, 1990:54-58. 82. Yang Y, Charlton B, Shimada A, Del Canto R, Fathman CG. Monoclonal T cells identified in early NOD islet infiltrates. Immunity 1996; 4:189-194. 83. Jorgenson JL, Esser U, Fazekas de St. Groth B, Reay PA, Davis MM. Mapping T-cell receptor-peptide contacts by vaiant peptide immunization of single chain transgenics. Nature 1992; 355:224-230. 84. Kurata A, Berzofsky JA. Analysis of peptide residues interacting with MHC molecule or T cell receptor. Can a peptide bind in more than one way to the same MHC molecule? J Immunol 1990; 144:4526-4535.

96

NOD Mice and Related Strains: Research Applications in Diseases

85. Bhayani H, Patterson Y. Analysis of peptide binding patterns in different MHC/T cell receptor complexes using pigeon cytochrome cspecific hybridomas. J Exp Med 1989; 170:1609-1625. 86. Peccound J, Dellabona P, Allen P, Benoist C, Mathis D. Delineation of antigen contact residues on an MHC class II molecule. EMBO J 1990; 9:4215-4223. 87. Kaufman DL, Clare-Salzler M, Tian J et al. Spontaneous loss of Tcell tolerance to glutamic acid decarboxylase in murine insulin-dependent diabetes. Nature 1993; 366:69-72. 88. Tisch R, Yang X-D, Singer SM, Liblau RS, Fugger L, McDevitt HO. Immune response to glutamic acid decarboxylase correlates with insulitis in non-obese diabetic mice. Nature 1993; 366:72-75. 89. Bergman B, Haskins K. Islet-specific T-cell clones from the NOD mouse respond to β-granule antigen. Diabetes 1994; 43:197-203. 90. Dallas-Pedretti A, McDuffie M, Haskins K. A diabetes-associated T-cell autoantigen maps to a teleomeric locus on mouse chromosome 6. Proc Natl Acad Sci USA 1995; 92:1386-1390. 91. Gelber C, Paborsky L, Singer S et al. Isolation of nonobese diabetic mouse T-lymphocytes that recognize novel autoantigens involved in the early events of diabetes. Diabetes 1994; 43:33-39. 92. Wegmann DR, Gill RG, Norbury-Glaser M, Schloot N, Daniel D. Analysis of the spontaneous T cell response to insulin in NOD mice. J Autoimmunity 1994; 7:833-843. 93. French MB, Allison J, Cram DS et al. Transgenic expression of mouse proinsulin II prevents diabetes in non-obese diabetic mice. Diabetes 1997; 46:34-39. 94. Aguilar-Diosdado M, Parkinson D, Corbett JA et al. Potential autoantigens in IDDM: expression of carboxypeptidase-H and insulin but not glutamate decarboxylase on the β-cell surface. Diabetes 1994; 43:418-425. 95. Serreze DV, Leiter EH. Transplantation analysis of beta cell destruction in (NOD x CBA)F1 bone marrow chimeras. Diabetologia 1990; 33:84-92. 96. Prochazka M, Serreze DV, Worthen SM, Leiter EH. Genetic control of diabetogenesis in NOD/Lt mice: development and analysis of congenic stocks. Diabetes 1989; 38:1446-1455. 97. Wicker LS, Miller BJ, Fischer PA, Pressey A, Peterson LB. Genetic control of diabetes and insulitis in the nonobese diabetic mouse. Pedigree analysis of a diabetic H-2nod/b heterozygote. J Immunol 1989; 142:781-784. 98. Wicker LS, Appel MC, Dotta F et al. Autoimmune syndromes in major histocompatability complex (MHC) congenic strains of nonobese diabetic (NOD) mice. The NOD MHC is dominant for insulitis and cyclophosphamide-induced diabetes. J Exp Med 1992; 176:67-77.

The Identity and Origins of Autoreactive T Lymphocytes in NOD Mice

97

99. Carrasco-Marin E, Shimizu J, Kanagawa O, Unanue ER. The class II MHC class I-Ag7 molecules from non-obese diabetic mice are poor peptide binders. J Immunol 1996; 156:450-458. 100. Serreze DV, Leiter EH. Defective activation of T suppressor cell function in Nonobese Diabetic mice. Potential relation to cytokine deficiencies. J Immunol 1988; 140:3801-3807. 101. Serreze DV, Leiter EH. Defective suppressor T cell activation in NOD/ Lt mice is associated with pathogenesis and homozygous expression of H2g7. Diabetes 1991; 40 (suppl. 1):49A. 102. Serreze DV, Hamaguchi K, Leiter EH. Immunostimulation circumvents diabetes in NOD/Lt mice. J Autoimmunity 1990; 2:759-776. 103. Choisch N, Harrison LC. Suppression of diabetes mellitus in the nonobese diabetic (NOD) mouse by an autoreactive (anti-I-Ag7) islet-derived CD-4+ T-cell line. Diabetologia 1993; 36:716-721. 104. Miyazaki T, Uno M, Uehira M et al. Direct evidence for the contribution of the unique I-Anod to the development of insulitis in nonobese diabetic mice. Nature 1990; 345:722-724. 105. Lund T, O’Reilly L, Hutchings P et al. Prevention of insulin-dependent diabetes mellitus in non-obese diabetic mice by transgenes encoding modified I-A β-chain or normal I-E α-chain. Nature 1990; 345:727-729. 106. Slattery RM, Kjer-Nielsen L, Allison J, Charlton B, Mandel T, Miller JFAP. Prevention of diabetes in non-obese diabetic I-Ak transgenic mice. Nature 1990; 345:724-726. 107. Singer SM, Tisch R, Yang X-D, McDevitt HO. An Abd transgene prevents diabetes in nonobese diabetic mice by inducing regulatory T cells. Proc Natl Acad Sci USA 1993; 90:9566-9570. 108. Uno M, Miyazaki T, Uehira M et al. Complete prevention of diabetes in transgenic NOD mice expressing I-E molecules. Immunology Letters 1991; 31:47-52. 109. Hanson MS, Serreze DV, Atkinson MA, Singh B, Leiter EH. Clonal diversion of T cell responses from Th1 to Th2: a possible mechanism of diabetes resistance in NOD-Ead transgenic mice. Autoimmunity 1995; 21:52. 110. Parish NM, Chandler P, Quartey-Papafio R, Simpson E, Cooke A. The effect of bone marrow and thymus chimerism between non-obese diabetic (NOD) and NOD-E transgenic mice, on the expression and prevention of diabetes. Eur J Immunol 1993; 23:2667-2675. 111. Slattery RM, Miller JFAP, Heath WR, Charlton B. Failure of a protective major histocompatability complex class II molecule to delete autoreactive T cells in autoimmune diabetes. Proc Natl Acad Sci USA 1993;90:10808-10810. 112. Serreze DV, Leiter EH. Development of diabetogenic T cells from NOD/Lt marrow is blocked when an allo-H-2 haplotype is expressed on cells of hematopoietic origin, but not on thymic epithelium. J Immunol 1991; 147:1222-1229.

98

NOD Mice and Related Strains: Research Applications in Diseases

113. Jacob CO, Aiso S, Michie SA, McDevitt HO, Acha-Orbea H. Prevention of diabetes in nonobese diabetic mice by tumor necrosis factor (TNF); similarities between TNF-α and interleukin 1. Proc Natl Acad Sci, USA 1990; 87:968-972. 114. Serreze DV, Gaskins HR, Leiter EH. Defects in the differentiation and function of antigen presenting cells in NOD/Lt mice. J Immunol 1993; 150:2534-2543. 115. Langmuir P, Bridgett M, Bothwell A, Crispe I. Bone marrow abnormalities in the non-obese diabetic mouse. Int Immunol 1993; 5:165-177. 116. Serreze DV, Gaedeke JW, Leiter EH. Hematopoietic stem cell defects underlying abnormal macrophage development and maturation in NOD/Lt mice: defective regulation of cytokine receptors and protein kinase C. Proc Natl Acad Sci USA 1993; 90:9625-9629. 117. Robinson P, Cai B, Keenan C, Clare-Salzler M. Macrophage products mediate suppression of the syngeneic mixed lymphocyte response in NOD mice. Autoimmunity 1993; 15 (suppl.):49. 118. Dinarello CA. Biology of interleukin 1. FASEB J 1988; 2:108-115. 119. Cheung DL, Hamilton JA. Regulation of human monocyte DNA synthesis by colony stimulating factors, cytokines, and cyclic adenosine monophosphate. Blood 1992; 79:1972-1981. 120. Bakouche O, Moreau JL, Lachman LB. Secretion of IL-1: role of protein kinase C. J Immunol 1992; 148:84-91. 121. Shapira L, Takashiba S, Champagne C, Amar S, Van Dyke TE. Involvement of protein kinase C and protein tyrosine kinase in lipopolysaccharide-induced TNF-α and IL-1β production by human monocytes. J Immunol 1994; 153:1818. 122. Goetzl EJ, An S, Zeng L. Specific suppression by Prostaglandin E2 of activation induced apoptosis of human CD4+8+ T lymphoblasts. J Immunol 1995; 154:1041-1047. 123. Singh B, Rabinovitch A. Influence of microbial agents on the development and prevention of autoimmune diabetes. Autoimmunity 1993; 15:209-213. 124. Bowman MA, Leiter EH, Atkinson MA. Autoimmune diabetes in NOD mice: a genetic programme interruptible by environmental manipulation. Immunol Today 1994; 15:115-120. 125. Gerling IC, Serreze DV, Christianson SW, Leiter EH. Intrathymic islet cell transplantation reduces beta-cell autoimmunity and prevents diabetes in NOD/Lt mice. Diabetes 1992; 41:1672-1676. 126. Clare-Salzler MJ, Brooks J, Chai A, Herle KV, Anderson C. Prevention of diabetes in nonobese diabetic mice by dendritic cell transfer. J Clin Invest 1992; 90:741-748. 127. Lepault F, Faveeuw C, Luan JJ, Gagnerault M-C. Lymph node Tcells do not optimally transfer diabetes in NOD mice. Diabetes 1993; 42:1823-1828.

The Identity and Origins of Autoreactive T Lymphocytes in NOD Mice

99

128. Zipris D, Lazarus AH, Crow AR, Hadazija M, Delovitch TL. Defective thymic T cell activation by concanavalin A and anti-CD3 in autoimmune Nonobese Diabetic Mice: Evidence of thymic T cell anergy that correlates with the onset of insulitis. J Immunol 1991; 146: 3763-3771. 129. Rapoport MJ, Lazurus AH, Jaramillo A, Speck E, Delovitch TL. Thymic T cell anergy in autoimmune nonobese diabetic mice is mediated by deficient T cell receptor regulation in the pathway of p21ras activation. J Exp Med 1993; 177:1221-1226. 130. Gill BM, Jaramillo A, Ma L, Laupland KB, Delovitch TL. Genetic linkage of thymic T-cell proliferative unresponsiveness to mouse chromosome 11 in NOD mice. Diabetes 1995; 44:614-619.

CHAPTER 4

The Immunopathogenic Roles of Antigen Presenting Cells in the NOD Mouse Michael Clare-Salzler

INTRODUCTION

I

n chapters 3 and 5 in this volume, David Serreze and Jean Francois Bach discuss the central role of effector and regulatory T cells in the pathogenesis of diabetes in NOD mice. Central to the activation of T cells, whether they be regulatory or effector, is their interaction with antigen presenting cells (APC). APC provide several critical factors for T cell activation which influence the diversity of T cell responses. Principal among these is the obligate interaction of the T cell receptor (TCR) with the APC peptide-majorhistocompatibility complex (MHC) which establishes clonality as well as the nature of T cell function. APC also express co-stimulatory molecules (e.g., CD80 and CD86), secrete cytokines (e.g., IL-1 , IL-10, and IL-12), and produce molecules such as prostanoids and nitric oxide that strongly influence T cell activation and function. Furthermore, there is great diversity in APC populations with regards to their capacity to activate Tc ells and promote specific T cell functions. Since APC provide a complex milieu of factors that strongly effect T cell activation, they are poised to play a critical role in developing the T cell repertoire. Genetic, NOD Mice and Related Strains: Research Applications in Diabetes, AIDS, Cancer and Other Diseases, edited by Edward Leiter and Mark Atkinson. © 1998 R.G. Landes Company.

102

NOD Mice and Related Strains: Research Applications in Diseases

immunologic, or environmental factors, which modify APC function in a manner that promotes autoreactive T cells, would therefore play a critical role in the pathogenesis of autoimmunity. This chapter will focus on studies which are beginning to define the role of APC populations and their associated molecules in the pathogenesis of autoimmunity in NOD mice.

THE CONTRIBUTION OF HEMATOPOIETICALLY DERIVED APC TO THE DEVELOPMENT OF DIABETES SUSCEPTIBILITY IN THE NOD MOUSE Studies have demonstrated that hematopoietically derived APC contribute to the development of diabetes in nonobese diabetic (NOD) mice. Although NOD APC expressing the NOD H2 g7 activate autoreactive T cells,1 they are unable to activate regulatory T cells in the syngeneic mixed lymphocyte response (SMLR).2 However, when developing T cells are forced to interact with hematopoietically derived cells, including APC which express diabetes-resistant MHC haplotypes the development of diabetes is blocked.3-5 In addition, treatment of NOD mice with immunomodulatory agents which upregulate APC function6-8 or activation of NOD T cells by potent APC presenting islet antigens9 effectively blocks the development of diabetes. Collectively, these results suggest that APC contribute to defective deletion of autoreactive T cells and to an inability to activate regulatory cells in the periphery. However, enhancing NOD APC function overrides the combined effects of MHC and non-MHC factors that lead to the development of T cell autoimmunity. The following sections discuss the potential contributions that the MHC and non-MHC factors make to NOD APC dysfunction. THE NOD H2g7 APC provide several factors that are important for the activation and development of T cell function. Principal amongst the APC molecules, critical for T cell activation and functional development, are the MHC antigens. The density of peptide-MHC complexes on APC, the genotype of the MHC, the affinity of the TCR for the MHCpeptide complex, and the stability of the MHC-peptide complex appears to strongly influence the nature of T cell activation.10-14 The T cell response may also be influenced by factors particular to the APC, which include the APC subpopulation presenting the peptide,15, 16 the level of expression of MHC expressed on the APC, and the level of

Immunopathogenic Roles of Antigen Presenting Cells in the NOD Mouse

103

APC co-stimulatory molecules such as CD80 and CD85.17,18 These aspects of the APC biology and MHC-peptide complex strongly influence the outcome of T cell activation, with higher levels of T cell activation in general stimulating a Th2 type response and lower levels promoting Thelpor1 lymphocyte activation.10-14 In addition to the quantity of MHC-peptide complexes on the APC, their stability also has important effects on the nature of the Tc ell responses. Multiple stimulations through the T cell receptor (TCR) appear to induce signals that promote apoptosis in the responding T cell, so called antigen or activation induced cell death (AICD).19-20 Therefore, increasing the longevity of a peptide-MHC complex on the surface of an APC, or the chronicity of APC exposure to an antigen source would allow for enhanced apoptosis of responding T cells. Critical to AICD and requiring a certain threshold of activation by APC is the expression of IL-2, the a chain of the high affinity IL-2 receptor (CD25),21-22 and the generation of the death signal through Fas-Fas ligand interactions.23-24 Mice which lack expression of Fas, Fas ligand, interleukin-2 (IL-2) or CD25, due to specific gene deletions or mutations, manifest impaired activation induced deletion of responding T cells and are highly prone to the development of autoimmune disease.25-27 Importantly, there appears to be differential sensitivity of Th1 and Th2 cells to the Fas induced cell death that occurs during T cell activation, with Th1 cells being more sensitive.28-29 This difference in Fas sensitivity may explain why in a number of reports, that T cells producing IL-4 persist while cells producing interferongamma (IFN-g) are lost following T cell activation. Supporting these concepts are studies in transgenic mice expressing a T cell receptor recognizing an encephalitogenic peptide of myelin basic protein (MBP) and a set of analogue peptides with various levels of MHC binding affinity.30 Immunization of EAE susceptible mouse strains with the analogue of MBP with the highest affinity protects animals from neurologic disease, apparently by inducing apoptosis in some responding T cells, while surviving TCR transgenic T cells are those which produce IL-4 and IL-10. In another study, the transfer of T cells expressing a transgenic TCR for hemagglutinin into mice transgenically expressing high levels of this antigen in hematopoietic cells leads to a transient expansion and subsequent apoptosis of the transferred T cells.31 In contrast to the previous study, the surviving transgenic T cells did not express a Th2 phenotype but were anergic and did not respond to hemagglutinin. Although these two

104

NOD Mice and Related Strains: Research Applications in Diseases

studies differ significantly in their design, they suggest that deletion of antigen-specific T cells is operational in developing peripheral tolerance. Whether tolerance mechanisms such as the emergence of Th2 regulatory cells or the development of anergy occur, may depend on factors including the genetic background, the chronicity of antigen exposure, the MHC allele, and the peptide stimulation of the TCR. A study of the NOD MHC class II antigen, the H2g7, reveals that this molecule is unstable in the presence of denaturing agents, binds peptides poorly and as a result, has a short half-life on the surface of the APC.32 The relative quantity of a peptide presented by an NOD APC in the context of H2g7 may therefore be limited in comparison to other MHC class II alleles which provide protection from diabetes. A decreased density of peptide-MHC complexes and a limitation in the life span of the complex on the NOD APC may therefore favor the development of Th1 responses as well as compromise deletion of autoreactive T cells. One would predict that tolerance for self antigens in the NOD could be enhanced if self antigens were chronically available, or available in greater quantities to APC to compensate for the limitation of the NOD H2g7 peptide binding. In addition, the presentation of self antigens by highly competent APC with endogenously greater stimulatory capacity (i.e., dendritic cells) would hypothetically enhance T cell tolerogenic mechanisms. In this regard, it is of interest in this regard, that treatment of NOD mice with high doses of intravenous glutamic acid decarboxylase (GAD65) or other target antigens blocks diabetes and is associated with the induction of a Th2 and loss of Th1 responses to this antigen.33 The evolution of Th2 responses and the loss of the Th1 responses in GAD treated NOD mice may occur as a result of a differential sensitivity of these T cell subset to FAS-induced cell death following T cell activation. Furthermore, peripheral tolerance is also enhanced by the transfer of highly potent NOD dendritic cells presenting islet antigens.9 Therefore, it appears that by providing conditions that promote quantitatively higher levels of T cell activation in the NOD mouse, tolerance is enhanced.

APC CO-STIMULATORY MOLECULES, CD80 AND CD86 The APC molecules CD80 and CD86 provide important costimulatory signals through CD28; a member of the immunoglobulin supergene family, expressed on naive T cells. Ligation of CD28 lowers the threshold for antigen-induced activation and results in sig-

Immunopathogenic Roles of Antigen Presenting Cells in the NOD Mouse

105

naling that leads to the stabilization of mRNAs for cytokines, particularly IL-2, which is important for the growth and differentiation of T cells.34 In addition, CD28 signaling promotes T cell survival by enhancing Bcl/XL expression.35 The lack of a co-stimulatory signal delivered through CD28, such as that provided by parenchyma cells expressing MHC antigens but lacking CD80 or CD86 expression blocks activation of naive T cells, leading to a state of anergy where Tc ells become refractory to subsequent activation.36 The role of CD28 in the immune response of normal and NOD mice has recently been experimentally addressed. Non-autoimmune B6x129 mice that lack CD28 expression due to a disruption of the CD28 gene (CD28–/– knock-out mice), mount an immune response to alloantigens and mitogens but manifest a reduction in T cell proliferation and IL-2 production.37 However, when the CD28–/– knock mouse, which do not develop autoimmune disease were backcrossed to the NOD, the onset of diabetes was accelerated.38 T cell proliferative responses and IL-2 production by NOD CD28–/– mice to nominal antigens was impaired, but the response to the autoantigen GAD was enhanced over that of the stock NOD mouse with higher levels of IFN-g and lower levels of IL-4 produced. These studies suggest that the absence of CD28 expression perturbs the Th1/Th2 balance and further exacerbates the autoimmune response. A second T cell molecular ligand for CD80/86, CTLA4, is expressed on T cells undergoing activation and has a 20- to 50-fold higher affinity for CD80/86 than CD28.39 However, instead of providing costimulation, CTLA4 downregulates T cell expression of IL-2, reduces Bcl-XL expression and limits T cell expansion.40 Coordinate with CTLA4 expression, in the late phase of an immune response is an upregulation of CD80 on APC, an observation which suggests a potential functional/temporal relationship of these molecules for regulation of an immune response. Unlike CD28 knock-out mice, those with gene disruption of CTLA4 readily develop a lymphoproliferative disorder, prolongation of T cell responses, and autoimmune disease.41 Although the role of CTLA4 has not been delineated in the NOD mouse, defective expression or function could contribute to diabetes by impairing the termination of immune responses to self antigens and reducing apoptosis in responding T cells. The roles of the APC ligands for CD28 and CTLA4, CD80/86, in determining the nature of T cells responses has also been assessed. Studies suggest that CD86 and CD80 differentially support the

106

NOD Mice and Related Strains: Research Applications in Diseases

activation of Th2 and Th1 responses, respectively.17 However, the relative contributions of each molecule to the polarization of the T cell response appears to be highly dependent on the conditions of T cell activation, i.e., antigen concentration and APC number.42 Under conditions where antigen concentrations are suboptimal, CD86 supports the generation of Th2 responses.42 The general effect of CD80 and CD86 on T cell activation therefore appears to enhance the magnitude of the T cell to an activating stimulus. As Th1 responses are less sensitive than Th2 responses to the magnitude of the activating stimulus, these molecules may play a more critical role in Th2 development when the TCR signal is quantitatively limited as may be the case for antigen presentation by NOD APC. To assess the effects of CD80 and CD86 on the development of diabetes, NOD mice have been treated with antibodies to these antigens. When NOD mice were treated with antibodies to CD80 beginning at the inception of the insulitis process (i.e., 4 weeks of age), the development of diabetes was accelerated.43 In contrast, treatment of NOD mice of the same age with anti-CD86 prevented the development of diabetes.43 Many interpretations are possible to explain the disease protection provided by antibodies to CD86, as they may induce anergy or favor the development of disease regulating T cells. Disease acceleration in the NOD caused by anti-CD80 treatment may be explained by its effects on limiting the interaction with CTLA4, thereby reducing signals that serve to terminate autoreactive T cell responses. The relationship between CD28, CTLA4, CD80 and CD86 in the activation, modulation and termination of T cell response remains complex. Results in the NOD mouse, however, suggest differential roles for CD80 and CD86 in the autoimmune response. Although further experimentation is required to add clarity to the roles of CD80 and CD86 in the autoimmune response of the NOD mouse, it appears that these APC molecules through their interaction with CD28 and CTLA4 are important in determining the nature of the immune response.

APC SUBPOPULATIONS In the NOD mouse, defects in APC appear to play a central role in the loss of tolerance to self. However, there are distinct APC subpopulations, including B cells, macrophages, and dendritic cells that differ markedly in their means of antigen uptake into the intracellu-

Immunopathogenic Roles of Antigen Presenting Cells in the NOD Mouse

107

lar compartment, in the level of MHC antigen and co-stimulatory molecule expression, in their tissue distribution, and in their capacity to activate naive and memory T cells. Because of the great functional diversity of these APC subpopulation their effects on tolerance mechanisms may differ substantially.

B LYMPHOCYTES B lymphocytes express membrane bound immunoglobulin receptors specific for an antigen. This receptor allows for a highly efficient uptake of soluble antigen and focusing of the antigenic repertoire presented. B lymphocytes circulate in the peripheral blood and reside predominantly in the lymph nodes and spleen. Of interest, the peripheral blood of NOD mice contains significantly fewer B220+ B lymphocytes than does the B6 mouse (E Leiter; The Jackson Laboratories, Bar Harbor, ME, unpublished observations). B cells express low levels of MHC class II antigens and co-stimulatory molecules but can be activated to express high levels of these molecules. Because resting B cells express lower levels of MHC antigen and co-stimulatory molecules, they function poorly as stimulators of naive T cells. Of potential importance for tolerance, resting B cells would not present soluble self antigens at a high density and would lack essential costimulatory molecules for activation of T cells. Presentation of self antigen by B cells would therefore favor the development of anergy and protect against autoimmune responses.44,45 In addition, Th2 responses are favored when activated B cells are used as APCs.15 Despite these attractive roles of the B cells as an APC supporting tolerance, recent reports suggest that mice deficient in B cells secondary to the inactivation of the Ig allele develop tolerance normally.46 In the NOD mouse, however, the development of B cell-deficient congenic mouse has convincingly demonstrated that B cells play an active role in the pathogenesis of diabetes as these mice are highly diabetes and insulitis resistant.47 Although the underlying mechanistic role of B cells role in the autoimmunity of NOD mice is not known, b cell antigen-reactive B cells could promote killing of b cells by ADCC, or function as an essential APC for the activation of autoreactive T cell populations. Future studies in the NOD will likely define the function of Bc ells as APC, and determine if susceptibility genes influence B cell function in a manner that promotes autoimmune disease.

108

NOD Mice and Related Strains: Research Applications in Diseases

MACROPHAGES Macrophages are widely distributed in tissues and lymphoid organs. Macrophages take up antigens primarily by phagocytosis and are generally poor stimulators of naive T cells, but activate primed lymphocytes.48 Macrophages, like dendritic cells, can acquire antigens in tissue sites and migrate to the draining lymph nodes. However, unlike dendritic cells, macrophages are dependent upon activation stimuli to upregulate APC function.49 Stimuli such as IFN-g provided by activated T cells, or bacterial products such as LPS, upregulate MHC antigen expression and induce co-stimulatory molecules (i.e., IL-1 and CD80/86 expression on macrophages). A lack of appreciable APC function in resting macrophages may be important because these cells are likely to process and present self antigens derived from dead or senescent cells or biochemically modified proteins such as oxidized or glycated proteins.50 Several studies support a role for macrophages in the pathogenesis of insulitis and diabetes in the NOD mouse. First, macrophage development in the NOD is abnormal51,52 as bone marrow derived macrophages do not mature normally when cultured with CSF-1 due to a defect in regulation of the CSF-1 receptor.51 A defect in the maturation of this cell population is further supported by the fact that NOD macrophages produce little IL-1 in response to LPS.2 These findings are of importance for tolerogenic mechanisms in the NOD as the production of IL-1 promotes Th2 cell activation.53 Because macrophage maturation is abnormal, other undefined abnormalities in APC function may be present as a result of this general defect and contribute to disease pathogenesis in the NOD. Macrophages are an important component of the insulitis lesion. Macrophages along with dendritic cells accumulate in close proximity to the islet of Langerhans before T cell infiltration begins.54 At the later stages of insulitis, a subpopulation of macrophages penetrates the islet while dendritic cells, which are often associated with T cells, remain outside the islet mass.54,55 The accumulation of macrophages appears to be a critical event in the development of insulitis as blocking their entry into the islet by treating NOD mice with anti-CD11b (Mac-1) prevents islet inflammation.56 In other studies, deletion of macrophages by injection of silica likewise prevents the development of diabetes and insulitis in the NOD mouse.57 Based on these studies, NOD macrophages appear to play an important part in the immunopathogenesis of insulitis and diabetes.

Immunopathogenic Roles of Antigen Presenting Cells in the NOD Mouse

109

The contribution(s) that NOD macrophages make to the insulitis/diabetes process, however, may be many and varied. For example, several studies demonstrate that isolated islets are killed in vitro by the products of activated macrophages such as monokines, oxygen radicals and nitric oxide.58-59 As macrophages are activated by the Th1 cytokine, IFN-g, and deactivated by the Th2 cytokine IL-10, the lymphocyte milieu may strongly influence the macrophage contribution to b cell destruction. In this regard, NOD macrophages in the islet may efficiently upregulate destructive insulitis as they produce high levels of IL-18 (IFNg-inducing factor) which promotes Th1 responses.60 Therefore, the accumulation of Th1 lymphocytes and macrophages within the islet may create a synergistic microenvironment that would favor the destruction of b cells. Macrophages may also play a role as APC in promoting the development of autoreactive T cell populations. Within the islet, macrophages, unlike dendritic cells or B lymphocytes, would recognize and phagocytize b cells undergoing apoptosis. Studies in NOD and NOD-scid mice indicate that b cells in these animals are prone to spontaneous apoptosis as detected by in situ end labeling techniques (TUNEL)61 in comparison to islets from control mice. These findings suggest that there may be a primary defect in the regulation of apoptosis in b cells. The phagocytosis of dying b cells by macrophages would potentially make them a rich source of APC presenting islet derived peptides. Therefore, macrophages may be an important APC within the islet for the activation of autoreactive T cells. Apart from functioning as APC, macrophages produce several products which strongly modify the outcome of T cell-APC interaction and T cell activation. Included in this long list of factors affecting T cells and APC are prostanoids. Prostanoids, particularly PGE2 and prostacyclin, are produced in large quantities by activated macrophages following the release of the arachidonic acid substrate by phopholipases and the expression of the inducible prostaglandin synthase 2 (PGS2) enzyme.62,63 A few published studies have demonstrated that prostaglandin metabolism in the NOD differs markedly from control mice. For example, T cell activation by mitogens is reduced in NOD spleen cells secondary to prostaglandin production.64 In other studies, purified NOD macrophages, particularly those from female NOD mice, produce significantly larger quantities of prostaglandin E 2 (PGE2) than

110

NOD Mice and Related Strains: Research Applications in Diseases

PGE2, PGD2, PGF2 (Isomerase, reductase)

➞

Leukotrienes

Lipoxygenase PGI(prostacyclin) Arachidonic ➞ PGH2➞ (Prostacyclin synthase) Acid Cyclooxygenase TXA2 (Thromboxane)

➞

➞

membrane Phospholipids

➞

PGS-2

➞

Phospholipase A2

(Thromboxane synthase)

PGS-1

Fig. 4.1. Arachidonic acid is released from plasma membrane pools by the action of phospholipase A2. Free arachidonic acid is metabolized to prostaglandins (PGH2) by the actions of the constitutively expressed cyclooxygenase PGS-1, or by PGS-2 following the induction of this gene by activating agents such as LPS. In the case of the NOD mouse PGS-2 is expressed constitutively during the estrus cycle of female mice. Prostaglandins, prostacyclins, and thromboxane are formed by the activity of specific synthases and reductases.

do macrophages from control female mice.65 Furthermore, placing NOD mice on a diet deficient in essential fatty acids reduces the membrane content of arachidonic acid, the substrate for prostaglandin production, enhances APC function, and significantly lowers diabetes incidence in diet, treated mice.66 We had noted that the presence of NOD peritoneal and splenic macrophages suppressed syngeneic T cell proliferation, whereas macrophages from control (BALB/c, C57BL/6) strains did not (unpublished data; manuscript in preparation). The negative effect of NOD macrophages on NOD T cells was particularly evident when lymphocytes were stimulated with highly potent syngeneic dendritic cells. In a series of experiments, we observed that the suppression of T cell activation by macrophages was predominantly mediated by prostaglandins as it was blocked by the addition of indomethacin. We confirmed that PGE2 production by NOD macrophages in vitro is significantly greater that that of control (C57BL/6, BALB/c, C57BL/10) macrophages. To determine why prostanoid production is enhanced in NOD macrophages, we examined the expression of PGS2, the inducible cyclooxygenase responsible for the production of large quantities of prostanoids by macrophages. We found that the mRNA and

Immunopathogenic Roles of Antigen Presenting Cells in the NOD Mouse

111

protein for this enzyme are spontaneously expressed at high levels in the macrophages of estrus NOD female mice, is not expressed in macrophages of sexually immature female or adult male NOD mice, nor is it expressed in macrophages of estrus female BALB/c or C57BL/6 control mice (Table 4.1). The effect of the estrus on NOD macrophage expression of PGS2 is quite remarkable and appears to be mediated through sex steroid hormones since estrogen and progesterone induce PGS2 expression in macrophages from NOD male and non-estrus female macrophages, but not in macrophages of control mice. We also demonstrated that the production of prostaglandins by the NOD macrophage is mediated by PGS2 as a specific inhibitor of this enyzme, NS-398, completely blocks prostaglandin production. Enhanced prostanoid production in the NOD mouse apparently promotes diabetes as treatment of these mice with indomethacin significantly delays the onset of diabetes and reduces the incidence by over 40%. Furthermore, NOD mice congenic for a segment of Chromosome 1 carrying a PGS-2 gene derived from B10 mice (NOD.B10 Chr1) (Linda Wicker and Laurence Peterson, Merck Laboratories) are characterized by normal expression of PGS-2 and a reduced incidence of diabetes. In contrast, B6 mice congenic for a segment of Chromosome 1 carrying the NOD PGS-2 gene (B6.C1t) (Mary Yui and Edward Wakeland, University of Florida) are characterized by spontaneous expression of this enzyme and the development of peri-insulitis. Table 4.1. Expression Pattern of PGS-2 in Peritoneal Macrophages of Estrus Female NOD, Congenic and Control Mouse Strains Mouse Strain

PGS-2 Expression

Insulitis

Diabetes Incidence

C57BL/6 BALB/c NOD NOD-scid

negative negative positive positive

0% 0% 80% at 30 wks of age 0%

NOD.B10 Chrm1 B6.CIt

negative positive

negative negative severe macrophage infiltration present peri-insulitis

approx. 40% 0%

Overall, these results suggest that aberrant PGS2 expression is responsible for enhanced prostanoid metabolism in NOD mice and that the expression of this enzyme is strongly influenced by the sex

112

NOD Mice and Related Strains: Research Applications in Diseases

steroid environment, perhaps explaining in part the gender differences in diabetes incidence. Furthermore, the data from congenic mice suggest a genetic basis for PGS-2 expression in the NOD mouse and strongly support the role of this enzyme in the immunopathogenesis of disease. The production of prostanoids by macrophages is potentially of great importance to tolerance mechanisms. First, prostaglandins reduce the APC stimulatory capacity by downregulating MHC class II antigens and co-stimulatory molecules.67 In addition, prostaglandins induce adenylate cyclase and cAMP production which potently inhibits IL-2 and CD25 gene expression;68 factors critical for T cell proliferation and AICD. In addition, cAMP inhibits the death signal generated through the Fas antigen in T cells, perhaps through its effects on RAS/RAF interaction.69 Overall, the effects of prostanoids markedly impair processes which are critical to T cell activation and AICD. Indeed a recent report suggests that prostanoids block AICD in T cells.70 Preliminary in vivo data from our laboratory (Xie and ClareSalzler) also support this concept as pre-treatment of NOD mice with indomethacin prior to Staphylococcal enterotoxin B immunization enhances antigen-reactive T cell deletion whereas the same treatment and immunization has no effect on T cell deletion in treated C57BL/6 control mice.

DENDRITIC CELLS Dendritic cells are widely distributed in tissues and in the lymphoid compartments. Immature dendritic cells (i.e., Langerhans cells which reside in tissue sites) readily take up large quantities of antigens primarily through macropinocytosis.71 From tissue sites, dendritic cells migrate via the afferent lymph to the draining lymph node where they mature into interdigitating cells, losing their capacity to process antigen, but markedly increasing the ability to activate T cells. Mature dendritic cells are 10-100 times more efficient in activating naive and primed T cells than are B cells or macrophages.48 The efficiency of DCs as APC lies at least in part in their capacity to express high constitutive levels of MHC class I and class II molecules as well as CD86, the ligand for the T cell co-stimulatory molecule CD28.72 CD80 is expressed at low levels on resting dendritic cells but its expression is upregulated on dendritic cells during the course of T cell activation.72

Immunopathogenic Roles of Antigen Presenting Cells in the NOD Mouse

113

Dendritic cells and macrophages are found in increased numbers around NOD islets before there is significant accumulation of T cells.54,55 During the later stages of insulitis, dendritic cells are also found in the same location, but are surrounded by clusters of T cells.55 The expression of surface molecules such as ICAMs, CD80/86 and MHC molecules on dendritic cells has not been assessed directly; however, NOD dendritic cells are as potent as dendritic cells from non-autoimmune control mice in stimulating syngeneic or allogenic T cell responses so that their function as APC appears to be intact (Clare-Salzler; manuscript in preparation). Dendritic cells appear to play an important role in the generation of tolerance in NOD mice. Transfer of large numbers of dendritic cells presenting islet antigens (i.e. those derived from the draining lymph node of the pancreas or pulsed in vitro with islet antigens) into young NOD female mice protects recipients against the development of diabetes.9 Therefore, it appears that dendritic cells present islet antigens in a highly efficient manner and induce tolerance in NOD mice. Why dendritic cells are not able to establish tolerance long term in vivo remains to be resolved, but may be secondary to acquired deficits in APC function. This, however, does not appear to be the case as highly purified NOD dendritic cells are highly potent APCs. A potential explanation may be that aberrant PGS2 regulation and prostaglandin production by NOD macrophages severely impairs tolerogenic mechanisms such as AICD or perturbs dendritic cell function in a manner that favors the accumulation of autoreactive T cells over time. Indeed, peritoneal macrophages from the NOD markedly impair dendritic cell clustering of T cells and their subsequent activation (Clare-Salzler; manuscript in preparation). Furthermore, the effects of NOD macrophages on T cell activation is reversed by the inclusion of indomethacin in cell cultures.

CONCLUSIONS The examination of APC function by several laboratories has provided a better understanding of the important role of APC populations and APC associated molecules in the immunopathogenesis of diabetes. As a whole, these studies support the notion that APC, with a limited capacity to activate T cells, predispose the host to the development of autoimmunity. The determination of the genetic, immunologic, and environmental factors which contribute to APC

114

NOD Mice and Related Strains: Research Applications in Diseases

dysfunction is important in developing an overall understanding of the development of the autoimmune response. Our recent finding of abnormal PGS-2 expression in peripheral blood monocytes of humans at high risk for diabetes, (Litherland and Clare-Salzler; manuscript in preparation) parallels that of NOD macrophages and suggests that these two species may share common immunologic defects predisposing to autoimmune diabetes. The delineation of factors affecting APC in the NOD mouse may therefore aid in identifying potential targets for which interventive therapies can be developed in humans. ACKNOWLEDGMENTS Cited work from the authors laboratory has been supported by grants from the National Institutes of Health (NIDDK), and from the Juvenile Diabetes Foundation International. REFERENCES 1. Haskins K, Portas M, Bergman B, Lafferty K. Pancreatic islet-specific T-cell clones from non-obese diabetic mice. Proc Natl Acad Sci USA 1989; 86:8000-8004. 2. Serreze DV, Leiter EH. Defective activation of T suppressor cell function in Nonobese Diabetic mice. Potential relation to cytokine deficiencies. J Immunol 1988; 140:3801-3807. 3. Serreze DV, Leiter EH. Development of diabetogenic T cells from NOD/Lt marrow is blocked when and allo-H-2 haplotype is expressed on cells of hematopoietic origin, but not on thymic epithelium. J Immunol 1991; 147:1222-1229. 4. Langmuir P, Bridgett M, Bothwell A, Crispe I. Bone marrow abnormalities in the non-obese diabetic mouse. Internat Immunol 1989; 5:169-177. 5. LaFace DW, Peck AB. Reciprocal allogeneic bone marrow transplantation between NOD mice and diabetes-nonsusceptible mice associated with transfer and prevention of autoimmune diabetes. Diabetes 1989; 38:894-901. 6. Serreze DV, Hamaguchi K, Leiter EH. Immunostimulation circumvents diabetes in NOD/Lt mice. J Autoimmun 1990; 2:759-776. 7. Jacob DO, Aiso S, Michie SA, McDevitt HO, Acha-Orbea H. Prevention of diabetes in nonobese diabetic mice by tumor necrosis factor (TNF):similarities between TNF-a and interleukin 1. Proc Natl Acad Sci USA 1990; 87:968-972. 8. Campbell IL, Oxbrow L, Harrison LC. Reduction in insulitis following administration of IFN-g and TNF- a in the NOD mouse. J Autoimmun 1991; 4:249-262.

Immunopathogenic Roles of Antigen Presenting Cells in the NOD Mouse

115

9. Clare-Salzler MJ, Brooks J, Chai A, Van Herle K, Anderson C. Prevention of Diabetes in Nonobese Diabetic mice by dendritic cell transfer. J Clin Invest 1992; 90:741-748. 10. Samson, MF, Smilek DE. Reversal of acute experimental autoimmune encephalomyelitis and prevention of relapses by treatment with myelin basic protein peptide analogue modified to form long-lived peptide-MHC complexes. J Immunol 1995; 155:2737-2746. 11. Fairchild PJ, Wildgoose R, Atherton E, Webb S, Wraith DC. An autoantigenic T cell epitope forms unstable complexes with class II MHC: a novel route for escape from tolerance induction. Int Immunol 1993; 5:1151-1158. 12. Hosken N, Shibuya K, Heath AW et al. The effect of antigen dose on CD4+ T helper cell phenotype development in a T cell receptor-abtransgenic model. J Exp Med 1995; 182:1579-1584. 13. Murray JS, Pfeiffer C, Madri J, Bottomly K. Major histocompatibility complex (MHC) control of CD4 T cell subset activation. II. A single peptide induces either humoral or cell-mediated responses in mice of distinct MHC genotype. Eur J Immunol 1992; 22:559-565. 14. Constant S, Pfeiffer C, Woodard A, Pasqualini T, Bottomly K. Extent of T cell receptor can determine the functional differentiation of naive CD4+ T cells. J Exp Med 1995; 182:1591-1596. 15. Day MJ, Tse AG, Puklavec M, Simmonds SJ, Mason DW. Targeting autoantigen to B cells prevents the induction of a cell-mediated autoimmune disease in rats. J Exp Med 1992; 175:655-659. 16. Macatonia Se, Hosken NA, Litton M et al. Dendritic cells produce IL-12 and direct the development of Th1 cells from naive CD4+ T cells. J Immunol 1995; 154:5071-5079. 17. Kuchroo VK, Das MP, Brown JA, Ranger SS, Zamvil SS, Sobel RA, Weiner HL, Nabavi N, Glimcher LH. B7-1 but not B7-2 co-stimulatory molecules activate differentially the Th1/Th2 developmental pathways:application to autoimmune disease therapy. Cell 1995; 80:707-718. 18. Freeman GJ, Boussiotis VA, Anumanthan A et al. B7-1 and B7-2 do not deliver identical costimulatory signals, since B7-2 but not B7-1 preferentially co-stimulates the initial production of IL-4. Immunity 1995; 2:523-532. 19. Lenardo MJ. Interleukin 2 programs mouse ab T lymphocytes for apopotosis. Nature 1991; 353:858-861. 20. Liblau RS, Pearson CI, Shokat K, Tisch R, Yang X-D, McDevitt HO. High-dose soluble antigen:peripheral T-cell proliferation or apoptosis. Immunol Rev 142:193-208. 21. Kneitz B, Herrmann T, Yonehara S, Schimpl A. Normal clonal expansion but impaired Fas-mediated cell death and anergy induction in interleukin-2-deficient mice. J Immunol 1995; 25:2571-2577.

116

NOD Mice and Related Strains: Research Applications in Diseases

22. Parijs JV, Biuckians A, Ibragimov A, Alt FW, Willerford DM, Abbas AK. Functional responses and apoptosis of CD25 (IL-2Ra) deficient T cells expressing a transgenic antigen receptor. J Immunol 1997; 158:3738-3745. 23. Brunner T, Mogil RJ, LaFace D et al. Cell autonomous Fas (CD95)/Fas-ligand interaction mediates activation-induced apoptosis in T-cell hybridomas. Nature 1995; 373:441-448. 24. Nagata S, Goldstein P. The Fas Death Factor. Science 1995; 267:1449-1455. 25. Watanabe-Fukunaga R, Brannan CI, Copeland NG, Jenkins NA, Nagata S. Lymphoproliferation disorder in mice explained by defects in Fas antigen that mediates apoptosis Nature 1992; 356:314-317. 26. Ludviksson BR, Gray B, Strober W, Ehrhardt RO. Dysregulated intrathymic development in the IL-2 deficient mouse leads to colitisinducing thymocytes. J Immunol 1997; 158:104-111. 27. Willerford DM, Chen J, Ferry JA et al. Interleukin-2 receptor a chain regulates the size and content of the peripheral lymphoid compartment. Immunity 1995:3. 28. Zhang X, Brunner T, Carter L et al. Unequal death in T helper cell (Th1) and Th2 effectors: Th1, but not Th2 effectors undergo Fas/ FasL-mediated apoptosis. J Exp Med 1997; 185:1837-1849. 29. Ramsdell F, Seaman MS, Meuller RE et al. Differential ability of Th1 and Th2 T cells to express Fas ligand and to undergo activation-induced cell death. Int Immunol 1994; 6:1545-1553. 30. Pearson CI, vanEwijk W, McDevitt HO. Induction of apoptosis and T helper 2 (Th2) responses correlates with peptide affinity for the major histocompatibility complex in self-reactive T cell receptor transgenic mice. J Exp Med 1997; 185:583-599. 31. Lanoue A, Bona C, von Boehmer H, Sarukhan A. Conditions that induce tolerance in mature CD4+ T cells. J Exp Med 1997; 185: 405-414. 32. Carrasco-Marin E, Shimizu J, Kanagawa O, Unanue ER. The class II MHC I-Ag–7 from non-obese diabetic mice are poor peptide binders. J Immunol 1996; 156:450-458. 33. Kaufman DL, Clare-Salzler M, Tian J, Forsthuber T, Ting GSP et al. Spontaneous loss of T-cell tolerance to glutamate decarboxylase in murine insulin-dependent diabetes. Nature 1993; 366:69-72. 34. Fraser JD, Irving BA, Crabtree GR, Weiss A. regulation of interleukin2 gene enhancer activity by the T cell accessory molecule CD28. Science 1991; 251:313-316. 35. Sperling AI, Auger JA, Ehst DC, Rulifson IC, Thompson CB et al. CD28/B7 interactions deliver a unique signal to naive T cells that regulates cell survival but not early proliferation. J Immunol 1996; 157:3903-3917. 36. Jenkins MK, Mueller D, Schwartz RH, Carding S, Bottomly K et al. Induction and maintenance of anergy in mature T cells. Adv Exp

Immunopathogenic Roles of Antigen Presenting Cells in the NOD Mouse

117

Med Biol. 1991; 292:167-76. 37. Green JM, Noel PJ, Sperling AI, Walunas TL, Gray GS et al. Absence of B7-dependent responses in CD28-deficient mice. Immunity 1995; 1:501-508. 38. Lenschow DJ, Herold KC, Rhee L, Patel B, Koons A et al. CD28/B7 regulation of Th1 and Th2 subsets in the development of autoimmune diabetes. Immunity 1996; 5:285-293. 39. Linsley PS, Greene JL, Brady W, Bajorath J, Ledbetter JA, Peach R. Human B7-1 (CD80) and B7-2(CD86) bind with similar avidities but different kinetics to CD28 and CTLA-4 receptors. Immunity 1994; 1:793. 40. Boise LH, Minn AJ, Noel PH et al. CD28 co-stimulation can promote T cell survival by enhancing the expression of Bcl-XL. Immunity 1995; 3:87-98. 41. Tivol EA, Boyd SD, Mckeon S et al. CTLA4Ig prevents lymphoproliferation and fatal multiorgan tissue destruction in CTLA-4-deficient mice. J Immunol 1997; 158:5091-5094. 42. Schweitzer AN, Borriello F, Wong RCK et al. Role of costimulators in T cell differentiation; Studies using antigen-presenting cells lacking expression of CD80 or CD86. J Immunol 1997; 158:2713-2722. 43. Lenschow DJ, Ho SC, Sattar H et al. Differential effects of anti-B7-1 and anti-B7-2 monoclonal antibody treatment on the development of diabetes in the nonobese diabetic mouse. J Exp Med 1995; 181: 1145-1155. 44. Eynon EE, Parker DC. Parameters of tolerance induction by antigen targeted to B lymphocytes. J Immunol 1993; 151:2985-2964. 45. Liblau RS, Singer SM, McDevitt HO. Th1 and Th2 CD4+ T cells in the pathogenesis of organ specific autoimmune diseases. Immunol Today. 1995; 16:34-38. 46. Vella AT, Scherer MT. Shultz L, Kappler JW, Marrack P. B cells are not essential for peripheral T-cell tolerance. Proc Natl Acad Sci USA 1996; 93:951-955. 47. Serreze DV, Chapman HD, Varnum DS et al. B lymphocytes are essential for the initiation of T cell-mediated autoimmune diabetes: analysis of a new “speed congenic” stock of NOD. Igm null mice. J Exp Med 1996; 184:2049-2053. 48. Inaba K, Steinman RM. Accessory cell T lymphocyte interactions. Antigen-dependent and independent clustering. J Exp Med 1986; 163:247-261. 49. Vasquez MA, Sicher SC, Proctor ML, Crowley JC, Lu CY. Differential regulation of Ia expression and antigen presentation by listeriolysin-producing versus non-producing strains lf listeria monocytogenese. J Leukoc Biol 1996; 59:683-690. 50. Vlassara H, Brownlee M, Cerami A. Novel macrophage receptor for glucose-modified protein is distinct from previously described scavenger receptors. J Exp Med 1986; 164:1301-1309.

118

NOD Mice and Related Strains: Research Applications in Diseases

51. Serreze DV, Gaedeke JW, Leiter EH. Hematopoietic stem-cell defects underlying abnormal macrophage development and maturation in NOD/Lt mice:defective regulation of cytokine receptors and protein kinase C. Proc Natl Acad Sci USA 1993; 90:9625-9629. 52. Serreze DV, Gaskins HR, Leiter EH. Defects in the differentiation and function of antigen presenting cells in NOD/Lt mice. J Immunol 1993; 150:2534-2543. 53. Greenbaum LA, Horowitz JB, Woods A, Pasqualini T, Reich E-P, Bottomly K. Autocrine growth of CD4+ T cells:differential effects of IL-1 on helper and inflammatory T cells. J Immunol 1988; 140: 1555-1560. 54. Jansen A, Homo-Delarche F, Hooijkaas H, Leenen PJ, Dardenne M, Drexhage H. Immunohistochemical characterization of monocytesmacrophages and dendritic cells involved in the initiation of the insulitis and b cell destruction in NOD mice. Diabetes 1994; 43:667-675. 55. Lo D, Reilly CR, Scott B, Liblau R, McDevitt HO, Burlky DC. Antigen presenting cells in adoptively transferred and spontaneous autoimmune diabetes. Eur J Immunol 1993; 23:1693-1698. 56. Hutchings P, Rosen H, O’Reilly L, Simpson E, Gordon S et al. Transfer of diabetes in mice prevented by blockade of adhesion-promoting receptor on macrophages. Nature 1990; 348:639-642. 57. Ihm S-H, Yoon J-W. Studies on autoimmunity for initiation of b cell specific cytotoxic effectors and insulitis in NOD mice. Diabetes 1990; 39:2730-2738. 58. Nerup J, Mandrup-Poulsen T, Molvig J et al. Mechanisms of pancreatic beta-cell destruction in type 1 diabetes. Diabetes Care 1988; 11suppl:16-23. 59. McDaniel ML, Kwon G, Hill Jr et al. Cytokines and nitric oxide in islet inflammation and diabetes. Proc Soc Exp Biol Med 1996; 211:24-32. 60. Rothe H, Jenkins NA, Copeland NG, Kolb H. Active stage of autoimmune diabetes is associated with the expression of a novel cytokine, IGIF, which is located near IDD2. J Clin Invest 1997; 99:469-474. 61. O’Brien BA, Harmon BV, Cameron DP, Allan DJ. Apoptosis is the mode of b cell death responsible for the development of IDDM in the nonobese diabetic (NOD) mouse. Diabetes 1997; 46:750-757. 62. Phillips TA, Kujubo DA, Herschman HR et al. The mouse macrophage activation associated marker protein p71/73 is an inducible prostaglandin endoperoxide synthase (cyclooxygenase). J Leuk Biol 1994; 270:1340-1344. 63. Masferrer JL, Reddy S, Zweifel BS et al. In vivo glucocorticoids regulate cyclooxygenase 2 but not cyclooxygenase 1 in peritoneal macrophages. J Pharm Exp Theraput 1994; 270:1340-1344. 64. Yokono K, Kawase Y, Nagata M et al. Suppression of concanavalin

Immunopathogenic Roles of Antigen Presenting Cells in the NOD Mouse

65.

66.

67.

68.

69. 70.

71. 72.

119

A-induced responses in splenic lymphocytes by activated macrophages in the non-obese diabetic mouse. Diabetologia 1989; 32:67-73. Letty MA, Coulaud J, Bens M et al. Enhanced metabolism of arachidonic acid by macrophages from the nonobese diabetic (NOD) mouse. Clin Immunol and Immunopath 1992; 64:188-196. Benhamou PY, Mullen Y, Clare-Salzler M et al. Essential fatty acid deficiency prevents autoimmune diabetes through a positive impact on antigen presenting cells and Th2 lymphocytes. Pancreas 1995; 11:26-37. Lu L, Pelus LM, Broxmeyer HE. Modulation of the expression of HLA-DR (Ia) antigen and the proliferation of human erythroid (BFU-3) and multipotential (CFU-GEMM) progenitor cells by prostaglandin E. Exp Hematol 1984; 12:741-748. Anastassiou ED, Paliogianni F, Balow JP et al. Prostaglandin E2 and other cyclic AMP-elevating agents modulate IL-2 and IL-2R (IL-2Ragene expression at multiple levels. J Immunol 1992; 148: 2845-2852. Cook SJ, McCormick F. Inhibition by cAMP of Ras-dependent activation of Raf. Science 1993; 262:1069-1072. Goetzl EJ, An S, Zeng L. Specific suppression by prostaglandin E2 of activation-induced apoptosis of human CD4+CD8+ T lymphoblasts. J Immunol 1995; 154:1041-1047. Lanzavecchia A. Mechanism of antigen uptake for presentation. Curr Opin Immunol 1996; 8:348-54. Inaba K, Witmer Pack M, Inaba M, Hathcock KS, Sakuta H et al. The tissue distribution of the B7-2 constimulator in mice; abundant expression on dendritic cells in situ and during maturation in vitro. J Exp Med 1994; 180:1849-1860

CHAPTER 5

The Natural History of Islet-Specific Autoimmunity in NOD Mice Jean-François Bach

T

he NOD mouse develops a strong and diversified immune response against islet cell autoantigens. This immune response involves effector cells that contribute to the β cell lesion at the origin of insulin dependent diabetes mellitus (IDDM) and regulatory cells that modulate the emergence of pathogenic cells. The aim of this chapter is to review the natural history of the islet specific response in order to define its relevance to the pathophysiology of the disease.

THE MULTI-FACETED ISLET SPECIFIC RESPONSE AUTOANTIBODIES Autoantibodies to several islet cell antigens have been described in the NOD mouse, including insulin,1-3 glutamic acid decarboxylase (GAD),4,5 peripherin6 and various ill-defined β cell antigens.3,7,8 Antiinsulin monoclonal antibodies,9-10 as well as monoclonal antibodies showing various unidentified specificities,9 have been derived from NOD mice. Using indirect immunofluorescence, islet cell antibodies (ICA) and islet cell surface antibodies (ICSA) have also been identified in these animals.3 These latter two autoantibodies are, however, far less well documented and clear-cut in the NOD mouse than in NOD Mice and Related Strains: Research Applications in Diabetes, AIDS, Cancer and Other Diseases, edited by Edward Leiter and Mark Atkinson. © 1998 R.G. Landes Company.

122

NOD Mice and Related Strains: Research Applications in Diseases

human IDDM. Indeed, most such descriptions involved detection by ELISA or tissue binding; two techniques which are often questionable with respect to their reproducibility and biological significance (i.e., difficult to distinguish immune autoantibodies from polyspecific low-affinity natural autoantibodies). This comment especially applies to an islet binding IgM autoantibody that appears early before insulitis.7 Consistent with a finding reported by Velloso for anti-GAD antibodies,1 we have been unable to detect anti-insulin or anti-GAD autoantibodies in NOD mice using (respectively) the radioimmunoassay and the radioligand assay successfully used for human diabetic sera (our unpublished data). However, some studies have reported positive findings utilizing NOD serum with radioimmunoassays for insulin and immunoprecipitation for GAD.2,5 A resolution of this discrepancy may involve comparing antibodies to GAD65 and GAD67 as well as autoantibodies of the various immunoglobulin (Ig)G isotypes. As far as the ontogeny of autoantibody production is concerned, anti-insulin antibodies have been observed as early as 5 weeks and anti-GAD antibodies at 50 days of age.1,5 One of the most persistent questions in islet autoantigen research involves quantification of GAD expression in NOD mouse islets. While experimentally limited, through their use of whole pancreas, some studies have suggested the RNA message in islets appears early and at a higher level in males than in females; this has opened the possibility that GAD induced self-tolerance is less prominent in females due to unfavorable hormonal regulation of GAD expression.4,12 At the protein level, GAD expression appears lower in all mouse strains tested compared to rat and human islets,11 and has most often been undetectable despite the use of numerous techniques including immunoprecipitation, Western blotting, and detection of enzymatic activity.4,11 Another important issue is that involving the pathogenic role of these autoantibodies. The fact that one can obtain diabetes transfer by injecting purified T cell populations from diabetic mice into NOD mice rendered agammaglobulinemic by perinatal anti-µ monoclonal antibody therapy argues against any pathogenic role, except perhaps for that involving autoantigen presentation.13 One may note, however, that the latter hypothesis would only apply at the macrophage level to Th1 cell-dependent IgG2a autoantibodies since NOD mice show defects in expression of FcγrII isoforms that mediate the binding of IgG1 and IgG2b antibodies.13a It is also interesting to note that the driving of IgG1 and IgG2b anti-hsp60 peptide autoantibodies in

The Natural History of Islet-Specific Autoimmunity in NOD Mice

123

NOD mice immunized against this peptide were subsequently found to be protected from diabetes (contrasting with the IgG2a autoantibodies present in non-immunized NOD mice) (Elias D, personal communication). T CELLS T cell proliferative responses to various β cell autoantigens (or peptides) have been demonstrated in young NOD mice.14-18 An early response to defined autoantigens is observed against GAD (three weeks) at the time of insulitis onset, but proliferative activity of T cells can be found in the spleen before that age (two weeks) with whole islet cell extracts.14,15,18 During the second month of life, the response to other defined autoantigens appears; notably that against insulin,14 carboxypeptidase H,14,15 heat shock protein (HSP) 60,14,15,19 and peripherin.15 An important question has been posed regarding the nature of the autoantigens that are recognized in the islets as early as two weeks of age that do not correspond to the aforementioned candidate autoantigens. The autoantigens detected by lymphocyte proliferation assays in the presence of living islet cells or islet cell extracts are truly islet specific,16,17 but not necessarily NOD mouse specific since spleen cells from NOD mice recognize islet cells from other strains and other mouse strains may also show some response to syngeneic islets.16 The same question can be posed for many of the T cell clones derived from the spleen or islets; cells that do not recognize current candidate autoantigens (with the exception of insulin) for a large proportion of islet (but not spleen) derived clones.21 This concern is highlighted in a report by Bergman and Haskins describing T cell clones specific to still undefined β granule antigens.22 However, a question remains regarding the representative nature of these clones among islet reactive T cells involved in the disease pathogenesis when one considers pathogenic clones are able to transfer the disease to nondiabetic syngeneic recipients.23 With respect to pathogenic T cells, as evaluated in transfer models by their capacity to induce diabetes in neonatal irradiated or NOD-scid recipients, the most effective appear at 14-18 weeks of age (i.e., 2-4 weeks before the typical onset of diabetes). Cells from earlier stages of life in NOD mice (e.g., 6-8-week-old), while capable of transferring disease, are far less rapid in their ability to transfer disease. The expression of this pathogenicity generally involves both CD4+

124

NOD Mice and Related Strains: Research Applications in Diseases

and CD8+ T cells when polyclonal T cells are used.24,25 CD4+ T cell clones alone can induce diabetes at least in young mice or NOD-scid mice.23,26-29 Tc ell clones (usually of CD4+ phenotype) which transfer disease in young mice (less than three weeks),27 even after CD8 monoclonal antibody therapy,28 transfer diabetes in NOD-scid mice but have never been shown to do so in irradiated recipients.29 Importantly, effector cells are potentially present as early as 5-6 weeks of age since cyclophosphamide administered at that age rapidly induces diabetes onset (i.e., 4-8 weeks before the acquisition of transfer capacity by splenocytes).30 It is also important to note that CD8+ T cell clones have also been reported to transfer disease in irradiated recipients;31 even when these clones were derived in special conditions from islet derived T cells incubated in the presence of B7 expressing β cells. Similarly, CD8+ T cells or clones have been reported to lyse chromium labeled β cells.31-34 Finally, it has not been experimentally possible to transfer diabetes with Th2 type clones or with clones from NOD T cell receptor (TCR) transgenics incubated in the presence of interleukin (IL)-4.35,36 The effective transfer of disease has also not been reported with the few GAD specific clones derived from untreated or GAD immunized NOD mice.37 In summary, it appears that both CD4+ and CD8+ T cells are required to achieve transfer of disease. However, CD4+ T cell clones, and to a lesser degree, CD8+ T cell clones, alone can also induce disease when injected in sufficient amounts in immunoincompetent (but usually non-irradiated) recipients. One may postulate that CD4+ and CD8+ T cells collaborate to generate effector cells, whether of the CD4 or CD8 phenotype, and that either of these two populations can achieve transfer alone when injected in large number as is made possible by the use of T cell clones or TCR transgenic mice. The respective functional contribution of CD4+ and CD8+ T cells in the diabetes pathogenesis will be discussed later in this chapter. THE TIME COURSE OF INSULITIS Insulitis is the hallmark pathologic lesion of the NOD mouse diabetes. It evolves through several stages that include peri-insulitis, peripheral insulitis, invading insulitis, and destructive insulitis. Periinsulitis is first observed at three weeks of age.38 It is controlled by a dominant gene located on Chromosome 1 (close to the Bcl2 gene)

The Natural History of Islet-Specific Autoimmunity in NOD Mice

125

that also codes for sialitis (i.e., an inflammatory lesion of the salivary gland).39 Peri-insulitis is also observed in H-2 congenic NOD mice with H-2k.40 Invading insulitis starts at 8-10 weeks, with destructive insulitis appearing not long before the onset of diabetes.41 Sequential quantitative immunohistochemical studies of the endocrine pancreas have indicated that the number of islets and the volume density of endocrine components were only reduced in mice showing fully invasive insulitis (i.e., heavy mononuclear cell infiltrate). This usually corresponds to clinically overt diabetes and less commonly to isolated glucose metabolism disturbance without glycosuria.41,42 There is still a global correlation between the intensity of the infiltrate and the β cell function in males in spite of moderate to heavy cellular infiltration in absence of diabetes in many mice. The functional significance of a report describing abnormalities of endothelial cells and microvasculature in the islet at an early age (2-3 weeks) before insulitis onset is unclear.43 INFILTRATE PHENOTYPE Several studies have documented the phenotype of mononuclear cells that infiltrate the islets at the different stages of insulitis. Lymphoid cells predominate and include both B and T cells.44-46 Among Tc ells, CD4+ cells tend to predominate over CD8+ cells at all stages (including prediabetes) and include a tendency for a progressive increase with age of the CD4/CD8 ratio.46-48 CD4- CD8- double negative T cells are also found in significant amounts in this lesion.48,49 Numerous activated T cells are detected as assessed by the presence of the interleukin-2 receptor (CD25).50,51 Studies of lymphoid cells isolated from the islets do not provide significant differences with respect to cryostat (i.e., histology) based studies.49,51 The analysis of macrophages has generated more controversial data. Some authors report an early influx of macrophages preceding the appearance of lymphoid cells.52,53 The triggering role of macrophages is also suggested by their capacity to lyse islet cells and by the prevention of diabetes offered through silica treatment.52 Other authors also reported the early presence of macrophages in the pancreas before T cell appearance;54 however, these were exclusively in the exocrine or perivascular areas at distance from the islets. It is only after T cells fully infiltrate the islets that macrophages are observed within the islets. Lastly, other authors have not reported the presence of macrophages at initial phases of insulitis.47

126

NOD Mice and Related Strains: Research Applications in Diseases

T CELL REPERTOIRE Meaningful data on the repertoire of islet infiltrating T cells have been generated using various techniques. At variance with data reported in some experimentally induced autoimmune diseases (notably experimental allergic encephalomyelitis where encephalitogenic T cell clones obtained by immunization with myelin basic protein use restricted Vβ and Vα TCR genes),55 no clear restriction of V gene usage has been observed in islet derived T cells analyzed by semiquantitative polymerase chain reaction (PCR), even at the early stages of insulitis.56 However, using anchored PCR, it was found in young NOD mice (4-8 weeks) that T cells derived from a single islet showed more restricted junctional (VDJ) sequence variability than overall pancreas derived cells, suggesting the possibility of intra-islet initial clonal expansion.57 One recent study has indicated at a very early age (two weeks) the predominance of a defined single T cell receptor.58 However, this postulated T cell monoclonality should be viewed with caution as very few cells are present at that stage in islets and technical problems may arise when major gene amplification is performed. The only two other studies pointing to repertoire restriction were based on diabetes prevention by anti-Vβ monoclonal antibodies. An anti-Vβ8 monoclonal antibody was found to prevent cyclophosphamide-induced diabetes, 59 but this result was not reproduced by another group.60 Another study indicated an anti-Vβ6 monoclonal antibody inhibited diabetes transfer in irradiated mice.61 These relatively negative results do not support the hypothesis of a single epitope driven restricted T cell response in NOD mice or a role for a superantigen. Rather, they suggest a polyclonality of the antiislet immune response. GENE EXPRESSION Cytokines Major efforts have recently been devoted to the study of cytokines produced by islet infiltrating cells, either by RT PCR, immunofluoresence or immunoenzymology. Interferon-gamma (IFNγ)+ and tumor necrosis factor (TNF)+ cells are readily seen at the stage of invading insulitis without IL-4+, IL-10+ and TGFβ+ cells.62-64 The latter cells are observed after treatment with complete Freund’s adjuvant (CFA)(i.e., IL-4) or oral insulin (i.e., IL-4, IL-10, TGFβ)62,65—two procedures that prevent the appearance of diabetes. However, slightly

The Natural History of Islet-Specific Autoimmunity in NOD Mice

127

different data have been obtained for CFA by Rabinovitch et al which shows depressed IFNγ and IL-2 expression, increased IL-10 expression, and no effect on IL-4 and IL1-β.66 Using intracellular immunofluorescence staining, Pilström et al reported the presence in prediabetic mice (6-12 weeks) of IL-1α, IL-6 and TNF (before stimulation by mitogens) as well as that of IFNγ (but not IL-2 and IL-4, after stimulation).63 In older mice (4-6 months) there was a clear Th1 profile with IL-2, IFNγ and TNF (after stimulation), but no IL-4 and very little IL-10.63 Another report has indicated high expression of TNFα and granzyme A (a serine protease associated with cytoplasmic granules of cytotoxic T cells) in islet infiltrating cells.67 It is important to note that most of these experiments were performed on islet grafts rather than on the whole initial pancreas in order to limit the proteolytic effect of exocrine pancreas enzymes. Therefore, these data should be considered as preliminary since a good, quantitative evaluation of intra-islet contents of cytokines has yet to be reported. This major pitfall may explain the surprising results reported by Anderson et al indicating a predominance of Th2 cytokines in CD4+ cells (IL-4, IL-5, IL-10) with IFNγ in CD8+ cells.68 A recent quantitative RT PCR analysis of cytokine genes expressed in islet-infiltrating leukocytes has recently appeared, showing diabetes resistance in NOD males associated with continued expression of Th2 cytokine genes in islet-infiltrating lymphocytes as contrasted to predominant IFNγ gene expression over time in lymphocytes in female islets (Fox C, Danska J. IL-4 expression at the onset of islet inflammation predicts nondestructive insulitis in Nonobese Diabetic mice. J Immunol 1997; 158:2414-2424). MHC Molecules Because of the well established role of class I and class II major histocompatibility complex (MHC) molecules in the presentation of antigens to T cells, several studies have been devoted to the evaluation of class I and class II molecule expression in the islets of NOD mice. Converging data have been reported indicating increased expression of class I; either in the natural course of the disease or after cyclophosphamide treatment.69-71 The reported decrease in MHC class I expression by lymphoid cells of NOD mice is still a matter of debate.72 Another controversy concerns the expression of class II MHC molecules in endocrine cells; a finding reported by some but not all authors.44,50,70,73,74 Indeed, some studies observed class II MHC expression exclusively on infiltrating macrophages and B cells.50,70,74 Finally,

128

NOD Mice and Related Strains: Research Applications in Diseases

MHC class II expression has been observed in endothelial cells after transfer of diabetogenic T cells.75 Taken together, these data argue against the primary role of aberrant MHC class II expression for triggering the islet-specific pathogenic response. Other Molecules Additional molecules of interest include the report of an increased synthesis of the reg protein, a β cell regeneration associated protein in the islets of diabetic mice at the onset of diabetes.76 Interpreting this study is, however, complicated by the use of pancreas and the non-endocrine expression of reg. Various adhesion and addressin molecules may also play a role in the homing and binding of Tc ells to islets.77,78 For example, cells mediating diabetes transfer lack Lselectin expression.79 The anti-MEL-14 monoclonal antibody that recognizes L-selectin as well as an anti-VLA-4 antibody can both be utilized to delay the onset of insulitis.80,81 A role for α4-integrin VCAM-1 in the homing of T cells in the islets is also suggested by the protective effect of a monoclonal antibody directed to this molecule in spontaneous and T cell transferred diabetes.82

THE NATURE OF THE β CELL LESION It had initially been thought that most of the failure of insulin production which defines insulin dependent diabetes (IDDM) was due to β cell destruction induced by the autoimmune process. This is true in the long-term after several years of diabetes in humans. However, data obtained in the NOD mouse has revealed that this β cell atrophy is preceded by a long period of islet inflammation that inhibits insulin production in a reversible manner. This mechanism is demonstrated by the partial but very spectacular recovery of insulin production after in vitro culture of islets from diabetic NOD mice and by the capacity of an anti-T cell monoclonal antibody to induce very rapid normalization of hyperglycemia in overtly diabetic NOD mice.83-85 The fact that anti-TCR, anti-CD4 and anti-CD8 monoclonal antibodies all reverse insulitis and hyperglycemia indicates that both CD4+ and CD8+ cells participate in the pancreatic islet inflammation. The observation that hemipancreatectomy in prediabetic NOD mice does not accelerate the age of onset of diabetes (Boitard C, in preparation) argues in the same direction. An interesting question is to determine when β cell atrophy starts and becomes biologically significant. The systematic study of β cell morphometry discussed above

The Natural History of Islet-Specific Autoimmunity in NOD Mice

129

addresses the insulin content but does not directly address the number of persisting β cells. Concerning the intimate mechanisms of the lesion induction, a debate persists regarding the nature of the involvement of T cells. The overall responsibility of CD8+ T cells in the pathogenesis of IDDM is indicated by transfer experiments described above which indicated that CD8+ cells are necessary for diabetes induction when transferring polyclonal T cells. This point has been demonstrated in several types of recipients: non-irradiated young mice,24 adult irradiated mice,25 athymic nude mice86,87 and NOD-scid mice.88,89 Purified CD4+ T cells from diabetic or prediabetic donors can transfer diabetes in NOD-scid mice, but the onset of diabetes is delayed compared to that induced by total T cells and is prevented by in vivo recipient treatment by an anti-CD8 monoclonal antibody.88 Such findings argue against the intrinsic capacity of polyclonal CD4+ cells to transfer diabetes alone in this model. This contention has also been demonstrated in a more direct fashion by the production from NOD islet-derived Tc ells of CD8+ cytotoxic T lymphocytes (CTL) following incubation with islet cells in the presence of IL-2.33 These CTL adhere to islets and induce their destruction as assessed by morphology and chromium release assay. Both CTL lines and clones could be derived in two substrains of NOD mice. These lines and clones revealed MHC class I restriction for either H-2Db or H-2Kd. The cells could not transfer diabetes in irradiated recipients alone, but could do so in the presence of spleen cells from diabetic mice previously depleted of CD8+ cells.34 Another study has demonstrated similar morphological islet lesions.90 In the same vein, CD8+ cytotoxic clones derived from NOD islet cell cultures in the presence of B7 expressing islets transfer diabetes in irradiated recipients.31 Lastly, islet infiltrating cells express both granzyme and perforin,67,91 two cytolytic proteins found in granules of CTL and NK cells. Transfer experiments performed in irradiated or athymic nude mice have shown that CD4+ T cells could create non-invasive insulitis when injected alone while CD8+ purified T cells did not home to islets in the absence of CD4+ cells.5,86,87 Contrary to these findings, islet reactive CD4+ T cell clones can transfer diabetes in young NOD mice and in NOD-scid mice.20-23,29 Additionally, using syngeneic islet grafts as a read-out, Wang et al have reported that β cell destruction is exclusively inhibited by anti-CD4 and not anti-CD8 antibody therapy, and does not require MHC compatibility at class I MHC as would be expected if CD8+ T cells had

130

NOD Mice and Related Strains: Research Applications in Diseases

induced β cell destruction.92 Charlton et al have reported that antiCD4 monoclonal antibody could prevent the onset of cyclophosphamide induced diabetes,93 but data obtained in our laboratory (L. Chatenoud, unpublished results) have indicated that either anti-CD4 or anti-CD8 antibodies can achieve such protection. In sum, it is difficult to reconcile all of these observations. It appears that both CD4+ T cell derived cytokines and CD8+ T cell mediated cytotoxicity can induce the β cell lesion. It remains to be determined which of these cells predominate in the spontaneous diabetes in unmanipulated NOD mice.

HYPOTHESES ON THE TRIGGERING OF THE β CELL SPECIFIC RESPONSE The driving role of β cell autoantigens in triggering the diabetogenic T cell response is indicated by experiments involving βc ell destruction induced by alloxan. NOD mice devoid of β cells lose their capacity to sustain the survival and expansion of diabetogenic T cells.94 The specificity of the autoantigens recognized by these Tc ells remains uncertain though in spite of the discovery of a number of candidate autoantigens (e.g., GAD, insulin and HSP 60). It is unclear whether one of these autoantigens plays a primary role while the other autoantigens become immunogenic through a mechanism of spread sensitization. The latter forms as the local inflammation induced by the initial β cell targeted reaction enhances the immunogenicity of the other β cell constituents. In any case autoantigen presentation remains under MHC gene control and its chronic nature most likely implicates a low or moderate Th2 type response; this poses the questions regarding the existence of a Th2 cell dysregulation and more generally, of thymus and T cell abnormalities in NOD mice. TRANSIENT EARLY PROTECTION FROM DIABETES BY CD4 IMMUNOREGULATORY T CELLS Although insulitis is first detected at three weeks of age, diabetes does not appear in female NOD mice before 3-4 months of age (and even later in males). It is at this very age (i.e., 3-4 weeks) that the transfer of diabetes by spleen cells from diabetic mice cannot be achieved without irradiating the recipient.24 Similar age dependence resistance has been reported for islet specific T cell clones.27 These data suggest the appearance of a “protector” cell subset. This subset appears CD4+ since administration of an anti-CD4 monoclonal anti-

The Natural History of Islet-Specific Autoimmunity in NOD Mice

131

body (plus adult thymectomy) can be substituted for irradiation in order to obtain diabetes transfer in adult mice.95 Other evidence for the existence of these regulatory cells is suggested by the acceleration of diabetes onset induced by thymectomy performed at three weeks of age and cyclophosphamide treatment.30,71,96-100 Cyclophosphamide induces diabetes within 2-4 weeks after one (or two) injection(s) of 200 mg/kg of the compound. The effect is observed in both males and females, and even in colonies with a very low disease incidence.98 The predominant action of cyclophosphamide in this effect on regulatory T cells is indicated by the fact that: 1) the effect is prevented by injection of non-drug treated NOD spleen cells following cyclophosphamide;98 2) the disease appears on fetal islet grafts placed after cyclophosphamide administration;98 and 3) lymphoid cells from male mice acquire the capacity to transfer diabetes after cyclophosphamide treatment.30,100 Cyclophosphamide induced diabetes is indeed due to T cells since its onset is prevented by monoclonal antibodies directed against CD3,101 CD4,93 and CD8 (Chatenoud L, unpublished observations). The drug revealed no diabetogenic effect in 11 control strains;98 however, little data are available on the precise T cell subset affected by the drug. In the thymus of treated animals, a transient increase of CD3+ cells and a decrease of CD4+ CD8+ double positive cells has been noted.99 The number of splenic T cells is also transiently reduced without a major bias in T cell subsets.99 Diabetes onset is associated with increased expression of IFN(and nitric oxide synthase) without increases of IL-2 and IL-4.102 Interestingly, the IL-12 message is detected before increased IFNβ expression at both the levels of spleen macrophages and islets.103 It is also important to note that diabetes onset can be accelerated in female NOD mice by various interventions in the B7-CD28/CTLA4 receptor ligand system: administration of anti-B7.1 monoclonal antibody,104 inactivation of the CD28 gene by homologous recombination (Bluestone J, personal communication), and transgenic expression of CTLA4-Ig (Bluestone J, personal communication). The direct evidence for the protector role of CD4+ T cells is provided by the prevention of diabetes transfer afforded by CD4+ T cells from prediabetic NOD mice.105 It is also interesting to note that the protective CD4+ T cells appear at 3-4 weeks of age and disappear after thymectomy performed at three weeks and spontaneously at 3-4 months.105 The demonstration of such T cell mediated suppression has been confirmed by other investigators in both irradiated recipients

132

NOD Mice and Related Strains: Research Applications in Diseases

and NOD mice.89,106-108 The nature of this protector subset is not known but indirect evidence argues in favor of its belonging to the Th2 subset. Specifically, the prevention of diabetes offered by anti-IFNβ monoclonal antibody therapy,109,110 IL-4111 and IL-10,112 as well as the absence of diabetes in transgenic NOD mice expressing IL-4 in β cells (Sarvetnick N, personal communication) suggest that these protector cells could indeed be of the Th2 phenotype. This contention has, however, not been proven by the demonstration of the Th2 cytokine production by the regulatory T cells. Indeed, no evidence for disease protection through the few Th2 clones that have been obtained from NOD spleen cells has been reported.35 In particular, the role of TGFγ producing cells in such a process cannot be excluded. Interesting differences have been noted in the effect of thymectomy according to gender. Acceleration of diabetes onset is only seen in females but can be provoked in males by concomitant castration.113 These findings indicate that the generation of the protector function is under the negative control of androgens. This interpretation fits with the identical capacity of cyclophosphamide to induce diabetes in male and female NOD mice.97,100 Several groups have attempted to generate T cell lines or T cell clones with suppressor activity. Kelley, Strom and collaborators have generated a CD8+ clone specific for islet cells that protects from the onset of diabetes in vivo in two accelerated models of the disease.114 This clone was shown to be immunosuppressive in vitro and to act through the production of a partially characterized soluble factor.115,116 It is not clear whether this factor is related to IL-10 or TGFβ messages which were found in the clone. Additional authors have reported the protective activity of other clones either of the CD4+ or CD8+ phenotype.117-119 Variable results have been obtained in terms of their cytokine profile (in one case intriguingly Th1).117,118 Little is known regarding the antigen specificity of these clones except in the case of a report indicating anti-MHC class II reactivity.120

THE SEARCH FOR EARLY THYMUS AND BONE MARROW ANOMALIES: AN NK1+ T CELL DEFECT The question must be raised regarding the location of the cellular origin of the autoimmune response that ultimately leads to β cell destruction. Converging arguments indicate that the central defect is at the level of the lymphoid tissues and not at the pancreas. Diabetes in NOD mice can be prevented by allogeneic bone marrow trans-

The Natural History of Islet-Specific Autoimmunity in NOD Mice

133

plantation.121 Conversely, the disease can be induced by concomitant bone marrow and thymus transplantation from NOD mice into (BALB/c x C57BL/6)F1 mice or BALB/c mice.122,123 Negative results have, however, been reported in CBA and C57BL/6 mice, and even in BALBc mice.123,124 It is not clear whether non-antigen specific anomalies at early stages of lymphoid cell or monocyte differentiation,125,126 or pecularities of the thymus stroma (discussed below) play a significant role in the predisposition to autoimmunity. These abnormalities could favor immune dysregulation (independently of intrathymic Tc ell education and selection). As described above, at approximately 4-5 weeks the NOD mouse mounts an islet-specific response (under MHC control) which includes both effector cells (as revealed by the induction of diabetes by treatment by cyclophosphamide at this age) and downregulatory CD4+ T cells.30,105 The latter cells progressively disappear at 10-12 weeks of age in most (but not all) mice, allowing room for effector mechanisms illustrated by the appearance of diabetogenic potential in spleen cells, development of invading and destructive insulitis and finally, diabetes (Fig. 5.1). A vital question that remains involves determination of whether this sequence of events is genetically programmed and expressed at an early age in the thymus. Dardenne et al and others have reported the presence of structural anomalies in the thymus of NOD mice.127-129 There is in both males and females a precocious decrease in the number of discrete medullary thymic epithelial cells subset; namely, those respectively defined by the expression of cytokeratins 3/10 and cytokeratin 19. In addition, cells bearing the TR.5 phenotype (normally restricted to the medulla) are detected in the NOD mouse thymic cortex. One also notes an early decrease in thymulin production in females, as compared to males. With regard to the extracellular matrix compartment, abnormally enlarged perivascular spaces are observed which, in addition, increase in size with age. These structures contain large amounts of T cells and, to a lesser extent, B cells. More recently, it has been shown that in intraperivascular spaces, T cells did not recirculate from the periphery. Rather, such cells derive from the medulla and accumulate in the perivascular spaces as a result of an arrest, or at least a retardation, in their emigration pathway.130 The genetic and cellular control mechanisms related to the appearance of giant perivascular spaces formation have been documented. Such results indicate a dominant inheritability of the abnormality and the potential influence of the hematopoietic stromal microenvironment.131

134

NOD Mice and Related Strains: Research Applications in Diseases

Fig. 5.1. Ontogeny of immune dysregulation in NOD mice. Immunological events in NOD mice: disease sensitivity to thymectomy (Tx), cyclo-phosphamide, administration of thymocytes and/or splenocytes, requirements for irradiation (Rx) to obtain diabetes transfer, and NK1+-like T cell phenotype.

More recently, we have demonstrated a defect of the NK1+ thymocyte subset in young NOD mice. A potential immunoregulatory function has been attributed to the discrete subset of MHC class I-restricted TCR-αβ+ mature thymocytes expressing an unusual Vα14 Vβ8.2-biased TCR repertoire.132 This T cell subset, which also selectively expresses the CD44 marker, is the main IL-4 producer in the thymus. NOD mice were found to present a marked deficit in the number of CD44+ TCR-αβ+ thymocytes from as early as three weeks of age.133 This numerical defect is associated with an even more spectacular functional deficiency as expressed by reduced IL-4 production. The abnormality involves both DN and CD4+ single-positive subsets. Its correction by in vitro incubation with IL-7133 indicates that the defect observed is not located at the NK1-like T cell precursor level, but perhaps at that of intrathymic IL-7 production. At three weeks of age, the abnormality is the most conspicuous. It is also (approximately) at this age that: 1) NOD mice are sensitive to disease acceleration following thymectomy (and not later on);96 2) these mice become resistant to diabetes transfer in the absence of irradiation;24 and 3) insulitis begins. Thus, it appears that three weeks of age is a major check-point in T cell ontogeny in NOD mice. The problem is posed regarding the causal relationship between these early events and the ultimate emergence of the disease.

The Natural History of Islet-Specific Autoimmunity in NOD Mice

135

The central question remaining involves a determination of whether this early abnormality in T cell ontogeny contributes to the immune dysregulation that promotes islet specific autoimmunity in NOD mice. The control of Th2 cell differentiation by NK1+ thymocytes is indicated by the absence of Th2 dependent IgE production by MHC class I (β2m–/–) deficient mice.134 One may hypothesize that similarly, the abnormality in NK1+-like T cell development demonstrated in our study leads to defective islet specific Th2 cells. The in vitro correction of this deficit by IL-7133 is interesting inasmuch as it appears at both the phenotypic and the functional level.

CONCLUSIONS The data discussed in this chapter indicate a precise chronology for the appearance of the various cell types presumably involved in the pathogenesis of IDDM. Effector cells appear as early as six weeks of age, but do not express their pathogenic potential before 10-12 weeks of age due to a suppressive effect mediated by CD4+ regulatory cells. The precise nature of these regulatory cells and the mechanisms explaining their disappearance at the time of diabetes onset are still unknown. There is, however, evidence suggesting that their premature decline could relate to a primary thymus abnormality at the level of the stroma and/or of the NK1+ T cell subset. REFERENCES 1. Michel C, Boitard C, Bach JF. Insulin autoantibodies in non-obese diabetic (NOD) mice. Clin Exp Immunol 1989; 75:457-460. 2. Maruyama T, Takei I, Asaba Y et al. Insulin autoantibodies in mouse models of insulin-dependent diabetes. Diabetes Res 1989; 11:61-65. 3. Pontesillio O, Carotenuto P, Gazada LS et al. Circulating lymphocyte populations and autoantibodies in non-obese diabetic (NOD) mice: a longitudinal study. Clin Exp Immunol 1989; 70:84-93. 4. Velloso LA, Erizirik DL, Karlsson FA et al. Absence of autoantibodies against glutamate decarboxylase (GAD) in the non-obese diabetic (NOD) mouse and low expression of the enzyme in mouse islets. Clin Exp Immunol 1989; 96:129-137. 5. De Aizpurua HJ, French MB, Chosich N et al. Natural history of humoral immunity to glutamic acid decarboxylase in non-obese diabetic (NOD) mice. J Autoimmun 1994; 7:643-653. 6. Boitard C, Villa MC, Becourt C et al. Peripherin: an islet antigen that is cross-reactive with nonobese diabetic mouse class II gene products. Proc Natl Acad Sci USA 1992; 89:172-176. 7. Shieh DC, Cornelius JG, Winter WE et al. Insulin-dependent diabetes in the NOD mouse model. I. Detection and characterization of

136

8.

9. 10.

11.

12.

13.

13a.

14.

15.

16. 17.

18.

19.

20. 21.

NOD Mice and Related Strains: Research Applications in Diseases

autoantibody bound to the surface of pancreatic beta cells prior to development of the insulitis lesion in prediabetic NOD mice. Autoimmunity 1993; 15:123-135. Karounos, DG Thomas JW. Recognition of common islet antigen by autoantibodies from NOD mice and humans with IDDM. Diabetes 1990; 39:1085-1090. Supon P, Stecha P, Haskin SK. Anti-islet cell antibodies from NOD mice. Diabetes 1990; 39:1366-1372. Pleau JM, Marche PN, Serrano MPI et al. Evidence for antigen driven selection in two monoclonal auto-antibodies derived from nonobese diabetic mice. Mol Immunol 1993; 30:1257-1264. Velloso LA, Kampe O, Eizirik DL et al. Human autoantibodies react with glutamic acid decarboxylase antigen in human and rat but not in mouse pancreatic islets. Diabetologia 1993; 36:39-46. Pleau JM, Esling A, Bach JF et al. Quantitative analysis of glutamate decarboxylase (GAD 65) gene expression in NOD mouse pancreas. CR Acad Sci III 1995; 318:129-133. Bendelac A, Boitard C, Bedossa P et al. Adoptive T cell transfer of autoimmune nonobese diabetic mouse diabetes does not require recruitment of host B lymphocytes. J Immunol 1988; 141:2625-2628. Luan JJ, Monteiro RC, Sautes C et al. Defective FcγRII gene expression in macrophages of NOD mice: genetic linkage with upregulation of IgG1 and IgG2b in serum. J Immunol 1996; 157:4707-4716. Kaufman DL, Clare-Salzler M, Tian NJ et al. Spontaneous loss of Tcell tolerance to glutamic acid decarboxylase in murine insulin-dependent diabetes. Nature 1993; 366:69-72. Tisch R, Yang XD, Singer SM et al. Immune response to glutamic acid decarboxylase correlates with insulitis in non-obese diabetic mice. Nature 1993:72-75. Burtless SS, Trembleau S, Drexler K et al. Absence of T cell tolerance to pancreatic islet cells. J Immunol 1993;149:2185-2193. Nagata M, Yokono K, Hatamori N et al. The presence of splenic T cells specific for islet cell antigens in nonobese diabetic mice. Clin Immunol Immunopathol 1989; 53:171-180. Gelber C, Paborsky L, Singer S et al. Isolation of nonobese diabetic mouse T-cells that recognize novel autoantigens involved in the early events of diabetes. Diabetes 1984; 43:33-39. Elias D, Markovits D, Reshef T et al. Induction and therapy of autoimmune diabetes in the non-obese diabetic (NOD/Lt) mouse by a 65-kDa heat shock protein. Proc Natl Acad Sci USA 1990; 87: 1576-1580. Haskins K, Portas M, Bradley B et al. T-lymphocyte clone specific for pancreatic islet antigen. Diabetes 1988; 37:1444-1448. Wegmann DR, Gill RG, Norbury-Glaser M et al. Analysis of the spontaneous T cell response to insulin in NOD mice. J Autoimmun 1994; 7:833-843.

The Natural History of Islet-Specific Autoimmunity in NOD Mice

137

22. Bergman B, Haskins K. Islet-specific T-cell clones from the NOD mouse respond to beta-granule antigen. Diabetes 1994; 43:197-203. 23. Wegmann DR, Norbury-Glaser M et al. Insulin-specific T cells are a predominant component of islet infiltrates in pre-diabetic NOD mice. Eur J Immunol 1994; 24:1853-1857. 24. Bendelac A, Carnaud C, Boitard C et al. Syngeneic transfer of autoimmune diabetes from diabetic NOD mice to healthy neonates. Requirement for both L3T4+ and Lyt-2+ T cells. J Exp Med 1987; 166:823-832. 25. Miller BJ, Appel MC, O’Neil JJ et al. Both the Lyt-2+ and L3T4+ T cell subsets are required for the transfer of diabetes in nonobese diabetic mice. J Immunol 1988; 140:52-58. 26. Haskins K, McDuffie M. Acceleration of diabetes in young NOD mice with a CD4+ islet-specific T cell clone. Science 1990; 249:1433-1436. 27. Peterson JD, Pike B, Dallas-Pedretti A et al. Induction of diabetes with islet-specific T-cell clones is age dependent. Immunology 1995; 85:455-460. 28. Bradley BJ, Haskins K, LaRosa FG et al. CD8 T cells are not required for islet destruction induced by a CD4+ islet-specific T-cell clone. Diabetes 1992; 41:1603-1608. 29. Peterson JD, Haskins K. Transfer of diabetes in the NOD-scid mouse by CD4 T-cell clones. Differential requirement for CD8 T-cells. Diabetes 1996; 45:328-336. 30. Yasunami R, Debray-Sachs M, Bach JF et al. Ontogeny of regulatory and effector T-cells in autoimmune NOD mice. In: Shafrir, E, ed. Frontiers in diabetes research. Lessons from animal diabetes III. London: Smith-Gordon, 1990:88-93. 31. Wong FS, Visintin I, Wen L et al. CD8 T cell clones from young nonobese diabetic (NOD) islets can transfer rapid onset of diabetes in NOD mice in the absence of CD4 cells. J Exp Med 1996:67-76. 32. Shimizu J, Kanagawa O, Unanue ER. Presentation of beta-cell antigens to CD4+ and CD8+ T cells of non-obese diabetic mice. J Immunol 1993; 151:1723-1730. 33. Nagata M, Yokono K, Hayakawa M et al. Destruction of pancreatic islet cells by cytotoxic T lymphocytes in nonobese diabetic mice. J Immunol 1989; 143:1155-1162. 34. Nagata M, Santamaria P, Kawamura T et al. Evidence for the role of CD8+ cytotoxic T cells in the destruction of pancreatic beta-cells in nonobese diabetic mice. J Immunol 1994; 152:2042-2050. 35. Healey D, Ozegbe P, Arden S et al. In vivo activity and in vitro specificity of CD4+ Th1 and Th2 cells derived from the spleens of diabetic NOD mice. J Clin Invest 1995; 95:2979-2985. 36. Katz JD, Benoist C, Mathis D. T helper cell subsets in insulin-dependent diabetes. Science 1995; 268:1185-1188. 37. Matsuba H, Jitsukawa T, Yamagata N et al. Establishment of rat glutamic acid decarboxylase (GAD)-reactive T-cell clones from NOD mice. Immunol Lett 1994; 42:101-103.

138

NOD Mice and Related Strains: Research Applications in Diseases

38. Fujino-Kurihara H, Fujita H, Hakura A et al. Morphological aspects on pancreatic islets of non-obese diabetic (NOD) mice. Virchows Arch. B. Cell Pathol Incl Mol Pathol 1985;49:107-120. 39. Garchon HJ, Bedossa P, Eloy L et al. Identification and mapping to chromosome 1 of a susceptibility locus for peri-insulitis in non-obese diabetic mice. Nature 1991; 353(6341):260-262. 40. Wicker LS, Appel MC, Dotta F et al. Autoimmune syndromes in major histocompatibility complex (MHC) congenic strains of nonobese diabetic (NOD) mice. The NOD MHC is dominant for insulitis and cyclophosphamide-induced diabetes. J Exp Med 1991; 176:67-77. 41. Debussche X, Lormeau B, Boitard C et al. Course of pancreatic beta cell destruction in prediabetic NOD mice: a histomorphometric evaluation. Diabete Metab 1992; 20:282-290. 42. Signore A, Procaccini E, Toscano AM et al. Histological study of pancreatic beta-cell loss in relation to the insulitis process in the nonobese diabetic mouse. Histochemistry 1994; 101:263-269. 43. Yamamoto K, Miyagawa JI, Hanafusa T et al. Endothelial and microvascular abnormalities in the islet of non-obese diabetic (NOD) mice: transmission and scanning electron microscopic studies. Biomed Res 1992; 13:259-267. 44. Miyazaki A, Hanafusa, T, Yamada K et al. Predominance of T lymphocytes in pancreatic islets and spleen of pre-diabetic non-obese diabetic (NOD) mice: a longitudinal study. Clin Exp Immunol 1985; 60:622-630. 45. O’Reilly LA, Hutchings PR, Crocker PR, Simpson E et al. Characterization of pancreatic islet cell infiltrates in NOD mice: effect of cell transfer and transgene expression. Eur J Immunol 1991; 21:1171-1180. 46. Kay TW, Campbell IL, Harrison LC. Characterization of pancreatic T lymphocytes associated with beta cell destruction in the non-obese diabetic (NOD) mouse. J Autoimmun 1991; 4:263-276. 47. Reddy S, Wu D, Swinney C et al. Immunohistochemical analyses of pancreatic macrophages and CD4 and CD8 T cell subsets prior to and following diabetes in the NOD mouse. Pancreas 1995; 11:16-25. 48. Formby B, Hosszufalusi N, Chan E et al. Quantitative and functional analyses of spleen and in situ islet immune cells before and after diabetes onset in the NOD mouse. Autoimmunity 1992; 12:95-102. 49. Goldrath AW, Barber L, Chen KE et al. Differences in adhesion markers, activation markers, and TcR in islet infiltrating vs. peripheral lymphocytes in the NOD mouse. J Autoimmun 1995; 8:209-220. 50. Signore A, Pozzilli P, Gale EA et al. The natural history of lymphocyte subsets infiltrating the pancreas of NOD mice. Diabetologia 1989; 32: 282-289. 51. Jarpe AJ, Hickman MR, Anderson JT et al. Flow cytometric enumeration of mononuclear cell populations infiltrating the islets of Langerhans in prediabetic NOD mice: development of a model of autoimmune insulitis for type I diabetes. Reg Immunol 1990; 3:305-317.

The Natural History of Islet-Specific Autoimmunity in NOD Mice

139

52. Kolb H, Burkar TV, Appel SB et al. Essential contribution of macrophages to islet cell destruction in vivo and in vitro. J Autoimmun 1990; 3 Suppl 1:117-120. 53. Jansen A, Homo-Delarche F, Hooijkaas H et al. Immunohistochemical characterization of monocytes-macrophages and dendritic cells involved in the initiation of the insulitis and beta-cell destruction in NOD mice. Diabetes 1994; 43:667-675. 54. Reddy S, LiuW, Elliott RB. Distribution of pancreatic macrophages preceding and during early insulitis in young NOD mice. Pancreas 1993; 8:602-608. 55. Acha-Orbea H. T-cell receptors in autoimmune disease. In: Bach, JF ed. Monoclonal Antibodies and Peptide Therapy in Autoimmune Diseases. New York: Marcel Dekker 1993:131-142. 56. Sarkuhan A, Gombert JM, Olivi M et al. Anchored polymerase chain reaction based analysis of the V-beta repertoire in the non-obese diabetic (NOD) mouse. Eur J Immunol 1994; 24:1750-1756. 57. Sarukhan A, Bedossa P, Garchon HJ et al. Molecular analysis of TCR junctional variability in individual infiltrated islets of non-obese diabetic mice: evidence for the constitution of largely autonomous T cell foci within the same pancreas. Int Immunol 1995; 7:139-146. 58. Yang Y, Charlton B, Shimada A et al. Monoclonal T cells identified in early NOD islet infiltrates. Immunity 1996;4: 189-194. 59. Bacelj A, Charlton B, Mandel TE. Prevention of cyclophosphamideinduced diabetes by anti-V beta 8 T-lymphocyte-receptor monoclonal antibody therapy in NOD/Wehi mice. Diabetes 1989; 38:1492-1495. 60. Taki T, Yokono K, Amano K et al. Effect of T-cell receptor V betaspecific monoclonal antibodies on cyclophosphamide-induced diabetes mellitus in non-obese diabetic mice. Diabetologia 1993; 36:391-396. 61. Edouard P, Thivolet C, Bedossa P et al. Evidence for a preferential V beta usage by the T cells which adoptively transfer diabetes in NOD mice. Eur J Immunol 1993; 23:727-733. 62. Hancock WW, Polanski M, Zhang J et al. Suppression of insulitis in non-obese diabetic (NOD) mice by oral insulin administration is associated with selective expression of interleukin-4 and -10, transforming growth factor-beta, and prostaglandin-E. Am J Pathol 1995; 147:1193-1199. 63. Pilstrom B, Bjork L, Bohme J. Demonstration of a TH1 cytokine profile in the late phase of NOD insulitis. Cytokine 1995; 7:806-814. 64. Rabinovitch A, Suarez-Pinzon WL, Sorensen O et al. IFN-gamma gene expression in pancreatic islet-infiltrating mononuclear cells correlates with autoimmune diabetes in nonobese diabetic mice. J Immunol 1995; 154:4874-4882. 65. Shehadeh NN, Larosa F, Lafferty KJ. Altered cytokine activity in adjuvant inhibition of autoimmune diabetes. J Autoimmun 1993; 6:291-300.

140

NOD Mice and Related Strains: Research Applications in Diseases

66. Rabinovitch A, Sorensen O, Suarez-Pinzon WL et al. Analysis of cytokine mRNA expression in syngeneic islet grafts of NOD mice: interleukin 2 and interferon gamma mRNA expression correlate with graft rejection and interleukin 10 with graft survival. Diabetologia 1994; 37:833-837. 67. Held W, Macdonald HR, Weissman IL et al. Genes encoding tumor necrosis factor alpha and granzyme A are expressed during development of autoimmune diabetes. Proc Natl Acad Sci USA 1990; 87:2239-2243. 68. Anderson JT, Cornelius JG, Jarpe A et al. Insulin-dependent diabetes in the NOD mouse model. II. Beta cell destruction in autoimmune diabetes is a TH2 and not a TH1 mediated event. Autoimmunity 1993; 15:113-122. 69. Calafiore R, Pietropaolo M, Basta G et al. Pancreatic beta-cell destruction in non-obese diabetic mice. Metabolism 1993; 42:854-859. 70. Mcinerney MF, Rath S, Janeway CA. Exclusive expression of MHC class II proteins on CD45+ cells in pancreatic islets of NOD mice. Diabetes 1991; 40:648-651. 71. Kay TW, Campbell IL, Oxbrow L et al. Overexpression of class I major histocompatibility complex accompanies insulitis in the nonobese diabetic mouse and is prevented by anti-interferon-gamma antibody. Diabetologia 1991; 34:779-785. 72. Faustman D, LI XP, Lin HY et al. Linkage of faulty major histocompatibility complex class I to autoimmune diabetes. Science 1991; 254:1756-1761. 73. Hanafusa T, Fujino-Kurihara H, Miyazaki A et al. Expression of class II major histocompatibility complex antigens on pancreatic B cells in the NOD mouse. Diabetologia 1987; 30:104-108. 74. Signore A, Cooke A, Pozzilli P et al. Class-II and IL2 receptor positive cells in the pancreas of NOD mice. Diabetologia 1987; 30:902-905. 75. Thivolet C, Bendelac A, Bedossa P et al. CD8+ T cell homing to the pancreas in the nonobese diabetic mouse is CD4+ T cell-dependent. J Immunol 1991; 146:85-88. 76. Baeza NJ, Moriscot CI, Renaud WP et al. Pancreatic regenerating gene overexpression in the nonobese diabetic mouse during active diabetogenesis. Diabetes 1996; 45:67-70. 77. Faveeuw C, Gagnerault MC, Lepault F. Expression of homing and adhesion molecules in infiltrated islets of Langerhans and salivary glands of nonobese diabetic mice. J Immunol 1994; 152:5969-5978. 78. Hanninen A, Taylor C, Streeter PR et al. Vascular addressins are induced on islet vessels during insulitis in nonobese diabetic mice and are involved in lymphoid cell binding to islet endothelium. J Clin Invest 1993; 92:2509-2515. 79. Lepault F, Gagnerault MC, Faveeuw C et al. Lack of L-selectin expression by cells transferring diabetes in NOD mice: insights into the mechanisms involved in diabetes prevention by Mel-14 antibody treatment. Eur J Immunol 1995; 25:1502-1507.

The Natural History of Islet-Specific Autoimmunity in NOD Mice

141

80. Yang XD, Karin N, Tisch R et al. Inhibition of insulitis and prevention of diabetes in nonobese diabetic mice by blocking L-selectin and very late antigen 4 adhesion receptors. Proc Natl Acad Sci USA 1993; 90:10494-10498. 81. Burkly LC, Jakubowski A, Hattori M. Protection against adoptive transfer of autoimmune diabetes mediated through very late antigen4 integrin. Diabetes 1994; 43:529-534. 82. Baron JL, Reich EP, Visintin I et al. The pathogenesis of adoptive murine autoimmune diabetes requires an interaction between alpha 4-integrins and vascular cell adhesion molecule-1. J Clin Invest 1994; 93:1700-1708. 83. Strandell E, Eizirik DL, Sandler S. Reversal of beta-cell suppression in vitro in pancreatic islets isolated from nonobese diabetic mice during the phase preceding insulin-dependent diabetes mellitus. J Clin Invest 1990; 85:1944-1950. 84. Sempe P, Bedossa P, Richard MF et al. Anti-alpha/beta T cell receptor monoclonal antibody provides an efficient therapy for autoimmune diabetes in nonobese diabetic (NOD) mice. Eur J Immunol 1991; 21:1163-1169. 85. Strandell E, Sandler S, Boitard C et al. Role of infiltrating T-cells for impaired glucose metabolism in pancreatic islets isolated from nonobese diabetic mice. Diabetologia 1992; 35:924-931. 86. Matsumoto M, Yagi H, Kunimoto K et al. Transfer of autoimmune diabetes from diabetic NOD mice to NOD athymic nude mice: the roles of T cell subsets in the pathogenesis. Cell Immunol 1993; 148:189-197. 87. Yagi H, Matsumoto M, Kunimoto K et al. Analysis of the roles of CD4+ and CD8+ T cells in autoimmune diabetes of NOD mice using transfer to NOD athymic nude mice. Eur J Immunol 1992; 22:2387-2393. 88. Christianson SW, Shultz LD, Leiter EH. Adoptive transfer of diabetes into immunodeficient NOD-scid/scid mice. Relative contributions of CD4+ and CD8+ T-cells from diabetic versus prediabetic NOD.NON-Thy-1a donors. Diabetes 1993; 42:44-55. 89. Rohane PW, Shimada A, Kim DT et al. Islet-infiltrating lymphocytes from prediabetic NOD mice rapidly transfer diabetes to NOD-scid/ scid mice. Diabetes 1995; 44:550-554. 90. Hayakawa M, Yokono K, Nagata M. Morphological analysis of selective destruction of pancreatic beta-cells by cytotoxic T lymphocytes in NOD mice. Diabetes 1991; 40:1210-1217. 91. Young LH, Peterson LB, Wicker LS et al. In vivo expression of perforin by CD8+ lymphocytes in autoimmune disease. Studies on spontaneous and adoptively transferred diabetes in nonobese diabetic mice. J Immunol 1989; 143:3994-3999. 92. Wang Y, Pontesilli O, Gill RG et al. The role of CD4+ and CD8+ T cells in the destruction of islet grafts by spontaneously diabetic mice. Proc Natl Acad Sci USA 1991; 88:527-531.

142

NOD Mice and Related Strains: Research Applications in Diseases

93. Charlton B, Mandel TE. Progression from insulitis to beta-cell destruction in NOD mouse requires L3T4+ T-lymphocytes. Diabetes 1988; 37:1108-1112. 94. Larger E, Becourt C, Bach JF et al. Pancreatic islet beta cells drive T cell-immune responses in the nonobese diabetic mouse model. J Exp Med 1995; 181:1635-1642. 95. Sempe P, Richard MF, Bach FJ et al. Evidence of CD4+ regulatory T cells in the non-obese diabetic male mouse. Diabetologia 1994; 37:337-343. 96. Dardenne M, Lepault F, Bendelac A et al. Acceleration of the onset of diabetes in NOD mice by thymectomy at weaning. Eur J Immunol 1989; 19:889-895. 97. Harada M, Makino S. Promotion of spontaneous diabetes in nonobese diabetes-prone mice by cyclophosphamide. Diabetologia 1984; 27:604-606. 98. Charlton B, Bacelj A, Slattery RM et al. Cyclophosphamide-induced diabetes in NOD/WEHI mice. Evidence for suppression in spontaneous autoimmune diabetes mellitus. Diabetes 1989; 38:441-447. 99. Zhang ZL, Georgiou HM, Mandel TE. The effect of cyclophosphamide treatment on lymphocyte subsets in the nonobese diabetic mouse: a comparison of various lymphoid organs. Autoimmunity 1993; 15:1-10. 100. Yasunami R, Bach JF. Anti-suppressor effect of cyclophosphamide on the development of spontaneous diabetes in NOD mice. Eur J Immunol 1988; 18:481-484. 101. Chatenoud L, Thervet E, Primo J et al. Anti-CD3 antibody induces long-term remission of overt autoimmunity in nonobese diabetic mice. Proc Natl Acad Sci USA 1994; 91:123-127. 102. Rothe H, Faust A, Schade U et al. Cyclophosphamide treatment of female non-obese diabetic mice causes enhanced expression of inducible nitric oxide synthase and interferon-gamma, but not of interleukin-4. Diabetologia 1994; 37:1154-1158. 103. Rothe H, Burkart V, Faust A et al. Interleukin-12 gene expression is associated with rapid development of diabetes mellitus in non-obese diabetic mice. Diabetologia 1996; 39:119-122. 104. Lenschow DJ, HO SC, Sattar H et al. Differential effects of anti-B7-1 and anti-B7-2 monoclonal antibody treatment on the development of diabetes in the nonobese diabetic mouse. J Exp Med 1995; 181:1145-1155. 105. Boitard C, Yasunami R, Dardenne M et al. T cell-mediated inhibition of the transfer of autoimmune diabetes in NOD mice. J Exp Med 1989; 169:1669-1680. 106. Hutchings PR, Cooke A. The transfer of autoimmune diabetes in NOD mice can be inhibited or accelerated by distinct cell populations present in normal splenocytes taken from young males. J Autoimmun 1990; 3:175-185.

The Natural History of Islet-Specific Autoimmunity in NOD Mice

143

107. Singer SM, Tisch R, Yang XD et al. An Abd transgene prevents diabetes in nonobese diabetic mice by inducing regulatory T cells. Proc Natl Acad Sci USA 1993; 90:9566-9570. 108. Slattery RM, Kjer-Neilsen L, Allison J et al. Prevention of diabetes in non-obese diabetic I-Ak transgenic mice. Nature 1990; 345:724-726. 109. Debray-Sachs M, Carnaud C, Boitard C et al. Prevention of diabetes in NOD mice treated with antibody to murine IFN gamma. J Autoimmun 1991; 4:237-248. 110. Campbell IL, Kay TW, Oxbrow L et al. Essential role for interferongamma and interleukin-6 in autoimmune insulin-dependent diabetes in NOD/Wehi mice. J Clin Invest 1991; 87:739-742. 111. Rapoport MJ, Jaramillo A, Zipris D et al. Interleukin 4 reverses T cell proliferative unresponsiveness and prevents the onset of diabetes in nonobese diabetic mice. J Exp Med 1993; 178:87-99. 112. Pennline KJ, Roque-Gaffney E, Monahan M. Recombinant human IL-10 prevents the onset of diabetes in the nonobese diabetic mouse. Clin Immunol Immunopathol 1994; 71:169-175. 113. Fitzpatrick F, Lepaul TF, Homo-Delarche F et al. Influence of castration, alone or combined with thymectomy, on the development of diabetes in the nonobese diabetic mouse. Endocrinology 1991; 129:1382-1390. 114. Pankewycz O, Strom TB, Rubin-Kelley VE. Islet-infiltrating T cell clones from non-obese diabetic mice that promote or prevent accelerated onset diabetes. Eur J Immunol 1991; 21:873-879. 115. Pankewycz OG, Guan JX, Benedict JF. A protective NOD islet-infiltrating CD8+ T cell clone, I.S. 2.15, has in vitro immunosuppressive properties. Eur J Immunol 1992; 22:2017-2023. 116. Diaz-Gallo C, Moscovitch-Lopatin M, Strom TB An anergic, isletinfiltrating T-cell clone that suppresses murine diabetes secretes a factor that blocks interleukin 2/interleukin 4-dependent proliferation. Proc. Natl Acad Sci USA 1992; 89:8656-8660. 117. Akhtar I, Gold JP, Pan LY et al. CD4+ beta islet cell-reactive T cell clones that suppress autoimmune diabetes in nonobese diabetic mice. J Exp Med 1995; 182:87-97. 118. Chosich N, Harrison LC. Suppression of diabetes mellitus in the nonobese diabetic (NOD) mouse by an autoreactive (anti-I-Ag7) isletderived CD4+ T-cell line. Diabetologia 1993; 36:716-721. 119. Reich EP, Scaringe D, Yagi J et al. Prevention of diabetes in NOD mice by injection of autoreactive T-lymphocytes. Diabetes 1989; 38:1647-1651. 120. Utsugi T, Nagata M, Kawamura T et al. Prevention of recurrent diabetes in syngenic islet-transplanted NOD mice by transfusion of autoreactive T lymphocytes. Transplantation 1994; 57:1799-1804. 121. Leiter EH, Serreze DV. Autoimmune diabetes in the nonobese diabetic mouse: suppression of immune defects by bone marrow transplantation and implications for therapy. Clin Immunol Immunopathol 1991; 59:323-334.

144

NOD Mice and Related Strains: Research Applications in Diseases

122. Stein PH, Rees MA, Singer A. Reconstitution of (BALB/c x B6)F1 normal mice with stem cells and thymus from nonobese diabetic mice results in autoimmune insulitis of the normal hosts’ pancreases. Transplantation 1992; 53:1347-1352. 123. Georgiou HM, Mandel TE. Induction of insulitis in athymic (nude) mice. The effect of NOD thymus and pancreas transplantation. Diabetes 1995; 44:49-59. 124. Ida T, Yasumizu R, Ohnishi Y et al. Analyses of development of insulitogenic T lymphocytes in NOD mice by transplantation of bone marrow, thymus, and pancreas. Transplantation 1990; 49:976-982. 125. Serrze DV, Gaedeke JW, Leiter EH. Hematopoietic stem-cell defects underlying abnormal macrophage development and maturation in NOD/Lt mice: defective regulation of cytokine receptors and protein kinase C. Proc Natl Acad Sci USA 1993; 90:9625-9629. 126. Langmuir PB, Bridgett MM, Bothwell et al. Bone marrow abnormalities in the non-obese diabetic mouse. Int Immunol 1993; 5:169-177. 127. Savino W, Boitard C, Bach FJ et al. Studies on the thymus in nonobese diabetic mouse. I. Changes in the microenvironmental compartments. Lab Invest 1991; 64:405-417. 128. Nabarra B, Andrianarison I. Thymus reticulum of autoimmune mice. 3. Ultrastructural study of NOD (non-obese diabetic) mouse thymus. Int J Exp Pathol 1991; 72:275-287. 129. O’reilly LA, Healey D, Simpson NE et al. Studies on the thymus of non-obese diabetic (NOD) mice: effect of transgene expression. Immunology 1994; 82:275-286. 130. Savino W, Carnaud C, Luan JJ et al. Characterization of the extracellular matrix-containing giant perivascular spaces in the NOD mouse thymus. Diabetes 1993; 42:134-140. 131. Colomb E, Savino W, Wicker L et al. Genetic control of giant perivascular space. Formation in the thymus of NOD mice. Submitted. 132. Vicari AP, Zlotnik A. Mouse NK1.1+ T cells: a new family of T cells. Immunol Today 1996; 17:71-76. 133. Gombert JM, Herbelin A, Tancrede-Bohin E et al. Early defect of immunoregulatory T cells in autoimmune diabetes. CR Acad Sci III 1996; 319:125-129. 134. Yoshimoto T, Bendelac A, Watson et al. Role of NK1.1+ T cells in a TH2 response and in immunoglobulin E production. Science 1995; 270:1845-1847.

CHAPTER 6

NOD Mice as a Model for Therapeutic Interventions in Human Insulin Dependent Diabetes Mellitus Mark A. Atkinson

A

s conveyed throughout this book, the nonobese diabetic (NOD) mouse serves as a model system for dissecting many aspects of insulin dependent diabetes mellitus (IDDM); a majority of which are thought to correlate with the disease as it is expressed in humans.1-4 While important to improving our understanding of the cause (s) and pathogenesis of this disease, these common features are also vital for this model to serve as a tool in identifying potential therapeutic modalities for the prevention of human IDDM.5-9 In addition to discussing the rationale for and history of preventing IDDM in humans, this chapter reviews the therapeutic manipulations that attenuate IDDM in NOD mice and the mechanisms by which they may operate.

THE PREVENTION OF IDDM IN HUMANS; PURPOSE AND HISTORICAL PERSPECTIVE Parenteral insulin treatment has been the therapeutic pillar for controlling the symptoms of IDDM for over 70 years. However, the physical benefits offered by this clinical exercise could be considered as merely “damage control”. While results of the Diabetes Care and NOD Mice and Related Strains: Research Applications in Diabetes, AIDS, Cancer and Other Diseases, edited by Edward Leiter and Mark Atkinson. © 1998 R.G. Landes Company.

146

NOD Mice and Related Strains: Research Applications in Diseases

Complications Trial (DCCT) indicate that microvascular complications can be diminished and/or delayed by intensive insulin therapy,10 such improvements in glycemic control would appear to only delay the inevitable morbidity and mortality associated with the disease.11,12 Insulin dependent diabetes remains a leading cause of adult-onset blindness, renal failure, heart disease, impotence, and non-traumatic amputations. This unfortunate clinical situation has lead a significant portion of IDDM research to focus on discovering a preventative/cure for the disease. After identifying the autoimmune basis for IDDM in the midto late-1970s, attempts at reversing the disease in humans were initiated and primarily directed at the use of immunosuppressive agents (e.g., cyclosporin, azathioprine, corticosteroids).13-17 These pharmacological agents had an extensive history of utilization in a variety of clinical situations; a major factor in the selection of such instruments in initial trials aimed at reversing IDDM. Multiple investigations demonstrated that treatment of new-onset IDDM patients with immunosuppressive agents lead to the continued ability of some patients to produce insulin (i.e., C-peptide) over an extended period of time; with anecdotal reports describing patients who obtained clinically significant remissions in their disease.13-17 While demonstrating a limited degree of promise, the usually minor yet sometimes serious side effects associated their long-term usage (e.g., nephrocytotoxicity, increased frequency of viral infections and cancer [B-cell lymphomas]) precluded their continued utilization and widespread acceptance.18,19 While disappointing in terms of finding a safe yet effective treatment for reversing the disease, the lessons learned from these experiences formed the basis for new questions which have directed efforts for the next generation of clinical trials. Today, three predominant questions are subject to considerable debate and attention in terms of preventing human IDDM: when should therapy be initiated, who should be subjected to treatment, and what agent should be utilized? With respect to timing, the aforementioned intervention studies made apparent the notion that the initiation of therapeutic regimens after the onset of clinical symptoms is most likely not optimal and suggested that effective methods for disease prevention may require therapy at a stage prior to the onset of symptoms.2 Human IDDM is preceded by a long (i.e., months to years) presymptomatic period with the clinical symptoms of disease arising only after the destruction of the majority of β cells.1-4 This

NOD Mice as a Model for Therapeutic Interventions in Human IDDM

147

preclinical period can be identified effectively through a combination of genetic, immunologic (e.g., islet cell autoantibodies), and metabolic markers for IDDM.20 Such markers can provide a critical tool to identify potential participants for therapeutic intervention trials prior to symptomatic IDDM. However, the level to which these markers truly identify those destined to develop IDDM has been the subject of considerable debate and lead to the second controversial issue: who should be subject to such interventions? 21 Although an individual’s risk for IDDM increases approximately 10-fold with a family history of the disease (i.e., estimated at 1 IDDM case per 300 individuals in the United States population versus 1 case per 25 firstdegree relatives of a proband with IDDM), some 80-90% of new IDDM cases occur in families with no family history of the disease.2 To provide a beneficial effect (i.e., preventing IDDM) for the greatest number of individuals will require that treatments be utilized in the general population; a group whose natural history of disease has not been subject to as intense investigation as relatives of probands with IDDM.21 Finally, which agents should be utilized within IDDM intervention trials remains unsettled. As the aforementioned use of immunosuppressive agents in asymptomatic individuals would at best be considered controversial, considerable effort has been directed at the identification of less-aggressive agents that selectively target cells involved in the autoimmune disease process or which only disable immunological components to a level that does not significantly impair the entire immune system function. Unlike immunosuppression pharmacology, analysis of such agents in human autoimmune disease does not have an extensive history.2,3 Fortunately, the NOD mouse serves as a vital tool to investigate each of these important questions and controversies.

THE NOD MOUSE AS A MODEL FOR PREVENTION OF IDDM IN HUMANS Relatively few characteristic differences exist between IDDM as expressed in humans and NOD mice; this is an important factor when analyzing agents for efficacy in preventing the disease in humans. The most striking difference is the pronounced female gender bias for disease in NOD mice; a situation unlike human IDDM where the disease expresses itself equally among both sexes. In addition, NOD mice do not display the severe diabetic ketoacidosis characteristic of untreated human IDDM patients. An increased percentage of T cells

148

NOD Mice and Related Strains: Research Applications in Diseases

(i.e., both CD4+ and CD8+ subsets) in the peripheral blood and lymphoid tissues of NOD mice also distinguishes these animals from humans with IDDM. NOD mice express endogenous defective retroviruses in their β cells.22-24 The islet cells of humans who have succumbed to IDDM at disease onset display evidence for a superantigen/endogenous retroviral-like process.25 However, the role of retroviruses in the pathogenesis of the two diseases is unclear and thus far appears unique. The sum of these differences is not great enough to diminish the value of this animal model in identifying a therapy for preventing IDDM in humans. However, when subject to such analyses, two major etiopathogenic distinctions between IDDM in humans and NOD mice must be considered: genetics and environment.

GENETIC AND ENVIRONMENTAL FACTORS: EFFECT ON INTERPRETATION OF OUTCOME MEASURES As the result of over 50 generations of brother-sister matings, NOD mice inherit the same gender-specific set of susceptibility genes. The penetrance of these genes can be analyzed under consistent and controlled environmental conditions (i.e., defined diet, temperature, freedom from exposure to pathogens). Consequently, the natural history of IDDM in a well-maintained specific pathogen free NOD colony is somewhat predictable and not normally subject to marked changes in the frequency of IDDM.26 Intervention studies in NOD mice can be designed where therapeutic regimens are initiated at birth, at a presymptomatic stage prior to the occurrence of insulitis (i.e., less than three weeks postpartum), before the onset of symptomatic disease (i.e., four to eight weeks postpartum) at a time when considerable numbers of β cells are still intact, or at the diagnosis of IDDM when the β cell damage has appreciated to the extent of overt hyperglycemia. By contrast, the genetic and environmental heterogeneity associated with the natural history of IDDM in humans is such that the age of disease onset is extremely broad; with symptoms occurring at any time from the first years of life to well beyond 50 years of age. Given the genetic heterogeneity of human populations, the development of IDDM is likely to reflect heterogeneous mixtures of susceptibility genes whose penetrances are responsive to different thresholds of intragenic and environmental influences. Such complexities have rendered it difficult for clinical investigators to develop simple mark-

NOD Mice as a Model for Therapeutic Interventions in Human IDDM

149

ers which accurately predict the time of symptomatic onset for humans destined to develop IDDM.21 For these reasons, studies analyzing therapeutic agents aimed at preventing IDDM in NOD mice must be carefully viewed for their functional as well as their practical applicability to therapeutic intervention in human disease. For example, agents utilized in NOD mice from birth at a time without β cell destruction may not be applicable to treatment of humans identified immediately prior to the onset of IDDM when significant β cell destruction has occurred. As previously indicated, “environment” clearly influences human IDDM; a situation which most likely extends to NOD mice. One survey of a majority of existing NOD mouse colonies indicated that the cumulative incidence of diabetes at 30 weeks of age is highly variable (e.g., 0-100% in females).26 While some of these differences in the colony frequency of IDDM may be explained by genetic divergence amongst substrains of NOD mice as they have undergone separation from their original source colony, a majority of these differences appear to be driven by variation in environmental factors (e.g., dietary chow constituents, cleanliness of colony, controlled breeding). For example, Caesarean transfer of pups from a conventional environment to a specific pathogen-free environment markedly increases the incidence of diabetes.27 The potential mechanism(s) whereby environmental agents influence NOD IDDM frequency are numerous and include those influencing genetic susceptibility. As the destruction of sufficient numbers of β cells to produce persistent hyperglycemia and glycosuria requires a complex interaction with numerous other genes that are not linked to the MHC locus, the penetrance of diabetes-susceptibility genes in the NOD mouse may therefore be influenced by agents in the extrinsic environment, including dietary components and microbial pathogens.28,29 The protective effects of exposure to microbial pathogens in NOD mice is of interest for many reasons. First, while epidemiological and anecdotal data exist to suggest that viruses may precipitate the autoimmunity that results in human IDDM,30,31 viral and bacterial infections have been reported to reduce rather than exacerbate the incidence of diabetes in NOD mice.26,32-35 These data in some ways contradict the paradigm of molecular mimicry wherein infection with a microbial antigen that is sufficiently similar to a self-antigen may provoke a pathogenic cross-reactive autoimmune response.36,37 In addition, while many current therapies for autoimmune disease

150

NOD Mice and Related Strains: Research Applications in Diseases

involve immunosuppression,18 the effect of microbial challenge on diabetes in the NOD mouse may occur through immunostimulation (discussed later).38-41

THERAPEUTIC INTERVENTIONS IN NOD MICE As of early 1996, over 100 articles were accessible by MEDLINE reference which pertained to investigations reporting the prevention or delay of IDDM in NOD mice. Unfortunately, a practical evaluation of the clinical relevance to human disease of these studies is varied; with reports associating alterations in the degree of insulitis, the ability to adoptively transfer IDDM, the capacity to avoid islet cell graft rejection, as well as altering the frequency of spontaneous or cyclophophamide-induced diabetes to equate with “disease prevention.” For the purpose of uncovering a potential therapeutic for preventing human IDDM, it could be argued that NOD studies with the most clinically relevant outcomes would include only those monitoring the spontaneous incidence of overt IDDM as other measures (e.g., insulitis) are uncertain predictors with respect to the development of IDDM. Hence, such criteria will be the limitation for discussion within this chapter. Notwithstanding this limitation, interventions which prevent IDDM in NOD mice are plentiful and remarkably diverse (Table 6.1). Indeed, the ease at which the incidence of IDDM can be altered in the NOD mouse has caused some to question this animal model in a simplistic fashion for the selection of therapeutic regimens to prevent IDDM in humans.42 However, this factor has previously been argued to attest for the utility of this model in dissecting the multiple components of a complex disease and the potential fragile nature by which the autoimmune disease may operate in human IDDM.39

Table 6.1. The A to Z’s of diabetes prevention in NOD mice Androgens Anesthesia Azathioprine Anti-B7-1 Bacille Calmette Gue’rin (BCG) Baculofin Blocking peptide of MHC class II

Interleukin-1 Interleukin-4 Interleukin-10 Islet cells - intrathymic Lactate dehydogenase virus (LDH) Lazaroid Linomide

NOD Mice as a Model for Therapeutic Interventions in Human IDDM

Table 6.1. The A to Z’s of diabetes prevention in NOD mice (cont.) Bone marrow transplantation Castration Anti-CD3 Anti-CD4 Anti-CD8 Cold exposure Anti-complement receptor Complete Freund’s adjuvant Anti-CTLA-4 Cyclosporin Cyclosporin A Deflazacort Dendritic cells from pancreatic lymph node Deoxysperogualin Diazoxide 1,25 dihydroxyl Vitamin D3 Elevated temperature Encephalomyocarditis virus (ECMV) Essential fatty acid deficient diets FK506 Glucose (neonatal) Glutamic acid decarboxylase -intraperitoneal, intrathymic, intravenous, oral Glutamic acid decarboxylase peptides —intraperitoneal, intrathymic, intravenous, oral Gonadectomy Heat shock protein 65 Heat shock protein peptide (p277) Anti-ICAM-1 Immobilization Anti-integrin alpha 4 Inomide Insulin -intraperitoneal, oral, subcutaneous, nasal Insulin B chain/B chain amino acids 9-23 -intraperitoneal, oral, subcutaneous, nasal Interferon-gamma Anti-interferon-gamma

Anti-LFA-1 Anti-L-selectin Lymphocyte choriomeningitis virus (LCMV) Anti-lymphocyte serum LZ8 MDL 29311 Anti-MHC class I Anti-MHC class II Mixed allogeneic chimerism Monosodium glutamate Murine hepatitis virus (MHV) Mycobacterium Natural antibodies Nicotinamide OK432 Overcrowding Pancreatectomy Pertussigen Poly [I:C] Pregestimil diet Probucol Prolactin Rampamycin Saline (repeated injection) Semi-purified diet (e.g., AIN-76) Silica Sodium fusidate Somatostatin Non-specific pathogen free conditions Streptococcal enterotoxins Superantigens Superoxide dismutasedesferrioxamine Anti-T cell receptor Anti-thy-1 Thymectomy (neonatal) Tolbutamide Tumor necrosis factor-alpha Vitamin E Anti-VLA-4

151

152

NOD Mice and Related Strains: Research Applications in Diseases

For the purpose of this chapter, therapeutic regimens which successfully interrupt the pathogenic process of spontaneous disease in NOD mice are grouped into six categories based on their presumed mechanism of function: 1) immunosuppression; 2) tolerance induction; 3) immunostimulation; 4) reducing metabolic activity; 5) dietary/hormonal manipulation; and 6) anti-inflammatory activity.

IMMUNOSUPPRESSION As expected when treating a disease of autoimmune origin, multiple methods defined as “immunosuppressive” have demonstrated an ability to retard or prevent the onset of IDDM in NOD mice. This extensive list includes cyclosporin, cyclosporin A, deoxysperogualin, azathioprine, sodium fusidate, rapamycin, FK506, LZ8 and linomide.4351 The exact pharmacological actions of these agents vary, yet each compound interferes with cellular division or expansion at some level of immunological ontogeny. As such, all treatments are classified as “nonspecific” in terms of their immunosuppressive capabilities. While not traditional “immunosuppressants”, treatments that interfere with the functional presentation of antigenic peptides to Tc ells could also be classified in this category for agents of disease prevention. At the antigen presenting cell (APC) level, protection from IDDM has been observed with silica mediated destruction of macrophages,52 antibodies against mouse MHC class II I-A molecules,53 and “blocking” peptides that compete for binding to I-Ag7.54 Likewise, treatments that compromise the effector functions or viability of lymphocytes can retard or circumvent the disease. Specifically, intervention studies utilizing an assortment of antibodies directed towards molecules involved in antigen presentation/recognition have demonstrated clinical utility. These include antibodies against the T-cell receptor (TCR) α and β chains, anti-lymphocyte serum, anti-CD3, antiCD4 (L3T4), and anti-Thy 1.2.55-62 Additional studies have shown that interference with the so-called second signal pathway (i.e., costimulation) for T cell activation can also have dramatic effects on NOD IDDM. Specifically, early (i.e., <10 weeks) treatment of NOD mice with anti-CLTA-4 (i.e., a soluble CD28 agonist) or anti-B7-2 (i.e., a CD28 ligand) monoclonal antibodies prevents IDDM.63 Similarly, antibodies directed against molecules involved in homing of lymphocytes/monocytes to the pancreas (e.g., anti-L-selectin, anti-VLA-4, anti-integrin alpha 4, anti-LFA-1, anti-ICAM-1) prevent IDDM when

NOD Mice as a Model for Therapeutic Interventions in Human IDDM

153

administered to prediabetic animals.64-67 Finally, antibodies targeting cytokines associated with lymphocyte/monocyte activation (e.g., antitumor necrosis factor[TNF]-α, anti-interferon [IFN]-β) prevent disease when administered to prediabetic mice.68,69 Interpretation of the significance of antibody based studies has not, however, been free of controversy as “irrelevant” control monoclonal antibodies to certain TCR Vβ clonotypes that are not present in NOD mice unexpectedly reduce the incidence of disease. However, this unusual form of protection may be mediated by nonspecific immunostimulation (discussed below) as a consequence of an antiglobulin immune response.70 Indeed, it is not uncommon for NOD mice to produce antibodies against the monoclonal antibodies being tested for therapeutic potency. Of additional concern is the clinical applicability of such agents. Indeed, the side effects of chronic administration, immunogenicity, and lack of selectivity may provide significant limitations in terms of their usage in preventing disease in humans.

TOLERANCE Tolerance to an antigen exists when the immune system makes no attempt to eliminate the agent from the body. At its basic level, autoimmunity is thought to result from a failure of the immune systemt o develop such non-reactivity (i.e., tolerance) to one’s selftissues. Antigen specific tolerance is thought to be induced by three distinct processes: clonal deletion, anergy, or induction of immunoregulatory cells. Clonal deletion occurs when potentially autoreactive lymphocytes are deleted either centrally in the thymus or in the peripheral immune system. Anergy occurs when circulating cells with potential autoreactivity are rendered non-reactive or functionless. The third mechanism of tolerance, control via immunoregulatory cell responses, is dependent on the production of a suppressive signal (e.g., the cytokine TGF-β) to downregulate the immune response. The research tools to monitor this latter mechanism far outweigh the former two, hence a majority of research reports to date have focused on measuring such processes. Regardless of the exact mechanism of tolerance, given the fundamental links between tolerance and autoimmunity, therapeutic regimens aimed at (re)establishment of tolerance to β cell autoantigens in IDDM would appear to be a primary target for designing methods

154

NOD Mice and Related Strains: Research Applications in Diseases

to prevent the disease.71 A number of therapies in NOD mice have been attributed to the induction of tolerance. A single injection of anti-CD3 monoclonal antibody into neonatal NOD mice is tolerogenic, prevents IDDM, and is capable of inducing clinical remissions in a majority of newly-diagnosed NOD mice.72 However, tolerance can also be achieved through the selective destruction or self-inactivation of autoreactive T cells without damage to the function of all T cells. Indeed, the agents (e.g., islet cells, β cell autoantigens) and routes of administration (e.g., thymic, intravenous, oral) by which tolerance to β cells can be achieved are numerous. Hence, a major research effort has been directed at uncovering a “primary” autoantigen for IDDM in NOD mice as well as in humans, with the hope that this finding would allow for the development of a therapeutic that induces “antigen specific tolerance” and results in the prevention of IDDM. Administration of a number of putative β cell autoantigens73 attenuates the onset of IDDM in NOD mice. Among these, the majority of attention has focused on insulin.74,75 Intraperitoneal, subcutaneous, or oral administration of insulin protects NOD mice from diabetes.76-81 Similarly, intraperitoneal injections of the insulin B chain, B chain amino acids 9-23, or the intranasal administration of the B chain amino acids 9-23 are as protective against IDDM as intact insulin.78,82 Unfortunately, the mechanism(s) underlying these therapies are known to varying degrees. When available, such studies have indicated that the route of administration of these agents may be a crucial variable in terms of a mechanism for tolerance induction and disease prevention. The intranasal insulin B chain 9-23 study demonstrated the induction of immunological tolerance to the inciting agent.82 However, analysis of intraperitoneal B chain 9-23 administration, rather than demonstrating a loss in cellular/humoral activity to insulin, provided evidence for a decrease in IFNγ production within the insulitis lesion.78 Depending on the dosage of antigen, oral administration of insulin appears to induce immunoregulatory tolerance and/or clonal anergy in T cells reactive to islet antigens. Specifically, oral administration of insulin into NOD mice induces the elaboration of immunoregulatory cytokines and a shift in the balance of cytokine expression from a T helper (Th)1 to a Th2 like pattern (i.e., enhanced interleukin [IL]-4, IL-10, prostaglandin-E, and TGF-β with diminished IL-2, IFNγ, or TNFα).80,81,83 The mechanism underlying subcutaneous prevention of disease is extremely contro-

NOD Mice as a Model for Therapeutic Interventions in Human IDDM

155

versial, with proposed mechanisms of reduced metabolic activity (discussed later) or some form of tolerance. Specifically, the autoimmune response in NOD mice appears dominated by the Th1 subset of T cells and the cytokine IFNγ; with the Th2 arm being under-represented (discussed in chapter 3). One additional theory proposes that insulin functions not only as a glucose regulatory hormone but also as a hormone for T cells; where it enhances lymphocyte function specifically by synergyzing with multiple cytokine/hormone pathways to enhance Th2 responses.74 Significant research attention has also been directed at the neuroendocrine enzyme glutamic acid decarboxylase (GAD).84 Such studies indicate thatprotection from IDDM is offered by intraperitoneal or intravenous administration of GAD when given to young NOD mice, and that the beneficial effect is achieved by an acquisition of T cell tolerance (i.e., alteration in immunoregulatory profile) to this β cell autoantigen.85-89 NOD mice given intrathymic administration of GAD at three weeks of age exhibit markedly reduced T cell proliferative responses to GAD and the animals remain free of IDDM.86 Whereas in one study GAD administration was protective, bovine serum albumin (BSA), an antigen of potential environmental significance,2 was ineffective in delaying the onset of diabetes in NOD mice.88 In addition, some of these studies suggested a primary autoantigenic nature of GAD versus other putative β cell autoantigens (i.e., insulin, heat shock protein [HSP]65, carboxypeptidase H) due to the superior ability of this molecule to prevent disease or a pronounced capacity to reduce autoimmune T cell activities.85,86 Heat shock proteins (e.g., HSP60, HSP65) also reportedly form a class of potential autoantigens in IDDM. Studies have suggested a pancreatic β cell molecule cross reactive with the 65 kDa HSP of Mycobacterium tuberculosis may form the basis of such an autoantigen and when administered correctly, it can effectively prevent IDDM in NOD mice.90 In addition, a peptide from the human variant of the HSP65 molecule, when administered to NOD mice, downregulates immunity to HSP65 and prevents the development of diabetes.91 A peptide of the 60 kDa HSP, designated p277, has been demonstrated as a target for diabetogenic T cell clones from NOD mice. Insulin dependent diabetes in NOD mice can be prevented by intraperitoneal administration of the p277 peptide; a process which tolerizes these animals against anti-p277 immunity.92 The proposed mechanism for this preventative effect is, however, somewhat unique in terms

156

NOD Mice and Related Strains: Research Applications in Diseases

of tolerance by the suggestion that p277 administration activates a network of anti-idiotypic cells which actively downregulate the effector cells which destroy β cells.91,92 In summary, numerous β cell autoantigens may induce tolerance in NOD mice to a level offering protection from disease. The mechanisms of their actions in terms of establishing tolerance appear to be multiple. In addition to the aforementioned antigen specific situations, the peripheral deletion of autoreactive clones has been achieved via vaccination with autoreactive CD4+ Vβ8+ T cells,93 T cell lines,94 and T cell clones specific for the 65 kDa HSP.91 The injection of dendritic cells or splenocytes appears to operate through the induction of regulatory tolerance or clonal anergy in T cells reactive to islet antigens. 95 In theory, thymic deletion of β cell autoreactive clones may be accomplished by intrathymic injection of islet cells into young NOD mice. Such therapies have shown the ability to prevent and/or delay the onset of IDDM in these animals;96,97 however, clear evidence for thymic deletion of autoreactive clones is limited.

IMMUNOSTIMULATION The classification of therapeutic modes of action outlined in this chapter are not rigid and through commonality in mechanistic action, may be subject to overlap. Such an example is the potential extension between the alterations in immunoregulation offered by tolerance induction (previous section) and that of immunostimulation. This category was included as several agents that non-specifically stimulate the immune system (e.g., poly (I:C), environmental pathogens) also reduce the frequency IDDM in NOD mice. As previously indicated, environmental agents appear especially effective as NOD mice exposed to bacteria or their products (e.g., Mycobacterium, Streptococcal enterotoxins, Bacillus Calmette-Guerin ([BCG], pertussigen, complete Freund’s adjuvant (CFA)) or viral pathogens (e.g., encephalomyocarditis virus, lymphocytic choriomeningitis virus, murine hepatitis virus) prevents and/or delays the onset of IDDM.32-35,98-105 The mechanism whereby these nonspecific immunostimulations prevent disease in NOD mice may involve a (partial) restoration in defective APC function, a correction in the balance of immunoregulatory cytokines, and/or the generation of an increase in the number of functional immunoregulatory cells capable of suppressing diabetogenesis.

NOD Mice as a Model for Therapeutic Interventions in Human IDDM

157

Indeed, one unusual characteristic of both NOD mice and individuals with IDDM is a depressed autologous mixed lymphocyte reaction (AMLR); a feature suggestive of a defect in peripheral immunoregulatory mechanisms. As discussed in chapters 3 and 4, defects in cytokine-elicited differentiation and maturation of APC from the bone marrow of NOD mice may result in an inefficient presentation of self-antigens and impair the tolerogenic capacity of these cells.106-108 Such a defect could explain why exposure of prediabetic NOD mice to a number of environmental pathogens imparts resistance to diabetes. Similarly, therapeutic manipulations may stimulate antigen processing and presentation by macrophages. Indeed, the protection associated with some of these treatments (e.g., IL-1, IL-2, mixed allogeneic chimerism) restores a near normal AMLR.106-109 Indeed, certain environmental stimuli may upregulate APC function through increased thymic deletion/peripheral anergization of autoreactive T cells, activation of immunoregulatory T cells in the periphery, or tolerization through a combination of these mechanisms. This concept is reinforced by the experimental findings that chronic administration of a variety of cytokines (e.g., TNFα, IL-10) or single injections of potent immunomodulators that upregulate endogenous cytokine expression, also circumvent diabetes.68,110-111

REDUCING β CELL METABOLIC ACTIVITY The concept that a reduction of β cell metabolic activity may be beneficial in terms of preventing IDDM dates back over a half-century; with attribution to Charles Best, a co-founder of pharmacologic insulin.112 A more recent study suggested that intensive insulin therapy in patients at the diagnosis of IDDM can result in a prolonged “honeymoon” period during which the endogenous insulin secreting capacity is maintained for an extended period.113 The first study in animal models to demonstrate the ability of daily low-dose insulin to prevent IDDM was in the BB rat.114 A later study indicated that this so-called “prophylactic insulin therapy”, when given from age four weeks and beyond, prevents diabetes and insulitis in NOD mice.76 Whether the mechanism of prophylactic insulin therapy is due to β cell rest,115 to the induction of tolerance (previously discussed), or an as yet unidentified mechanism is unknown.74,75 In association with the former hypothesis, suppression of insulin secretion is thought to reduce diabetes by causing the β cells to be less visible to the immune surveillance system, or less susceptible to inflammatory damage.

158

NOD Mice and Related Strains: Research Applications in Diseases

Evidence in accordance with the concept of β cell rest includes observations that chronic treatment with diazoxide or somatostatin (i.e., inhibitors of endogenous insulin production) prevent IDDM in NOD mice.79 By contrast, tolbutamide, a potent insulin-stimulator, also reduced the incidence of IDDM.116 Likewise, neonatal glucose treatments (i.e., insulin stimulatory), rather than exacerbating IDDM, lead to disease prevention.117 Other studies, which may relate to metabolism or β cell function, have extended beyond insulin to monitor the effects of altering the action of gamma-amino butyric acid (GABA); a molecule involved in paracrine signaling within the pancreatic islet.84 Specifically, baculofin, a GABA B-receptor agonist, delays the onset of IDDM in NOD mice.118 In summary, whereas the similar effects of oral, intraperitoneal and intranasal administration of whole insulin does provide evidence that prophylactic insulin therapy protects via insulin tolerization or immunospecific stimulation, significant evidence exists to suggest that a reduction in metabolic activity of the β cells provides a beneficial effect in terms of preventing IDDM. Rather than an exclusive mechanism, it is more likely that depending on its route of administration, each of the aforementioned mechanisms may simultaneously contribute to protection from β cell destruction.79

DIETARY/HORMONAL MANIPULATION Research aimed at identifying whether dietary factors represent a catalyst for the induction of IDDM in humans are controversial, difficult to perform, and subject to marked differences in interpretation.119,120 Spontaneous animal models for IDDM provide a unique tool to identify diabetogenic dietary agents and/or determine a role for diet in the pathogenesis of the disease.121 The first such study of animals (i.e., BB rats), published over a decade ago, implicated skim milk from cows.122 Multiple investigations have since shown that the natural history and frequency of IDDM in these animal models can be modified through many forms of dietary manipulation. Indeed, feeding NOD mice semi-purified diets (e.g., AIN-76, Pregestimil®) based on hydrolyzed casein or other amino acid sources prevents or delays the onset of IDDM while feeding them chows comprised of mixed-cereals results in an increased frequency of IDDM.123,125 The major difficulty in interpreting dietary studies of animals is the high experimental variability observed by some authors. However, the majority of controversy has surrounded the role

NOD Mice as a Model for Therapeutic Interventions in Human IDDM

159

of dietary cow’s milk in IDDM. The diabetogenic effect of cow’s milk components is in fact minor in comparison to the major effect of plant components (e.g., soy, wheat and other plant associated products) in chows.120 In addition, immunity to BSA in animal models does not appear to correlate with risk for IDDM, nor does attenuation of anti-BSA immunity provide therapeutic value in terms of disease prevention.88,125,126 These animal models also suggest that shortterm exposure to a diabetogenic agent is not in itself sufficient to provide a trigger which ultimately results in IDDM, and that the diabetogenic potential of food is not restricted to infancy in that firsttime exposure until the equivalent of adolescence can result in a high frequency of IDDM.120 The mechanism by which dietary manipulation modulates the frequency of IDDM in animals is unknown but may include the absence of a diabetogenic dietary agent or a low stimulatory effect of the diet on the metabolism of β cells (i.e., the aforementioned β cell rest); a process which some hypothesize is a method for diabetes prevention.1 Diets deficient in essential fatty acids are protective from IDDM and may be related to enhanced immunological activities (e.g., enhanced TNF-α, IL-1, IL-4 production).127 Others analyzing NOD mice fed Pregestimil® have not observed marked changes in peripheral immune responses.126 While animal models for IDDM have revealed that diet can be a disease modulating variable, they fall far short of representing human IDDM in that they are not subject to genetic variation, constantly changing environmental exposures and conditions (e.g., parasitic, bacterial and viral infections), and alterations in diet composition. Alterations of sex hormones in NOD mice also influence the frequency of IDDM. These include treatment of female NOD mice with androgens or testosterone,128, 129 male and female NOD mice with prolactin,130 castration of male mice129,131,132, or ovarectomy of female mice.129 The mechanisms underlying hormonal therapy are unclear; with proposals involving alteration of the development or function of immune system cells,128 glucocorticoid and/or androgen inducing insulin resistance,132 or related to an animal’s weight.129

ANTI-INFLAMMATORY AGENTS Multiple studies, both in vivo and in vitro, have suggested that destruction of β cells may occur through damage induced by oxygen (O 2 – . , H 2 O 2 ) and nitrogen (NO . ) free radicals produced by

160

NOD Mice and Related Strains: Research Applications in Diseases

macrophages. Other reports suggest that inflammatory cytokines (e.g, IL-1, TNF-α) in addition to having direct cytotoxic effects on β cells, may further induce these cells to produce toxic levels of nitric oxide; a process that results in cellular suicide. Indeed, evidence to support this contention derives from the extensive list of agents which target such processes and prevent IDDM in NOD mice.133-142 Administration of nicotinamide, superoxide dismutasedesferrioxamine and vitamin E have the effect of reducing such oxidative processes and have demonstrated the ability to decrease β cell destruction and IDDM in NOD mice.133-136 Other anti-oxidants (probucol, probucol plus deflazacort [a corticosteriod], lazaroid, MDL 29311 [a probucol analog], 1,25-dihydroxyl vitamin D3 and monosodium glutamate) also prevent disease in NOD mice.137-142 Indeed, only one report utilizing aminoguanidine (a competitive inhibitor of nitric oxide synthase) failed to support a role for nitric oxide production in the pathogenesis of β cell destruction through its inability to prevent IDDM in NOD mice.143

HUMAN TRIALS FOR IDDM PREVENTION Through research insights gained through studies of diabetes interventions in the NOD mouse, a number of the aforementioned observations have lead to the initiation of clinical trials in humans aimed at preventing IDDM. As previously indicated, insulin therapy prevents diabetes in both rodent models for IDDM. A pilot trial in humans reported promising preliminary results for prophylactic insulin therapy in delaying and/or preventing IDDM.144 The U.S. National Institutes of Health has formed a Diabetes Prevention Trial (DPT-1) which seeks to address the question of whether this form of therapy is capable of preventing IDDM in humans at increased-risk for the disease (due to a family history of IDDM and anti-islet cell autoantibodies). The ability of so-called “isohormonal therapy” to prevent disease is not limited to preventing IDDM; rather it is the focus of attention for potential therapeutic value in treating a number of autoimmune endocrine diseases.75 As anti-inflammatory agent inhibitors demonstrated promise in preventing IDDM in NOD mice, oral nicotinamide has been used in pilot studies of humans for the purpose of preventing IDDM in asymptomatic individuals at increased risk for IDDM.145-147 The preliminary results of these studies, with one exception,148 were encouraging and have formed the basis for a large European trial (ENDIT—

NOD Mice as a Model for Therapeutic Interventions in Human IDDM

161

European Nicotinamide Diabetes Intervention Trial) for the prevention of the disease.149 The beneficial effects of this drug may derive from prevention of β cell damage by oxygen or nitrogen (e.g., nitric oxide) free radicals. While both of these agents have shown promise and are considered safe at the prescribed dosages, the potential side effects offered by these therapies— hypoglycemia (prophylactic insulin therapy) and liver dysfunction (nicotinamide)—have lead some to seek even more benign therapies. Hence another strategy under consideration for clinical trials in humans is that involving the oral induction of tolerance to islet-cell proteins that have been implicated as autoantigens in antiβ cell immunity. As previously indicated, the onset of diabetes was both attenuated and protracted in NOD mice fed with porcine insulin.77 Therefore, the DPT-1 trial also includes this form of therapy for individuals deemed at increased-risk for IDDM. As already addressed, the prevention of IDDM in NOD mice can be achieved through immune-enhancement therapy. This approach, which at first appears paradoxical, is based on the hypothesis that the essential problem underlying IDDM is the inability to maintain peripheral immunological tolerance to pancreatic islets. Thus, any nonspecific immunization event that enhances general tolerogenic mechanisms may be sufficient to prevent disease. Similar to the observations that carefully timed doses of CFA prevents diabetes in NOD mice, one study suggests similar promise utilizing BCG in humans.150 Indeed, further human trials that will test this hypothesis by actively immunizing individuals with BCG are underway. Considerable interest has been generated by the observation that intrathymic transplantation of islet cells at birth in rodent models of IDDM prevents both insulitis and diabetes. As tolerance is mediated, in part, by interactions between maturing thymocytes and self-antigens presented by thymic stromal cells, the beneficial effect of intrathymic transplantation may result from the specific modulation of diabetogenic T cells that mature in a thymic microenvironment enriched for islet cell antigens. While the mechanism remains unclear, intrathymic transplantation of islet cells has been proposed as a site for surgical implantation in patients with long-standing IDDM.151 Once the autoantigens that elicit human IDDM are identified, autoantigen-immunization in early life may result in tolerance to the antigen and prevention of IDDM. Thus, a method for the induction

162

NOD Mice and Related Strains: Research Applications in Diseases

of specific tolerance to islet-cell antigen in humans would appear to be promising. While subject to high variability, multiple studies have demonstrated that diet influences the frequency of IDDM in NOD mice. However, our knowledge of dietary agents which may be diabetogenic and how these agents, were they to exist, contribute to the pathogenesis of IDDM is unclear. The process of identifying food diabetogens in humans may prove difficult if not impossible when one considers the variety and complexity of food constituents, the timing and quantity in which they are consumed, and the potential for interaction with other environmental agents. Some studies have implicated cow’s milk as having diabetogenic potential. Based on this information and the reports of anti-cow’s milk immunity in humans with IDDM, the American Academy of Pediatrics has modified its infant nutrition guidelines to promote breast-feeding for the purpose of preventing IDDM. However, the reports of anti-cow’s milk immunity in humans are controversial and the most recent epidemiological studies have not supported a role for infant feeding practices as a risk factor for the disease. Despite the controversial nature of diet in the pathogenesis of IDDM, clinical trials aimed at cow’s milk avoidance are ongoing.152 However, due to the result of the design of such trials, information regarding their efficacy will not be available for an extended period of time (i.e., greater than a decade).

CONCLUSIONS Improvements in our understanding of the genetic, metabolic and immunologic mechanisms underlying IDDM in humans have allowed for the identification of individuals destined to develop IDDM. Such individuals should benefit from clinical interventions aimed at attenuating and/or interrupting the autoimmune disease process. Given the significant side effects observed with immunosuppression, safe yet effective immunotherapies appear to be required to prevent the disease. A combination of methodologies in association with the identification of the autoantigens involved in the disease process as well as the application of different therapies to individuals based on their preexisting level of β cell destruction may increase the likelihood of identifying a successful treatment. In addition to providing a model to study the pathogenesis and natural history of IDDM in a system with characteristics similar to human IDDM, the NOD mouse

NOD Mice as a Model for Therapeutic Interventions in Human IDDM

163

has and will continue to serve as a valuable tool to test intervention protocols that could be used to prevent the disease in humans. ACKNOWLEDGMENTS This writing, and experiments by the author cited herein, have been supported by grants from The National Institutes of Health, The American Diabetes Association, and The Juvenile Diabetes Foundation. REFERENCES 1. Castano L, Eisenbarth GS. Type 1 diabetes: a chronic autoimmune disease of human, mouse and rat. Annu Rev Immunol 1990; 8:647-679. 2. Atkinson MA, Maclaren NK. The pathogenesis of insulin dependent diabetes mellitus. N Engl J Med 1994; 7:62-71. 3. Bach JF. Insulin-dependent diabetes mellitus as an autoimmune disease. Endocrin Rev 1994; 15:516-542. 4. Tisch R, McDevitt H. Insulin-dependent diabetes mellitus. Cell 1996; 85:291-297. 5. Kikutani H, Makino S. The murine autoimmune diabetes model: NOD and related strains. Adv Immunol 1992; 52:285-322. 6. Leiter EH, Serreze DV. Antigen presenting cells and the immunogenetics of autoimmune diabetes in NOD mice. Reg Immunol 1992; 4:263-273. 7. Rossini AA, Handler ES, Mordes JP et al. Animal models of human disease. Human autoimmune diabetes mellitus: lessons from BB rats and NOD mice—Caveat Emptor. Clin Immunol Immunopathol 1995; 74:2-9. 8. Serreze DV, Leiter EH. Genetic and Pathogenic basis of autoimmune diabetes in NOD mice. Curr Opin Immunol 1994; 6:900-906. 9. Jaramillo A, Gill B, Delovich TL. Insulin dependent diabetes in the non-obese diabetic mouse: a disease mediated by T cell anergy. Life Sciences 1994; 15:1163-1177. 10. The Diabetes Control and Complications Trial Research Group: The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993; 310:361-368. 11. Diabetes Epidemiology Research International Mortality Study Group: International evaluation of cause-specific mortality and IDDM. Diabetes Care 1991; 14:55-60. 12. Portuese E, Orchard T. Mortality in insulin-dependent diabetes. In: Diabetes in America. National Institutes of Health Publication 95-1468. 1995:221-232.

164

NOD Mice and Related Strains: Research Applications in Diseases

13. Stiller CR, Dupre J, Gent M et al. Effects of cyclosporin immunosuppression in insulin-dependent diabetes mellitus of recent onset. Science 1984; 223:1362-1367. 14. Assan R, Feutren G, Debray-Sachs M et al. Metabolic and immunological effects of cyclosporin in recently diagnosed type 1 diabetes mellitus. Lancet 1985; 1:67-71. 15. Silverstein J, Maclaren N, Riley W et al. Immunosuppression with azathioprine and prednisone in recent-onset insulin-dependent diabetes mellitus. N Engl J Med 1988; 319:599-604. 16. Canadian-European Randomized Control Trial Group: Cyclosporininduced remission of IDDM after early intervention: association with enhanced insulin secretion. Diabetes 1988; 37:1574-1582. 17. Bougneres PF, Landais P, Boisson C et al. Limited duration of remission of insulin dependency in children with recent overt type 1 diabetes treated with low-dose cyclosporin. Diabetes 1990; 39: 1264-1272. 18. Fathman CG, Myers B. Cyclosporin therapy for autoimmune disease. N Engl J Med 1992; 326:1693-1695. 19. Lipton RB, LaPorte RE, Becker DJ et al. Cyclosporin therapy for prevention and cure of IDDM: epidemiological perspectives of risks and benefits. Diabetes Care 1990; 13:776-784. 20. Palmer JP. Predicting IDDM. Use of humoral immune markers. Diabetes Rev 1:104-115. 21. Bingley PJ, Bonifacio E, Gale EAM. Can we really predict IDDM? Diabetes 1993; 42:213-220. 22. Gaskins H, Prochazka M, Hamaguchi K et al. Beta cell expression of endogenous xenotropic retrovirus distinguishes diabetes-susceptible NOD/Lt from resistant NON/Lt mice. J Clin Invest 1992; 90: 2220-2227. 23. Pak CY, Jun HS, Lee M et al. Beta cell-specific expression of retroviral mRNAs and group-specific antigen and the development of beta cellspecific autoimmunity in non-obese diabetic mice. Autoimmunity 1995; 20:19-24. 24. Tsumura H, Wang JZ, Ogawa S et al. The character of endogenous retroviruses in pancreatic beta-cells of NOD mice. Lab Anim Sci 1994; 44:47-51. 25. Conrad B, Weldmann E, Trucco G et al. Evidence for superantigen involvement in insulin-dependent diabetes mellitus aetiology. Nature 1994; 371:351-355. 26. Pozzilli P, Signore A, Williams AJK et al. NOD mouse colonies around the world—recent facts and figures. Immunol Today 1993; 14: 193-196. 27. Suzuki T, Yamada T, Takao T et al. In: Rygaard NBJ, Graem N, Sprang-Thomsen M, eds. Immune-Deficient Animals in Biomedical Research. Basel: Karger, 1984:112-116. 28. Scott FW, Marliss EB. Conference summary: diet as an environmental factor in development of insulin-dependent diabetes mellitus. Can

NOD Mice as a Model for Therapeutic Interventions in Human IDDM

165

J Physiol Pharmacol 1991; 69:311-319. 29. Hoorfar H, Buschard K, Dagnaes-Hansen F. Prophylactic nutritional modification of the incidence of diabetes in autoimmune non-obese diabetic (NOD) mice. Br J Nutr 1993; 69:597-607. 30. Maclaren NK, Atkinson MA. Is insulin dependent diabetes mellitus environmentally induced? New Engl J Med 1992; 327:348-349. 31. Szopa TM, Titchener PA, Portwood ND et al. Diabetes mellitus due to viruses—some recent developments. Diabetologia 1993; 36:687-695. 32. Oldstone MB. Prevention of type I diabetes in nonobese diabetic mice by virus infection. Science 1988; 239:500-502. 33. Hermitte L, Vialettes B, Naquet P et al. Paradoxical lessening of autoimmune processes in non-obese diabetic mice after infection with the diabetogenic variant of encephalomyocarditis virus. Eur J Immunol 1990; 20:1297-1303. 34. Wilberz S, Partke HJ, Dagnaes-Hansen F et al. Persistent MHV (mouse hepatitis virus) infection reduces the incidence of diabetes mellitus in non-obese diabetic mice. Diabetologia 1991; 34:2-5. 35. Takei I, Asaba Y, Kasatani T et al. Suppression of development of diabetes in NOD mice by lactate dehydrogenase virus infection. J Autoimmun 1992; 5:665-673. 36. Oldstone MBA. Molecular mimicry in autoimmune disease. Cell 1990; 50: 819-820. 37. Wucherpfennig KW, Stromenger JL. Molecular mimicry in T cellmediated autoimmunity: viral peptides activate human T cell clones specific for myelin basic protein. Cell 1995; 80:695-705. 38. Serreze DV Hamaguchi K, Leiter EH. Immunostimulation circumvents diabetes in NOD/Lt mice. J Autoimmun 1990; 2:759-776. 39. Bowman MA, Leiter EH, Atkinson MA. Prevention of diabetes in the NOD mouse: implications for therapeutic intervention in human disease. Immunol Today 1994; 15:115-120. 40. Rabinovich A. Immunoregulatory and cytokine imbalances in the pathogenesis of IDDM. Therapeutic intervention by immunostimulation? Diabetes 1994; 43:613-621. 41. Rabinovich A, Singh B. Immune system ‘stimulation’ can prevent autoimmune diabetes. In: Lessons from Animal Diabetes. Shafrir E, London: ed, Smith Gordon, 1994:41-48. 42. Gale EAM, Bingley PJ. Can we prevent IDDM? Diabetes Care 1994; 17:339-344. 43. Mori Y, Suko M, Okudaira H et al. Preventive effects of cyclosporin on diabetes in NOD mice. Diabetologia 1986; 4:244-247 44. Miyagawa J, Yamamoto K, Hanatusa T et al. Preventive effect of a new immunosuppressant FK-506 on insulitis and diabetes in nonobese diabetic mice. Diabetologia 1990; 33:503-505. 45. Nicoletti F, Zaccone P, Di Marco R et al. Effects of sodium fusidate in animal models of insulin-dependent diabetes mellitus and septic shock. Immunology 1995; 85:645-650.

166

NOD Mice and Related Strains: Research Applications in Diseases

46. Gross DJ, Sidi H, Weiss L et al. Prevention of diabetes mellitus in non-obese diabetic mice by Linomide, a novel immunomodulating drug. Diabetologia 1994; 37:1195-1201. 47. Kino K, Misumoto K, Sone T et al. An immunomodulating protein Ling Zhi-8 (LZ-8) prevents insulitits in non-obese diabetic mice. Diabetologia 1990; 33:713-718. 48. Califiore R, Basta G, Falorni A et al. Preventative effects of azathioprine (AZA) on the onset of diabetes mellitus in NOD mice. J Endocrinol Invest 1993; 16:869-873. 49. Baeder WL, Sredy J, Sehgal SN et al. Rapamycin prevents the onset of insulin-dependent diabetes mellitus (IDDM) in NOD mice. Clin Exp Immunol 1992; 89:174-178. 50. Kurasawa K, Koike T, Matsumura R et al. The immunosuppressant FK-506 prevents progression of diabetes in nonobese diabetic mice. Clin Immunol Immunopathol 1990; 57:274-279. 51. Strandell E, Sandler S. Protection against hyperglycemia in female nonobese diabetic mice treated with 15-deoxyspergualin. Autoimmunity 1992; 14:159-165. 52. Lee KU, Amano K, Yoon JW. Evidence for initial involvement of macrophage in development of insulitis in NOD mice. Diabetes 1988; 37:989-991. 53. Boitard C, Bendelac A, Richard MF et al. Prevention of diabetes in nonobese diabetic mice by anti-IA monoclonal antibodies: transfer of protection by splenic T cells. Proc Natl Acad Sci USA 1988; 85:9719-9723. 54. Hurtenbach U, Lier E, Adorini L et al. Prevention of autoimmune diabetes in non-obese diabetic mice by treatment with a class II major histocompatibility complex-blocking peptide. J Exp Med 1993; 177:1499-1504. 55. Harada M, Makino S. Suppression of overt diabetes in NOD mice by anti-thymocyte serum or anti-Thy 1, 2 antibody. Jikken Dobutso 1986; 35:501-504. 56. Maki T, Ichikawa T, Blanco R et al. Long-term abrogation of autoimmune diabetes in nonobese diabetic mice by immunotherapy with anti-lymphocyte serum. Proc Natl Acad Sci USA 1992; 89: 3434-3438. 57. Sempe P, Bedossa P, Richard MF et al. Anti-alpha/beta T cell receptor monoclonal antibody provides an efficient therapy for autoimmune diabetes in nonobese diabetic (NOD) mice. Eur J Immunol 1991; 21:1163-1169. 58. Shizuru JA, Taylor-Edwards C, Banks BA et al. Immunotherapy of the nonobese diabetic mouse: treatment with an antibody to T-helper lymphocytes. Science 1988; 240:659-662. 59. Charlton B, Mandel TE. Recurrence of insulitis in the NOD mouse after early prolonged anti-CD4 monoclonal antibody treatment. Autoimmunity 1989; 4:1-7.

NOD Mice as a Model for Therapeutic Interventions in Human IDDM

167

60. Kurasawa K, Sakamoto A, Maeda T et al. Short term administration of anti-L3T4 moAb prevents diabetes in NOD mice. Clin Exp Immunol 1993; 91:376-380. 61. Hayward AR, Shriber M, Cooke A et al. Prevention of diabetes but not insulitis in NOD mice injected with antibody to CD4. J Autoimmun 1993; 6:301-310. 62. Chatenoud L, Thervet E, Primo et al. Anti-CD3 antibody induces long-term remission of overt autoimmunity in nonobese diabetic mice. Proc Natl Acad Sci USA 1994; 91:123-127. 63. Lenschow DJ, Ho SC, Asattar H et al. Differential effects of anti-B71 and anti-B7-2 monoclonal antibody treatment on the development of diabetes in the nonobese diabetic mouse. J Exp Med 1995; 181:1145-1155. 64. Xang XD, Karin N, Tisch R et al. Inhibition of insulitiis and prevention of diabetes in nonobese diabetic mice by blocking L-selectin and very late antigen 4 adhesion receptors. Proc Natl Acad Sci USA 1993; 90:10494-10498. 65. Hasegawa Y, Yokono K, Taki T et al. Prevention of autoimmune insulin-dependent diabetes in non-obese diabetic mice by anti-LFA-1 and anti-ICAM-1 mAb. Int Immunol 1994; 6:831-838. 66. Xang XD, Michie SA, Tisch R et al. Cell adhesion molecules: a selective therapeutic target for alleviation of IDDM. J Autoimmun 1994; 7:859-864. 67. Xang XD, Michie SA, Tisch R et al. A predominant role of integrin alpha 4 in the spontaneous development of autoimmune diabetes in nonobese diabetic mice. Proc Natl Acad Sci USA 1994; 91: 12604-12608. 68. Xang XD, Tisch R, Singer SM et al. Effect of tumor necrosis factor alpha on insulin-dependent diabetes mellitus in NOD mice. The early development of autoimmunity and the diabetogenic process. J Exp Med 1994; 180:995-1004. 69. Campbell I, Kay TWH, Oxbrow L et al. Essential role for interferongamma and interleukin-6 in autoimmune insulin-dependent diabetes. J Clin Invest 1991; 87:739-742. 70. Andersson A, Ekstrand-Hammarstrom B, Eriksson B et al. Neonatal treatment with monoclonal natural antibodies restores a normal pattern of VH gene utilization in the non-obese diabetic mouse. Int Immunol 1994; 6:623-630. 71. Parish NM, Hutchings PR, O’Reilly L et al. Tolerance induction as a therapeutic strategy for the control of autoimmune endocrine disease in mouse models. Immunl Rev 1995; 144:269-300. 72. Hayward AR, Shreiber M. Neonatal injection of CD3 antibody into nonobese diabetic mice reduces the incidence of insulitis and diabetes. J Immunol 1989; 143:1555-1559. 73. Atkinson MA, Maclaren NK. Islet cell autoantigens in insulin-dependent diabetes. J Clin Invest 1993; 92:1608-1616.

168

NOD Mice and Related Strains: Research Applications in Diseases

74. Gladstone P, Nepom GT. The prevention of IDDM. Injecting insulin into the cytokine network. Diabetes 1995; 44:859-862. 75. Schloot N, Eisenbarth GS. Isohormonal therapy of endocrine autoimmunity. Immunol Today 1995; 16:289-294. 76. Atkinson M, Maclaren N, Luchetta R et al. Insulitis and diabetes in NOD mice reduced by prophylactic insulin therapy. Diabetes 1990; 39:933-937. 77. Zhang ZJ, Davidson L, Eisenbarth G et al. Suppression of diabetes in nonobese diabetic mice by oral administration of porcine insulin. Proc Natl Acad Sci USA 1991; 88:10252-10256. 78. Muir A, Peck A, Clare-Salzler MC et al. Insulin immunization of nonobese diabetic mice induces a protective insulitis characterized by diminished intra-islet interferon-γ transcription. J Clin Invest 1995; 95:628-634. 79. Bowman MA, Campbell L, Darrow BL et al. Immunological and metabolic effects of prophylactic insulin therapy in the NOD-scid/scid adoptive transfer model of IDDM. Diabetes 1996; 45:205-208. 80. Hancock WW, Polanski M, Zhang J et al. Suppression of insulitis in non-obese diabetic (NOD) mice by oral insulin administration is associated with selective expression of interleukin-4 and -10, transforming growth factor- β , and prostoglandin-E. Am J Pathol 1995; 147:1193-1199. 81. Bergerot I, Fabien N, Mayer A et al. Active suppression of diabetes after oral administration of insulin is determined by antigen dosage. Ann NY Acad Sci 1996; 13:362-362. 82. Daniel D, Wegmann DR. Protection of nonobese diabetic mice from diabetes by intranasal or subcutaneous administration of insulin peptide B-(9-23). Proc Natl Acad Sci USA 1996; 93:956-960. 83. Weiner HL, Friedman A, Miller A et al. Oral tolerance: immunologic mechanisms and treatment of animal and human organ-specific autoimmune diseases by oral administration of autoantigens. Annu Rev Immunol 1994; 12:809-837. 84. Ellis TE, Atkinson MA. The clinical significance of an autoimmune response against glutamate decarboxylase. Nature Med 1995; 2:148-153. 85. Kaufman D, Clare-Salzler M, Tian J et al. Spontaneous loss of T-cell tolerance to glutamic acid decarboxylase in murine insulin-dependent diabetes. Nature 1993; 366:69-72. 86. Tisch R, Yang X-D, Singer S et al. Immune response to glutamic acid decarboxylase correlates with insulitis in non-obese diabetic mice. Nature 1993; 366:72-75. 87. Elliott JF, Quin HY, Bhatti S et al. Immunization with the larger isoform of mouse glutamic acid decarboxylase (GAD67) prevents autoimmune diabetes in NOD mice. Diabetes 1994; 43:1494-1499. 88. Petersen JS, Karlsen AE, Markholst H et al. Neonatal tolerization with glutamic acid decarboxylase but not with bovine serum albumin de-

NOD Mice as a Model for Therapeutic Interventions in Human IDDM

169

lays the onset of diabetes in NOD mice. Diabetes 1994; 43:1478-1484. 89. Pleau JM, Fernandex-Saravia F, Esling A et al. Prevention of autoimmune diabetes in nonobese diabetic female mice by treatment with recombinant glutamic acid decarboxylase (GAD 65). Clin Immunol Immunopathol 1995; 76:90-95. 90. Elias D, Markovits D, Reshef T et al. Induction and therapy of autoimmune diabetes in the non-obese diabetic (NOD/LT) mouse by a 65-kDa heat shock protein. Proc Natl Acad Sci USA 1990; 87: 1576-1580. 91. Elias D, Reshef T, Birk OS et al. Vaccination against autoimmune mouse diabetes with a T-cell epitope of the human 65-kDa heat shock protein. Proc Natl Acad Sci USA 1991; 88:3088-3091. 92. Elias D, Cohen IR. Peptide therapy for diabetes in NOD mice. Lancet 1994; 343:704-706. 93. Formby B, Shao T. T cell vaccination against autoimmune diabetes in nonobese diabetic mice. Ann Clin Lab Sci 1993; 23:137-147. 94. Reich EP, Scaringe D, Yagi J et al. Prevention of diabetes in NOD mice by injection of autoreactive T-lymphocytes. Diabetes 1989; 38:1647-1651. 95. Clare-Salzler MJ, Brooks J, Chai A et al. Prevention of diabetes in nonobese diabetic mice by dendritic cell transfer. J Clin Invest 1992; 90:741-748. 96. Gerling IC, Serreze DV, Christianson S et al. Intrathymic islet cell transplantation reduces beta-cell autoimmunity and prevents diabetes in NOD/Lt mice. Diabetes 1992; 41:1672-1676. 97. Charlton B, Taylor-Edwards C, Tisch R et al. Prevention of diabetes and insulitis by neonatal intrathymic islet administration in NOD mice. J Autoimmun 1994; 7:549-560. 98. Sadelain MW, Qin HY, Lauzon J et al. Prevention of type I diabetes in NOD mice by adjuvant immunotherapy. Diabetes 1990; 39:583-589. 99. McInerney MF, Pek SB, Thomas DW. Prevention of insulitis and diabetes onset by treatment with complete Freund’s adjuvant in NOD mice. Diabetes 1991; 40:715-725. 100. Ulaeto D, Lacy PE, Kipnis DM et al. A T-cell dormant state in the autoimmune process of nonobese diabetic mice treated with complete Freund’s adjuvant. Proc Natl Acad Sci USA 1992; 89:3927-3931. 101. Qin HY, Sadelain MW, Hitchon C et al. Complete Freund’s adjuvant-induced T cells prevent the development and adoptive transfer of diabetes in nonobese diabetic mice. J Immunol 1993; 150: 2072-2780. 102. Harada M, Kishimoto Y, Makino S. Prevention of overt diabetes and insulitis in NOD mice by a single BCG vaccination. Diabetes Res Clin Pract 1990; 8:85-89. 103. Toyota T, Satoh J, Oya K et al. Streptococcal preparation (OK-432) inhibits development of type I diabetes in NOD mice. Diabetes 1986; 35:496-499.

170

NOD Mice and Related Strains: Research Applications in Diseases

104. Kawamura T, Nagata M, Utsugi T et al. Prevention of autoimmune type I diabetes by CD4+ suppressor T cells in superantigen-treated non-obese diabetic mice. J Immunol 1993; 151:4362-4370. 105. Huang SW, Luchetta R, Peck AB. Pertussigen treatment retards, but fails to prevent, the development of type 1, insulin-dependent diabetes mellitus (IDDM) in NOD mice. Autoimmunity 1991; 9:311-317. 106. Serreze DV, Gaskins HR, Leiter EH. Defects in the differentiation and function of antigen presenting cells in NOD/Lt mice. J Immunol 1993; 150:2534-2543 107. Langmuir P, Bridgett M, Bothwell A et al. Bone marrow abnormalities in the non-obese diabetic mouse. Int Immunol 1993; 5:169-177. 108. Rapoport M, Zipris D, Lazarus A et al. Interleukin 4 reverses T cell proliferative unresponsiveness and prevents the onset of diabetes in nonobese diabetic mice. J Exp Med 1993; 178:87-99. 109. Li H, Kaufman CL, Boggs SS et al. Mixed allogeneic chimersim induced by a sublethal approach prevents autoimmune diabetes and reverses insulitis in non-obese diabetic (NOD) mice. J Immunol 1996; 156:380-388. 110. Pennline KJ, Roque-Gaffney E, Monahan M. Recombinant human IL-10 prevents the onset of diabetes in the nonobese diabetic mouse. Clin Immunol Immunopathol 1994; 71:169-175. 111. Jacob CO, Aiso S, Michie SA et al. Prevention of diabetes in nonobese diabetic mice by tumor necrosis factor (TNF); Similarities between TNFα and interleukin 1. Proc Natl Acad Sci 1990; 87:968-972. 112. Haist RE, Campbell J, Best CH. The prevention of insulin dependent diabetes. N Engl J Med 1940; 223:607-615. 113. Shah SC, Malone JI, Simpson NE. A randomized trial of intensive insulin therapy in newly diagnosed insulin-dependent diabetes mellitus. N Engl J Med 1989; 320:550-554. 114. Godfredson GF, Buschard K, Frandsen EK. Reduction of diabetes incidence of BB Wistar rats by early prophylactic insulin treatment of diabetes-prone animals. Diabetologia 1985; 28:933-935. 115. Buschard K. The functional state of the beta cells in the pathogenesis of insulin-dependent diabetes mellitus. Autoimmunity 1991; 10:65-69. 116. Williams AJ, Beales PE, Krug J et al. Tolbutamide reduces the incidence of diabetes mellitus, but not insulitis, in the non-obese-diabetic mouse. Diabetologia 1993; 36:487-492. 117. Bock T, Kjaer TW, Jorgensen M et al. Reduction of diabetes incidence in NOD mice by neonatal glucose treatment. Acta Pathol Microbiol, Immunol Scand 1991; 99:989-992. 118. Beales PE, Hawa M, Williams AJ et al. Baculofen, a gammaaminobutyric acid-B receptor agonist, delays diabetes onset in the non-obese diabetic mouse. Acta Diabetol 1995; 32:53-56. 119. Ellis TM, Atkinson MA. Early infant diets and insulin-dependent diabetes. Lancet 1996; 347:1464-1465. 120. Scott FW, Norris JM, Kolb H. Milk and type 1 diabetes: examining the evidence and broadening the focus. Diabetes Care 1996;

NOD Mice as a Model for Therapeutic Interventions in Human IDDM

171

19:379-383. 121. Elliott RB, Bibby NJ. Dietary triggers of diabetes in the NOD mouse. In: Shafrir E, ed. Frontiers in Diabetes Research. Lessons from Animal Diabetes III. London: Smith-Gordon, 1990: 195-197. 122. Elliott RB, Martin JM. Dietary protein: a trigger of insulin-dependent diabetes in the BB rat? Diabetologia 1984; 26:297-299. 123. Elliott RB, Reddy SN, Bibby NJ et al. Dietary prevention of diabetes in non-obese diabetic mouse. Diabetologia 1988; 31:62-64. 124. Scott FW, Elliott RB, Kolb H. Diet and autoimmunity: prospects of prevention of type 1 diabetes. Diabetes Nut Metab 1989; 2:61-73. 125. Coleman DL, Kuzava JE, Leiter EH. Effect of diet on incidence of diabetes in nonobese diabetic mice. Diabetes 1990; 39:432-436. 126. Hermitte L, Atlan Gepner C, Payan MJ et al. Dietary protection against diabetes in NOD mice: lack of a major change in the immune system. Diabete Metab 1995; 21:261-268. 127. Benhamou PY, Mullen Y, Clare-Salzler M et al. Essential fatty acid defeciency prevents autoimmune diabetes in nonobese diabetic mice through a positive impact on antigen-presenting cells and Th2 lymphocytes. Pancreas 1995; 11:26-37. 128. Fox HS. Androgen treatment prevents diabetes in nonobese diabetic mice. J Exp Med 1992; 175:1409-1412. 129. Hawkins T, Gala RR, Dumbar JC. The effect of neonatal sex hormone manipulation on the incidence of diabetes in nonobese diabetic mice. Proc Soc Exp Biol Med 1993; 202:201-205. 130. Hawkins TA, Gala RR, Dunbar JC. Prolactin modulates the incidence of diabetes in male and female NOD mice. Autoimmunity 1994; 18:155-162. 131. Fitzpatrick F, Lepault F, Homo-Delarche F et al. Influence of castration, alone or combined with thymectomy, on the development of diabetes in the nonobese diabetic mouse. Endocrinology 1991; 129:1382-1390. 132. Amrani A, Jafarian-Tehrani M, Mormede P et al. Interleukin-1 effect on glycemia in the non-obese diabetic mouse at the pre-diabetic stage. J Endocrinol 1996; 148:139-148. 133. Yamada K, Nonaka K, Hanafusa T et al. Preventive and therapeutic effects of large-dose nicotinamide injections on diabetes associated with insulitis. An observation in nonobese diabetic (NOD) mice. Diabetes 1982; 31:749-753. 134. Reddy S, Bibby NJ, Elliott RB. Early nicotinamide treatment in the NOD mouse: effects on diabetes and insulitis suppression and autoantibody levels. Diabetes Res 1990; 15:95-102. 135. Reddy S, Bibby NJ, Wu D et al. A combined casein-free-nicotinamide diet prevents diabetes in the NOD mouse with minimum insulitis. Diabetes Res Clin Pract 1995; 29:83-92 136. Beales PE, Williams AJ, Albertini MC et al. Vitamin E delays diabetes onset in the non-obese diabetic mouse. Horm Metab Res 1994; 26:450-452.

172

NOD Mice and Related Strains: Research Applications in Diseases

137. Uehara Y, Shimizu H, Sato N et al. Probucol partially prevents development of diabetes in NOD mice. Diabetes Res 1991; 17:131-134. 138. Rabinovitch A, Suarez WL, Power RF. Combination therapy with an antioxidant and a corticosteroid prevents autoimmune diabetes in NOD mice. Life Sci 1992; 51:1937-1943. 139. Rabinovich A, Suarez WL, Power RF. Lazaroid antioxidant reduces incidence of diabetes and insulitis in nonobese diabetic mice. J Lab Clin Med 1993; 121:603-607. 140. Mathiew C, Laureys J, Sobis H et al. 1,23 Dihydroxy vitamin D3 prevents insulitis in NOD mice. Diabetes 1992; 41:1491-1495. 141. Nakajima H, Tochin Y, Fujino-Kurihata H et al. Decreased incidence of diabetes mellitus by monosodium glutamate in the non-obese diabetic (NOD) mouse. Res Commun Chem Pathol Pharmacol 1985; 50:251-257. 142. Corbett JA, Mikhael A, Shimizu J et al. Nitric oxide production in islets from nonobese diabetic mice: aminoguanidine-sensitive and resistant stages in the immunological diabetic process. Proc Natl Acad Sci USA 1993; 90:8992-8995. 143. Bowman M.A, Simell OG, Peck AB et al. Pharmacokinetics of aminoguanidine administration and effects on the diabetes frequency in non-obese diabetic mice. J Pharm Exp Therap 1996; 279:790-794. 144. Keller RJ, Eisenbarth GS, Jackson RA. Insulin prophylaxis in individuals at high risk of type I diabetes. Lancet 1993; 341:927-928. 145. Lewis MC, Canafx DM, Sprafka J et al. Double blind randomized trial of nicotinamide on early onset diabetes. Diabetes Care 1992; 15:121-123. 146. Chase HP, Butler-Simon N, Garg S et al. A trial of nicotinamide in newly diagnosed patients with type 1 (insulin-dependent) diabetes mellitus. Diabetologia 1990; 33:444-446. 147. Elliott RB, Chase HP. Prevention or delay of type 1 (insulin-dependent) diabetes mellitus in children using nicotinamide. Diabetologia 1991; 34:362-365. 148. Dumont-Jerskowitz R, Jackson RA, Soeldner JS et al. Pilot trial to prevent type 1 diabetes: progression to overt IDDM despite oral nicotinamide. J Autoimmun 1989; 2:733-737. 149. Pociot F, Reimers JI, Andersen HU. Nicotinamide—biological actions and therapeutic potential in diabetes prevention. Diabetologia 1993; 36:574-576. 150. Shehadeh N, Calcinaro F, Bradley BJ et al. Effect of adjuvant therapy on development of diabetes in mouse and man. Lancet 1994; 434:706-707. 151. Posselt AM, Barker CF, Tomaszewski JE et al. Induction of donorspecific unresponsiveness by intrathymic islet transplantation. Science 1990; 249:1293-1295. 152. Akerblom HK, Savilahti E, Saukkonen TT et al. The case for elimination of cow’s milk in early infancy in the prevention of type 1 diabetes: the Finnish experience. Diab Metab Rev 1993; 9:269-278.

The Use of NOD/LtSz-SCID/SCID Mice in Biomedical Research

173

CHAPTER 7

The Use of NOD/LtSz-scid/scid Mice in Biomedical Research Dale L. Greiner, Leonard D. Shultz

T

he severe combined immunodeficiency (scid) mutation in mice was discovered by Bosma et al in 1983 while investigating immunoglobulin (Ig) isotypes in C.B-17 strain mice.1 C.B-17 mice are an Ig congenic partner of BALB/c mice, deriving a portion of Chromosome 12 containing the Ig heavy chain isotype (Igh-1b) from the C57BL/Ka strain of mice.2 During the course of their studies, the investigators observed mice deficient in serum Ig, and selective breeding rapidly generated a congenic C.B-17-scid/scid strain of mice. The scid mutation was found to segregate as a single, autosomal recessive trait on Chromosome 16 linked to mahoganoid, a recessive coat color marker.3 Recently, the catalytic subunit of a DNA-dependent protein kinase encoded by a gene termed “protein kinase, DNA activated catalytic polypeptide” (prkdc) has been identified as the candidate gene for the scid mutation.4-8

PHENOTYPIC CHARACTERISTICS OF scid MICE IMMUNODEFICIENCY Over 1300 studies utilizing mice homozygous for the scid mutation have been reported since the mutation was first described in 1983. These studies have been based on the severely deficient humoral and cell mediated immunity observed in mice homozygous for the scid NOD Mice and Related Strains: Research Applications in Diabetes, AIDS, Cancer and Other Diseases, edited by Edward Leiter and Mark Atkinson. © 1998 R.G. Landes Company.

174

NOD Mice and Related Strains: Research Applications in Diseases

mutation due to their total absence of mature T or B cells.9 The lack of mature lymphocytes renders the mice severely immunodeficient, and allows the ready acceptance of lymphohemopoietic grafts for analysis of lymphocyte development and function in an in vivo environment where the host adaptive immune system does not contribute to the lymphoid function under study. It also allows the mice to accept foreign grafts of hemopoietic tissues and solid organs for subsequent study in an in vivo environment. The lack of mature antigenreceptor lymphocytes in scid/scid (hereafter referred to as scid) mice results from the inability of developing lymphocytes to undergo productive T cell receptor (TCR) or Ig chain rearrangement during ontogeny.10,11 The inability of antigen receptor recombination to occur is due to a defect in a DNA recombinase enzyme system.12 The defect in lymphocyte development occurs in thymocytes at the CD4–CD8– stage of differentiation.13,14 In B cells, the developmental arrest occurs at the B220+, cytoplasmic µ–, surface µ– stage of differentiation.15 These stages of lymphocyte development occur just before TCR expression in thymocytes and rearrangement of Ig light chains in B cells, respectively. This is at the variable-diversity-joining (VDJ) rearrangement step of differentiation.10,11 Despite the presence of the defect in VDJ recombinase activity, some productive rearrangements have been noted to occur, and result in pauciclonal expansion of receptor bearing lymphocytes with limited antigen receptor diversity.16 This phenomena is described as “leakiness”, and occurs to varying extent in scid mice as they age.17,18 Leakiness in the B cell compartment is easily monitored by analysis of the presence of serum Ig and in the T cell compartment by the ability of scid mice to reject allogeneic skin grafts.16 The “leaky” phenotype interferes with the utility of the scid mouse for the developmental, functional and grafting studies that are performed in these mice. The cause of the “leaky” phenotype is unknown, but it appears to be strongly influenced by the environment16 and strain background17-19 of the scid mouse.

DEFECTIVE DNA REPAIR Although the primary focus of the scid mutation has been on its effect in immune system development, the mutation results in a more generalized phenotype. Cells of all lineages examined in scid mice are deficient in their ability to repair double-stranded DNA breaks induced by ionizing radiation or radiomimetic drugs. Cells from scid mice have been found to be 2- to 3-fold more sensitive to ionizing

The Use of NOD/LtSz-SCID/SCID Mice in Biomedical Research

175

radiation as compared to cells from wild-type parental mice. This has been hypothesized to result from defects in DNA strand rejoining as a result of defective recombinase enzyme activity and the documented functional defect in DNA repair in scid mice.18,20-24 C.B-17-scid mice are extremely sensitive to whole body irradiation. While the scid mice will not survive doses above 400 rad gamma irradiation, the congenic +/+ mice readily survive much higher irradiation doses.

ROBUST INNATE IMMUNITY In contrast to the severe deficiency in their adaptive immune system, C.B-17-scid mice have a normal to enhanced innate immune system. The scid mutation blocks T and B cell differentiation at the antigen receptor rearrangement point of differentiation, but does not alter myeloid, erythroid or natural killer cell (NK) development. C.B-17-scid mice display normal granulocyte and macrophage function. Macrophages are readily activated following infection with Listeria monocytogenes,25 and antigen presenting activity of C.B-17-scid splenocytes is normal.26 NK cell activity in C.B-17-scid mice is elevated in relation to that observed in the parental C.B-17 wild type strain.18 The red blood cell (RBC) count is normal, and hematocrit levels are within a normal range and comparable to those observed in C.B-17 wild type mice.18 CARE AND MANAGEMENT OF scid MOUSE COLONIES HOUSING REQUIREMENTS A number of scid mouse research colonies at The Jackson Laboratory are housed under specific pathogen free (SPF) conditions in filter top cages. Although many animal facilities house scid mice in microisolators or laminar flow ventilated cages pressurized individually, this is not required in an SPF facility. The scid mice are maintained on acidified water, and treated on three consecutive days per week with trimethoprim-sulfamethoxazole (40 mg of trimethoprim and 200 mg of sulfamethoxazole per 5 ml of suspension, 0.125 ml of suspension per 4 ml of water per mouse per day). The trimethoprimsulfamethoxazole treatment prevents infection with Pneumocystis carinii, a common pathogen often observed in immuno-incompetent hosts.27 In rooms free of Pneumocystis carinii, trimethoprimsulfamethoxazole treatment is not required. It should be cautioned, however, that sentinels brought into a mouse room, although free of

176

NOD Mice and Related Strains: Research Applications in Diseases

infectious agents typically screened for in SPF facilities, may still be carrying Pneumocystis carinii. This can be easily determined by histological staining of lung tissue with Gomori’s methenamine-silver stain.28 Autoclaved cages and food are used for housing of scid mice, and the cages are changed once/week. Although scid mice lack an adaptive immune system, they appear to be as resistant as athymic nu/nu mice to most infectious diseases caused by opportunistic microorganisms, and require no additional protective measures above that employed for the maintenance of nu/nu mice. BREEDING The scid mutation does not alter the breeding capability of the inbred strain upon which the scid mutation is backcrossed. C.B-17-scid mice are easily bred under SPF conditions, and provide excellent fostering of their young. The average litter size of C.B-17-scid mice is comparable to that of C.B-17 wild type mice. Backcrossing the scid mutation onto the BALB/cBy, NOD/Lt, C57BL/6J, C3H/HeJ, and DBA/ 2J strains of mice does not result in alteration of the breeding capabilities or fostering characteristics of the scid congenic stock as compared to the wild type stocks of mice. For example, the average litter size of NOD/LtSz-scid mice, like the NOD/Lt +/+ mice, averages 8-12 offspring and the mothers display excellent nurturing characteristics. The breeding characteristics contrast sharply with that observed in nu/nu mice, where the athymic nu/nu mother often displays poor nurturing characteristics in the care of her offspring. This is further highlighted by comparing C57BL/6J-nu/nu and C57BL/6J-scid mice. The C57BL/6J-nu/nu mice have small litters and poor nurturing characteristics, while the C57BL/6J-scid litter sizes are comparable to that of C57BL/6J wild type mice and the mothers equivalent in the care of their offspring.

UTILITY OF C.B-17-scid MICE AS RESEARCH TOOLS LIMITATIONS OF C.B-17-scid MICE AS RECIPIENTS OF DIABETOGENIC NOD CELLS AND AS HOSTS FOR HUMAN LYMPHOHEMOPOIETIC CELLS As discussed above, the availability of C.B-17-scid mice has facilitated research in numerous areas. However, there are a number of limitations associated with the use of C.B-17-scid mice. Although these mice lack an adaptive immune system, they still possess a robust innate immune system. They exhibit normal to elevated NK cell and

The Use of NOD/LtSz-SCID/SCID Mice in Biomedical Research

177

myeloid function, and generate increased hemolytic complement activity. The presence of an enhanced innate immune system facilitates the ease of housing and handling of the mouse stock, but hinders the engraftment of normal and malignant allogeneic and xenogeneic lymphohemopoietic tissues. Regarding the use of C.B-17-scid mice in murine diabetes, this strain of mice differs from the prototypic mouse model of insulin dependent diabetes mellitus (IDDM), the NOD/Lt mouse, in both the major histocompatibility complex (MHC) class I and class II antigens. These histocompatibility differences render the C.B-17-scid mice inappropriate as hosts for adoptive transfer of NOD/Lt cells for studies of diabetes. C.B-17-scid mice accept human lymphohemopoietic cell grafts. They support engraftment with normal human peripheral blood lymphocytes (Hu-PBL-SCID),29 human fetal thymus tissue (SCID-Hu),27 and human stem cells.30 Xenogeneic chimeric C.B-17-scid mice have been widely used as small animal models for the study of human lymphohemopoietic cell development and function. However, the engraftment levels observed in these xenogeneic scid mouse-human chimeras are extremely low and variable, hindering the utility of these potentially important model systems. DEVELOPMENT OF NOD/LtSz-scid MICE To attempt to address the shortcomings of the C.B-17-scid mouse as a recipient strain for human lymphohemopoietic cells and tumors, and as adoptive recipients of NOD/Lt cells for the study of diabetes, the scid mutation was backcrossed onto the NOD/Lt strain background. NOD/Lt wild type mice spontaneously develop an autoimmune form of T cell-mediated diabetes with many disease characteristics that are similar to those observed in human IDDM.31 The NOD/Lt strain also has defects in NK cell function,32,33 antigen presenting cell (APC) development and function,33-36 and genetically lacks C5,37 resulting in a deficiency of hemolytic complement. It was reasoned that these deficiencies in innate immunity would make an NODscid mouse an improved recipient of human lymphohemopoietic cells and tumors. It was also reasoned that NOD/Lt strain mice homozygous for the scid mutation would fail to develop IDDM since the mouse would be deficient in T cells and the development of diabetes is T cell dependent.31 This latter characteristic would facilitate the study of the T cells in IDDM. The recently generated NOD/LtSz-scid mouse provides an adoptive recipient that is histocompatible with NOD/Lt

178

NOD Mice and Related Strains: Research Applications in Diseases

strain mice.18 Moreover, NOD/LtSz-scid mice can be induced to express IDDM by the transfer of purified T cell populations or T cell clones in the absence of the confounding effects of a host adaptive immune system or of histocompatibility differences.

CHARACTERISTICS OF NOD/Lt/Sz-scid MICE INNATE IMMUNITY There are many differences in innate immunity between the C.B17-scid strain and the NOD/LtSz-scid strain of mice. These are summarized in Table 7.1 and detailed below. Leakiness The NOD/Lt inbred strain of mouse displays multiple defects in innate immunity. Backcrossing of the scid mutation to the NOD/Lt inbred strain of mouse resulted in the retention of the deficiencies in innate immunity in the new congenic stock. NOD/LtSz-scid mice, as expected, lack mature T and B cells, and remain insulitis and diabetes-free throughout life.18 In contrast to C.B-17-scid mice which develop circulating levels of serum Ig at a frequency of close to 90% by 200 days of age, less than 10% of NOD/LtSz-scid mice develop detectable serum Ig by this age.18 The low level of leakiness in serum Ig was confirmed at the T cell level by the observation that NOD/LtSzscid mice are able to retain MHC class I-disparate allogeneic skin grafts for several months. Peripheral Lymphoid Tissues NOD/LtSz-scid mice display decreased percentages of lymphocytes, with an increase in percentages of peripheral blood neutrophils, monocytes and eosinophils.18 As a result, peripheral lymphoid organs are devoid of functional lymphocytes, and consist mostly of stromal cells, granulocytes and monocytes. The deficiency of lymphocytes also results in greater than a 4-fold reduction of nucleated spleen cells in NOD/LtSz-scid mice as compared with NOD/Lt strain mice, and over a 2-fold reduction in bone marrow cell counts. Surprisingly, NOD/ LtSz-scid mice display reduced erythrocyte counts, and increased erythrocyte mean cell volumes. This results in the expression of a mild macrocytic anemia, even though the hematocrit levels are normal.

The Use of NOD/LtSz-SCID/SCID Mice in Biomedical Research

179

Table 7.1. Comparision of lymphohemopoietic and innate immune parameters between the NOD/LtSz-scid stock and the C.B-17-scid stock of mice Parameter

C.B-17-scid mice

NOD/LtSz-scid mice

Lifespan Macrophage Development Antigen Presenting Cell Function Natural Killer Cell Activity Hemolytic Complement Age-Dependent Serum Immunoglobulin Production Red Blood Cells

>2 years Normal Normal Elevated Present >70% by 200 days of age Normal

Growth of Human CEM Tumor Cells Engraftment of Human Peripheral Lymphocytes Engraftment of Human Hemopoietic Stem Cells

Delayed

8.5 months Impaired Defective Severely Reduced Absent <7% by 200 days of age Mild Macrocytic Anemia Rapid

Low

High

Low

High

Myeloid Development and Function NOD/Lt mice are deficient in APC development and function,33-36 and this characteristic has been found to include defective interleukin (IL)-1 secretion.33 The defect is readily revealed by culturing NOD/Lt macrophages with lipopolysaccharide (LPS). Congenic NOD/LtSz-scid mice also display the macrophage defect in IL-1 secretion. This observation confirms the defect in APC activity in NOD/LtSz-scid mice, and suggests that myeloid development is impaired, similar to that observed in the parental stock of NOD/Lt mice. Natural Killer Activity Natural killer cells are thought to be important in natural resistance to bone marrow grafts and to lymphohemopoietic cell grafts.38-40 Previous studies of C.B-17 wild type and C.B-17-scid mice have demonstrated that backcrossing the scid mutation onto the wild type strain increases NK cell activity.18 However, NOD/Lt mice are severely depressed in NK cell activity, and NK cell activity remains low in NOD/LtSz-scid strain mice.18 NK cell activity in NOD/LtSz-scid mice can be demonstrated following treatment with the interferon inducer, polyinosinic acid-polycytidylic acid (poly I:C), and this activity is

180

NOD Mice and Related Strains: Research Applications in Diseases

unaffected by treatment with anti-NK1.1 monoclonal antibody. This result suggests that the NK cells are present at low levels in NOD/ LtSz-scid mice and are NK1.1-negative.18 Hemolytic Complement Deficiency Approximately 30% of mouse strains lack hemolytic complement due to a genetic deficiency in C5.2,37 C5 is part of the membrane attack complex, and is required for the hemolytic activity of the complement pathway. C5 deficiency is caused by a deletion in the hemolytic complement (Hc) locus which maps to Chromosome 2.41 Since the scid mutation maps to Chromosome 16,3 development of the NOD/LtSz-scid congenic stock of mice did not result in the loss of the unlinked Hc locus on the NOD/Lt background. Analysis of C.B17-scid mice demonstrated increased hemolytic complement activity compared to C.B-17 wild type mice, while the NOD/LtSz-scid mouse lacked detectable hemolytic complement.18 THYMIC LYMPHOMAGENESIS NOD/LtSz-scid mice have a mean life span of approximately 260 days of age.18 The cause of death in most NOD/LtSz-scid mice is due to outgrowth of thymic lymphomas. Thus, long-term experiments using NOD/LtSz-scid mice as adoptive hosts of NOD cells or of human cells are constrained by the presence of the thymic lymphomas in almost 70% of the mice by 40 weeks of age and their early death due to the thymic tumor.42 Although NOD/Lt mice develop a variety of neoplasms, including non-thymic lymphomas, thymic lymphomas are rarely observed.43 The high incidence of thymic lymphomas in NOD/LtSz-scid mice has been postulated to result from the expression of a NOD mouse-unique endogenous ecotropic murine leukemia provirus locus.42 This locus is located on the proximal region of Chromosome 11, and is termed Emv30.42,44 The Emv-30 provirus is thought to synergize with the scid-imparted block in thymocyte development, leading to activation of the NOD-unique Emv30, resulting in thymomagenesis. This hypothesis was tested directly by producing a stock of Emv30 nullNOD/LtSz-scid mice by congenic replacement of the proximal end of Chromosome 11 with genetic material derived from the closely related NOR/Lt strain.45 These Emv30nullNOD/LtSz-scid mice still developed thymic lymphomas, but the rate of thymic development was significantly delayed, with a reduction from 76% to 20% of tumors greater than 170 mg in 25-week-

The Use of NOD/LtSz-SCID/SCID Mice in Biomedical Research

181

old females.45 The Emv30nullNOD/LtSz-scid mice retained their ability to express diabetes following the adoptive transfer of NOD/Lt T cells, and also retained the enhanced ability to support the engraftment of human peripheral blood leukocytes that is observed in NOD/LtSz-scid mice (see below).

NOD/LtSz-scid MICE IN DIABETES RESEARCH ADOPTIVE TRANSFER OF DIABETES INTO NOD/LtSz-scid MICE One of the important uses of the NOD/LtSz-scid mouse is as an adoptive recipient of cells from NOD/Lt mice. This model system allows the study of diabetes induction in an immuno-incompetent host without the confounding effects of a background host adaptive immune system. Previous hosts used for the adoptive transfer of NOD diabetes were irradiated NOD recipients,46 or young pre-diabetic recipients.47 In the former case, the contribution of radio-resistant T and B cells to disease induction, and in the latter case, the contribution of the developing immune system of young NOD mice, could not be excluded as contributing factors. NOD-nu/nu mice have also been used as adoptive recipients of diabetogenic cells from NOD mice, but NOD-nu/nu mice have normal levels of B lymphocytes and have circulating serum Ig.48,49 In these adoptive transfer model systems, it is impossible to dissect the relative contribution of the adoptively transferred cells from the contribution of the host B cells or of extrathymic T cells in diabetes induction. These caveats have been overcome by utilization of the NOD/LtSz-scid mouse as the adoptive recipient of NOD cells. The absolute requirement for donor NOD T cells in the adoptive transfer of diabetes was first demonstrated by Christianson et al using NOD/LtSz-scid mice.50 In this study, the ability of pre-diabetic and diabetic T cell subsets from NOD/Lt mice to transfer diabetes was compared. It was observed that T cells from diabetic NOD/Lt mice were able to induce insulitis within 10 days and diabetes within 24 days in NOD/LtSz-scid recipients. In contrast, when cells were obtained from young pre-diabetic donors, the time to diabetes onset was increased to 85 days.50 Injection of T cell fractions enriched in CD4+ T cells from diabetic donors were able to transfer diabetes, but purified CD8+ T cells were not. If the CD8+ T cells were rigorously excluded from the CD4+ T cell population by injection of enriched CD4+ T cells followed by treatment of the NOD/LtSz-scid recipient

182

NOD Mice and Related Strains: Research Applications in Diseases

with anti-CD8 antibody, CD4+ T cells from diabetic, but not pre-diabetic NOD/Lt donors were able to transfer diabetes. The lack of contribution of host-origin T and B cells in the induction of diabetes in the NOD/LtSz-scid recipients was excluded by analysis of serum Ig levels and the use of congenic NOD.NON-Thy-1a donors to allow the identification of the T cells in the scid recipients as derived from donor, and not due to “leakiness” in the host. This initial characterization of NOD/LtSz-scid mice as adoptive recipients of NOD cells for the study of diabetes has led to the widespread utilization of this model to study the immunopathogenesis and immunoregulation of NOD diabetes.51-62 Rohane et al51 have shown that islet-infiltrating cells from prediabetic NOD mice can rapidly transfer diabetes to NOD/LtSz-scid mice. Using time-dependent kinetics of diabetes induction following transfer of spleen cells from pre-diabetic versus diabetic mice to NOD/LtSz-scid mice, it was shown that the recipients of islet-infiltrating cells were similar to recipients of splenocytes from diabetic mice in their time to onset of diabetes. This adoptive transfer model was extended to demonstrate that CD4+, but not CD8+, T cells from young, pre-diabetic NOD mice could delay the rapid transfer of disease by either islet-infiltrating cells or spleen cells from diabetic mice.51 This observation suggests that the adoptive transfer system of diabetogenic NOD T cells into NOD/LtSz-scid mice can be used to study immunoregulatory interactions in disease pathogenesis, as well as for the characterization of the autoreactive effector T cell population. This was initially suggested by Gerling et al,52 who demonstrated that diabetes in NOD mice could be prevented by intrathymic islet cell transplantation into 4-week-old females. Prevention of IDDM was the result of tolerance induction, since splenocytes from untreated 24-week-old non-diabetic NOD would rapidly transfer diabetes to NOD/LtSz-scid recipients, while splenocytes from recipients injected intrathymically with islets failed to adoptively transfer disease. In a separate series of studies, the mechanism of tolerance induction was examined. Following intrathymic injection of islet cells, NOD mice that failed to develop diabetes still demonstrated the presence of diabetogenic effector cells in the peripheral lymphoid tissues, although the potency of these effector cell populations suggested that they had not been activated to the same degree as those from mice that did not receive islet cells intrathymically.52

The Use of NOD/LtSz-SCID/SCID Mice in Biomedical Research

183

Peterson and Haskins tested the diabetogenic potential of two cloned CD4+ T cell populations, each expressing a Vβ4 TCR. Both could transfer diabetes to young NOD mice, but only one was able to adoptively transfer diabetes to NOD/LtSz-scid mice.53 This observation indicated that there may be at least two subsets of CD4+ diabetogenic T cells in NOD mice which differ in their requirements for CD8+ Tc ells for IDDM induction.53 A similar observation was reported by Daniel et al.54 Of six CD4+ insulin-specific T cell clones that could accelerate diabetes in young NOD mice, only one was able to induce diabetes upon transfer to NOD/LtSz-scid mice. Moreover, islet specific T cell lines and clones could be easily derived from islet isografts that were placed in NOD/LtSz-scid mice that were adoptive recipients of diabetogenic effector cells from NOD donors. 55 Further studies used the NOD/LtSz-scid recipient to demonstrate that CD45RBlowCD4+ cells may be either pathogenic or protective.56,63 A recent study by Wong et al used the NOD/LtSz-scid mouse to demonstrate that CD8+ T cell clones derived from NOD islets were diabetogenic and could transfer diabetes in the absence of CD4+ T cells.57 Prior to the development and utilization of NOD/LtSz-scid mice, the ability to demonstrate the adoptive transfer of diabetes in the complete absence of CD4+ cells was not possible. Bowman et al58 used the adoptive transfer model of IDDM into NOD/LtSz-scid mice to study the effect of prophylactic insulin therapy on disease incidence in the adoptive recipients. When initiated at the time of adoptive transfer of cells, both high and low doses of insulin retarded the onset of IDDM. However, when initiated two weeks after adoptive transfer of cells, insulin therapy was ineffective, but treatment with somatostatin at this time point did provide a beneficial effect. Shimizu et al59 have recently used the NOD/LtSz-scid mouse to investigate antigen presentation by islet APC in the presence (i.e., NOD) or absence (i.e., NOD/LtSz-scid) of autoimmune islet damage. It was observed that islet APC, even in NOD/LtSz-scid mice, contained diabetogenic peptides on their I-Ag7 molecules, and that upon islet damage following injection of islet specific T cell clones into the NOD/ LtSz-scid mice, the intra-islet APC activity increased. In addition, the diabetogenic peptides were observed in splenic APC following this acute injury to the islets.59 These results, and results from other laboratories, suggest that the adoptive transfer system using

184

NOD Mice and Related Strains: Research Applications in Diseases

NOD/LtSz-scid/scid mice as recipients can be used to investigate the modulation, as well as the identity of the diabetogenic effector cells.60,61 The above observations clearly demonstrate the utility of using the NOD/LtSz-scid mouse as an adoptive recipient of NOD/Lt lymphocytes for the study of IDDM, and as a valuable tool for differentiation of the contribution of the donor lymphocytes versus the host immune system for the expression of the disease. These observations also confirmed the previously reported ability of pre-activated CD4+ cells to adoptively transfer diabetes to immuno-incompetent NOD/LtSz-scid recipients in the absence of donor or host CD8+ Tc ells, and demonstrated that cloned T cells with a single T cell receptor specificity can induce diabetes in the NOD/LtSz-scid recipients.

INDUCTION OF CHEMICAL DIABETES IN NOD/LtSz-scid MICE Administration of multiple low doses of streptozocin (MD-STZ), a pancreatic β-cell toxin, induces hyperglycemia in male mice of many strains. Numerous lines of evidence have suggested that diabetes in MD-STZ is the result of a T cell mediated autoimmune response. The disease is attenuated in mice homozygous for the nu mutation, and antibodies to T cells can prevent disease induction.64 The contribution of an intact immune system in MD-STZ pathogenesis was investigated using NOD/LtSz-scid mice. These immunodeficient mice were found to be highly susceptible to five doses of as little as 25 mg/kg of streptozotocin (STZ), a dose comparable to that required for the induction of hyperglycemia in the immunocompetent NOD/Lt strain of mice.65 The pathogenesis of autoimmune diabetes in NOD/Lt mice and MD-STZ diabetes in NOD/LtSz-scid mice was demonstrated to be distinct. Splenocytes from autoimmune, spontaneously diabetic NOD mice readily transferred IDDM with insulitis to NOD/LtSz-scid recipients, while splenocytes from MD-STZ diabetic NOD-scid mice failed to transfer either insulitis or diabetes. These results using the NOD/LtSz-scid mouse clearly indicated that hyperglycemia in the MDSTZ model of diabetes can be the result of chemical toxicity induced in the absence of functional T cells, and is dependent on strain background. Susceptibility of NOD/LtSz-scid mice to a single high dose of streptozocin has also been investigated. Induction of hyperglycemia

The Use of NOD/LtSz-SCID/SCID Mice in Biomedical Research

185

by this method is important in the establishment of islet transplant models where the NOD/LtSz-scid mouse would be used as a recipient of both target islets and immune cells. We have observed that a single dose of 155 mg/kg of streptozotocin to either male or female NOD/LtSz-scid mice induces hyperglycemia (>400 mg/dl) within 7-10 days. These mice are subsequently able to accept syngeneic, allogeneic, and/or xenogeneic islet grafts (unpublished observations). Adoptive transfer of allogeneic or xenogeneic immunocompetent lymphocytes into the islet-bearing mice will allow the pathogenesis of islet graft rejection to be investigated in the absence of host immune system contribution.

NOD/LtSz-scid MICE AS ISLET GRAFT RECIPIENTS As noted above, NOD/LtSz-scid mice can be readily rendered hyperglycemic following the injection of multiple low doses or a single high dose of streptozocin. This now enables NOD/LtSz-scid mice to function as a universal recipient of islet grafts and immune cells to study the pathogenesis of diabetes and graft rejection. In previous studies, C.B-17-scid mice have been used to examine the rejection of islet allografts and xenografts.66-68 In these reports, chemically diabetic C.B-17-scid mice were transplanted with islets from C57BL/6 mice or allogeneic CBA islets. The C.B-17-scid mice were then injected with spleen cells from untreated Igh-1b congenic BALB/c mice, or from BALB/c mice tolerized to allogeneic C57BL/6 islet grafts by transplantation of C57BL/6 islet grafts pre-cultured to remove APC activity. C.B-17-scid mice receiving BALB/c spleen cells from mice tolerant to C57BL/6 islets promptly rejected the CBA, but not the C57BL/6 islet grafts. C.B-17-scid mice also accept xenogeneic rat islet grafts, and these mice bearing xenogeneic islet grafts have been used as a model to examine the cellular requirements for xenograft islet rejection.66-68 The results using C.B-17-scid mice suggest that this model paradigm may also be useful for investigation of the immune rejection of functioning human islets in vivo. When combined with the enhanced human cell engraftment observed in NOD/LtSz-scid mice (described below), this model system may be useful for investigating the interaction of a mature or developing human immune system with human islets in vivo in a small animal model.

186

NOD Mice and Related Strains: Research Applications in Diseases

HUMAN CELL ENGRAFTMENT IN NOD/LtSz-scid MICE ENGRAFTMENT OF MATURE HUMAN LYMPHOCYTES First described in 1988, intraperitoneal injection of mature human peripheral blood leukocytes into C.B-17-scid mice (termed HuPBL-SCID) results in engraftment of the human cells.29 Human cell engraftment typically represents 1-5% of the cells present in the recipient spleen and blood 4-8 weeks after transfer of the cells. Engraftment levels then decrease with time, with few to no human cells detectable by 12-16 weeks after injection. Human cell engraftment can be increased by irradiation and/or treatment of the recipient with anti-asialo GM1, suggesting a role for NK cells in the resistance of the C.B-17-scid mouse to the engrafted human cells.69-71 This model system has been used extensively to examine secondary human immune responses in vivo, to analyze immune activity of lymphocytes obtained from autoimmune patients, and to investigate human-specific infectious diseases such as human immunodeficiency virus-1 (HIV-1). In the specific example of diabetes, Dyrberg et al demonstrated that adoptive transfer of lymphocytes from diabetic patients resulted in the production of islet related autoantibodies following challenge of the C.B-17-scid recipient with rat islets.72 The engraftment of human peripheral blood lymphocytes in C.B-17-scid mice, however, appears to be predominated almost exclusively by the expansion of antigen specific cells following challenge and by a strong human anti-mouse MHC class II reaction.73,74 Although a few reports have suggested that a human primary immune response can be generated in Hu-PBL-SCID mice,69,75-77 this remains extremely difficult to reliably reproduce. Based on the observation that NOD/LtSz-scid mice display numerous defects in innate immunity,18 including deficiencies in NK cell activity and macrophage function, it has been reasoned that NOD/LtSz-scid mice would be an improved host for human lymphocyte engraftment. Engraftment of human spleen cells in NOD/LtSzscid mice were initially compared to that observed in C.B-17-scid mice. In contrast to the low levels of human cell engraftment observed in C.B-17-scid mice in the spleen and peripheral blood 4-8 weeks after injection of human spleen cells, high levels of engraftment were observed in NOD/LtSz-scid mice.78 The increased levels of engraftment, up to 90% of the spleen cells being of human origin, were associated with elevated levels of human Ig in the serum, migration of high numbers of human cells to peripheral lymphoid and non-lymphoid tis-

The Use of NOD/LtSz-SCID/SCID Mice in Biomedical Research

187

sues, and evidence of stable engraftment for up to 24 weeks. In addition, human anti-mouse red blood cell antibodies were observed, and antibodies against filarial antigens could be induced following challenge with Brugia malayi L3 larvae. These results suggested that NOD/LtSz-scid mice were much better recipients of human lymphocytes than were C.B-17-scid mice, and represented an improved model for study of human lymphocytes in scid mice. To generalize this observation, human peripheral blood mononuclear cell (PBMC) engraftment was compared between NOD/LtSzscid mice and a number of other genetic stocks of scid mice, including C.B-17-scid, C3H/HeJ-scid, C57BL/6J-scid, NK1.1-depleted C57BL/6-scid, and DBA/2-scid mice.79 In all cases, human PBMC engraftment in NOD/LtSz-scid mice was 5-10-fold higher than in any of the other genetic stocks of scid mice examined. The increased engraftment levels in NOD/LtSz-scid mice were associated with histological evidence of infiltration of human lymphocytes into the lung and liver, and elevated levels of human Ig in the circulation.79 These results demonstrated that ablation of NK cell activity (NK-depleted C57BL/6J-scid mice), loss of hemolytic complement (DBA/2J-scid mice) or defects in lipopolysaccharide (LPS) responsiveness (C3H/HeJ-scid mice) failed to increase engraftment levels of human cells in mice expressing the scid mutation on other genetic backgrounds to the levels observed in NOD/LtSz-scid mice. Combinations of these defects, or abnormalities of NOD/LtSz-scid mice not yet described, are most likely important in the improved engraftment of mature human cells in this mouse strain.

ENGRAFTMENT OF HEMOPOIETIC STEM CELLS Based on the improved engraftment of mature human lymphoid cells in NOD/LtSz-scid mice, investigators have tested the ability of human hemopoietic stem cells to engraft in this model system. The ability of human bone marrow to engraft in irradiated C.B-17-scid mice was first described in 1988.30 The engraftment levels obtained were low, and represented only a few percent of host bone marrow cells. Differentiation of the stem cells into mature progeny could be driven by the administration of exogenous human cytokines, including c-kit ligand and PIXY321 (a fusion protein of IL-3 and granulocyte-macrophage colony-stimulating factor) and erythropoietin.30 The authors observed that the growth factor-treated mice contained both multipotential and committed human myeloid and erythroid

188

NOD Mice and Related Strains: Research Applications in Diseases

progenitors. This model has been extended to examine engraftment of primitive hematopoietic cells from β-thalassemia and sickle cell anemia patients.80 In this report, NOD/LtSz-scid mice were included as recipients, and the engraftment levels in the bone marrow of these mice increased to at least 40% in the animals tested. NOD/LtSz-scid mouse recipients of human bone marrow consistently contained higher levels of human cell engraftment than were observed in C.B-17-scid mice.80 In a second model of human hemopoietic stem cell engraftment, investigators have injected human cord blood into irradiated scid mice.81 Cord blood is a source of human CD34+CD38– cells, a population of primitive hemopoietic stem cells.82,83 In addition, cord blood injected mice do not require administration of exogenous human growth factors for engraftment of the human stem cells. This is thought to be due to the presence of immature T cells in the cord blood inoculum that secrete the necessary human cytokines for proliferation and differentiation of the human stem cells. Again, in cord blood injected C.B-17-scid mice, engraftment levels of human cells in the bone marrow of the host were low.81 Even in irradiated BALB/c-scid mice carrying transgenes for human IL-3, granulocyte-macrophage-colony stimulating factor (GM-CSF), and c-kit ligand, engraftment levels of human cord blood cells were low, and could only be detected in 50% of the engrafted mice.84 However, injection of cord blood into irradiated NOD/LtSz-scid mice resulted in engraftment levels of 5-95% in all 11 animals examined in this report. In our laboratories, we have demonstrated that NOD/LtSz-scid mice are improved recipients of human cord blood cells as compared directly with C.B-17-scid mice, and that human cell engraftment in the bone marrow reaches levels of 80-90% in many NOD/LtSz-scid recipients within six weeks after injection.85 Engraftment levels in C.B-17-scid mice were 5- to 10-fold lower following injection of the same cord blood cells. The engraftment of human cord blood cells in the bone marrow of NOD/LtSz-scid mice was associated with the presence of human CD34+ cells, and engraftment was observed to persist for at least 15 weeks. In addition, it was shown that human cord blood cells would engraft in unconditioned NOD/LtSz-scid recipients following multiple day injections of unfractionated cord blood cells.85 This latter observation suggests that irradiation preconditioning for engraftment of human hemopoietic stem cells in scid mice may not

The Use of NOD/LtSz-SCID/SCID Mice in Biomedical Research

189

be necessary, an especially important observation since scid mice have defects in DNA repair and increased susceptibility to ionizing radiation.20-24 There is increasing use of NOD/LtSz-scid mice as recipients of human stem cells. Confirmation that engraftment of human stem cells is more robust in NOD/LtSz-scid mice has appeared,86-89 and the injection of purified human CD34+ cord blood progenitor cells has been initiated using NOD/LtSz-scid mice as the preferred recipients.87,88,90 This model has also been extended to examine the proliferative potential of cells capable of initiating human acute leukemia. Human acute myeloid leukemia cells transplanted into irradiated NOD/LtSz-scid mice in the presence of stem cell factor, IL-3 and granulocyte-macrophage colony stimulating factor readily expanded in the adoptive recipients.91 This may represent an excellent model system for the study of leukemia initiating cells in an in vivo environment, and provide a pre-clinical model for the analysis of efficacy of treatment modalities.

EPSTEIN BARR VIRUS (EBV)-RELATED TUMORS Immunocompromised patients such as those undergoing chronic immunosuppression or patients with acquired immunodeficiency (AIDS) often develop EBV-induced B cell lymphomas.92 The Hu-PBL-SCID mouse readily develops EBV-induced B cell lymphomas when a higher number of human cells (≥50 x 106) are injected.93,94 The majority of mice eventually succumb to these tumors. Since human cells engraft at higher levels in NOD/LtSz-scid mice than in C.B-17-scid mice,78,79,85 it was of interest to determine the incidence of EBV-related tumors in this murine model. Upon injection of 50x10 6 human PBMC into NOD/LtSz-scid mice and into C.B-17-scid mice, it was observed that the incidence of EBV-related tumors in the NOD/LtSz-scid mice was significantly lower.95 This was interpreted to suggest that greater engraftment of CD8+ T cells resulted in enhanced virus-specific cytotoxic T cell activity, impeding the development of the EBV-induced tumors. This also suggests that the enhanced engraftment of human cells observed in NOD/LtSz-scid mice might be increased even more by injection of higher numbers of human cells without the associated increase in EBV-induced tumors that are observed in C.B-17-scid mice when higher numbers of human cells are injected.

190

NOD Mice and Related Strains: Research Applications in Diseases

HUMAN IMMUNODEFICIENCY VIRUS-1 (HIV-1) INFECTION OF HU-PBL-NOD-scid MICE The development and optimization of a small animal model for the investigation of HIV-1 remains an important goal of researchers studying the infectivity and pathogenesis of this virus. The utility of the Hu-PBL-SCID model for HIV-1 infection studies was demonstrated soon after the first description of the model system.96 The caveats associated with this model include the low levels of human cell engraftment that are observed, and the low incidence of infectivity that is present following injection of HIV-1 into the human cell engrafted C.B-17-scid mice.97 In addition, the C.B-17-scid mouse has normal macrophage function, elevated NK cell activity, and circulating hemolytic complement, all of which may be important in viral toxicity in vivo. To investigate the effect of host strain background on human PBMC engraftment, Hesselton et al studied the infectivity rate of human PBMC engrafted NOD/LtSz-scid mice.79 Corresponding with the 5- to 10-fold higher levels of human cell engraftment observed in the NOD/LtSz-scid mice as compared to C.B-17-scid mice, the authors also observed an increase in the frequency of infection following inoculation of HIV-1. Four weeks after infection, almost 80% of the human cell-engrafted NOD/LtSz-scid mice harbored replicating virus, while only 40% of the C.B-17-scid mice that received the same cells and virus stock had detectable levels of replicating virus in their spleens.79 These studies were performed with primary isolates of HIV-1 obtained from pediatric patients rather than from laboratory tissue culture-adapted strains of virus which demonstrate enhanced growth rates in Hu-PBL-SCID mice. In additional studies, it has been observed that injection of cytopathic strains of HIV-1 into human cell-engrafted NOD/LtSz-scid mice results in a significant decrease of human cell engraftment at four weeks post infection, with a loss of both CD4+ and CD8+ T cells (Hesselton et al, personal communication). These results suggest that investigation of human HIV-1 pathogenesis, antiviral drug efficacy, and HIV-1-related gene therapy protocols might be facilitated by the utilization of the NOD/LtSz-scid mouse model. The NOD/LtSz-scid mouse has also been used as a source of fetal thymus for human stem cell/murine fetal thymic organ cultures. When human stem cells are engrafted in scid mice, few to no developing T cells of human origin have been observed.98 To attempt to study the infectivity and pathogenesis of HIV-1 on developing human T

The Use of NOD/LtSz-SCID/SCID Mice in Biomedical Research

191

cells in vitro, human cord blood cells were seeded onto NOD/LtSzscid fetal thymi, and infected with a non-cytopathic strain of HIV-1.99 The developing human T cells in the xenogeneic chimeric thymic organ were readily infected with the virus, and infected cultures were dramatically diminished in the number of human T cells that developed. This in vitro model system promises to be useful for analyses of the effects of HIV-1 on developing human T cells in a defined organ culture system.

NOD/LtSz-scid MICE AS MODELS FOR HUMAN TUMOR GRAFTS Athymic nu/nu mice and C.B-17-scid mice have been used as recipients of human tumors, organs and tissues due to their inability to mount adaptive immune responses to the human xenografts.100-105 To investigate whether the NOD/LtSz-scid mouse might have an enhanced ability to support growth of xenogeneic tumors, NOD/LtSz-scid mice and C.B-17-scid mice were injected with the human T lymphoblastoid cell tumor line, CEM.18 As was observed when the engraftment of human lymphocytes was compared between NOD/LtSz-scid mice and C.B-17-scid mice, the CEM T cell tumor also showed enhanced growth in NOD/LtSz-scid mice. By 4 weeks after the injection of 1 x 106 tumor cells, the spleen and peripheral blood of the NOD/LtSz-scid mouse had 4- to 5-fold higher numbers of human cells than did C.B-17-scid mice.18 This observation suggests that engraftment of many of the human tumors, organs, and tissues that have been difficult to study in C.B-17-scid mice may be more easily engrafted and investigated in NOD/LtSz-scid mice. Coupled with the enhanced ability of NOD/LtSz-scid mice to support human lymphoid cell engraftment, this mouse model may now allow the in vivo investigation of the interaction of the human immune system with the target tissue or tumor of choice. USE OF NOD/LtSz-scid MICE IN PARASITIC RESEARCH The study of human specific parasites in murine models has been hampered by the presence of a host immune system, and the specificity of the parasite for tissues or cells in the host species. The development of the C.B-17-scid mouse circumvented in part these obstacles, by eliminating the host adaptive immune system, and allowing the engraftment of the human target tissues. The former characteristic is clearly illustrated by studies that have utilized C.B-17-scid mice in the

192

NOD Mice and Related Strains: Research Applications in Diseases

investigation of the human lymphatic filarial parasite, Brugia malayi. Following injection of infective B. malayi L3 larvae, this nematode has been found to grow into adults, mate, and produce microfilaria in C.B-17-scid mice.106 Since B. malayi fails to infect immunocompetent mice, the apparent human specificity of the parasite is in part due to the ability of the murine adaptive immune system to eliminate the L3 larvae. It has also allowed the components of the murine adaptive immune system that are important in the resistance of normal mice to the human filarial parasite to be investigated.107,108 Based on the numerous defects in innate immunity in NOD/LtSz-scid mice as compared to C.B-17-scid mice, it was postulated that the human parasite would show enhanced growth in NOD/LtSz-scid mice. However, when B. malayi was injected into both strains of scid mice, the NOD/LtSzscid mouse showed increased resistance to the growth of the filarial nematode, and few to no adult worms were observed surviving in these mice (Rajan et al, personal communication). These observations suggest that either NOD/LtSz-scid mice have an enhanced innate immune parameter function different than that present in C.B-17-scid mice to which B. malayi is particularly susceptible, or that NOD/LtSz-scid mice might lack a growth factor important for the survival of B. malayi in vivo. The determination of the difference in susceptibility of NOD/LtSz-scid mice and C.B-17-scid mice to infection with B. malayi may provide important insights into the growth requirements of the parasite, innate immune function of NOD/LtSzscid mice, and additional differences between C.B-17-scid and NOD/LtSz-scid mice that may be important in their relative ability to support human cell engraftment and their susceptibility to other infectious agents. A second parasite of particular interest is that of Plasmodium falciparum. This is the causative agent of malaria in man, and is the most pathogenic of the four human Plasmodial species that have been identified.109 The target cell for this organism is the human red blood cell (RBC), and the specificity of this parasite-target interaction prevents the establishment of a small animal model where murine cells can be infected with this organism. Attempts to establish significant levels of circulating human red blood cells in C.B-17-scid mice beyond a few hours have failed.110 Injection of human RBC into NOD/LtSz-scid mice results in sustained RBC levels in the circulation that can be maintained by daily intra-peritoneal human RBC supple-

The Use of NOD/LtSz-SCID/SCID Mice in Biomedical Research

193

mental injections. When the initial injection of human RBCs are infected with the human malarial parasite P. falciparum, parasites could be detected for up to seven days in the circulation.111 When splenectomized NOD/LtSz-scid mice were used to prevent the splenic trapping and removal of human RBCs from the circulation, the injection of infected RBCs resulted in the circulation of mature sexual stage parasites (gametocytes). The infected NOD/LtSz-scid mice were able to transmit the mature gametocytes to Anopheline mosquitoes feeding on the mice, resulting in the development of oocysts in the mosquito midguts.111 These results demonstrate that the NOD/LtSz-scid mouse may be an enhanced model for study of human blood-specific parasites, and may also provide an excellent model for the study of human red blood cell disorders of lymphohemopoietic origin. As noted above, β-thalassemia and sickle cell anemia have recently been studied using the C.B-17-scid mouse.80 While low levels of human cell engraftment in the bone marrow was observed in C.B-17-scid mice, higher levels were observed in NOD/LtSz-scid mice.

NEW MODELS OF NOD/LtSz-IMMUNODEFICIENT MICE NOD/LtSz-scid-B2mnull mice: NOD/LtSz-scid mice, although deficient in NK cell activity, nonetheless display detectable NK cell function that can be elevated by administration of poly I:C.18 In addition, in the human PBL engraftment studies, a major component of the engrafted cells are postulated to be xenoreactive to MHC antigens.73,74 To address these limitations, genetic crosses were performed to produce NOD/LtSz mice doubly homozygous for the scid mutation and the β2microglobulinnull (B2mnull) allele. Since β2m is required for cell surface expression of MHC class I, B2mnull mice are MHC class I negative.112,113 As before, the innate immune defects apparent in the NOD/Lt parental strain were observed in the NOD/LtSz-scid-B2mnull mutant strain of mouse.114 In these mice, no NK cell activity could be detected, even after stimulation with poly I:C. Upon engraftment of human PBL, these mice were found to support elevated levels of human T cell engraftment as compared with NOD/LtSz-scid mice. The increased engraftment was associated with a major increase in the percentage and numbers of human CD4+ T cells.114 This increase led to a normalization of the CD4:CD8 ratio of human T cells in NOD/LtSz-scid-B2mnull mice. Thus, the lack of MHC class I expression led to an increase in the expansion of human CD4+ T cells. The

194

NOD Mice and Related Strains: Research Applications in Diseases

higher levels of human CD4+ T cells and the normalization of the CD4:CD8 ratio suggests that this system may be an excellent model for studies of immune responses and HIV pathogenesis. In addition to the higher levels of human T cell engraftment, NOD/LtSz-scid-B2mnull mice also display two additional characteristics not observed in NOD/LtSz-scid mice. First, due to the lack of β2m, NOD/LtSz-scid-B2mnull mice also rapidly clear IgG.114 B2m appears to be required for normal IgG circulation, and the absence of β2m leads to a defective Fc receptor for IgG that is hypothesized to normally protect IgG from catabolism via a salvage pathway.115,116 Second, as in all β2m deficient mice,117,118 NOD/LtSz-scid-B2mnull mice displayed a parenchymal iron overload.114 The high levels of iron in the liver of NOD/LtSz-scid-B2mnull mice resemble the human syndrome of familial hemochromatosis.119 The use of the NOD/LtSz-scid-B2mnull mouse as a model for human hemochromatosis may allow insights into the pathogenesis and etiology of this human disorder in which little of the pathogenetic mechanisms are currently known. NOD-RAG-2null mice: Another approach to generate an immunodeficient NOD mouse has been to backcross the RAG-2null locus onto the NOD strain.120 These mice, termed NOD/Bom-rag2null, do not develop functionally mature T or B lymphocytes, and fail to develop either insulitis or diabetes. Adoptive transfer of spleen cells from diabetic NOD donors rapidly induced diabetes in the NOD/LtSz-scidB2mnull mice recipients.120 It will be important to determine the innate immune functions, the development of thymic lymphomas, and the ability of this new genetic stock of mice to support human cell engraftment.

CONCLUSIONS The above observations demonstrate that the NOD/LtSz-scid mouse may be an improved model for many of the experimental systems currently using the C.B-17-scid mouse. The NOD/LtSz-scid mouse will be of significant importance in the study of the autoimmune syndrome of NOD/Lt mice, and in the model systems that utilize engraftment of human lymphohemopoietic cells into scid mice. The defects in innate immunity, including lack of hemolytic complement, abnormalities in APC development and function, and severe deficiencies in NK cell activity, render the NOD/LtSz-scid mouse a superior recipient for both murine autoimmune effector cells and

The Use of NOD/LtSz-SCID/SCID Mice in Biomedical Research

195

human lymphohemopoietic cells. In addition, the defects in innate immunity increase the utility of this murine stock of mice in studies of the growth of human tumors and human-specific parasites. Constraints associated with the utility of the NOD/LtSz-scid mouse continue to be the high incidence of thymic lymphoma development, and the resultant short lifespan of the animal as result of these tumors. However, continuing efforts to identify the cause of the thymic lymphomagenesis, and to address the issues of irradiation sensitivity, anti-murine xenoreactivity by engrafted human cells, and the lack of easily demonstrable human primary immune responses in human cell engrafted NOD/LtSz-scid mice are resolving these concerns. REFERENCES 1. Bosma GC, Custer RP, Bosma MJ. A severe combined immunodeficiency mutation in the mouse. Nature 1983; 301:527-530. 2. Anonymous. Mouse genome database (MGD), mouse genome informatics project, The Jackson Laboratory, Bar Harbor, Maine. World Wide Web (http://www.informatics.jax.org). 1996. 3. Bosma GC, Davisson MT, Ruetsch NR et al. The mouse mutation severe combined immune deficiency (scid) is on Chromosome 16 [published erratum appears in Immunogenetics 1989;29(3):224]. Immunogenetics 1989; 29:54-57. 4. Kirchgessner CU, Patil CK, Evans JW et al. DNA-dependent kinase (p350) as a candidate gene for the murine SCID defect. Science 1995; 267:1178-1183. 5. Blunt T, Gell D, Fox M, Taccioli GE et al. Identification of a nonsense mutation in the carboxyl-terminal region of DNA-dependent protein kinase catalytic subunit in the scid mouse. Proc Natl Acad Sci USA 1996; 93:10285-10290. 6. Fried LM, Koumenis C, Peterson SR et al. The DNA damage response in DNA-dependent protein kinase-deficient SCID mouse cells: replication protein A hyperphosphorylation and p53 induction. Proc Natl Acad Sci USA 1996; 93:13825-13830. 7. Jeggo PA, Jackson SP, Taccioli GE. Identification of the catalytic subunit of DNA dependent protein kinase as the product of the mouse scid gene. Curr Top Microbiol Immunol 1996; 217:79-89. 8. Miller RD, Hogg J, Ozaki JH et al. Gene for the catalytic subunit of mouse DNA-dependent protein kinase maps to the scid locus. Proc Natl Acad Sci USA 1995; 92:10792-10795. 9. Bosma MJ, Carroll AM. The SCID mouse mutant: definition, characterization, and potential uses. Annu Rev Immunol 1991; 9:323-350. 10. Lieber MR, Hesse JE, Lewis S. et al. The defect in murine severe combined immune deficiency: joining of signal sequences but not coding segments in V(D)J recombination. Cell 1988; 55:7-16.

196

NOD Mice and Related Strains: Research Applications in Diseases

11. Malynn BA, Blackwell TK, Fulop GM et al. The scid defect affects the final step of the immunoglobulin VDJ recombinase mechanism. Cell 1988; 54:453-460. 12. Taylor PC. The use of SCID mice in the investigation of human autoimmune diseases. 1994:1-127. 13. Habu S, Kimura M, Katsuki M et al. Correlation of T cell receptor gene rearrangements to T cell surface antigen expression and to serum immunoglobulin level in scid mice. Eur J Immunol 1987; 17:1467-1471. 14. Shores EW, Sharrow SO, Uppenkamp I et al. T cell receptor-negative thymocytes from SCID mice can be induced to enter the CD4/CD8 differentiation pathway. Eur J Immunol 1990; 20:69-77. 15. Hardy RR, Kemp JD, Hayakawa K. Analysis of lymphoid population in scid mice; detection of a potential B lymphocyte progenitor population present at normal levels in scid mice by three color flow cytometry with B220 and S7. Curr Top Microbiol Immunol 1989; 152:19-25. 16. Bosma GC, Fried M, Custer RP et al. Evidence of functional lymphocytes in some (leaky) scid mice. J Exp Med 1988; 167:1016-1033. 17. Nonoyama S, Smith FO, Bernstein ID et al. Strain-dependent leakiness of mice with severe combined immune deficiency. J Immunol 1993; 150:3817-3824. 18. Shultz LD, Schweitzer PA, Christianson SW et al. Multiple defects in innate and adaptive immunologic function in NOD/LtSz-scid mice. J Immunol 1995; 154:180-191. 19. Mosier DE, Stell KL, Gulizia RJ et al. Homozygous scid/scid;beige/ beige mice have low levels of spontaneous or neonatal T cell-induced B cell generation. J Exp Med 1993; 177:191-194. 20. Hendrickson EA, Qin XQ, Bump EA et al. A link between doublestrand break-related repair and V(D)J recombination: the scid mutation. Proc Natl Acad Sci (USA) 1991; 88:4061-4065. 21. Biedermann KA, Sun JR, Giaccia AJ et al. scid mutation in mice confers hypersensitivity to ionizing radiation and a deficiency in DNA double-strand break repair. Proc Natl Acad Sci (USA) 1991; 88:1394-1397. 22. Tanaka T, Yamagami T, Oka Y et al. The scid mutation in mice causes defects in the repair system for both double-strand DNA breaks and DNA cross-links. Mutation Research 1993; 288:277-280. 23. Fulop GM, Phillips RA. The scid mutation in mice causes a general defect in DNA repair. Nature 1990; 347:479-482. 24. Disney JE, Barth AL, Shultz LD. Defective repair of radiation-induced chromosomal damage in scid/scid mice. Cytogenetics Cell Genetics 1992; 59:39-44. 25. Bancroft GJ, Schreiber RD, Unanue ER. T cell-independent macrophage activation in scid mice. Curr Top Microbiol Immunol 1989; 152:235-242.

The Use of NOD/LtSz-SCID/SCID Mice in Biomedical Research

197

26. Czitrom AA, Edwards S, Phillips RA et al. The function of antigenpresenting cells in mice with severe combined immunodeficiency. J Immunol 1985; 134:2276-2280. 27. McCune JM, Namikawa R, Kaneshima H et al. The SCID-hu mouse: murine model for the analysis of human hematolymphoid differentiation and function. Science 1988; 241:1632-1639. 28. Vargas SL, Hughes WT, Wakefield AE et al. Limited persistence in and subsequent elimination of Pneumocystis carinii from the lungs after P. carinii pneumonia. J Infect Dis 1995; 172:506-510. 29. Mosier DE, Bulizia RJ, Baird SM et al. Transfer of a functional human immune system to mice with severe combined immunodeficiency. Nature 1988; 335:256-262. 30. Kamel-Reid S, Dick JE. Engraftment of immune-deficient mice with human hematopoietic stem cells. Science 1988; 242:1706-1709. 31. Bowman MA, Leiter EH, Atkinson MA. Prevention of diabetes in the NOD mouse: Implications for therapeutic intervention in human disease. Immunol Today 1994; 15:115-120. 32. Miyake K, Medina K, Ishihara K et al. A VCAM-like adhesion molecule on murine bone marrow stromal cells mediates binding of lymphocyte precursors in culture. J Cell Biol 1991; 114:557-565. 33. Serreze DV, Leiter EH. Defective activation of T suppressor cell function in nonobese diabetic mice. Potential relation to cytokine deficiencies. J Immunol 1988; 140:3801-3807. 34. Serreze DV, Gaedeke JW, Leiter EH. Hemopoietic stem-cell defects underlying abnormal macrophage development and maturation in NOD/Lt mice: Defective regulation of cytokine receptors and protein kinase C. Proc Natl Acad Sci (USA) 1993; 90:9625-9629. 35. Serreze DV, Gaskins HR, Leiter EH. Defects in the differentiation and function of antigen presenting cells in NOD/Lt mice. J Immunol 1993; 150:2534-2543. 36. Jacob CO, Aiso S, Michie SA et al. Prevention of diabetes in nonobese diabetic mice by tumor necrosis factor (TNF): Similarities between TNF-α and interleukin. Proc Natl Acad Sci USA 1990; 87:968-972. 37. Baxter AG, Cooke A. Complement lytic activity has no role in the pathogenesis of autoimmune diabetes in NOD mice. Diabetes 1993; 42:1574-1578. 38. Cudkowicz G, Hochman PS. Do natural killer cells engage in regulated reactions against self to ensure homeostasis? Immunol Rev 1979; 44:13-41. 39. Kiessling R, Hochman PS, Haller O et al. Evidence for a similar or common mechanism for natural killer cell activity and resistance to hemopoietic grafts. Eur J Immunol 1977; 7:655-663. 40. Yu YY, Kumar V, Bennett M. Murine natural killer cells and marrow graft rejection. Ann Rev Immunol 1992; 10:189-213. 41. Wetsel RA, Fleischer DT, Haviland DL. Deficiency of the murine fifth complement component (C5). A 2-base pair gene deletion in a 5'-exon. J Biol Chem 1990; 265:2435-2440.

198

NOD Mice and Related Strains: Research Applications in Diseases

42. Prochazka M, Gaskins HR, Shultz LD et al. The nonobese diabetic scid mouse: model for spontaneous thymomagenesis associated with immunodeficiency. Proc Natl Acad Sci USA 1992; 89:3290-3294. 43. Kikutani H, Makino S. The murine autoimmune diabetic model: NOD and related strains. Adv Immunol 1992; 51:285-322. 44. Prochazka M, Gaskins HR, Leiter EH. Emv-30: a novel ecotropic MuLV present in NOD/Lt and NON/Lt strains. Mouse Genome 1991; 89:280. 45. Serreze DV, Leiter EH, Hanson MS et al. Emv30null NOD-scid mice. An improved host for adoptive transfer of autoimmune diabetes and growth of human lymphohematopoietic cells. Diabetes 1995; 44:1392-1398. 46. Wicker LS, Miller BJ, Mullen Y. Transfer of autoimmune diabetes mellitus with splenocytes from nonobese diabetic (NOD) mice. Diabetes 1986; 35:855-860. 47. Bendelac A, Carnaud C, Boitard C et al. Syngeneic transfer of autoimmune diabetes from diabetic NOD mice to healthy neonates. Requirement for both L3T4+ and Lyt-2+ T cells. J Exp Med 1987; 166:823-832. 48. Yagi H, Matsumoto M, Kunimoto K et al. Analysis of the roles of CD4+ and CD8+ T cells in autoimmune diabetes of NOD mice using transfer to NOD athymic nude mice. Eur J Immunol 1992; 22: 2387-2393. 49. Matsumoto M, Yagi H, Kunimoto K et al. Transfer of autoimmune diabetes from diabetic NOD mice to NOD athymic nude mice: the roles of T cell subsets in the pathogenesis. Cell Immunol 1993; 148:189-197. 50. Christianson SW, Shultz LD, Leiter EH. Adoptive transfer of diabetes into immunodeficient NOD-scid/scid mice. Relative contributions of CD4+ and CD8+ T-cells from diabetic versus prediabetic NOD. NON-Thy-1a donors. Diabetes 1993; 42:44-55. 51. Rohane PW, Shimada A, Kim DT et al. Islet-infiltrating lymphocytes from prediabetic NOD mice rapidly transfer diabetes to NOD-scid/scid mice. Diabetes 1995; 44:550-554. 52. Gerling IC, Atkinson MA, Leiter EH. The thymus as a site for evaluating the potency of candidate beta cell autoantigens in NOD mice. J Autoimmunity 1994; 7:851-858. 53. Peterson JD, Haskins K. Transfer of diabetes in the NOD-scid mouse by CD4 T-cell clones. Differential requirement for CD8 T-cells. Diabetes 1996; 45:328-336. 54. Daniel D, Gill RG, Schloot N et al. Epitope specificity, cytokine production profile and diabetogenic activity of insulin-specific T cell clones isolated from NOD mice. Eur J Immunol 1995; 25:1056-1062. 55. Wegmann DR, Shehadeh N, Lafferty KJ et al. Establishment of isletspecific T-cell lines and clones from islet isografts placed in spontaneously diabetic NOD mice. J Autoimmunity 1993; 6:517-527.

The Use of NOD/LtSz-SCID/SCID Mice in Biomedical Research

199

56. Shimada A, Rohane P, Fathman CG et al. Pathogenic and protective roles of CD45RBlowCD4+ cells correlate with cytokine profiles in the spontaneously autoimmune diabetic mouse. Diabetes 1996; 45:71-78. 57. Wong FS, Visintin I, Wen L et al. CD8 T cell clones from young nonobese diabetic (NOD) islets can transfer rapid onset of diabetes in NOD mice in the absence of CD4 cells. J Exp Med 1996; 183:67-76. 58. Bowman MA, Campbell L, Darrow BL et al. Immunological and metabolic effects of prophylactic insulin therapy in the NOD-scid/ scid adoptive transfer model of IDDM. Diabetes 1996; 45:205-208. 59. Shimizu J, Carrasco-Marin E, Kanagawa O et al. Relationship between β cell injury and antigen presentation in NOD mice. J Immunol 1995; 155:4095-4099. 60. Moriyama H, Yokono K, Amano K et al. Induction of tolerance in murine autoimmune diabetes by transient blockade of leukocyte function-associated antigen-1/intercellular adhesion molecule-1 pathway. J Immunol 1996; 157:3737-3743. 61. Grewal IS, Grewal KD, Wong FS et al. Local expression of transgene encoded TNF alpha in islets prevents autoimmune diabetes in nonobese diabetic (NOD) mice by preventing the development of auto-reactive islet-specific T cells [see comments]. J Exp Med 1996,184:1963-1974. 62. Tian J, Atkinson MA, Clare-Salzler M et al. Nasal administration of glutamate decarboxylase (GAD65) peptides induces Th2 responses and prevents murine insulin-dependent diabetes. J Exp Med 1996; 183:1561-1567. 63. Shimada A, Charlton B, Rohane P et al. Immune regulation in type 1 diabetes. J Autoimmunity 1996; 9:263-269. 64. Kolb H. IDDM: Lessons from the low-dose streptotozin model in mice. Diabetes Reviews 1993; 1:116-126. 65. Gerling IC, Friedman H, Greiner DL et al. Multiple low-dose streptozocin-induced diabetes in NOD-scid/scid mice in the absence of functional lymphocytes. Diabetes 1994; 43:433-440. 66. Coulombe M, Lafferty KJ, Gill RG. Nature of tolerance induction to peripheral (extrathymic) islet allografts. Transplant Proc 1994; 26:720-721. 67. Wolf L, Gill RG. Xenoreactive T-cell lines initiate pancreatic islet graft destruction in vivo. Transplant Proc 1993; 25:440-441. 68. Gill RG, Wolf L, Daniel D et al. CD4+ T cells are both necessary and sufficient for islet xenograft rejection. Transplant Proc 1994; 26:1203. 69. Sandhu J, Shpitz B, Gallinger S et al. Human primary immune response in SCID mice engrafted with human peripheral blood lymphocytes. J Immunol 1994; 152:3806-3813. 70. Shpitz B, Chambers CA, Singhal AB et al. High level functional engraftment of severe combined immunodeficient mice with human peripheral blood lymphocytes following pretreatment with radiation and anti-asialo GM1. J Immunol Meth 1994; 169:1-15.

200

NOD Mice and Related Strains: Research Applications in Diseases

71. Cavacini LA, Giles-Komar J, Kennel M et al. Effect of immunosuppressive therapy on cytolytic activity of immunodeficient mice: implications for xenogeneic transplantation. Cell Immunol 1992; 144:296-310. 72. Petersen JS, Marshall MO, Baekkeskov S et al. Transfer of type 1 (insulin-dependent) diabetes mellitus associated autoimmunity to mice with severe combined immunodeficiency (SCID). Diabetologia 1993; 36:510-515. 73. Tary-Lehmann M, Lehmann PV, Schols D et al. Anti-SCID mouse reactivity shapes the human CD4+ T cell repertoire in hu-PBL-SCID chimeras. J Exp Med 1994; 180:1817-1827. 74. Tary-Lehmann M, Saxon A, Lehmann PV. The human immune system in hu-PBL-SCID mice. Immunol Today 1995; 16:529-533. 75. Chargui J, Dye D, Blomberg J et al. he humanized severe combined immunodeficient mouse as a model for primary human humoral response against HIV1 peptides. J Immunol Meth 1995; 181:91-100. 76. Ifversen P, Martensson C, Danielsson L et al. Induction of primary antigen-specific immune reponses in SCID-hu-PBL by coupled T-B epitopes. Immunol 1995; 84:111-116. 77. Markham RB, Donnenberg AD. Effect of donor and recipient immunization protocols on primary and secondary human antibody responses in SCID mice reconstituted with human peripheral blood mononuclear cells. Infect Immun 1992; 60:2305-2308. 78. Greiner DL, Shultz LD, Yates J et al. Improved engraftment of human spleen cells in NOD/LtSz-scid/scid mice as compared with C.B-17-scid/scid mice. Am J Path 1995; 146:888-902. 79. Hesselton RM, Greiner DL, Mordes JP et al. High levels of human peripheral blood mononuclear cell engraftment and enhanced susceptibility to human immunodeficiency virus type 1 infection in NOD/LtSz-scid/scid mice. J Inf Dis 1995; 172:974-982. 80. Larochelle A, Vormoor J, Lapidot T et al. Engraftment of immunedeficient mice with primitive hematopoietic cells from beta-thalassemia and sickle cell anemia patients: implications for evaluating human gene therapy protocols. Human Mol Genet 1995; 4:163-172. 81. Vormoor J, Lapidot T, Pflumio F et al. Immature human cord blood progenitors engraft and proliferate to high levels in severe combined immunodeficient mice. Blood 1994; 83:2489-2497. 82. Moore T, Huang S, Terstappen LW et al. Expression of CD34 on murine and human pluripotent hematopoietic stem cells. J Immunol 1994; 153:4978-4987. 83. Peault B, Weissman IL, Baum C et al. Lymphoid reconstitution of the human fetal thymus in SCID mice with CD34+ precursor cells. J Exp Med 1991;174:1283-1286. 84. Bock TA, Orlic D, Dunbar CE et al. Improved engraftment of human hematopoietic cells in severe combined immunodeficient (SCID) mice carrying human cytokine transgenes. J Exp Med 1995; 182: 2037-2043.

The Use of NOD/LtSz-SCID/SCID Mice in Biomedical Research

201

85. Lowry PA, Shultz LD, Greiner DL et al. Improved engraftment of human cord blood stem cells in NOD/LtSz-scid/scid mice following irradiation or multiple day injections into unirradiated recipients. Biol Blood Marrow Transplant 1996; 2:15-23. 86. Torbett BE, Conners K, Shao LE et al. High level multilineage engraftment of human hematopoietic cells is maintained in NOD-scid mice. Blood 1995; 86(suppl. 1):112a. 87. Cashman JD, Lapidot T, Wang JCY et al. Kinetic evidence of the regeneration of multi-lineage hematopoiesis from primitive cells in normal human bone marrow transplanted into immunodeficient mice. Blood 1997 (In Press). 88. Larochelle A, Vormoor J, Hanenberg H et al. Identification of primitive human hematopoietic cells capable of repopulating NOD-scid mouse bone marrow: implications for gene therapy. Nature Medicine 1996; 2:1329-1337. 89. Pflumio F, Izac B, Katz A et al. Phenotype and function of human hematopoietic cells engrafting immune-deficient CB17-severe combined immunodeficiency mice and nonobese diabetic-severe combined immunodeficiency mice after transplantation of human cord blood mononuclear cells. Blood 1996; 88:3731-3740. 90. Keller G, Hogan C, McBurney E et al. Transplantation of CD34+ human (hu) umbilical cord blood progenitor cells into non-obese diabetic (NOD)/severe combined immunodeficiency (scid) mice. Blood 1995; 86 (suppl. 1):573a. 91. Bonnet D, Dick JE. Purification and characterization of a cell initiating human acute leukemia after transplantation into NOD-scid mice. Blood 1995; 86 (suppl. 1):435a. 92. Sullivan JL. Epstein-Barr virus and lymphoproliferative disorders. Sem Hematol 1988; 25:269-279. 93. Mosier DE, Baird SM, Kirven MB. EBV-associated B-cell lymphomas following transfer of human peripheral blood lymphocytes to mice with severe combined immune deficiency. Curr Top Microbiol Immunol 1990; 166:317-323. 94. Mosier DE, Picchio GR, Kirven MB et al. EBV-induced human B cell lymphomas in hu-PBL-SCID mice. AIDS Research & Human Retroviruses 1992; 8:735-740. 95. Hesselton RM, Greiner DL, Mordes JP et al. High levels of human peripheral blood mononuclear cell engraftment and enhanced susceptibility to human immunodeficiency virus-1 infectionin NOD/LtSzscid/scid mice. J Infect Dis 1995; 96. Hesselton RM, Cromwell M, Shultz LD et al. EBV-associated lymphoproliferative disease in human PBMC-reconstituted NOD/LtSz-scid/scid mice. Cold Spring Harbor Conference on EpsteinBarr Virus and Associated Diseases 1994; 1:130. 97. Mosier DE, Gulizia RJ, Baird SM et al. Human immunodeficiency virus infection of human-PBL-SCID mice. Science 1991; 251:791-794.

202

NOD Mice and Related Strains: Research Applications in Diseases

98. Hesselton RM, Koup RA, Cromwell MA et al. Human peripheral blood xenografts in the SCID mouse: Characterization of immunologic reconstitution. J Infect Dis 1993; 168:630-640. 99. Dick JE, Pflumio F, Lapidot T. Mouse models for human hematopoiesis. Sem Immunol 1991; 3:367-378. 100. Greiner DL, Shultz LD, Deluca D et al. HIV-1 infectivity of human T cells in a human/murine chimeric fetal thymic organ culture system. In Vivo 1996; 10:33-37. 101. Kawata A, Yoshida M, Okazaki M et al. Establishment of new SCID and nude mouse models of human B leukemia/lymphoma and effective therapy of the tumors with immunotoxin and monoclonal antibody: marked difference between the SCID and nude mouse models in the antitumor efficacy of monoclonal antibody. Cancer Research 1994; 54:2688-2694. 102. Byers HR, Etoh T, Lee KW et al. Organ-specific metastases in immunodeficient mice injected with human melanoma cells: a quantitative pathological analysis. Melanoma Research 1993; 3:247-253. 103. Garofalo A, Chirivi RG, Scanziani E et al. Comparative study on the metastatic behavior of human tumors in nude, beige/nude/xid and severe combined immunodeficient mice. Invasion & Metastasis 1993; 13:82-91. 104. Oakley CS, Welsch MA, Zhai YF et al. Comparative abilities of athymic nude mice and severe combined immune deficient (SCID) mice to accept transplants of induced rat mammary carcinomas: enhanced transplantation efficiency of those rat mammary carcinomas that have elevated expression of neu oncogene. Int J Cancer 1993; 53:1002-1007. 105. Volpe R, Kasuga Y, Akasu F et al. The use of the severe combined immunodeficient mouse and the athymic “nude” mouse as models for the study of human autoimmune thyroid disease. [Review]. Clin Immunol Immunopath 1993; 67:93-99. 106. Xie X, Brunner N, Jensen G et al. Comparative studies between nude and scid mice on the growth and metastatic behavior of xenografted human tumors. Clinical & Experimental Metastasis 1992; 10:201-210. 107. Nelson FK, Greiner DL, Shultz LD et al. The immunodeficient scid mouse as a model for human lymphatic filariasis. J Exp Med 1991; 173:659-663. 108. Rajan TV, Nelson FK, Shultz LD et al. CD8+ T lymphocytes are not required for murine resistance to human filarial parasites. J Parasitol 1992; 78:744-746. 109. Rajan TV, Shultz LD, Yates J et al. B lymphocytes are not required for murine resistance to the human filarial parasite, Brugia malayi. J Parasitol 1995; 81:490-493. 110. Miller LH, Howard RJ, Carter R et al. Research toward malaria vaccines. Science 1986; 234:1349-1356.

The Use of NOD/LtSz-SCID/SCID Mice in Biomedical Research

203

111. Ishihara C, Tsuji M, Hagiwara K et al. Transfusion with xenogeneic erythrocytes into SCID mice and their clearance from the circulation. J Vet Med Science 1994; 56:1149-1154. 112. Moore JM, Kumar N, Shultz LD et al. Maintenance of the human malarial parasite, Plasmodium falciparum, in scid mice and transmission of gametocytes to mosquitoes. J Exp Med 1995; 181:2265-2270. 113. Koller BH, Marrack P, Kappler JW et al. Normal development of mice deficient in b2M, MHC class I proteins, and CD8+ T cells. Science 1990; 248:1227-1230. 114. Zijlstra M, Bix M, Simister NE et al. Beta2-microglobulin deficient mice lack CD4-8+ cytolytic T cells. Nature 1990; 344:742-746. 115. Christianson SW, Greiner DL, Hesselton RM et al. Enhanced human CD4+ T cell engraftment in β2-microglobulin-deficient NOD-scid mice. J Immunol 1997; 158:3578-3586. 116. Junghans RP, Anderson CL. The protection receptor for IgG catabolism is the b2-microglobulin-containing neonatal intestinal transport receptor. Proc Natl Acad Sci USA 1996; 93:5512-5516. 117. Ghetie V, Hubbard JG, Kim JK et al. Abnormally short serum halflives of IgG in beta 2-microglobulin-deficient mice. Eur J Immunol 1996; 26:690-696. 118. Rothenberg BE, Voland JR. Beta2 knockout mice develop parenchymal iron overload: A putative role for class I genes of the major histocompatibility complex in iron metabolism. Proc Natl Acad Sci USA 1996; 93:1529-1534. 119. de Sousa M, Reimao R, Lacerda R et al. Iron overload in beta 2-microglobulin-deficient mice. Immunol Lett 1994; 39:105-111. 120. Feder JN, Gnirke A, Thomas W et al. A novel MHC class I-like gene is mutated in patients with hereditary haemochromatosis. Nature Genetics 1996; 13:399-408. 121. Soderstrom I, Bergman ML, Colucci F et al. Establishment and characterization of RAG-2 deficient non-obese diabetic mice. Scand J Immunol 1996; 43:525-530.

204

NOD Mice and Related Strains: Research Applications in Diseases

205

Index

INDEX A Acquired immunodeficiency syndrome (AIDS), 189 Activation induced cell death (AICD), 103, 112-113 Activation threshold, 74 Adoptive transfer, 177, 181-183, 185-186, 194 Alloxan, 2, 4, 22, 130 alloxan resistant (ALR) mice, 2, 4, 22 alloxan susceptible (ALS), 2, 4, 22 -induced diabetes, 4 Antigen administration intra-peritoneal, 9 intravenous, 21, 84, 154-155 oral, 154, 158, 160-161 prophylactic, 89-90, 157-158, 160-161, 183 thymic, 154-155 Antigen presenting cells, 6, 10, 12, 15, 24, 43, 73, 101 Apoptosis, 11, 14, 50, 57, 74, 89, 103, 105, 109 Autoantibodies, 6, 13-16, 18, 22, 59, 121-123, 147, 160, 186 Autoantigen, 6, 15, 45, 54, 59, 76, 79-80, 83-84, 87, 89-90, 105, 121-123, 130, 153-156, 161-162 Autologous mixed lymphocyte reaction, 156 Azathioprine, 146, 150, 152

B β cell, 6, 9-11, 14-15, 21-22, 24, 45, 48-50, 52, 58-60, 71, 75-90, 106-107, 109, 112, 121, 123-125, 128-130, 132-133, 146, 148-149, 153-162, 174-175, 178, 181-182, 189 B lymphocytes, 25, 78, 107, 109, 181, 194 B7, 77, 124, 129, 131, 150, 152 Bacille Calmette Gue’rin, 150 Backcross, 18, 21, 38-39, 45, 49, 51-53, 55, 58, 60-61, 105, 176-179, 194 BB rat, 8-9, 20, 157-158 Bcl/XL, 105 Bcl2, 14, 54, 55, 57, 124

Bone marrow, 12-13, 42, 71, 73, 87-88, 108, 132-133, 150, 157, 178-179, 187-188, 193 Bovine serum albumin, 155 Breeding, 2-5, 7, 18, 21-22, 38-41, 149, 173, 176 Brugia malayi, 187, 192

C C.B-17-scid, 173, 175-180, 185-194 Caesarean transfer, 149 Carboxypeptidase, 83, 123, 155 Castration, 8, 132, 151, 159 Cataract Shionogi, 3, 20 CD-1 mice, 22 CD25, 103, 112, 125 CD28, 104-106, 112, 131, 152 CD3, 11, 21, 72, 131, 151-152, 154 CD4, 6, 10, 13, 17, 57-58, 71, 73-78, 80-85, 87-88, 123-125, 127-135, 148, 151-152, 156, 174, 181-184, 190, 193 CD8, 6, 10, 20-21, 25, 46-48, 51, 71, 73, 75, 77, 78, 80-82, 85, 124-125, 127-132, 148, 151, 174, 181-184, 189-190, 193 CD80, 101-102, 104-106, 108, 112 CD86, 101, 104-106, 112 Chemical diabetes, 184 Clones, 10, 75-77, 80-82, 84-85, 123-124, 126, 129-130, 132, 155-156, 178, 183 Co-stimulatory molecules, 101-102, 107-108, 112 Colitis, 17 Complete Freund’s adjuvant, 19, 126, 151, 156 Concanavalin A, 14, 21 Congenic mouse, 107 Congenic stocks, 18, 22, 26, 39, 45, 49, 51, 53 Constitutive, 17, 43, 47-48, 110, 112 CTLA-4, 54, 151 Cyclophosphamide induced diabetes, 130-131 Cyclosporin, 146, 151-152 Cytokines, 10, 44, 58-59, 74, 101, 104, 126-127, 130, 152, 154, 156-157, 159, 187-188

206

NOD Mice and Related Strains: Research Applications in Diseases

D Deafness, 17, 39 Delayed type hypersensitivity, 74 Dendritic cells, 73, 88-89, 104, 106, 108-110, 112-113, 151, 156 Developmental biology, 18 Diabetes prevention trial, 160 Diet, 8, 16, 109, 148, 151, 158-159, 162 DNA repair, 174-175, 189

Housing requirements, 175 Human cell, 47, 180, 185-195 Human immunodeficiency virus (HIV), 186, 190 Human islet, 185 Human leukocyte antigen, 40 Human peripheral blood mononuclear cell, 187 Hyperglycemia, 3, 8-9, 128, 148-149, 184-185

E

I

Effector cell, 10, 44-45, 57, 71-80, 83, 8690, 101, 121, 124, 133, 135, 152, 155, 182-184, 194 Engraftment, 177, 181, 185, 186-191 Environment, 3, 7-8, 39-41, 48-49, 149 Epstein Barr virus (EBV), 189 European Nicotinamide Diabetes Intervention Trial, 160

I-A, 14, 43-45, 73, 75-76, 86-87, 152, 183 I-E, 15-16, 24, 44-45, 73, 75, 87 Idd1, 13, 23, 43, 55, 57 Idd2, 55, 57-58 Idd3, 13, 53-56 Idd4, 55, 57 Idd5, 54-55, 57 Idd6, 55, 57-58, 60 Idd7, 55, 60-61 Idd8, 52, 55, 60 Idd9, 55 Idd10, 13, 54-56 Idd11, 55 Idd12, 53, 55 Idd13, 46, 55 Idd14, 55, 57 Idd15, 55, 58 Idd16, 46, 55, 58 IDDM1, 43, 57, 59 IDDM2, 57, 59 IDDM3, 57 IDDM4, 57 IDDM5, 57 IDDM6, 57 IDDM7, 57 IDDM8, 57 IDDM9, 57 IL-1, 12-13, 56-57, 74, 88, 108, 157, 159, 179 IL-2, 13-14, 74, 87, 103, 105, 112, 127, 129, 131, 154, 157 IL-2 receptor, 103 IL-3, 13, 187-189 IL-4, 10, 13-14, 103, 105, 126-127, 131-132, 134, 159 IL-7, 10, 134-135 IL-10, 13, 101, 103, 109, 126-127, 132, 154, 157 IL-12, 101, 131

F Fas, 11, 14, 78, 103, 112 Fas ligand, 11, 103 Fc, 42, 54, 194

G Gender dimorphism, 7 Gene rearrangements, 78, 81, 83 Genetic polymorphisms, 39, 42 Genetic susceptibility, 41 Glutamic acid decarboxylase, 6, 45, 59, 83, 104, 121, 151, 155 Glycosuria, 8-9 Gonadectomy, 7-8 Granulocyte-macrophage colonystimulating factor, 187

H H2ct, 21, 46, 50, 55 H2g7, 41-51, 53-58, 60-61, 102, 104 Harderian gland, 11 Heat shock protein, 44, 83, 123, 151, 155 Hemochromatosis, 194 Hemolytic anemia, 18 Hemolytic complement, 177, 179-180, 187, 190, 194 Hormones, 110, 159

207

Index

Immunodeficiency, 1, 10, 173, 186, 189-190 Immunoglobulin, 41, 48, 52, 173 Immunostimulation, 150, 152-153, 156 Immunosuppression, 147, 150, 152, 189 Indomethacin, 110, 113 Innate immunity, 175, 177-178, 186, 192, 194 Insulin, 1, 6, 9-10, 13, 15, 37, 59-60, 71, 83-85, 121-123, 127-130, 145-146, 151, 154-155, 157-161, 177, 183 Insulin dependent diabetes mellitus, 1, 6, 37, 121, 145-146, 155, 177 Insulitis, 1, 3-4, 6-7, 10-11, 15-16, 19, 21-22, 24, 26, 44-46, 48-51, 53-58, 106-109, 111-112, 122-127, 148, 150, 154, 157, 161, 178, 181, 184, 194 Interferon gamma, 12-13, 22, 44, 58, 140 Interleukin, 42, 103, 124-125, 150, 154, 179 Intra-islet, 6, 10, 49, 183 IQI, 22 Islet, 3, 6, 8-10, 22, 24, 42-43, 45, 48-49, 54, 58-59, 75-77, 80-86, 89, 102, 104, 108-109, 112-113, 121-133, 135, 147148, 150, 154, 156, 158, 160-161, 182183, 185-186

K Ketoacidosis, 9, 147 Kilham rat virus, 6

L L-selectin, 151-152 Lacrimal gland, 11, 15-16 Leakiness, 174, 178, 182 Lipopolysaccharide, 56 Litters, 18-19, 38, 176 Lymphoaccumulation, 6, 10-11, 14, 17, 21-22, 48-49 Lymphohemopoietic, 179 Lymphomas, 11, 18, 25, 146

M Macrophage, 6, 12-14, 42-43, 48, 56, 58, 106, 108,-113, 152, 157, 159 Mammary tumor virus (Mtv), 51 Mel-14, 21 Memory, 107

Metabolic activity, 152, 154, 157-158 MHC, 1, 13, 15-16, 18, 21, 23-25, 39-41, 43-47, 49-51, 53-54, 56, 58-61, 102-108, 112, 149 MHC class I, 21, 25, 43, 46-48, 55 MHC class II, 14-15, 21, 23, 44, 46, 54, 104, 107, 112, 150-152 Mouse strains, 38, 41, 103, 111 Myelin basic protein, 103, 126

N Natural killer cell, 14, 42, 60, 175, 179 Negative selection, 44, 74 Neoplasia, 18 Nerup hypothesis, 41, 46 Nicotinamide, 151, 160-161 Nitric oxide, 101, 109, 131, 160-161 Non cataract, 20 Nonobese diabetic, 1, 3, 40, 102, 145 Nonobese nondiabetic, 3, 19

O Original description, 15 Oxygen radicals, 109

P p277, 151, 155 Pancreas, 10-11, 15, 19, 22, 43, 49, 113, 122, 125-128, 133, 152 Parasite, 191-194 Peri-insulitis, 6, 49, 54-55, 57 Phenotype, 3, 24, 39, 45, 49, 56, 60, 103, 174 Plasmodium falciparum, 192 Pneumocystis carinii, 175-176 Polymerase chain reaction, 80, 126 Polymorphisms, 37, 38, 39, 42, 47-48 Positive selection, 73-74 Pregestimil, 8, 151, 158-159 Prostaglandin E 2, 109 Protein kinase C, 72 Protein kinase, DNA activated catalytic polypeptide, 173

208

NOD Mice and Related Strains: Research Applications in Diseases

R Receptors, 13, 107 Red blood cell, 175, 179, 187, 192-193

S Salivary gland, 11, 15, 20, 125 Severe combined immunodeficiency, 1, 76, 173 Shionogi Research Laboratories, 3 Sialoadenitis, 15-16, 22 Simple sequence repeat polymorphisms, 20 Sjögren’s syndrome, 15 Soy, 159 Specific pathogen free, 148, 151, 175 Speed congenic, 52-53 Streptozotocin, 17, 184-185 Submandibular salivary glands, 15 Sulfamethoxazole, 175 Suppression, 45, 110, 132, 147, 150, 157 Susumu Makino, 1 Syngeneic mixed leukocyte reaction, 87

T-lymphoaccumulation, 10-11, 22, 48-49 Threshold liability, 49 Thymectomy, 11, 131-132, 134, 151 Thymus, 10-12, 14, 20, 42, 44, 59-60, 71-74, 86, 130-135, 153, 177, 190 Thyroid, 11, 16 Thyroiditis, 16-17 Time course, 15 Tolerance, 9, 23, 47, 54, 59-60, 74, 79, 83, 86, 104, 106-107, 111, 113, 122, 152-157, 161, 182 Transplantation, 133, 150, 161, 182, 185 Transporter associated with antigen processing, 20 Tumor necrosis factor (TNF), 126, 151, 153 Tumors, 18, 177, 180, 189, 191, 194-195

V Variable, 41, 49, 149, 154, 159 Variable number tandem repeat (VNTR), 59 Virus, 6, 42, 51-52, 73, 149-151, 156, 186, 189-191

T T cell receptor, 40-41, 51, 53, 71, 101, 103, 124, 126, 151, 174, 184 T cells, 71-90, 101-105 T helper, 154 T lymphocyte, 54, 57

W Wheat, 159