Animal Models in Diabetes: A Primer (Frontiers in Animal Diabetes Research)

ANIMAL MODELS OF DIABETES

Frontiers in Animal Diabetes Research Each volume of this series will be topic oriented with timely and liberally referenced reviews and provide in depth coverage of basic experimental diabetes research. Edited by Professor Anders A.F.Sima, Wayne State University, Detroit, USA and Professor Eleazar Shafrir, Hadassah University Hospital, Jerusalem, Israel. Volume 1 Chronic Complications in Diabetes: Animal Models and Chronic Complications edited by Anders A.F.Sima Volume 2 Animal Models of Diabetes: A Primer edited by Anders A.F.Sima and Eleazar Shafrir Volumes in preparation Insulin Signaling: From Cultured Cells to Animal Models edited by George Grunberger and Yehiel Zick Muscle Metabolism in Animal Models in Diabetes edited by Harriet Wallberg-Henriksson and Juleen R.Zierath This book is part of a series. The publisher will accept continuation orders which may be cancelled at any time and which provide for automatic billing and shipping of each title in the series upon publication. Please write for details.

ANIMAL MODELS OF DIABETES A PRIMER

Edited by

Anders A.F.Sima Departments of Pathology and Neurology Wayne State University Detroit, USA and

Eleazar Shafrir Department of Biochemistry Hadassah University Hospital Jerusalem, Israel

harwood academic publishers Australia • Canada • France • Germany • India • Japan • Luxembourg Malaysia • The Netherlands • Russia • Singapore • Switzerland

This edition published in the Taylor & Francis e-Library, 2005. “To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.” Copyright © 2001 OPA (Overseas Publishers Association) N.V. Published by license under the Harwood Academic Publishers imprint, part of The Gordon and Breach Publishing Group. All rights reserved. No part of this book may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and recording, or by any information storage or retrieval system, without permission in writing from the publisher. Amsteldijk 166 1st Floor 1079 LH Amsterdam The Netherlands British Library Cataloguing in Publication Data ISBN 0-203-30473-X Master e-book ISBN

ISBN 0-203-34329-8 (Adobe eReader Format) ISBN: 90-5823-096-1 (Print Edition) ISSN: 1029-841X

CONTENTS

Preface to the Series

vii

Preface

viii

Contributors

ix

1

Autoimmune Diabetes Mellitus in the BB Rat John P.Mordes, Rita Bortell, Herman Groen, Dennis Guberski, Aldo A.Rossini and Dale L.Greiner

1

2

The NOD Mouse and its Related Strains Hiroshi Ikegami and Susumu Makino

38

3

Obesity/Diabetes in Mice with Mutations in the Leptin or Leptin Receptor Genes Lieselotte Herberg and Edward H.Letter

56

4

The Zucker Diabetic Fatty (ZDF) Rat Richard G.Peterson

95

5

KK and KKAy Mice Shigehisa Taketomi and Hitoshi Ikeda

113

6

The Obese Spontaneously Hypertensive Rat (SHROB, Koletsky Rat): A Model of Metabolic Syndrome X Richard J.Koletsky, Jacob E.Friedman and Paul Ernsberger

126

7

Characteristics of Wistar Fatty Rat Hiroyuki Odaka, Yasuo Sugiyama and Hitoshi Ikeda

141

8

The New Zealand Obese Mouse: A Polygenic Model of Type 2 Diabetes Sofianos Andrikopoulos, Anne W.Thorburn and Joseph Proietto

152

9

The NSY Mouse: An Animal Model of Human Type 2 Diabetes Mellitus with Polygenic Inheritance Hironori Ueda, Hiroshi Ikegami, Masao Shibata and Toshio Ogihara

164

10

The Goto-Kakizaki Rat Claes-Göran Östenson

174

11

The OLETF Rat Kazuya Kawano, Tsukasa Hirashima, Shigehito Mori, Zhi-wei Man and Takashi Natori

188

vi

12

The JCR:LA-cp Rat: An Animal Model of Obesity and Insulin Resistance with Spontaneous Cardiovascular Disease J.C.Russell and S.E.Graham

199

13

The Neonatally Streptozotocin-Induced (n-STZ) Diabetic Rats, a Family of NIDDM Models Bernard Portha, M.H.Giroix, P.Serradas, J.Movassat, D.Bailbe and M.Kergoat

216

14

Galactosemic Animal Models W.Gerald Robison Jr.

237

15

The Rhesus Monkey (Macaca mulatto): A Unique and Valuable Model for the Study of Spontaneous Diabetes Mellitus and Associated Conditions Noni L.Bodkin

269

16

Psammomys obesus: Primary Insulin Resistance Leading to Nutritionally Induced Type 2 Diabetes Ehud Ziv and Rony Kalman

286

17

The C57BL/6J Mouse as a Model of Diet-Induced Type 2 Diabetes and Obesity Ann E.Petro and Richard S.Surwit

301

Index

313

PREFACE TO THE SERIES

Diabetes has been declared a major global health hazard by the WHO. Over the last few decades there has been an alarming increase in the incidence of diabetes particularly in densely populated areas such as India, China, southeast Asian countries and Arab nations. Even in North America and Europe the incidence of diabetes increases by 5% a year. The direct and indirect costs associated with diabetes are enormous. In the US they amounted to $137 billion in 1997 or a seventh of the total health care costs in this country. To avert this rapidly evolving global epidemic, it behoves the international biomedical community and responsible federal agencies and interest groups to intensify research into the causes of this disease and its complications, and to rapidly increase public awareness of the disease through education. Major advances have been made in diabetes research in animal models, contributing enormously to the understanding of etiopathology of this disease and its dreaded chronic complications. In particular factors in the areas of immunology, insulin signal transduction and insulin action as well as pathogenetic mechanisms involved in the development of the chronic complication have become clearer. The new knowledge gained is only slowly being translated to the benefit of the patients and to serve as a basis for the development of new therapeutic modalities. The accumulation of this scattered information and ongoing publication of data from the interdisciplinary and critical reviews on diabetes in various animals is our fundamental motive. It is our hope that this book series on Frontiers in Animal Diabetes Research will be an efficient vehicle for communicating extensive up-to-date review articles by the leading world experts in the field. Each volume will be topic oriented with timely and liberally referenced reviews. It will fill a gap in the spectrum of diabetes related journals and publications in as far as it will focus on all aspects of basic experimental diabetes research. As such we hope it will provide a valuable reference source for graduate students, research fellows, basic academic and pharmacological researchers as well as clinic investigators. Anders A.F.Sima Eleazar Shafrir

PREFACE

Since Oscar Minkowski discovered, about 110 years ago, that the removal of the pancreas causes diabetes in dogs, many other animal models with spontaneous diabetes or nutritionally induced diabetes have been used to obtain a better understanding of this disease. This holds true for both type 1 and type 2 of the disease. Because of the continuing increase of diabetes to epidemic proportions, mainly of type 2, its pathophysiology in many respects is still largely unknown and animal models of type 2 diabetes are of great importance for future research in diabetology. Much that we know about the pathological processes, immune derangements, insulin secretion and insulin signaling abnormalities, as well as nutritional influences in diabetes, has been derived from studying diabetes in animals. An effective diabetes model demonstrates tissue functions which have been compromised by genetic mutation or environmental effects often associated with longstanding complications. Studies on models of diabetes became fundamental, since laboratory investigation in human subjects is limited by the availability of tissues and the long duration of observations required for the basic approach to the study of cellular changes. In addition, there is an urgent need for preventive and curative diabetes research. This book is intended to provide a review of the characteristics of the more commonly used animal species with various diabetic syndromes which were developed and extensively investigated during the last few decades. Animal models have been included which are readily available, reasonably well described and proven to be of value in the research of both types of diabetes. It is hoped that this extensively referenced book will be helpful to established investigators, as well as to graduate students, young investigators and pharmaceutical scientists working on the development of antidiabetic and preventive modalities in various areas of diabetes and its complications.

CONTRIBUTORS

Sofianos Andrikopoulos Department of Medicine University of Melbourne Royal Melbourne Hospital Parkville, Victoria 3050 Australia D.Bailbe Lab. Physiopathologie Nutrition CNRS ESA 7059 Université Paris 7/D. Diderot 2 place Jussieu 75251 Paris Cedex 05 France Noni L.Bodkin Obesity and Diabetes Research Center Department of Physiology School of Medicine University of Maryland 10 South Pine Street MSTF 6–00

x

Baltimore, Maryland 21201 USA Rita Bortell Department of Medicine University of Massachusetts Medical Center Worcester, Massachusetts 01605 USA Paul Ernsberger Departments of Nutrition and Medicine Case Western Reserve University School of Medicine Cleveland, Ohio 44106–4935 USA Jacob E.Friedman Departments of Nutrition and Medicine Case Western Reserve University School of Medicine Cleveland, Ohio 44106–4935 USA M.H.Giroix Lab. Physiopathologie Nutrition CNRS ESA 7059 Université Paris 7/D. Diderot 2 place Jussieu 75251 Paris Cedex 05 France S.E.Graham Department of Surgery University of Alberta Edmonton, Alberta Canada Dale L.Greiner Department of Medicine University of Massachusetts Medical Center Worcester, Massachusetts 01605 USA Herman Groen Department of Medicine University of Massachusetts Medical Center Worcester, Massachusetts 01605 USA Dennis Guberski

xi

Department of Medicine University of Massachusetts Medical Center Worcester, Massachusetts 01605 USA lieselotte Herberg Diabetes Research Institute Heinrich-Heine University of Düsseldorf Düsseldorf Germany Tsukasa Hirashima Tokushima Research Institute Otsuka Pharmaceutical Co., Ltd. Otsuka Japan Hitoshi Ikeda Pharmaceutical Research Laboratories Takeda Chemical Industries Ltd. Yodogawa-ku, Osaka 532–8686 Japan Hiroshi Ikegami Department of Geriatric Medicine Osaka University Medical School 2–2 Yamadaoka Suita, Osaka 565 Japan Rony Kalman Diabetes Research Unit Hadassah-Hebrew University Medical Center Jerusalem Israel Kazuya Kawano Tokushima Research Institute Otsuka Pharmaceutical Co., Ltd. Otsuka Japan M.Kergoat Merck-Lipha Centre de Recherché 91380 Chilly-Mazarin

xii

France Richard J.Koletsky Departments of Nutrition and Medicine Case Western Reserve University School of Medicine Cleveland, Ohio 44106–4935 USA Edward H.Leiter The Jackson Laboratory Bar Harbor, Maine 04609 USA Susumu Makino AC Center Shionogi Laboratories Osaka Japan Zhi-wei Man Tokushima Research Institute Otsuka Pharmaceutical Co., Ltd. Otsuka Japan John P.Mordes Department of Medicine University of Massachusetts Medical Center Worcester, Massachusetts 01605 USA Shigehito Mori Tokushima Research Institute Otsuka Pharmaceutical Co., Ltd. Otsuka Japan J.Movassat Lab. Physiopathologie Nutrition CNRS ESA 7059 Université Paris 7/D. Diderot 2 Place Jussieu 75251 Paris Cedex 05 France Takashi Natori Tokushima Research Institute Otsuka Pharmaceutical Co., Ltd.

xiii

Otsuka Japan Hiroyuki Odaka Pharmaceutical Research Division Takeda Chemical Industries Ltd. Yodogawa-ku, Osaka 532–8686 Japan Toshio Ogihara Department of Geriatic Medicine Osaka University Medical School 2–2 Yamadaoka Suita, Osaka 565 Japan Claes-Göran Östenson Department of Molecular Medicine The Endocrine and Diabetes Unit Karolinska Institute and Hospital SE-171 76, Stockholm Sweden Richard G.Peterson Indiana University School of Medicine, Anatomy MS 5035, 635 N.Barnhill Drive Indianapolis, Indiana, 46202–5120 USA Ann E.Petro Duke University Medical Center Durham, North Carolina 27710 USA Bernard Portha Lab. Physiopathologie Nutrition CNRS ESA 7059 Université Paris 7/D. Diderot 2 place Jussieu 75251 Paris Cedex 05 France Joseph Proietto Department of Medicine University of Melbourne Royal Melbourne Hospital Parkville, Victoria 3050

xiv

Australia W.Gerald Robison, Jr. National Eye Institute National Institutes of Health Bethesda, Maryland 20892–2735 USA Aldo A.Rossini Department of Medicine University of Massachusetts Massachusetts Medical Center Worcester, Massachusetts 01605 USA J.C.Russell Department of Surgery University of Alberta Edmonton, Alberta Canada P.Serradas Lab. Physiopathologie Nutrition CNRS ESA 7059 Université Paris 7/D. Diderot 2 place Jussieu 75251 Paris Cedex 05 France Masao Shibata Department of Health Aichi-Gakuin University College of General Education Iwasaki, Nishincho Aichi-gun, Aichi 470–01 Japan Yasuo Sugiyama Pharmaceutical Research Division Takeda Chemical Industries Ltd. Yodogawa-ku, Osaka 532–8686 Japan Richard S.Surwit Duke University Medical Center Durham, North Carolina 27710 USA

xv

Shigehisa Taketomi Discovery Research Laboratories Takeda Chemical Industries Ltd. Yodogawa-ku, Osaka 532–8686 Japan Anne W.Thorburn Department of Medicine University of Melbourne Royal Melbourne Hospital Parkville, Victoria 3050 Australia Hironori Ueda Department of Geriatric Medicine Osaka University Medical School 2–2 Yamadaoka Suita, Osaka 565 Japan Ehud Ziv Diabetes Research Unit Hadassah—Hebrew University Medical Center Jerusalem Israel

1. AUTOIMMUNE DIABETES MELLTTUS IN THE BB RAT JOHN P.MORDES, RITA BORTELL, HERMAN GROEN, DENNIS GUBERSKI, ALDO A.ROSSINI and DALE L.GREINER Department of Medicine, University of Massachusetts Medical Center, Worcester, MA 01605, USA

INTRODUCTION Since last reviewed in detail by us in 1992, more than 250 studies of the BB rat have appeared. This literature testifies to the continuing scientific interest in this animal as a model of human Type 1 diabetes mellitus, even among proponents of the NOD mouse (Atkinson and Leiter, 1999). Many lines of evidence document that Type 1 or insulin-dependent diabetes (IDDM) in humans is a disease of autoimmunity. These include observations of pancreatic insulitis, islet autoantibodies, reappearance of disease after syngeneic pancreas allografts, induction of disease in bone marrow allograft recipients, successful immunoprophylaxis with cyclosporin, and major histocompatibility complex (MHC) associations. Analogous evidence suggests that the hyperglycemic syndrome of the BB rat is a similar disorder. SPONTANEOUSLY DIABETIC BB RATS The Original BioBreeding Colony The spontaneously diabetic BB rat was discovered in 1974 in a colony of outbred Wistar rats at BioBreeding Laboratories in Ottawa, Canada (Nakhooda et al., 1977). Many breeding colonies have subsequently been established (Table 1). The cumulative frequency of spontaneous diabetes among the original BioBreeding animals was about 10%. Selective breeding has increased the cumulative frequency to

2

AUTOIMMUNE DIABETES MELLITUS IN THE BB RAT

>90% in some colonies. These animals are all described as “diabetes prone” BB rats to distinguish them from “diabetes resistant” or “control” sublines developed later. Genetic Heterogeneity Among BB Bats from Different Sources As detailed below, all spontaneously diabetic BB rats share the class II RT1U rat MHC, develop pancreatic insulitis with selective beta cell destruction, and are Table 1 Pedigree of BB rat originating from the bio-breeding laboratory Ottawa Ontario CN, spontaneous mutation 1974. U.S. Colony Founded 1977 in Worcester, MA

MRC Colony Founded in Ottawa Ontario Canada 1978

BB/Wor Rat Diabetes Prone Lines

BB/Wor-Derived Colonies: Location

Year of Derivation and designation

Diabetes Prone BB Lines

dpBB-Derived Colonies: Location

Designation

BBBA/Wor// Brm F70

Gentofe, Denmark (Hagedorn) Toronto Ontario (Hospital for Sick Children) NIH Reference Colony

1983 BB/H

dpBB

Edinburgh, UK

dpBB/E

1984

Edinburgh, UK

dpBB/Ed

1983–1998

Univ. Greifscoald, Karlsburg, Germany Philadelphia, USA

BBDP/Wor//Brm F70

BBNB//Wor/ Brm F71

BBPA/Wor//Brm F71

M&B (Formerly Møllegaard) Brussells Belgium Memphis, TN U.S.A. Denver CO, USA, Bucharest Romania Westmead Australia Pittsburgh, PA USA None

1987 BBDP/ Wor//Mol 1988 BBDP/Wor

dpBB/OK dpBB/Ph

1989 BBDP/ Wor//Utm 1986 BBDP/Wor 1996 BBDP/Wor 1986 1988

This work was supported in part by grants DK25306 (AAR), DK36024 (DLG), and DK41235 (JPM) from the National Institutes of Health, and by grant DFN98.501 (HG) from the Diabetes Foundation of the Netherlands. The contents of this publication are solely the responsibility of the authors and do not necessarily represent the officiai views of the National Institutes of Health.

JOHN P.MORDES ET AL.

U.S. Colony Founded 1977 in Worcester, MA BB/Wor Rat Diabetes Prone Lines BBIWor Diabetes Resistant Lines BBDR/Wor// Brm F71

BBVB/Wor// Brm F71

BB/Wor-Derived Colonies: Location

3

MRC Colony Founded in Ottawa Ontario Canada 1978 Year of Derivation and designation

Diabetes Prone BB Lines

dpBB-Derived Colonies: Location

Designation

MRC Resistant BB Rats Gentofe, Denmark Hagedorn 1983 Westmead Australia 1986 Pittsburgh, PA USA 1988 Denver CO, USA, 1986 Brussells Belgium, 1988 M&B (Formerly Møllegaard) 1987 None

BBDR/Wor

drBB (BBc, BBn) BBDR/Ed

BBDR/Wor BBDR/Wor BBDR/Wor//Mol

Partial listing of BB rat sublines and their origins. All BB rats were derived from the outbred Wistar rats discovered at the BioBreeding Laboratory in Canada in 1974. A subset of these animals formed the nucleus of two colonies that later became the source of all tertiary colonies. BB/Wor rats have been inbred for 70+ generations. The tertiary colonies of Worcester-origin were derived from one of three sublines, as indicated. Diabetes prone dpBB rats have remained outbred, and the tertiary colonies from these Ottawa animals have undergone various degrees of inbreeding. BBDR/Wor animals were derived from diabetes prone BBBA/Wor forebears in the fifth generation of inbreeding. The BBVB/Wor rat was derived directly from outbred forebears from the original BioBreeding colony. Additional information is available from the website of the Institute for Laboratory Animal Research (ILAR) at http://www2.nas.edu/ilarhome/ and from Festing (Festing, 1993).

severely lymphopenic. Despite these uniform features, however, “BB rats” are not all identical. Table 1 summarizes some of the history of the BB rat. The original outbred BioBreeding animals gave rise to two broad classes of contemporary BB rats. The first class is derived from the colony established in 1977 in Worcester, MA, USA and sponsored for two decades by the National Institutes of Health (NIH). The second class is derived from a colony established in 1979 in Ottawa, Canada from BioBreeding progenitors under the sponsorship of the Canadian Medical Research Council (MRC). The Worcester facility developed inbred sublines of diabetes prone “BB/Wor” rats. In the 1980s, it supplied nuclear breeders for several tertiary colonies around the world. The Ottawa colony has remained outbred. It has been the source of several other tertiary colonies that in turn have been partially or competely inbred (See Table 1). All available BB rats are thought to belong to one of these two progenitor classes. For BB/Wor derived animals, Table 1 indicates both when and from which subline tertiary colonies were derived. Available data for the Ottawa-derived colonies are less detailed. All secondary and tertiary colonies

4

AUTOIMMUNE DIABETES MELLITUS IN THE BB RAT

obviously vary with respect to source of the original breeders and duration of genetic isolation; not surprisingly they also vary with respect to frequency and severity of disease and immunological characteristics (Crisá et al., 1992; Guberski, 1994). The genetic heterogeneity of the various outbred and inbred “BB rats” implicit in this history is well documented (Prins et al., 1991). These authors investigated the extent of the heterogeneity among 26 distinct lines (24 inbred, 2 outbred) of the BB rat by analyzing 19 protein markers. They observed polymorphisms in 9 markers and used them to define 7 distinct haplotypes. In the decade since the appearance of that report, additional divergence may have occurred. In addition to genetic drift, environmental factors may also affect BB rat colonies (Crisá et al., 1992). As will be discussed in detail below, the incidence of diabetes in BB rats can be modified by environmental factors including diet and infectious agents. These factors can affect experimental outcomes. In addition, breeding programs based on the identification of diabetic rats may select for environmentally activated (or suppressed) genes if they contribute locally to the appearance (or suppression) of diabetes. In summary, there is significant genotypic and phenotypic variation among lines of BB rats. Interpretations and comparisons of results obtained using “the BB rat” should take into account the origin and status of the colonies from which the experimental animals were obtained. Nomenclature An additional complication in interpreting BB rat data is inconsistent nomenclature. Some Institute for Laboratory Animal Research (ILAR) designations are listed in Table 1, but for the most part these have not been used in the literature. Few reports use the newest designations. For the sake of consistency with the published literature, this review generally uses designations employed by the authors of original reports. DIABETES RESISTANT BB RATS Important assets in the study of autoimmunity in the BB rat are lines of diabetes resistant animals, the origins of some of which are listed in Table 1. In the Ottawa colony, resistant animals (usually designated BBdr but sometimes BBc or BBn) are outbred and non-lymphopenic. A diabetes-resistant (DR) subline of the BB rat based on animals obtained from the Ottawa colony has also been established in Edinburgh, U.K (Joseph et al., 1993). These BB-DR/Ed rats are unusual in that, like diabetes-prone BB-DP/Ed rats, they are lymphopenic. In the Worcester colony, there are two sublines of diabetes resistant BB/Wor rats: BBDR/Wor and BBVB/ Wor (Crisá et al., 1992; Guberski, 1994). BBDR/Wor animals were derived from diabetes prone BBBA/Wor forebears in the fifth generation of inbreeding. The BBVB/Wor rat was derived directly from outbred forebears from the original BioBreeding colony. Both the BBDR/Wor and BBVB/Wor sublines have been continuously selected for resistance to diabetes (Guberski, 1994). What is now officially designated as the BBDR/Wor rat was frequently referred to as the DR-BB/Wor rat in the literature. Unlike diabetes prone BB/ Wor rats, diabetes resistant BB/Wor rats are not lymphopenic (Crisá et al., 1992). They have been used in the majority of studies of disease resistance in the BB rat. The BBDR/Wor rat has been inbred for >70 generations. When housed in a viral antibody free (VAF) vivarium, none (0%) become spontaneously diabetic (Guberski, 1994). Important features that distinguish these animals from DP-BB/Wor rats are summarized in Table 2. One of the most useful characteristics of the DR-BB/Wor rat is its propensity to become diabetic in response to environmental perturbation. As will be discussed, it is clear that diabetes resistant BB/Wor rats harbor populations of diabetogenic effector cells that are normally held in check by

JOHN P.MORDES ET AL.

5

populations of regulatory cells, among them cells that express the RT6+ phenotype (Crisá et al., 1992). BBDR/Wor rats are susceptible to collagen-induced arthritis (Watson et al., 1990) and inducible lymphocytic thyroiditis (Crisá et al., 1992). OBTAINING BB RATS Information on the status of BB rat lines and colonies is available from the International Index of Laboratory Animals (Festing, 1993) and the ILAR web site (http://www2.nas.edu/ilarhome/). The web site provides contact information for many BB rat colonies. The major American supplier of BB rats is Biomedical Research Models, Inc., Worcester, MA (http://www.brm.com); the major European vendor is M&B, A/S, Ry, Denmark (http://www.m-b.dk/). At this writing, the approximate costs, exclusive of shipping, for weanling diabetes prone and diabetes resistant rats from BRM, Inc. are USD26 and USD21, respectively. From M&B, the costs are roughly comparable. The production facility for BB rats at BRM, Inc merits comment because of the well defined status of its animals. This company was formed when the NIH decided to privatize the colony that had been maintained under their sponsorship at the University of Massachusetts Medical Center in Worcester (Guberski, 1994). The “BB/Wor//Brm” animals in this colony are inbred, have clearly defined Table 2 Selected clinical and immunological characteristics of diabetes prone and diabetes resistant

Cumulative frequency of diabetes Age at onset Insulitis Diabetic rats Nondiabetic rats Thyroiditis Peripheral lymphocytes Lymphocyte Subsets CD4+, CD5+ OX22, OX32 (CD45R+) CD8+, RT6+ RT6+ intraepithelial Lymphocytes NK Cell Activity Environmental Perturbants KRV infection LCMV infection Hydrolysed casein diet

Diabetes Prone

Diabetes Resistant

>90% >85% of cases between 55 and 120 days of age

0% in VAF vivaria Inducible in animals if treatment begun by 30 days

100% 50% 5% in BBVB/Wor 100% in BBDR/Wor Lymphopenic

100% 0% in VAF vivaria ~20% in animals given poly I:C + anti-RT6 mAb Nonlymphopenic

Low Low Very low

Normal Normal Normal

Present at reduced levels Increased

Present Low

No effect Reduces IDDM incidence Reduces IDDM

Induces diabetes Not known No effect

Partial listing of characteristics that distinguish diabetes prone and diabetes resistant BB/Wor rats. Susceptible and resistant animals that originated from the Ottawa colony (see Table 1) may differ. Citations are provided in the text.

6

AUTOIMMUNE DIABETES MELLITUS IN THE BB RAT

pedigrees, and are the progenitors of many other colonies worldwide (Table 1). They are housed in viral antibody free (VAF) conditions, and available for shipment worldwide. Diabetes prone BB rats in the BRM, Inc. colony are available from three of the six sublines listed in Table 1: BBDP/Wor, BBNB/Wor, and BBDR/ Wor (Guberski, 1994). Two lines of diabetes resistant BB rats, BBDR/Wor and BBVB/Wor, are also available from BRM. All BB/Wor animals have been inbred for >70 generations. The diabetes prone BB/Wor rat has also been crossed with the Zucker fatty rat to yield two coisogenic lines, the BBZDP/Wor rat, an obese animal with autoimmune features, and the BBZDR/Wor, a model of NIDDM available from BRM. MAINTAINING AND USING BB RATS BB rats should be obtained from a VAF production facility and maintained in microisolators in VAF or specific-pathogen free quarters. A laminar flow hood should be used as a workstation for animal husbandry and for experimental procedures. These precautions are necesssary because, as detailed below, the frequency of both spontaneous and induced diabetes in BB rats is modified by viral infection. For example, lymphocytic choriomeningitis virus (LCMV) prevents dia betes in diabetes prone rats, and Kilham rat virus (KRV) can induce IDDM in resistant animals (Crisá et al., 1992). We recommend that all materials taken into rooms housing BB rats be autoclaved. Non-autoclavable materials and the exposed surfaces of tubes, equipment, and other objects should be disinfected with a solution of Clidox™ or its equivalent. Personnel handling BB rats should wear sterile outer garments, including bonnets, masks, gloves, and shoe covers. Frequent disinfection of handling areas, racks, and other objects in the facility with fresh Clidox™ is recommended to the facility free of contaminating viruses. The pathogen status of BB rat vivaria should be monitored by serological testing of sentinel animals housed without microisolation in the same rooms. We routinely add some used bedding from randomly selected cages of BB rats to the sentinel cages at regular intervals. Because DP-BB rats are lymphopenic, it is preferable to use DR-BB rats or immunocompetent non-BB rats as sentinels. Procedures for detecting diabetes and basic experimental protocols for use with BB rats are available (Whalen et al., 1996). Diabetic BB rats require daily treatment with insulin and may require additional treatments for ketonemia; detailed therapeutic protocols have been published (National Research Council, 1996). Breeding diabetes prone BB rats requires special procedures. Mating diabetic male BBDP rats with female BBDP rats that have been prophylactically transfused to prevent diabetes (see page 11) enhances both fertility and nursing (Guberski et al., 1992). CLINICAL FEATURES AND PATHOLOGY Clinical autoimmunity and organ pathology are similar in spontaneously diabetic BB rats and in DR-BB/ Wor rats that are induced to become hyperglycemic. Clinical Diabetes in BB Rats Like human IDDM, spontaneous diabetes in BB rats appears during adolescence, typically between 60 and 100 days of age (Crisá et al., 1992). Incidence in males and females is similar. Insulin response to glucose challenge may be normal up to 10 days before disease onset; it then declines as beta cell mass declines (Teruya et al., 1993). Clinical onset is abrupt, and hyperglycemia is accompained by weight loss, hypoinsulinemia, and ketonuria. Fatal ketoacidosis ensues unless exogenous insulin is given. The

JOHN P.MORDES ET AL.

7

cumulative frequency of diabetes through 120 days of age generally ranges between 60 and 90% depending on the colony; it is >85% in the Worcester colony (Guberski, 1994). In a study of old BB/Mol rats that had not developed spontaneous diabetes, it was observed that they nonetheless harbored populations of autoreactive cells capable of inducing disease in adoptive recipients (MacKay, 1995). Hyperglucagonemia and other endocrine abnormalities characteristic of IDDM (e.g. decreased growth hormone secretion) are present (Crisá et al., 1992). The clinical diabetic syndrome observed after induction in the DR-BB/Wor rat has not been reported in as much detail. It is, however, similarly characterized by susceptibility to ketoacidosis in the absence of exogenous insulin. Pancreatic Insulitis Intense mononuclear infiltration within and around the islets of Langerhans (“insulitis”) is the characteristic histopathological lesion of spontaneous diabetes in BB rats at the time of onset (Nakhooda et al., 1977; Crisá et al., 1992). Serial pancreatic biopsies have shown that the lesions start as early as 2–3 weeks before overt diabetes and rapidly progress to complete, selective destruction of islet beta cells. In chronically diabetic animals, “end stage” islets with few or no inflammatory cells are observed. In contrast, islet alpha, delta, and pancreatic polypeptide (PP) cell numbers and morphology appear to be preserved. Sub-populations of lymphoid cells present at different stages of insulitis have been extensively characterized (Crisá et al, 1992). Macrophages are among the earliest, possibly the first, of the cellular elements observed (Hanenberg et al, 1989). Alternatively, it has been suggested that the earliest infiltrating elements may include dendritic cells, which may have enhanced antigen presenting capability in the BB rat (Tafuri et al., 1993). Both CD4+ and CD8+ T cells, natural killer (NK) cells and, to a lesser extent, B cells subsequently infiltrate the islets. Insulitis has also been studied immunocytochemically in resistant BB rats in which diabetes was induced by immunological perturbation (Jiang et al, 1990). This approach permitted kinetic analyses of islet pathology in relationship to induction by administration of a cytotoxic anti-RT6 antibody (See page 24). The study revealed a prodromal period of 10 days during which no morphologic abnormalities of the pancreas were detected. This was followed by a second phase of early insulitis in which a few islets were infiltrated by both macrophages and T cells. The lesions rapidly progressed, and by day 18 insulitis was generalized and intense. Pancreatic ductular lymphocytic infiltration (clusters of lymphocytes around small ductules and venules of the pancreas) can also antedate the appearance of insulitis (Bone et al, 1990). Both the early periductular lesion and infiltrative insulitis may be found in animals that do not become diabetic. Periductular lesions can also be found in F1 crosses of BB rats with other non-diabetes prone inbred strains, although these animals never become diabetic (Colle, 1990). The relationship of this lesion to insulitis remains unclear. Hyper-expression of class I antigens occurs on islet and exocrine cells in diabetic BB rats (Bone et al, 1990). Class II antigen expression is not seen on BB rat pancreatic beta cells (Timsit et al, 1989). In situ hybridization studies have demonstrated the presence of tumor necrosis factor (TNF), IL-1 and IL-6 mRNA in cells at the site of lymphocytic infiltration in newly diabetic BB rats (Jiang and Woda, 1991). RT-PCR has been used to measure mRNA encoding T helper (Th) type 1 and 2 cytokines present in inflamed islets (Zipris et al, 1996; Kolb et al, 1996). In both DP and RT6-depleted DR-BB/Wor rats, interferon-gamma (IFN-) mRNA was present in islets before and during disease onset. IL-2 and IL-4 mRNAs were minimal or undetectable in infiltrated islets but present in activated peripheral T cells (Zipris et al., 1996). IL-10 mRNA was present at low abundance. The data suggest that insulitis in the BB rat is a Th1 type inflammatory response. Consistent with this interpretation, it was found that mRNA encoding the p40 chain of IL-12 was

8

AUTOIMMUNE DIABETES MELLITUS IN THE BB RAT

also present before and during disease onset. Th1 lymphocytes appear to predominate over Th2 lymphocytes in these inflammatory lesions. Nitric oxide is a candidate beta cell cytotoxic agent. The inducible NO synthase (iNOS) has been found in pancreatic lesions of adult diabetes-prone BB rats using both RT-PCR and immunohistochemistry (Kleemann et al., 1993; Kolb et al., 1996). Expression was absent in normal Wistar rats, young DP-BB rats without insulitis, and in DR-BB rats. Parallel staining for ED1+ macrophages showed restriction of iNOS expression to areas of islet infiltration by macrophages, suggesting the possibility of iNOS involvement in islet destruction (Kleemann et al., 1993). Pancreatic Endothelium Activation and selective increase in the leakiness of islet capillaries and postcapillary venules at the onset of spontaneous diabetes have been observed (De Paepe et al., 1992). Other studies have described an inducible venular defect specific to the pancreas (Desemone et al., 1990) that can be prevented by anti-inflammatory agents (Kitagawa et al., 1993). The leakiness defect is also seen in other, non-diabetic rat strains, and its relationship to diabetes pathogenesis remains uncertain. Endothelium could participate in the BB rat autoimmune process in other ways. Hyper-expression of class I and induction of class II MHC molecules on pancreatic endothelium occurs early in the disease (Bone et al., 1990). Anti-endothelial cell autoantibodies also develop in spontaneously diabetic BB/Wor rats and in DR-BB/ Wor rats during the course of diabetes induction (Doukas et al., 1996). Thyroiditis Autoimmune thyroiditis is more common among humans with IDDM than in the general population. Lymphocytic infiltration of the thyroid has also been described in both diabetic and non-diabetic DP-BB/ Wor animals, but the lesion does not progress to frank hypothyroidism in the absence of dietary or other manipulation (Reimers et al., 1996; Mooij et al., 1993; Mori et al., 1998). Pathologically, thyroiditis is associated with infiltration of both dendritic cells and lymphocytes (Simons et al., 1998). RT-PCR analyses of cytokine mRNA in the thyroid glands of RT6-depleted DR-BB/Wor rats revealed a Th1-type cytokine profile similar to that observed in inflamed islets (Zipris, 1996). Thyroiditis can also be adoptively transferred from DP-BB rats to MHC compatible naïve recipients using splenocytes (McKeever et al, 1990), and it can be prevented by the transfusion of normal MHCcompatible lymphocytes (Burstein et al., 1989). Lines of thyroid-reactive cells have also been derived from diabetes prone animals (Allen and Thupari, 1996). Like IDDM, thyroiditis can also be induced in diabetes resistant BB/Wor rats (Thomas et al., 1991), and DR-BB/Wor thymocytes can adoptively transfer thyroiditis (Kimura and Davies, 1996; Whalen et al., 1994). In early reports, thyroiditis was observed in about 59% of diabetic and 11% of non-diabetic DP-BB/Wor rats. It is MHC-associated (Awata et al., 1995). The prevalence of lymphocytic thyroiditis varies among different diabetes prone sublines in the Worcester colony (Rajatanavin et al., 1991). At ~110 days of age, for example, the prevalence was 100% in BBNB/Wor rats but only 4.9% in BBBE/Wor animals. This observation, together with the low rate of concordance of diabetes and thyroiditis (Pettersson et al., 1995), suggests that these two hereditary autoimmune diseases are not tightly linked genetically. Understanding the relationship between IDDM and lymphocytic thyroiditis in the BB rat remains a major unsolved problem.

JOHN P.MORDES ET AL.

9

Extrapancreatic Abnormalities and Diabetic Complications Extra-pancreatic diseases other than thyroiditis have also been described in both diabetic and non-diabetic BB rats as reviewed by us previously (Crisá et al., 1992). Some have sialadenitis. The most prominent abnormalities involve the lymphoid system. Lymph nodes of young animals show variable degrees of paracortical and medullary replacement by plasmacytoid lymphocytes. In older rats, B cell lymphomas occur, the frequency being higher in long-term diabetic rats (15%) than in non-diabetic rats (~1%). The observation of similar lymphoproliferative lesions in other autoimmune diseases and in chronic graft vs. host disease suggests that antigen or autoantigen driven clonal expansion of B cells might play a role in the development of these malignancies (Friedman et al., 1991; Shirai et al., 1991). In a necropsy study of BB/E rats, the increased frequency of lymphoma was reportedly associated with a translocation of the c-myc oncogene (Meehan et al., 1993). Diabetes prone BB rats also have an increased susceptibility to pulmonary infections; sterile granulomas in lymph nodes, kidney and pancreas; and prostatic atrophy. As is the case in humans with IDDM, chronically hyperglycemic BB rats gradually develop secondary systemic complications (Crisá et al., 1992). Pathological changes affecting retina (Chakrabarti and Sima, 1997), myocardium (Giles et al., 1998), kidney (Chakrabarti et al., 1989), gut (Young et al., 1995), gonads and sexual function (McVary et al., 1997), bone metabolism (Verhaeghe et al., 1990), vascular endothelium (Lindsay et al., 1997), and autonomic and peripheral nerves (Mohseni and Hildebrand, 1998; Sima and Sugimoto, 1999) have been reported. An advantage of the BB rat in the study of complications is its expression of aldose reductase, which is present at only low levels in the NOD mouse (Yagihashi et al., 1990). BB rats have also been used to model the effects of IDDM on pregnancy (Lea et al., 1996). Thymic Epithelial Defects Given the immunological defects involved in the development of autoimmune diabetes in BB rats, the thymus has been a major focus of study. The expression class II MHC antigen by BB rat thymic epithelial cells is abnormal. In both diabetes prone and diabetes resistant BB/Wor rats, areas of the thymic cortex lack class II MHC expressing epithelial cells (Rozing et al., 1989), and some regions of thymic cortex and medulla lack any thymic epithelium (Doukas et al., 1994). Thymic epithelial defects first appear at 4 weeks of age. The genetic predisposition to thymic epithelial defects in the BB/Wor segregates as an autosomal dominant trait (Doukas et al., 1994), but does not segregate with the diabetes phenotype (DLG, JPM, unpublished observations). HUMORAL IMMUNITY IN THE BB RAT Several different autoantibodies can be detected in the serum of BB rats (Crisá et al., 1992). As is the case with human IDDM, the degree to which abnormal humoral immunity actually contributes to the pathogenic process is unclear. Antibodies directed against the surface of islet cells (ICSA) appear 1–2 months before diabetes onset. The majority are capable of complement mediated cytolytic activity against pancreatic islets in vitro, but no direct pathogenic role of these islet-specific antibodies has been demonstrated in vivo. They may be secondary to the destruction of islet cells. Islet cell cytoplasmic antibodies (ICA) have not been conclusively demonstrated in BB rats (Crisá et al., 1992). DP-BB rats circulate autoantibodies reactive against lymphocytes, gastric parietal cells, smooth * BBBE/Wor//Brm rats are no longer available.

10

AUTOIMMUNE DIABETES MELLITUS IN THE BB RAT

muscle and thyroid colloid, but not adrenal tissue. Anti-lymphocyte antibodies can be present before diabetes onset and their presence predicts the development of spontaneous diabetes in Ottawa BBdp rats (Bertrand et al., 1994). Anti-endothelial cell autoantibodies that appear to be capable of inducing endothelial leakiness are detectable in both untreated DP-BB/ Wor rats and RT6-depleted DR-BB/Wor rats before diabetes onset (Doukas et al., 1996). Autoantibodies important in the prediction of human IDDM may or may not have counterparts in BB rats. Anti-glutamic acid decarboxylase (GAD) antibodies are reportedly absent in BB/d (Davenport et al., 1995) and other BB rats (Mackay et al., 1996), but may be present in the BB/OK strain (Ziegler et al., 1994). Another human IDDM autoantibody, IA-2, is either absent (DeSilva et al., 1996) or present only at low titer (Myers et al., 1998) in BB rats. BB rat autoantibodies against heat shock protein 65 (hsp-65) (Mackay et al., 1996) are absent. Insulin autoantibodies (IAA) have been reported in BB/W/D rats (Wilkin et al., 1986), but their presence could not be confirmed in DP or DR-BB/Wor rats (Markholst et al., 1990). The possibility that age or strain specific effects could account for the different results remains open. The pathogenic significance of these autoantibodies and the B lymphocytes secreting them is not clear. An expanded population of CD5+ B cells may be present in DP-BB rats (Shockett and Woodland, 1988). A phenotypically similar CD5+ B cell population appears early in ontogeny in normal humans and secretes antibodies with a pattern of cross-reactive idiotypes directed against some of the autoantigens listed above (Casali and Notkins, 1989). An immunoregulatory role early in ontogeny has been postulated for this subset

JOHN P.MORDES ET AL.

11

Figure 1 The teeter-totter or equilibrium hypothesis of IDDM expression. Depicted are the diabetes prone and the diabetes resistant BB rats, together with the perturbants known to alter the balance between autoreactive (A) and regulatory (R) cells. Also indicated are the genetic loci associated with the DP and DR strains.

of B-1 lymphocytes (Casali and Notkins, 1989). In the rat, however, the presence of a phenotypically distinct B-1 subpopulation is disputed (Vermeer et al., 1994). An alternative hypothesis is that the pattern of islet specific autoantibodies in the BB rat is due to a defect in antibody response to T cell-dependent but not to T-independent antigens. CELLULAR AUTOIMMUNITY IN THE BB RAT: A MATTER OF IMMUNOLOGICAL BALANCE We have previously conceptualized the etiology of autoimmunity in general, and IDDM in particular, using the analogy of a teeter-totter (Mordes et al., 1996). Our working hypothesis holds that the expression of diabetes depends on the relative balance between autoreactive (TA) cells and “regulatory” (TR) cells that should normally prevent beta cell destruction (Figure 1). The data supporting this concept come principally from studies of the BB/Wor rat and provide a framework for our review of the immunology of these animals. The top half of Figure 1 depicts the absence of self-tolerance, organ destruction, and the development of autoimmune diabetes mellitus. On the left is shown the spontaneously diabetic BB rat. The presence of autoreactive cell populations is inferred from the observation that spleen cells from such animals are capable of the adoptive transfer of the disorder (Koevary et al., 1983; Crisá et al., 1992). The absence of a regulatory cell population that might hold such effector populations in check in these lymphopenic animals

12

AUTOIMMUNE DIABETES MELLITUS IN THE BB RAT

was first inferred from experiments showing that lymphocyte transfusions prevent disease in lymphopenic, RT6-deficient DP rats if RT6+ donor T cells become engrafted (Rossini et al., 1986; Burstein et al., 1989). This is depicted schematically in the bottom left of the figure as the restoration of a protective equilibrium. The other line of evidence supporting this hypothesis come from diabetes resistant BB rats, depicted on the right of the figure. Although they never develop spontaneous disease, adoptive transfer experiments show that they nonetheless harbor autoreactive cells capable of recognizing and destroying normal beta cells (Greiner et al., 1987; McKeever et al., 1990). The hypothesis, depicted in the lower part of the figure, that they are protected from diabetes by the presence of a regulatory population, has been confirmed by demonstrating that >50% of young non-VAF DR-BB/Wor rats become diabetic if they are depleted in vivo of RT6+ T cells (Greiner et al., 1987). This effect is readily observed in conventionally housed DR rats that have been exposed to common rat viruses, whereas viral antibody free (VAF) DR rats require coadministration of an immune system activator such as polyinosinic-polycytidylic acid (poly I:C) (Thomas et al., 1991; Guberski et al., 1991). Cyclophosphamide and low dose irradiation can also precipitate diabetes in this animal (Crisá et al., 1992). CELLULAR ONTOGENY IN THE BB RAT The Lymphopenia of Spontaneously Diabetic BB Rats It has been recognized for some time that the peripheral lymphoid system in DPBB rats exhibits phenotypic and functional abnormalities (Crisá et al., 1992). The most striking is profound T cell lymphopenia characterized by severe reduction in the number of CD4+ T cells and nearly complete absence of the CD8+ T cell subset. As detailed below, the phenotypic T cell abnormalities in DP rats can be attributed to severely reduced life span of peripheral T cells. The first signs of lymphopenia can be detected in the thymus (Groen et al., 1996; Plamondon et al., 1990). The lymphopenia of BB rats is inherited as an autosomal recessive trait that is discussed in the section on Genetics, page 25. Thymocyte Development in the BB Rat The importance of intrathymic events in BB rat diabetes was recognized in very early thymectomy studies (Like et al., 1982). Subsequent experiments demonstrated that thymocytes from DR-BB/Wor rats can adoptively transfer diabetes to athymic recipients (Whalen et al., 1995). These observations demonstrate that the DR-BB rat thymus harbors abnormal cell populations predisposed to autoreactivity and localize the developmental defect leading to diabetes to an abnormal intrathymic selection process. Studies involving bone marrow and thymus transplantation document that prothymocytes in DP rats have an intrinsic defect leading to the lymphopenia, abnormal phenotypic distribution, and impaired proliferative capacity of T cells (Crisá et al., 1992). Other studies have demonstrated that bone marrow derived thymic antigen presenting cells (APCs) play an additional role in the development of the T lymphopenia (Georgiou et al., 1988; Georgiou and Bellgrau, 1989). Early studies on DP thymi had revealed no major differences between DP-BB, DR-BB, and normal rats with respect to absolute numbers of thymocytes and to the expression and distribution of T cell markers in the thymus. However, DP thymi show a 50% reduction in the numbers of mature thymocytes as compared to control rats in the TcRhiCD4−CD8+ population (Plamondon et al., 1990). This observation has been extended by showing that the TcRhi CD4+ CD8+ precursors of these thymocytes are reduced in absolute number (Groen et al, 1996). The density of cell surface CD8 is slightly decreased on DP rat thymocytes

JOHN P.MORDES ET AL.

13

(Groen et al, 1996). Speculatively, these observations may relate to altered cytokine expression in the thymi of DP-BB rats (Bieg et al., 1997). These findings could have implications for the efficiency of positive and negative selection and therefore for the generation of autoreactive T cells in DP rats (Groen et al., 1996; Bellgrau and Lagarde, 1990). Another abnormality observed in both DR-BB and DP-BB thymi is a marked reduction in numbers of medullary B lymphocytes (Tullin et al., 1997). Since B lymphocytes play an important role in thymic negative selection, this defect may contribute to the generation of autoreactivity in DP and DR rats (Tullin et al., 1997). A final defect may be a reduction of purine nucleoside phosphorylase (PNP) in the thymus of certain BB rats, but this observation has not been pursued (Wu and Marliss, 1991). BB Rat T Cell Maturation In the rat, T cell maturation can be traced by expression of CD90 (Thy-1), RT6, and CD45RC. CD90 is expressed on stem and progenitor bone marrow cells, thymocytes, and immature peripheral T and B cells (Thiele et al., 1987; Powrie and Mason, 1990; Ritter et al, 1978). CD90 expression is lost during the early stages of post-thymic T cell development (Kampinga et al, 1997; Hosseinzadeh and Goldschneider, 1993). RT6 is a post-thymic T cell alloantigen expressed by the majority of mature peripheral T cells (Bortell et al., 1999). CD45RC is expressed by B cells, ~75% of CD4+ T cells, and most CD8+ T cells (Powrie and Mason, 1990). Recent thymic emigrants (RTE) in the rat are characterized by expression of CD90 and the absence of RT6 and CD45RC (Kampinga et al., 1997; Hosseinzadeh and Goldschneider, 1993). RTEs become CD90−RT6+ CD45RC+ after 7–11 days in the periphery. In the DP-BB rat, the percentage of CD90+ T cells is significantly increased (Groen et al., 1995). In contrast to normal rat strains (Groen et al., 1993), <10% of DP-BB rat peripheral T cells express detectable levels of RT6 (Greiner et al., 1986) or CD45RC (Groen et al., 1989). These phenotypic data suggest that the peripheral T cell pool in DP rats is immature. The basis for the near total defect in peripheral RT6 expression is not known. It is not due to a defect in the gene encoding RT6 (Crisá et al., 1993), and there is abundant RT6 expression by DP rat intraepithelial lymphocytes (IELs) (Waite et al., 1996; Fangmann et al., 1991). It could be due to mRNA instability, altered gene regulation, or reduced life span (Sarkar et al., 1992). RTEs and T Cell Life Span in BB Rats Although the relative percentage of RTEs in DP rats is increased, the absolute number of RTEs is decreased (Groen et al., 1995), due at least in part to reduced thymic output (Zadeh et al., 1996). The number of CD8+ RTEs is reduced to a greater extent than the number of CD4+ RTEs (Groen et al., 1996). In addition to impaired emigration, peripheral T cell number is further compromised during the transition from the immature CD90 + to the mature CD90− phenotype (Groen et al., 1996; Zadeh et al., 1996), particularly in the CD8+ compartment (Groen et al., 1996). The observed loss of cells from blood and lymphoid organs is interpreted as evidence of early cell death in DP rats. It is now known that the majority of cultured DP-BB T cells undergo apoptosis within 24 hours (Iwakoshi et al., 1998; Ramanathan et al., 1998). In addition, DPBB rats harbor a population of large peripheral T cells that appear to be activated. Some exhibit a pre-apoptotic α TcRlowB220+CD4−CD8− phenotype (Iwakoshi et al., 1998) and undergo apoptosis in the liver. Apoptotic cells comprise ~80% of intrahepatic T cells in DP rats, but only ~20% in DR-BB/Wor rats. Many of the intrahepatic apoptotic T cells are RTEs (Iwakoshi et al., 1998). The cause of increased apoptosis in DP-BB/Wor rat RTEs is

14

AUTOIMMUNE DIABETES MELLITUS IN THE BB RAT

unknown, but antigen-activation can save some DP rat T cells from early cell death (Ramanathan et al., 1998). It is possible that autoantigen-activation may in part account for the presence of autoreactive T cells in BB rats (Bellgrau et al., 1994). Metabolic Abnormalities in Lymphocytes and Macrophages Energy metabolism in splenocytes from diabetes prone BB rats is reportedly abnormal. The production of ammonia, glutamate, aspartate, and CO2 from glutamine is increased, observations that have been speculatively attributed to the activation of splenocyte subsets (Wu et al., 1991). A general state of marked metabolic activation and an abnormally low cytosolic redox state in response to mitogen have also been reported (Field et al., 1990). Analyses of peritoneal macrophages from spontaneously diabetic BB rats have identified enhanced glucose metabolism and respiratory bursts (Wu and Marliss, 1993). Increased macrophage glucose metabolism was observed in diabetes-prone BBdp rats that were 75–80 but not 50 days of age (Wu et al., 1991). These observations have been interpreted as indicative of macrophage activation, but it is not known if any of these lymphocyte or macrophage metabolic parameters play any role in the abnormal maturation and function of DP lymphocytes. NK Cells, NK T Cells, and Intraepithelial Lymphocytes NK Cells DP-BB rat CD8+CD5− NK cells are relatively increased in number (Woda et al., 1986). In contrast, DR-BB rats are relatively deficient in NK cells when compared with non-BB strain rats (Woda et al, 1986). A role for CD8+CD5− cells in IDDM expression in DP-BB rats was suggested by in vivo antibody depletion experiments. Injection of anti-CD8 monoclonal antibody (mAb) resulted in a reduction of CD8+CD5− cells from ~5% to <0.5%, and reduced diabetes frequency from 61% to 12% (Like et al., 1986). Because DP rats have < 1.0% CD8+ T cells, the data were interpreted to suggest that NK cells are important mediators of autoimmunity in BB rats. However, more recent studies using an NK cell-specific mAb against the NKR-P1 receptor (clone 3.2.3) failed to reduce the frequency of diabetes in DP-BB rats (Ellerman et al., 1993). In these experiments, the antibody was shown to be active in vivo as evidenced by its ability to clear the circulation of phenotypically identifiable NK cells, to deplete in vitro splenic NK cell activity, and to reduce the intra-islet accumulation of 3.2.3+ cells (Ellerman et al., 1993). NK T Cells The early studies of NK cells in the BB rat were undertaken before the distinction between NK and NK T cells was fully appreciated. They suggest that NK cells may not be critical in regulating IDDM in the DPBB rat, but do not exclude a role for NK T cells. In the rat, NK T cells are NKR-P1+ αβTcR+, and CD8+ (Brissette-Storkus et al., 1994; Badovinac et al., 1998). Cells with this phenotype appear to be deficient in the DP-BB/Wor rat (DLG, JPM unpublished observation). The potential importance of a defect in NK T cell number in the pathogenesis of BB rat diabetes remains unknown.

JOHN P.MORDES ET AL.

15

Intraepithelial Lymphocytes As in mice and humans, there is a large population of TcR+ and CD8+ IELs in the rat (Todd et al., 1999). Rat IELs do not express CD2, but do express very high levels of RT6 (Todd et al., 1999). DP-BB rats, although deficient in peripheral RT6+ T cells, have detectable RT6+ IELs; the level of RT6 on DP-BB rat IELs is lower than on normal rat IELs (Waite et al, 1996). GENERATION OF AUTOREACTIVE CELLS IN THE BB RAT Origin of Autoreactive Cells The section on Cellular Autoimmunity (page 11) has reviewed the evidence for the existence of autoreactive cells in both DP- and DR-BB rats. The cellular defect responsible for their generation is still not understood. The data reviewed above suggest that they may be among the activated short-lived populations of RTEs in the DP-BB rat. We and others have demonstrated by adoptive transfer that the predisposition to IDDM in the BB rat resides in bone marrow cells (Crisá et al., 1992). Reciprocal bone marrow and thymus transplants have been used to demonstrate the presence of a thymic microenvironmental defect of bone marrow origin (Georgiou and Bellgrau, 1989). The data suggested that intrathymic antigen presenting cells are abnormal in BB rats. The abnormality appears to generate a limited peripheral TcR chain variable region repertoire associated with diabetes expression (Gold and Bellgrau, 1991). As noted above, there are epithelial defects in the thymus of both DP-BB and DR-BB rats (Doukas et al., 1994), and intrathymic tolerance induction defects in DR-BB rats (Battan et al., 1994). These data suggest that the thymus of BB rats may generate large numbers of autoreactive thymocytes. This suggestion is supported by early thymectomy studies (Crisá et al., 1992) and by studies showing induction of IDDM in athymic histocompatible WAG-rnu/ rnu rats after injection of DR-BB thymocytes (Whalen et al., 1995). Consistent with the equilibrium hypothesis (Figure 1), IDDM in the latter study was inducible only if development of regulatory RT6+ T cells in adoptive recipients was prevented. Phenotype The phenotype of the autoreactive cell population has been partially characterized. Depletion studies using antibodies directed against CD8 and NKR-P1 have implicated a CD8+ NKR-P1−α/βTcR+ T cell population in DP-BB/Wor rats (Ellerman et al., 1993). Additional data suggest that this population may in part be comprised of long-lived CD8low T cells (Bellgrau and Lagarde, 1990). In the DR-BB rat, depletion analysis has implicated a CD8+RT6− cell (Woda et al., 1991). Adoptive transfer methodologies have suggested that, as in the NOD mouse (Christianson et al., 1993), CD4+RT6− and CD8+RT6− cells acting synergistically are required for efficient disease induction (Whalen et al., 1994). These observations raise the possibility that different cell types are involved in the initiation of autoreactivity vs. the actual implementation and amplification of beta cell killing by recruited cytotoxic cell populations. Finally, it has been observed that CD4+ CD45RChigh (clone OX22hlgh) T cells are capable of initiating a wasting graft vs. host-like disease (Powrie and Mason, 1990), and preliminary data from our laboratory suggest that CD4+ CD45RChigh T cells from BB rats adoptively transfer disease (Whalen, B, unpublished observations).

16

AUTOIMMUNE DIABETES MELLITUS IN THE BB RAT

Identity and Specificity A diabetes-inducing class II restricted T cell line has been reported (Ellerman and Like, 1995), but refinement of the autoreactive phenotype has been impaired by the absence of any data generated by using a depleting anti-CD4 monoclonal antibody, the absence of rat knockouts, and the absence of cell clones. For these same reasons, in part, the identity of the actual target of the disease-initiating autoreactive T cell population remains unknown. Many of the candidate human and NOD mouse islet autoantigens including GAD and insulin have been assessed for their ability to prevent disease (see below), but the data fail to implicate any of these proteins. Other incompletely characterized antigens may play a role in the disease (Karges et al., 1996; Ko et al., 1991; Jun and Yoon, 1994). The observation that islets prepared from normal rats and transplanted into DP animals can be the target of autoimmune attack suggests that BB rat beta cells are not antigenically unique (Gottlieb et al, 1990). Cytokines The cytokines present in the insulitis lesion have been identified as Th1-type predominant (See page 7). The cytokine profile of the initiator of autoreactivity is uncertain. It is known, however, that activated CD8+ cells produce IFN-γ, and anti-IFN-γ antibody can prevent IDDM in the DP-BB rat (Nicoletti et al., 1997). Paradoxically, however, administration of recombinant IFN- also prevents the disease (Nicoletti et al, 1998). Mechanism of Beta Cell Destruction Beta cell killing in IDDM may occur either as the result of direct interaction with the autoreactive cell population or indirectly via inflammatory mediators released by lymphocytes or macrophages. In vitro studies have failed to demonstrate conclusively direct cytotoxicity by BB rat T cells. Although BB rat lymphoid cells are cytotoxic to islets in vitro, the killing appears to be nonspecific, due to NK cells and not T cells (Kitagawa et al, 1991). In contrast, there is abundant evidence that several non-specific inflammatory mediators may participate in beta cell destruction in BB rats. These include the cytokines IL-1, TNF-α, TNF-β, IFN-γ, and oxygen free radicals, including O2−, H2O2, and NO. Their role in islet destruction has recently been reviewed (Rabinovitch and Suarez-Pinzon, 1998). Whether the BB rat islet or pancreas is deficient in enzyme systems like superoxide dismutase and glutathione peroxidase that are protective against oxidative stress is unclear (Bellmann et al, 1997). Mechanism by Which Autoreactive Cells are Generated Little information is available concerning the mechanism by which autoreactive cells escape intrathymic selection processes in the BB rat. As noted above, cell surface expression of CD8 on DP-BB rat T cells is low and may interfere with normal selection (Bellgrau and Lagarde, 1990). It is also known that CD4+ BB rat T cells are resistant to activation induced cell death in response to superantigens suggesting another mechanism by which autoreactive BB rat T cells could inappropriately survive (Sellins et al., 1996). In addition, the survival of both CD4+ and CD8+ T cells in DP-BB rats has been shown to be dependent on antigen activation, suggesting a mechanism by which autoantigen activated cells survive to populate the periphery (Ramanathan et al, 1998). Very recently it has been observed that organ culture of adult DR- and DP-BB/Wor rat thymi, respectively, recapitulates the normal and abnormal T cell development seen in vivo

JOHN P.MORDES ET AL.

17

Figure 2 Schematic diagram of the thymic developmental abnormalities in the BB rat that lead to an imbalance between regulatory and effector cell populations.

in these animals (Whalen et al., 1999). This technique may prove useful for the eventual identification of the process by which autoreactive thymocytes are generated. Figure 2 summarizes schematically the elements implicated in the development of autoreactivity in the BB rat. REGULATORY CELLS IN THE BB RAT The concept of peripheral regulatory or “suppressor” cells has been controversial, but it has gained a significant measure of validation through analyses of the BB rat. Those analyses have demonstrated clearly that a subset of rat T cells expressing the RT6 surface marker has the ability to regulate the expression of autoimmunity. The regulatory function of RT6+ T cells in autoimmunity was first demonstrated in studies of the DP-BB/ Wor rat (Crisá et al., 1992). It was observed that they could be protected from spontaneous IDDM by transfusions of whole blood. It was later recognized that the lymphopenia of DP-BB rats is in large measure due to severe deficiency of peripheral RT6+ T cells. Prevention of IDDM by transfusion appears to be

18

AUTOIMMUNE DIABETES MELLITUS IN THE BB RAT

mediated by a population of RT6+CD4+ T cells. A single transfusion of as few as 50×106 lymphocytes can prevent IDDM if RT6+ T cells become engrafted. The role of the RT6+ T cell subset in the pathogenesis of rat autoimmunity was confirmed in analyses of the DR-BB/Wor rat (Crisá et al., 1992). Treatment of DR-BB rats with a cytotoxic anti-RT6.1 mAb leads to autoimmune diabetes and thyroiditis within two to four weeks (Mordes et al., 1996). This result has been interpreted to suggest that the DR-BB rat shares the genetic predisposition of the DP animal to autoreactivity, but is kept free of disease by its population of RT6+ “suppressor T cells.” The hypothesis that an RT6+ T cell subset acts as suppressor cells received additional support from adoptive transfer studies. Injection of T lymphocytes from diabetic RT6-depleted DR-BB rats into histocompatible athymic rnu/rnu recipients rapidly induced diabetes that was preventable by coadministration of RT6+ T cells (Whalen et al., 1994). The phenotype of the protective cell subset has been identified as RT6+CD4+CD45low (unpublished observations). The RT6+CD45RClow subset is virtually absent in DP rats (Groen et al., 1997). Recently, it has been documented that, like the autoreactive cells, the regulatory T cells in BB rats originate in the thymus (Whalen et al., 1995). Transfer of DR-BB thymocytes to histocompatible rnulrnu recipients results in the generation of both RT6− and RT6+ T cell populations. These adoptive recipients do not develop IDDM. If, however, recipients are treated with an anti-RT6.1 mAb to deplete the regulatory RT6.1+ T cell subset emerging from the thymocyte inoculum, the recipients eventually develop both IDDM and thyroiditis (Whalen et al., 1995). The result is consistent with the hypothesis that the expression of autoimmunity in rats is a function of RT6− effector cells acting in the absence of RT6+ regulatory cells. The regulatory cell phenotype identified in the BB rat is similar to that of a regulatory population in a PVG rat model of IDDM induced by thymectomy and irradiation. CD4+ RT6+ CD45RClow/− T cells have been shown to prevent IDDM in PVG.RT1u rats (Fowell and Mason, 1993). The origin of the regulatory cell population in this model has been traced back to the thymus (Seddon et al., 1996; Saoudi et al., 1996; Seddon and Mason, 1997). These cells are associated with high levels of mRNA encoding IL-4 (Fowell and Mason, 1993). Figure 2 summarizes schematically the elements implicated in the development of regulatory cell activity in the BB rat. RT6 AS A POTENTIAL MEDIATOR OF REGULATION The studies summarized above have established that the RT6 alloantigen serves as a marker for regulatory cells that can prevent IDDM in the DP-BB rat. Conversely, the RT6− cell subset in the BB rat contains effector populations that can adoptively transfer diabetes to histocompatible athymic recipients. Expression of the RT6 alloantigen appears uniquely to distinguish effector from regulatory cell populations. No comparable marker is presently available in any other species (Greiner et al., 1997; Bortell et al., 1999). Because of this singular distinction, the expression of RT6 has been studied extensively in the BB rat model of IDDM. These data suggest that RT6 protein itself may play a role in the regulatory activity. Nomenclature and Genetics of Rat RT6 Two alleles of the rat RT6 gene located on chromosome 1, RT6a and RT6b, have been reported (Crisá et al., 1992). RT6a encodes the RT6.1 protein and RT6b encodes the RT6.2 protein. Each allele of RT6 segregates as an autosomal dominant with variable penetrance. The majority of lymphoid cells in the offspring of a cross between homozygous RT6a and RT6b parents co-express both RT6.1 and RT6.2 on their surface, but a subset of RT6.2+ RT6.1− cells can be detected (Thiele et al., 1993).

JOHN P.MORDES ET AL.

19

Recognition that RT6 proteins are ADP-ribosyltransferases (ARTs) has established RT6 as a member of the “ART” family (Koch-Nolte et al., 1996). As a result, a new nomenclature has been proposed (Haag and Koch-Nolte, 1997). For the rat, ART2a and ART2b will replace RT6a and RT6b, respectively. Most published literature still refers to the protein products of the RT6a (ART2a) and the RT6b (ART2b) alleles as RT6.1 and RT6.2, respectively. RT6 and Immunomodulation How RT6+ T cells regulate immune function is not known but is the subject of growing interest (Bortell et al., 1999). Cell biological studies of RT6 together with newer biochemical and molecular analyses suggest at least four potential mechanisms by which cells expressing RT6 could modulate immune reactivity. The simplest hypothesis is that RT6 merely marks T cells that regulate immune system activity indirectly. This hypothesis does not invoke mechanisms that depend on RT6 protein per se. The mechanism could involve RT6+ T cell-mediated deviation of the cytokine profile of effector RT6− T cells, resulting in a Th2-type (protective) rather than a Th1-type (destructive) cellular response (Racke et al., 1994). Second, cell signaling mediated by RT6 could modulate the immune system. RT6 could modify signals received by the cell through the TcR or other receptors and influence the type and quantity of cytokine receptors expressed. Alternatively, RT6-mediated signals could induce cell differentiation. The observations that RT6 co-immunoprecipitates with members of the src tyrosine kinase family and that cross-linking cell surface RT6 induces the expression of IL-2 and/or IL-4 receptors support these possibilities (Rigby et al., 1996). Third, the NAD+ glycohydrolase activity of RT6 may influence the function of activated T cells. Recombinant RT6.2 protein synthesized in mammalian cells demonstrated NAD+ glycohydrolase activity, cleaving NAD+ and releasing free ADP-ribose and nicotinamide (Takada et al., 1994). Supporting the hypothesis that RT6 enzymatic activity could subserve an immunoregulatory function, it has been observed that addition of the enzyme’s substrate, NAD+, to stimulated rat RT6+ T lymphocyte cultures inhibits cell proliferation (Rigby et al., 1996). Fourth, the ADP-ribosyltransferase activities of RT6 may be important. Rat RT6 undergoes auto-ADPribosylation that could modulate the signal transduction properties or other activities of the protein itself (Haag et al., 1995); Maehama et al., 1995; Rigby et al, 1996). Finally, it has been observed that soluble RT6 circulates in readily detectable amounts in both BB and non-BB rats (Waite et al, 1993). The diabetes prone BB rat circulates less RT6.1 than does any other strain, including the diabetes resistant BB/Wor line, and injections of depleting anti-RT6.1 antibody rapidly eliminate soluble RT6 from the circulation of diabetes resistant BB rats. The existence of soluble RT6 suggests the possibility that circulating RT6 protein might possess immunomodulatory properties. MODULATION OF DIABETES EXPRESSION IN THE BB RAT The frequency and age at onset of diabetes in the spontaneously diabetic BB rat can be modified by both immunomodulatory therapeutics and by environmental perturbation. Many of these interventions have parallels in studies of the NOD mouse (Atkinson and Leiter, 1999), but many of the more than 125 preventive agents that are effective in the latter are ineffective in the rat. Both immunological therapeutics and environmental agents can induce autoimmunity in diabetes resistant BB rats.

20

AUTOIMMUNE DIABETES MELLITUS IN THE BB RAT

Prevention of Spontaneous Diabetes Prevention of IDDM in the BB rat has been targeted at all stages of the disease process. These include the generation of the autoreactive cell, the immunodeficiency associated with absence of RT6+ cells, homing of the autoreactive cell to the pancreas, and the terminal cytotoxic events that lead to beta cell destruction. Immunosuppression Diabetes in the BB rat, being a T cell dependent autoimmune disorder, is preventable by a variety of immunosuppressive agents (Crisá et al., 1992). These include anti-lymphocyte serum, radiation, cyclosporin, fusidic acid (Nicoletti et al, 1996), mycophenolate mofetil (Hao et al., 1993), tacrolimus (formerly FK-506) (Lieberman et al., 1993), and the adenosine deaminase inhibitor, 2'-deoxycoformycin (dCF) (Thliveris et al, 1997). Other reagents targeted at macrophages including silica (Oschilewski et al, 1985) and deoxyspergualin (Di Marco et al, 1996) are also effective. Antibodies directed against various T cell subsets are also effective. These include anti-CD8 and antiNKR-P1(Ellerman et al., 1993) and also anti-CD2, which may act by depleting CD4+ T cells, preventing the activation of effector cells, or by blocking CD2/ligand interaction between effector and target cells (Barlow and Like, 1992). Immunomodulation by Cytokines, Cells, Antigens, and other Factors Cytokines

IDDM in BB rats can be prevented or retarded by TNF- (Satoh et al., 1990), lymphotoxin (Takahashi et al., 1993), and IFN- (Nicoletti et al., 1998; Sobel and Newsome, 1997), but paradoxically anti-IFN- is also effective (Nicoletti et al., 1997). Treatment with IL-2 is not effective (Burstein et al., 1987). Islet expression of IFN-α reportedly precedes the onset of BB rat diabetes (Huang et al., 1994), but, again paradoxically, exogenous IFN-α reduces the frequency of the disease (Sobel et al, 1998). Some of these inchoate observations do give some additional insight into pathogenesis of BB rat diabetes. TNF-α production, for example, has been associated with defective CD45RC expression in DP-BB rats (Rothe et al., 1992), involves defective PGE2 feedback inhibition (Rothe et al, 1994), and is corrected by administration of complete Freund’s adjuvant (CFA). CFA is, in turn, capable of preventing IDDM in the DP-BB rat (Rabinovitch et al., 1996). Cells

Cell-based strategies to prevent diabetes in the BB rat have focused either on the restoration of defective peripheral regulation using either bone marrow cells or transfusions of MHC compatible spleen cells (See page 11) or cell lines (Nagata and Yoon, 1994). Intrathymic islets prevent diabetes in the DP-BB rat, presumably by affecting intrathymic selection processes (Posselt et al, 1992; Leiter and Gill, 1994). Antigens

Various strategies to prevent diabetes in humans based on parenteral or oral tolerization with antigen have been proposed. Some of these have been tested in the BB rat, largely with negative results. GAD and bovine serum albumin are candidate autoantigens in humans and effective in the prevention of NOD mouse diabetes (Atkinson and Leiter, 1999), but are ineffective in the BB rat (Petersen et al, 1997). Parenteral insulin treatment prevents BB rat diabetes (Crisá et al, 1992). In the original report (Gotfredsen et al, 1985),

JOHN P.MORDES ET AL.

21

it was noted that, despite the discontinuation of insulin administration at 140 days of age, most treated DP rats remained normoglycemic until 230 days of age. The protective effect is probably not due simply to hypoglycemia since treatment with the insulin secretion inhibitor diazoxide also decreases the incidence of overt diabetes by about 50% (Vlahos, Seemayer and Yale, 1991). Immune deviation has been proposed as an explanatory mechanism (Kolb et al, 1997), but because large doses of insulin are needed, it remains to be documented that parenteral insulin has actually tolerized treated animals rather than inducing hypoglycemia and a state of “hypo-antigenic beta cell rest.” Orally administered insulin with or without oral IFN- as adjuvant, a strategy effective in preventing NOD mouse diabetes (Atkinson and Leiter, 1999), is ineffective in the BB rat (Mordes et al, 1996) and may be deleterious (Bellmann et al, 1998). This result highlights important differences between BB rats and NOD mice with respect to immunological responses that are of potential relevance to human IDDM and IDDM prevention trials. Lastly, the BB rat has been used to investigate the question as to whether vaccination of humans may contribute to expression of IDDM in susceptible individuals, but these studies have not yet yielded conclusive information. Administration of human vaccines to DP-BB rats less than 2 weeks of age prevented diabetes, whereas administration of pertussis vaccine starting at 8 weeks of life was associated with increased diabetes (Classen, 1996). Other Interventions

Recombinant platelet-activating factor inhibitor (Lee et al., 1999; Jobe et al., 1993), sulfonylurea-type oral hypoglycemic agents (Cheta et al., 1995; Hosszufalusi et al., 1994), and low dose poly I: C (Sobel et al., 1998) prevent BB rat diabetes by unknown mechanisms. Oral administration of N-omega-nitro-Larginine methyl ester (an inhibitor of NO synthase) from 30 to 150 days of age reduces the incidence of IDDM in diabetes-prone BB/Ed rats (Lindsay et al., 1995). Hydroxyethyl starch deferoxamine, an iron chelator, delays diabetes in BB rats (Roza et al, 1994), as do orally administered free radical scavengers (Murthy et al., 1992), possibly by interference with hydroxyl radicals involved in the final killing of beta cells. The anti-inflammatory drug tetrandrine is reported to prevent diabetes in BB rats alone (Lieberman et al., 1992) and in combination with tacrolimus (Lieberman et al., 1993). Most reports suggest that nicotinamide does not prevent BB rat diabetes, but oral administration may be partially efficacious (Crisá et al., 1992). Retinol deficiency reduces diabetes and insulitis in DP-BB/Wor rats by an unknown mechanism (Driscoll et al., 1996). Environmental Manipulation: Diet and Infection Diet

Dietary modification, in particular with respect to the source of protein, can reduce the frequency of IDDM in the BB rat (Scott et al, 1997). Diabetes can be substantially reduced by the substitution of l-amino acids, soy protein, or hydrolyzed casein for intact protein in the diet (Li et al., 1995). In the case of the lamino acid based diet, the protective effect was abrogated by milk protein supplementation (Elliott and Martin, 1984). This last observation is of particular interest given the hypothesis that exposure to cow’s milk proteins triggers IDDM in genetically susceptible humans (Dosch, 1993). Additional studies have indicated, however, that although dietary modification can reduce spontaneous IDDM expression in DP-BB rats, the agent of protection is not elimination of cow’s milk protein (Malkani et al., 1997).

22

AUTOIMMUNE DIABETES MELLITUS IN THE BB RAT

Efforts have been made to determine the mechanisms by which the source of dietary protein affects diabetes incidence, but the issue remains unresolved (Scott et al., 1997). In a more controlled framework, it has been shown that a diet lacking in essential fatty acids, which is known to inhibit macrophage migration and function, can prevent spontaneous diabetes in BB rats (Lefkowith et al., 1990). The inhibitor of glutamine metabolism, acividin, also prevents BB rat diabetes, possibly by targeting the metabolic machinery of activated, glutamine-dependent cells (Misra et al., 1996). Infection

Infection is not a prerequisite for spontaneous diabetes in BB rats; gnotobiotic animals develop the disease (Crisá et al., 1992). Infection with lymphocytic choriomeningitis virus (LCMV) prevents the disease (Shyp et al., 1990). Kilham rat virus, which induces IDDM in the DR-BB rat (see below) does not accelerate the disease in the DP-BB rat (Guberski et al., 1991). Induction of IDDM in Resistant Animals As noted previously, conventionally housed DR-BB/Wor rats become diabetic if they are depleted in vivo of RT6+ regulatory T cells (Greiner et al., 1987). Viral antibody free (VAF) DR rats require coadministration of an immune system activator such as polyinosinic-polycytidylic acid (poly I: C) (Thomas et al., 1991; Guberski et al., 1991). Cyclophosphamide and low dose irradiation can also precipitate diabetes in this animal (Crisá et al., 1992). Perhaps the most important perturbant that induces diabetes in the resistant rat, because of its relevance to human IDDM, is infection. Infection with Kilham rat virus (KRV, a parvovirus) induces diabetes in ~30% of DR-BB rats (Chung et al., 1997; Guberski et al., 1991) without infecting beta cells (Brown et al., 1993) or causing lymphopenia (Guberski et al., 1991; Gaertner et al., 1993). Infection in conjunction with either poly I: C or depletion of RT6+ regulatory cells induces disease in essentially all DR rats (Ellerman et al., 1996). With one exception (a congenic RT1U/U LEW.WRl: rat) (Ellerman et al., 1996), KRV, despite its ubiquity and documented ability to infect all tested rat strains, is not known to induce autoimmunity in any strain of rat other than the BB (Jacoby and Ball-Goodrich, 1995). The effect of infection appears to be specific, as infection with sialodacryoadenitis virus (Thomas et al, 1991) and rat cytomegalovirus (JPM, unpublished observations) does not induce IDDM. The observations made with KRV cast light on the role of poly I: C in the inductive method that combines it with anti-RT6 mAb. Poly I: C is a synthetic double-stranded polyribonucleotide. Presumably due to structural resemblance to double-stranded viral RNA, poly I: C elicits immune responses that mimic viral infection. These include stimulation of type I interferon production by various cells, IL-1 production by monocytes, and activation of NK cells, B cells, and endothelial cells (Doukas et al., 1994). Treatment of DR-BB rats with low doses of poly I: C (5 µg/g) induces diabetes in ~20% of animals (Thomas et al., 1991), a rate similar to that observed when DR-BB rats are infected with KRV (Guberski et al., 1991). Higher doses of poly I: C (10 µg/g) induce diabetes in nearly all DR rats (Sobel et al., 1992). Parenthetically, high dose poly I: C accelerates diabetes in the DP-BB rat (Ewel et al., 1992; Sobel et al., 1995), an effect associated with increased levels of IFN-α (Sobel et al., 1994), whereas at lower doses it is protective (Sobel et al., 1998).

JOHN P.MORDES ET AL.

23

Animal Models of Induced Type 1 Diabetes It can be argued that analyses of induced IDDM in diabetes resistant BB rats are of limited relevance because of their dependence on perturbation with viral infection, poly I: C, or depletion of RT6+ regulatory cells. Perturbants, however, do not generate autoreactivity. They only facilitate its expression and detection. Poly I: C does not induce diabetes in animals, e.g. WF rats, that do not have the genetic predisposition of the BB rat (see below) (Ellerman and Like, 1998). In addition, it is clear that human IDDM reflects the interaction of a genetic predisposition with environmental determinants (Yoon, 1990). Arguably the diabetes resistant BB rat provides a faithful model of “real-world” human IDDM, in particular with respect to its response to KRV infection. Alternative rodent models of induced type 1 IDDM are based on beta cell toxins like alloxan and streptozotocin (STZ). For example, mice given multiple small subdiabetogenic doses of STZ develop pancreatic insulitis, selective beta cell destruction, and diabetes after a delay of several days. Many studies suggest that the immune system is involved in this form of diabetes (Herold et al., 1997; Kolb and Kröncke, 1993). In contrast to the induced diabetes of the diabetes resistant BB rat, however, study of the pathogenesis of low-dose STZ diabetes suggests strongly that the immunological process is dependent on chemical alteration of the beta cell. For example, adoptive transfer of low-dose STZ diabetes requires pretreatment of recipients with “priming” doses of STZ (Kolb and Kröncke, 1993). In addition, scid/ scid mice that lack both T and B cells are susceptible to low-dose STZ diabetes induction (Gerling et al., 1994), emphasizing that on at least some genetic backgrounds the combination of reagent and protocol that was employed is inherently beta cell cytotoxic. It is the requirement for beta cell injury that most clearly distinguishes the toxinbased models of type 1 diabetes from the diabetes resistant BB rat. The susceptibility of diabetes resistant BB rats to environmental agents like streptozotocin and alloxan has yet to be investigated in detail. Prevention of Induced Diabetes in the DR-BB Rat A diet low in essential fatty acids, which interferes with macrophage function (Lefkowith et al., 1990) and liposomes that target macrophages (Chung et al., 1997) both prevent the induction of diabetes in the DR-BB rat. Casein-free diets are ineffective (Malkani et al., 1997). Insulin treatment prevents clinical diabetes in the RT6-depleted diabetes resistant BB rat, but the treatment appears to depend on beta cell rest, not tolerance induction, and does not prevent the development of autoreactive cell populations that cause thyroiditis and adoptively transfer diabetes (Gottlieb et al., 1991). GENETIC BASIS OF IDDM SUSCEPTIBILITY IN RATS The inheritance of autoimmune diabetes in the BB rat is polygenic. Like IDDM in humans and NOD mice, one susceptibility locus maps to the MHC, but specific gene defects remain to be identified. Recent data suggest, however, that inheritance of IDDM in the BB rat might involve fewer major genes than in humans or NOD mice, and that the availability of resistant strains may offer strategic analytical advantages. iddm2 Inheritance of diabetes in all BB rats involves at least one gene, designated iddm2, associated with the MHC chromosome 20 (Crisá et al., 1992). This region is comprised of the RT1B and RT1D loci, homologous to mouse I-A and I-E; these are flanked by the RT1A, RT1E, and RT1C class I loci. BB rats expresses RT1Au

24

AUTOIMMUNE DIABETES MELLITUS IN THE BB RAT

RT1Bu RT1Du. IDDM expression is independent of class I haplotype, but requires at least one class II RT1u allele. The RT1A locus is not involved in diabetes expression. The rat RT1D locus is homologous to human HLA-DR. Intercross studies indicate that the u allele of the BB rat is not a unique diabetogenic variant; u alleles derived from normal rat strains also confer susceptibility. Sequence analysis of RT1B and RT1D β chains in BB rats shows that they do differ from those of LEW and SD rats and from NOD mouse class II genes (Chao et al., 1989). When the BB rat RT1B sequences were compared with 1) the RT1B allele of SD rats and 2) the I-A allele of the NOD mouse, the NOD displayed a higher degree of homology with the BB rat sequences. Additional studies of the MHC showed that not all u/u lymphopenic animals develop diabetes, and that the presence of only one u allele greatly reduces disease expression. In one analysis, (DP×LEW) F1×DP backcross rats yielded 75% expression of IDDM in lymphopenic u/u homozygotes, and only 7% expression of IDDM in lymphopenic u/l heterozygotes (Jacob et al., 1992). The region around position 57 of the class II β chains of non-diabetes prone LEW and BUF rats has been found to be identical to that found in both DP and DR-BB rats (Crisá et al., 1992). The enrichment for DQβ alleles with uncharged amino acids at position 57 found in diabetic Caucasian humans and NOD mice is not found in the BB rat (Chao et al, 1989). However, these specific sequence differences do not exclude the possibility that differential efficiency in peptide presentation, as proposed by Nepom, et al. (Nepom and Kwok, 1998), underlies the role of the RT1u/u class II MHC in the rat. iddm1 An autosomal recessive locus (termed iddm1) determining T cell lymphopenia (lyp) is strongly associated with the development of spontaneous autoimmune diabetes in all diabetes prone BB rats (Guberski et al., 1989; Markholst et al., 1991). This locus has been mapped to chromosome 4 (Jacob et al., 1992). The mechanism of T cell lymphopenia in the DP-BB rat is not yet known. Lymphopenia and diabetes susceptibility can be independently inherited traits (Like et al., 1986; Colle, 1990), but deficiency in peripheral T cells is tightly linked with the spontaneous disease expression (Fuks et al, 1988; Guberski et al, 1989; Markholst et al, 1991). In studies of crosses between DP-BB and DR-BB rats, diabetes segregated as a single, autosomal recessive trait and was always accompanied by lymphopenia (Markholst et al., 1991). The lyp locus has also been associated indirectly with excessive production of NO by BB rat macrophages (Lau et al., 1998). Recent studies confirm the association of diabetes with lymphopenia, but not all lymphopenic animals develop diabetes (Jacob et al., 1992). Because spontaneous diabetes in the BB rat appears to depend on the presence of lyp/iddm1, it has been hypothesized that lyp itself might explain completely the genetic predisposition to diabetes (Markholst et al., 1991; Jacob et al., 1992). Available data suggest, however, that autoreactive T cells could be generated, but not detected, in the absence of lyp. As discussed below, there is evidence that lyp/iddm1 does not determine the latent predisposition of BB rats to autoimmunity (Martin et al, 1999). iddm3 In two separate reports, the presence of a resistance gene has been inferred from crosses between spontaneously diabetic BB and resistant, non-lymphopenic non-BB rats. In one study, the RT1u and lyp genes were placed on the ACI strain background (Colle et al, 1992). A total of 72 backcross animals were followed to 140 days of age in a clean facility. Among animals from the N4, N5 and N6 generations that were homozygous for RT1u and lyp, none developed diabetes. The second study reported results from F2

JOHN P.MORDES ET AL.

25

offspring of (DP-BB×F344)F1 hybrids. The authors found 6 F2 rats homozygous for both RT1u and lyp (Jacob et al, 1992). Because none of these became diabetic, they proposed that there was a third, diabetesmodifying locus that conferred resistance. They designated this locus iddm3. Proof of its existence was sought by Klöting et al who discovered a candidate location on chromosome 18 in the BB/OK rat (Klöting et al., 1995), but this has not been confirmed (Jackerott et al, 1997). iddm4 and iddm5 We have recently reported two new IDDM susceptibility loci in the BB/Wor rat (Martin et al., 1999) in a backcross of the DP-BB/Wor rat to the histocompatible WF rat. Neither WF nor (WF×DP) F1 animals develop spontaneous IDDM. However, 95% of (WF×DP) F1 rats and a fraction of (WF×DP)×WF backcross animals readily developed IDDM after treatment with poly I: C and anti-RT6.1 monoclonal antibody (mAb). Loci were mapped on chromosomes 4 (iddm4) and 13 (iddm5) with significant linkage to IDDM and insulitis. The iddm4 susceptibility locus is linked to, but not identical to lyp. The data suggest that the diabetogenic allele of iddm4 also contributes to IDDM susceptibility in DR-BB rats. Of great interest, the iddm4 gene is located in a large region containing several major autoimmunity loci in the rat. One locus, cia3, contributes to susceptibility to autoimmune arthritis (Remmers et al., 1996). A closely linked and perhaps identical locus, Aia3, is associated with adjuvant-induced arthritis (Kawahito et al., 1998). Other nearby autoimmunity genes on chromosome 4 include Aia2 and a locus involved in susceptibility to experimental autoimmune uveitis (Sun et al, 1997). It is likely that other autoimmune diseases are affected by genes in this interval (Remmers et al, 1996). These findings raise the possibility that iddm4 is the same as one of these autoimmunity genes. Interestingly, the BBDR/Wor rat is susceptible to the induction of collagen arthritis (Watson et al, 1990). Genetic Hypothesis We hypothesize that expression of IDDM in BB rats involves several genetic elements. The first maps to the RT1u/u MHC, which is strongly associated with susceptibility to autoimmunity (Ellerman et al., 1996; Colle et al., 1986; Fowell and Mason, 1993; Ellerman and Like, 1998). The second element is iddm4, which, because it maps to the identical region on rat chromosome 4 in both DR- and DPBB rats, may be the same gene in both sublines. The iddm4 gene is associated with early events in pathogenesis, perhaps the initiation of insulitis (Martin et al., 1999). The third element is lyp/iddm1. We have proposed that the role of lyp/iddm1 in diabetes is to cause lymphopenia, which unmasks the genetic predisposition to autoreactivity in DP rats that is associated with iddm2 and iddm4. The lyp-dependent expression of spontaneous autoimmune IDDM in DP rats could be due to the absence of RT6+ regulatory cells from birth (Mordes et al., 1996). These relationships are summarized in Figure 1. THE BB RAT AS A MODEL OF HUMAN AUTOIMMUNE DIABETES We have often reiterated the view expressed by Renold and Shafrir (Renold et al., 1988) that animal models of disease must be interpreted with caution. It cannot be overemphasized that extrapolation from rodent to human can be misleading. The

26

AUTOIMMUNE DIABETES MELLITUS IN THE BB RAT

Table 3 Experimental in nmunotherapies for autoimmunity in BB rats

Selective immunosuppression with monoclonal antibodies Cytokine therapy Immunostimulation High dose Poly I: C Low dose Poly I: C CFA BCG Ciamexone Peptide immunization Miscellaneous Parenteral Nicotinamide Oral Nicotinamide Parenteral insulin Oral insulin Vitamins Androgen NO inhibitors

DP-BB Rat

Induced DR-BB rat diabetes

Prevention by antibodies against CD8, CD2, ASGM1, CD3 Prevention by TNF-α, lymphotoxin, INFγ, IFNα,

Prevention by anti-CD8

Accelerates onset Prevents diabetes Prevents disease Prevents diabetes Probably ineffective GAD, HSP65 ineffective

Induces diabetes Induces diabetes Unknown Unknown Unknown Unknown

Does not prevent diabetes May reduce frequency Prevents IDDM at doses that cause hypoglycemia Ineffective Unknown Does not prevent diabetes Delay onset

Unknown Unknown Same as DP

Unknown

Ineffective Unknown Unknown Unknown

Partial listing of immunotherapeutics assayed in the BB rat. Source citations can be found in the text. Poly I: C, polyinosinic polycytidylic acid; CFA, complete Freund’s adjuvant; HSP, heat shock protein; BCG, Bacille Calmette Guerin; IFN, interferon; ASGM1, asialoGM1; ALS, anti-lymphocyte serum; IL2r, interleukin 2 receptor; NO, nitric oxide. Many immunosuppressive drugs are effective in preventing DP-BB rat diabetes.

large number of effective preventives for NOD mouse diabetes (Mordes et al, 1996) that are inapplicable to human IDDM are a case in point. And it is important to recall that spontaneously diabetic BB rats are lymphopenic, whereas humans with IDDM are not. That being said, however, it is useful to recall that hallmarks of IDDM include insulitis, autoantibodies, recurrence of disease in transplanted islets, adoptive transfer by bone marrow allografts, prevention by cyclosporin, and MHC associations. The data reviewed here suggest that the hyperglycemic syndrome of the BB rat shares essentially all of these characteristics. In addition, like human IDDM, it is relatively difficult to prevent by non-immunosuppressive methods (Table 3). There is reason to hope that data derived from this animal will provide useful guidance in the development of human therapies. Another reason to focus interest on these animals is the availability of BB rat strains that are phenotypically normal, nonlymphopenic, free of spontaneous disease in VAF vivaria, but susceptible to IDDM in response to perturbation. In conclusion, we suggest that the predisposition to autoimmune disease may be common, but that its expression depends on an appropriate concatenation of MHC haplotype, susceptibility genes, and environmental perturbation that are well modeled in the BB rat.

JOHN P.MORDES ET AL.

27

REFERENCES Allen, E.M. and Thupari, J.N. (1996) The pathogenicity of spontaneously-occurring thyroglobulin-reactive T lymphocytes from BB/WOR rats. Autoimmunity, 23, 35–44. Atkinson, M. and Leiter, E.H. (1999) The NOD mouse model of insulin dependent diabetes: as good as it gets? Nature Med., In press. Awata, T., Guberski, D.L. and Like, A.A. (1995) Genetics of the BB rat: Association of autoimmune disorders (diabetes, insulitis and thyroiditis) with lymphopenia and major histocompatibility complex class II. Endocrinology, 136, 5731–5735. Badovinac, V., Boggiano, C., Trajkovic, V., Frey, A.B., Vujanovic, N.L., Gold, D.P., et al. (1998) Rat NKR-P1+ CD3+ T cells: selective proliferation in interleukin-2, diverse T-cell-receptor-Vb repertoire and polarized interferongamma expression. Immunology, 95, 117– 125. Barlow, A.K. and Like, A.A. (1992) Anti-CD2 monoclonal antibodies prevent spontaneous and adoptive transfer of diabetes in the BB/Wor rat. Am. J. Pathol., 141, 1043–1051. Battan, R., Mordes, J.P., Abreau, S., Greiner, D.L., Handler, E.S. and Rossini, A.A. (1994) Evidence that intrathymic islet transplantation does not prevent diabetes or subsequent islet graft destruction in RT6-depleted diabetes resistant BioBreeding/Worcester rats. Transplantation, 57, 731–736. Bellgrau, D. and Lagarde, A.C. (1990) Cytotoxic T-cell precursors with low-level CD8 in the diabetes-prone Biobreeding rat: implications for generation of an autoimmune T-cell repertoire. Proc. Natl. Acad. Sci. USA, 87, 313–317. Bellgrau, D., Redd, J.M. and Sellins, K.S. (1994) Peculiar T-cell signaling does not preclude positive selection in the diabetes-prone BB rat. Diabetes, 43, 47–52. Bellmann, K., Hui, L., Radons, J., Burkart, V. and Kolb, H. (1997) Low stress response enhances vulnerability of islet cells in diabetes-prone BB rats. Diabetes, 46, 232–236. Bellmann, K., Kolb, H., Rastegar, S., Jee, P. and Scott, F.W. (1998) Potential risk of oral insulin with adjuvant for the prevention of Type I diabetes: a protocol effective in NOD mice may exacerbate disease in BB rats. Diabetologia, 41, 844–847. Bertrand, S., Vigeant, C. and Yale, J.-F. (1994) Predictive value of lymphocyte antibodies for the appearance of diabetes in BB rats. Diabetes, 43, 137–142. Bieg, S., Moller, C., Olsson, T. and Lernmark, A. (1997) The lymphopenia (lyp) gene controls the intrathymic cytokine ratio in congenic BioBreeding rats. Diabetologia, 40, 786–792. Bone, A.J., Walker, R., Varey, A.-M., Cooke, A. and Baird, J.D. (1990) Effect of cyclosporin on pancreatic events and development of diabetes in BB/Edinburgh rats. Diabetes, 39, 508– 514. Bortell, R., Kanaitsuka, T., Stevens, L.A., Moss, J., Mordes, J.P., Rossini, A.A., et al. (1999) The RT6 (ART2) family of ADP-ribosyltransferases in rat and mouse. Mol. Cell. Biochem., In press. Brissette-Storkus, C., Kaufman, C.L., Pasewicz, L., Worsey, H.M., Lakomy, R., Ildstad, S.T., et al. (1994) Characterization and function of the NKR-P1dim/T cell receptor-alpha beta+ subset of rat T cells. J.Immunol., 152, 388–396. Brown, D.W., Welsh, R.M. and Like, A.A. (1993) Infection of peripancreatic lymph nodes but not islets precedes Kilham rat virus-induced diabetes in BB/Wor rats . J.Virol., 67, 5873–5878. Burstein, D., Handler, E.S., Schindler, J., Seals, J., Mordes, J.P. and Rossini, A.A. (1987) Effect of interleukin-2 on diabetes in the BB/Wor rat. Diabetes Res., 5, 163–167. Burstein, D., Mordes, J.P., Greiner, D.L., Stein, D., Nakamura, N., Handler, E.S., et al. (1989) Prevention of diabetes in the BB/Wor rat by a single transfusion of spleen cells: Parameters that affect the degree of protection. Diabetes, 38, 24–30. Casali, P. and Notkins, A.L. (1989) CD5+ B lymphocytes, polyreactive antibodies and the human B cell repertoire. Immunol, Today, 10, 364–368. Chakrabarti, S., Ma, N. and Sima, A.A.F. (1989) Reduced number of anionic sites is associated with glomerular basement membrane thickening in the diabetic BB-rat. Diabetologia, 32, 826–828.

28

AUTOIMMUNE DIABETES MELLITUS IN THE BB RAT

Chakrabarti, S. and Sima, A.A.F. (1997) Endothelin-1 and endothelin-3-like immunoreactivity in the eyes of diabetic and non-diabetic BB/W rats. Diabetes Res. Clin. Pract., 37, 109– 120. Chao, N.J., Timmerman, L., McDevitt, H.O. and Jacob, C.O. (1989) Molecular characterization of MHC class II antigens (beta 1 domain) in the BB diabetes-prone and -resistant rat. Immunogenetics, 29, 231–234. Cheta, D.M., Lim, J., Chan, E.K., Kunakorn, T. and Charles, M.A. (1995) Glimepirideinduced prevention of diabetes and autoimmune events in the BB rat: Revised. Life Sci., 57, 2281–2290. Christianson, S.W., Shultz, L.D. and Leiter, E.H. (1993) 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, 42, 44–55. Chung, Y.H., Jun, H.-S., Kang, Y., Hirasawa, K., Lee, B.-R., Van Rooijen, N., et al. (1997) Role of macrophages and macrophage-derived cytokines in the pathogenesis of Kilham rat virus-induced autoimmune diabetes in diabetesresistant BioBreeding rats. J.Immunol., 159, 466–471. Classen, J.B. (1996) The timing of immunization affects the development of diabetes in rodents. Autoimmunity, 24, 137–145. Colle, E. (1990) Genetic susceptibility to the development of spontaneous insulin-dependent diabetes mellitus in the rat. Clin. Immunol. Immunopathol., 57, 1–9. Colle, E., Fuks, A., Poussier, P., Edouard, P. and Guttmann, R.D. (1992) Polygenic nature of spontaneous diabetes in the rat: Permissive MHC haplotype and presence of the lymphopenic trait of the BB rat are not sufficient to produce susceptibility. Diabetes, 41, 1617–1623. Colle, E., Guttmann, R.D., Fuks, A., Seemayer, T.A. and Prud’homme, G.J. (1986) Genetics of the spontaneous diabetic syndrome. Interaction of MHC and non-MHC-associated factors. Mol. Biol Med., 3, 13–23. Crisá, L., Mordes, J.P. and Rossini, A.A. (1992) Autoimmune diabetes mellitus in the BB rat. Diabetes/Metab. Rev., 8, 9–37. Crisá, L., Sarkar, P., Waite, D.J., Haag, F., Koch-Nolte, F., Rajan, T.V., et al. (1993) An RT6a gene is transcribed and translated in lymphopenic diabetes-prone BB rats. Diabetes, 42, 688–695. Davenport, C., Lovell, H., James, R.F.L. and Todd, I. (1995) Brain-reactive autoantibodies in BB/d rats do not recognize glutamic acid decarboxylase. Clin. Exp. Immunol., 101, 127– 135. De Paepe, M.E., Corriveau, M., Tannous, W.N., Seemayer, T.A. and Colle, E. (1992) Increased vascular permeability in pancreas of diabetic rats: Detection with high resolution protein A-gold cytochemistry. Diabetologia, 35, 1118–24. Desemone, J., Majno, G., Joris, I., Handler, E.S., Rossini, A.A. and Mordes, J.P. (1990) Morphological and physiological characteristics of pancreas-specific venular permeability induced by Monastral blue B. Exp. Mol. Pathol., 52, 141–153. DeSilva, M.G., Jun, H.S., Yoon, J.W., Notkins, A.L. and Lan, M.S. (1996) Autoantibodies to IA-2 not detected in NOD mice or BB rats. Diabetologia, 39, 1237–1238. Di Marco, R., Zaccone, P., Magro, G., Grasso, S., Lunetta, M., Barcellini, W., et al. (1996) Synergistic effect of deoxyspergualin (DSP) and cyclosporin A (CsA) in the prevention of spontaneous autoimmune diabetes in BB rats. Clin. Exp. Immunol., 105, 338–343. Dosch, H.-M. (1993) The possible link between insulin dependent (juvenile) diabetes mellitus and dietary cow milk. Clin. Biochem., 26, 307–308. Doukas, J., Cutler, A.H. and Mordes, J.P. (1994) Polyinosinic: polycytidylic acid is a potent activator of endothelial cells. Am. J. Pathol., 145, 137–147. Doukas, J., Majno, G. and Mordes, J.P. (1996) Anti-endothelial cell autoantibodies in BB rats with spontaneous and induced IDDM. Diabetes, 45, 1209–1216. Doukas, J., Mordes, J.P., Swymer, C., Niedzwiecki, D., Mason, R., Rozing, J., et al. (1994) Thymic epithelial defects and predisposition to autoimmune disease in BB rats. Am. J. Pathol., 145, 1517–1525. Driscoll, H.K., Chertow, B.S., Jelic, T.M., Baltaro, R.J., Chandor, S.B., Walker, E.M., et al. (1996) Vitamin A status affects the development of diabetes and insulitis in BB rats. Metabolism, 45, 248–253. Ellerman, K.E. and Like, A.A. (1995) A major histocompatibility complex class II restriction for BioBreeding Worcester diabetes-inducing T cells. J. Exp. Med., 182, 923–930.

JOHN P.MORDES ET AL.

29

Ellerman, K.E. and Like, A.A. (1998) Diabetes susceptibility genes are widely distributed in the rat. Diabetes, 47 (Suppl. 1), A231–A231. Ellerman, K.E., Richards, C.A., Guberski, D.L., Shek, W.R. and Like, A.A. (1996) Kilham rat virus triggers T-celldependent autoimmune diabetes in multiple strains of rat. Diabetes, 45, 557–562. Ellerman, K.E., Wrobleski, M., Rabinovitch, A. and Like, A. (1993) Natural killer cell depletion and diabetes mellitus in the BB/Wor rat (revisited). Diabetologia, 36, 596–601. Elliott, R.B. and Martin, J.M. (1984) Dietary protein: A trigger of insulin dependent diabetes in the BB rat? Diabetologia, 26, 297–299. Ewel, C.H., Sobel, D.O., Zeligs, B.J. and Bellanti, J.A. (1992) Poly I:C accelerates development of diabetes mellitus in diabetes-prone BB rat. Diabetes, 41, 1016–1021. Fangmann, J., Schwinzer, R., Hedrich, H.-J., Klöting, I. and Wonigeit, K. (199D Diabetesprone BB rats express the RT6 alloantigen on intestinal intraepithelial lymphocytes. Eur. J.Immunol., 21, 2011–2015. Festing, M.F.W. (1993) International Index of Laboratory Animals, pp. 1–238. Leicester, UK: MRC Toxicology Unit, University of Leicester. Field, C.J., Wu, G., Métroz-Dayer, M.-D., Montambault, M. and Marliss, E.B. (1990) Lactate production is the major metabolic fate of glucose in splenocytes and is altered in spontaneously diabetic BB rats. Biochem. J., 272, 445–452. Fowell, D. and Mason, D. (1993) Evidence that the T cell repertoire of normal rats contains cells with the potential to cause diabetes. Characterization of the CD4+ T cell subset that inhibits this autoimmune potential. J. Exp. Med., 177, 627–636. Friedman, D.F., Cho, E.A., Goldman, J., Carmack, C.E., Besa, E.L., Hardy, R.R., et al. (1991) The role of clonal selection in the pathogenesis of autoreactive B cell lymphoma. J. Exp. Med., 174, 525–537. Fuks, A., Colle, E., Ono, S., Prud’homme, G.J., Seemayer, T. and Guttmann, R.D. (1988) Immunogenetic studies of insulin-dependent diabetes in the BB rat. In Frontiers in diabetes research: Lesson from animal diabetes II, edited by E. Shafrir and A.E.Renold, pp. 29–33. London: John Libbey. Gaertner, D.J., Jacoby, R.O., Johnson, E.A., Paturzo, F.X., Smith, A.L. and Brandsma, J.L. (1993) Characterization of acute rat parvovirus infection by in situ hybridization. Virus Res., 28, 1–18. Georgiou, H.M. and Bellgrau, D. (1989) Thymus transplantation and disease prevention in the diabetes-prone BioBreeding rat. J.Immunol., 142, 3400–3405. Georgiou, H.M., Lagarde, A.C. and Bellgrau, D. (1988) T cell dysfunction in the diabetesprone BB rat. A role for thymic migrants that are not T cell precursors. J. Exp. Med., 167, 132–148. Gerling, I.C., Friedman, H., Greiner, D.L., Shultz, L.D. and Leiter, E.H. (1994) Multiple lowdose streptozocin-induced diabetes in NOD-scid/scid mice in the absence of functional lymphocytes. Diabetes, 43, 433–440. Giles, T.D., Ouyang, J., Kerut, E.K., Given, M.B., Allen, G.E., McIlwain, E.F., et al. (1998) Changes in protein kinase C in early cardiomyopathy and in gracilis muscle in the BB/ Wor diabetic rat. Am.J.Physiol. Heart Circ. Physiol., 274, H295—H307. Gold, D.P. and Bellgrau, D. (1991) Identification of a limited T-cell receptor β chain variable region repertoire associated with diabetes in the BB rat. Proc. Natl. Acad. Sci. USA, 88, 9888–9891. Gotfredsen, C.F., Buschard, K. and Frandsen, E.K. (1985) Reduction of diabetes incidence of BB Wistar rats by early prophylactic insulin treatment of diabetes-prone animals. Diabetologia, 28, 933–935. Gottlieb, P.A., Berrios, J.P., Mariani, G., Handler, E.S., Greiner, D., Mordes, J.P. and Rossini, A.A. (1990) Autoimmune destruction of islets transplanted into RT6-depleted diabetesresistant BB/Wor rats. Diabetes, 39, 643–645. Gottlieb, P.A., Handler, E.S., Appel, M.C., Greiner, D.L., Mordes, J.P., et al. (199D Insulin treatment prevents diabetes in RT6-depleted diabetes resistant BB/Wor rats. Diabetologia, 34, 296–300. Greiner, D.L., Handler, E.S., Nakano, K., Mordes, J.P. and Rossini, A.A. (1986) Absence of the RT-6 T cell subset in diabetes-prone BB/W rats. J.Immunol., 136, 148–151. Greiner, D.L., Malkani, S., Kanaitsuka, T., Bortell, R., Doukas, J., Rigby, M., et al. (1997) The T cell marker RT6 in a rat model of autoimmune diabetes. Adv. Exp. Med. Biol., 419, 209–216.

30

AUTOIMMUNE DIABETES MELLITUS IN THE BB RAT

Greiner, D.L., Mordes, J.P., Handler, E.S., Angelillo, M., Nakamura, N. and Rossini, A.A. (1987) Depletion of RT6.1+ T lymphocytes induces diabetes in resistant BioBreeding/ Worcester (BB/W) rats. J. Exp. Med., 166, 461–475. Groen, H., Klatter, F.A., Brons, N.H.C., Mesander, G., Nieuwenhuis, P. and Kampinga, J. (1996) Abnormal thymocyte subset distribution and differential reduction of CD4+ and CD8+ T cell subsets during peripheral maturation in diabetes-prone BioB reeding rats. J. Immunol, 156, 1269–1275. Groen, H., Klatter, F.A., Brons, N.H.C., Wubbena, A.S., Nieuwenhuis, P. and Kampinga, J. (1995) High-frequency, but reduced absolute numbers of recent thymic migrants among peripheral blood T lymphocytes in diabetes-prone BB rats. Cell. Immunol, 163, 113–119. Groen, H., Klatter, F.A., Van Petersen, A.S., Pater, J.M., Nieuwenhuis, P. and Kampinga, J. (1993) Composition of rat CD4+ resting memory T-cell pool is influenced by major histocompatibility complex. Transplant. Proc., 25, 2782–2783. Groen, H., Pater, J.M., Klatter, F.A., Nieuwenhuis, P. and Rozing, J. (1997) Expression of RT6, but not CD45RC, is disturbed on immature peripheral T cells in the BB rat. Adv. Exp. Med. Biol., 419, 253–256. Groen, H., van der Berk, J.M.M.M., Nieuwenhuis, P. and Kampinga, J. (1989) Peripheral T cells in diabetes prone (DP) BB rats are CD45R-negative. Thymus, 14, 145–150. Guberski, D.L. (1994) Diabetes-prone and diabetes-resistant BB rats: Animal models of spontaneous and virally induced diabetes mellitus, lymphocytic thyroiditis and collagen-induced arthritis. ILAR News, 35, 29–37. Guberski, D.L., Butler, L., Kastern, W. and Like, A.A. (1989) Genetic studies in inbred BB/ Wor rats: Analysis of progeny produced by crossing lymphopenic diabetes-prone rats with nonlymphopenic diabetic rats. Diabetes, 38, 887–893. Guberski, D.L., Manzi, S.M. and Like, A.A. (1992) Correction of lymphopenia by injection of diabetes-resistant splenocytes prevents diabetes and improves breeding efficiency in diabetes-prone BB/Wor rats. Contemp. Topics Lab. Animal Sci., 31, 13–13. Guberski, D.L., Thomas, V.A., Shek, W.R., Like, A.A., Handler, E.S., Rossini, A.A., et al. (1991) Induction of type 1 diabetes by Kilham’s rat virus in diabetes resistant BB/Wor rats. Science, 254, 1010–1013. Haag, F., Andresen, V., Karsten, S., Koch-Nolte, F. and Thiele, H.-G. (1995) Both allelic forms of the rat T cell differentiation marker RT6 display nicotinamide adenine dinucleotide (NAD)-glycohydrolase activity, yet only RT6. 2 is capable of automodification upon incubation with NAD. Eur.J.Immunol., 25, 2355–2361. Haag, F. and Koch-Nolte, F. (1997) The vertebrate gene family of mono (ADP-ribosyl) transferases: Proposal for a unified nomenclature. Adv. Exp. Med. Biol., 419, 459–462. Hanenberg, H., Kolb-Bachofen, V., Kantwerk-Funke, G. and Kolb, H. (1989) Macrophage infiltration precedes and is a prerequisite for lymphocytic insulitis in pancreatic islets of pre-diabetic BB rats. Diabetologia, 32, 126–134. Hao, L., Chan, S.-M. and Lafferty, K.J. (1993) Mycophenolate mofetil can prevent the development of diabetes in BB rats. Ann. NY Acad. Sci., 696, 328–332. Herold, K.C., Vezys, V., Koons, A., Lenschow, D., Thompson, C. and Bluestone, J.A. (1997) CD28/B7 costimulation regulates autoimmune diabetes induced with multiple low doses of streptozotocin. J. Immunol., 158, 984–991. Hosseinzadeh, H. and Goldschneider, I. (1993) Recent thymic emigrants in the rat express a unique antigenic phenotype and undergo post-thymic maturation in peripheral lymphoid tissues. J. Immunol., 150, 1670–1679 Hosszufalusi, N., Reinherz, L., Takei, S., Chan, E. and Charles, M.A. (1994) Glipizide-induced prevention of diabetes and autoimmune events in the BB rat. J. Autoimmun., 7, 753–761. Huang, X., Hultgren, B., Dybdal, N. and Stewart, T.A. (1994) Islet expression of interferon-α precedes diabetes in both the BB rat and streptozotocin-treated mice. Immunity, 1, 469– 476. Iwakoshi, N.N., Goldschneider, I., Tausche, F., Mordes, J.P., Rossini, A.A. and Greiner, D.L. (1998) High frequency apoptosis of recent thymic emigrants in the liver of lymphopenic diabetes prone BioBreeding rats. J. Immunol., 160, 5838–5850. Jackerott, M., Hornum, L., Andreasen, B.E. and Markholst, H. (1997) Segregation of autoimmune Type 1 diabetes in a cross between diabetic BB and Brown Norway rats. J. Autoimmun., 10, 35–41. Jacob, H.J., Pettersson, A., Wilson, D., Mao, Y., Lernmark, A. and Lander, E.S. (1992) Genetic dissection of autoimmune type I diabetes in the BB rat. Nature Genet., 2, 56–60.

JOHN P.MORDES ET AL.

31

Jacoby, R.O. and Ball-Goodrich, L.J. (1995) Parvovirus infections of mice and rats. Semin. Virol., 6, 329–337. Jiang, Z., Handler, E.S., Rossini, A.A. and Woda, B.A. (1990) Immunopathology of diabetes in the RT6-depleted diabetes-resistant BB/Wor rat. Am. J. Pathol, 137, 767–777. Jiang, Z. and Woda, B.A. (1991) Cytokine gene expression in the islets of the diabetic BioBreeding/Worcester rat. J. Immunol., 146, 2990–2994. Jobe, L.W., Ubungen, R., Goodner, C.J., Baskin, D.G., Braquet, P. and Lernmark, A. (1993) Protection from BB rat diabetes by the platelet-activating factor inhibitor BN50730. Autoimmunity, 16, 259–266. Joseph, S., Diamond, A.G., Smith, W., Baird, J.D. and Butcher, G.W. (1993) BB-DR/ Edinburgh: A lymphopenic, nondiabetic subline of BB rats. Immunology, 78, 318–328. Jun, H.S. and Yoon, J.-W. (1994) Initiation of autoimmune type 1 diabetes and molecular cloning of a gene encoding for islet cell-specific 37kd autoantigen. Adv. Exp. Med. Biol., 347, 207–220. Kampinga, J., Groen, H., Klatter, F.A., Pater, J.M., Van Petersen, A.S., Roser, B., et al. (1997) Post-thymic T-cell development in the rat. Thymus, 24, 173–200. Karges, W., Pietropaolo, M., Ackerley, C.A. and Dosch, H.M. (1996) Gene expression of islet cell antigen p69 in human, mouse and rat. Diabetes, 45, 513–521. Kawahito, Y., Cannon, G.W., Gulko, P.S., Remmers, E.F., Longman, R.E., Reese, V.R., et al. (1998) Localization of quantitative trait loci regulating adjuvant-induced arthritis in rats: Evidence for genetic factors common to multiple autoimmmune diseases . J. Immunol., 161, 4411–4419. Kimura, H. and Davies, T.F. (1996) Thyroid-specific T cells in the Wistar rat. 3. Induction of anergy by a syngeneic thyroid cell line. Autoimmunity, 24, 47–53. Kitagawa, Y., Desemone, J. and Mordes, J.P. (1993) Pancreas-specific venular labeling by Monastral blue B in the BB rat: modulation by prostaglandins and their inhibitors. Immunopharm., 25, 229–238. Kitagawa, Y., Greiner, D.L., Reynolds, C.W., Ortaldo, J.R., Toccafondi, R., Handler, E.S., et al. (1991) Islet cells but not thyrocytes are susceptible to lysis by NK cells. J.Autoimmun., 4, 703–716. Klöting, I., Vogt, L. and Serikawa, T. (1995) Locus on chromosome 18 cosegregates with diabetes in the BB/OK rat subline. Diabète Metab., 21, 338–344. Kleemann, R., Rothe, H., Kolb-Bachofen, V., Xie, Q., Nathan, C., Martin, S., et al. (1993) Transcription and translation of inducible nitric oxide synthase in the pancreas of pre-diabetic BB rats. FEBS Lett., 328, 9–12. Ko, I.Y, Ihm, S.H. and Yoon, J.W. (1991) Studies on autoimmunity for initiation of Beta-cell destruction. VIII. Pancreatic Beta-cell dependent autoantibody to a 38 kilodalton protein precedes the clinical onset of diabetes in BB rats. Diabetologia, 34, 548–554. Koch-Nolte, F., Haag, F., Kastelein, R. and Bazan, F. (1996) Uncovered: the family relationship of a T-cell-membrane protein and bacterial toxins. Immunol. Today, 17, 402–405. Koevary, S., Rossini, A.A., Stoller, W., Chick, W. and Williams, R.M. (1983) Passive transfer of diabetes in the BB/W rat. Science, 220, 727–728. Kolb, H. and Kröncke, K.-D. (1993) IDDM: Lesson from the low-dose streptozocin model in mice. Diab. Rev., 1, 116–126. Kolb, H., Wörz-Pagenstert, U., Kleemann, R., Rothe, H., Rowsell, P., Rastegar, S. and Scott, F.W. (1997) Insulin therapy of prediabetes suppresses TH1 associated gene expression in BB rat pancreas. Autoimmunity, 26, 1–6. Kolb, H., Wörz-Pagenstert, U., Kleemann, R., Rothe, H., Rowsell, P. and Scott, F.W. (1996) Cytokine gene expression in the BB rat pancreas: natural course and impact of bacterial vaccines. Diabetologia, 39, 1448–1454. Lau, A., Ramanathan, S. and Poussier, P. (1998) Excessive production of nitric oxide by macrophages from DP-BB rats is secondary to the T-lymphopenic state of these animals. Diabetes, 47, 197–205. Lea, R.G., McCracken, J.E., McIntyre, S.S., Smith, W. and Baird, J.D. (1996) Disturbed development of the preimplantation embryo in the insulin-dependent diabetic BB/E rat. Diabetes, 45, 1463–1470. Lee, E.S., Jiang, J.J., Sund, G.C., Simonson, W.T., Graham, J., Dietsch, G., et al. (1999) Recombinant human plateletactivating factor acetylhydrolase reduces the frequency of diabetes in the diabetes-prone BB rat. Diabetes, 48, 43–49.

32

AUTOIMMUNE DIABETES MELLITUS IN THE BB RAT

Lefkowith, J., Schreiner, G., Cormier, J., Handler, E.S., Driscoll, H.K., Greiner, D., et al. (1990) Prevention of diabetes in the BB rat by essential fatty acid deficiency. Relationship between physiological and biochemical changes. J. Exp. Med., 171, 729–743. Leiter, E.H. and Gill, R.G. (1994) Prevention of autoimmune diabetes in the BB rat by intrathymic islet transplantation at birth. J. Endocrinol Invest., 17, 595–599. Li, X.B., Scott, F.W., Park, Y.H. and Yoon, J.W. (1995) Low incidence of autoimmune type I diabetes in BB rats fed a hydrolysed casein-based diet associated with early inhibition of non-macrophage-dependent hyperexpression of MHC class I molecules on beta cells. Diabetologia, 38, 1138–1147. Lieberman, I., Heil, B.V., Seow, W.K. and Thong, Y.H. (1993) Synergy between tetrandrine and FK506 in prevention of diabetes in BB rats. Life Sci., 53, PL453-PL458. Lieberman, I., Lentz, D.P., Trucco, G.A., Seow, W.K. and Thong, Y.H. (1992) Prevention by tetrandrine of spontaneous development of diabetes mellitus in BB rats. Diabetes, 41, 616– 619. Like, A.A., Biron, C.A., Weringer, E.J., Byman, K., Sroczynski, E. and Guberski, D.L. (1986) Prevention of diabetes in BioBreeding/Worcester rats with monoclonal antibodies that recognize T lymphocytes or natural killer cells. J. Exp. Med., 164, 1145–1159. Like, A.A., Guberski, D.L. and Butler, L. (1986) Diabetic BioBreeding/Worcester (BB/Wor) rats need not be lymphopenic. J. Immunol, 136, 3254–3258. Like, A.A., Kislauskis, E., Williams, R.M. and Rossini, A.A. (1982) Neonatal thymectomy prevents spontaneous diabetes mellitus in the BB/W rat. Science, 216, 644–646. Lindsay, R.M., Peet, R.S., Wilkie, G.S., Rossiter, S.P., Smith, W., Baird, J.D., et al. (1997) In vivo and in vitro evidence of altered nitric oxide metabolism in the spontaneously diabetic, insulin-dependent BB/Edinburgh rat. Br. J. Pharmacol., 120, 1–6. Lindsay, R.M., Smith, W., Rossiter, S.P., McIntyre, M.A., Williams, B.C. and Baird, J.D. (1995) Nomega-nitro-Larginine methyl ester reduces the incidence of IDDM in BB/E rats. Diabetes, 44, 365–368. Mackay, I.R., Bone, A., Tuomi, T., Elliott, R., Mandel, T., Karopoulos, C., et al. (1996) Lack of autoimmune serological reactions in rodent models of insulin dependent diabetes mellitus. J. Autoimmun., 9, 705–711. MacKay, P. (1995) Adoptive transfer of diabetes to and from old normoglycaemic BB rats. Diabetologia, 38, 145–152. Maehama, T., Nishina, H., Hoshino, S., Kanaho, Y. and Katada, T. (1995) NAD+-dependent ADP-ribosylation of T lymphocyte alloantigen RT6.1 reversibly proceeding in intact rat lymphocytes. J. Biol. Chem., 270, 22747–22751. Malkani, S., Nompleggi, D., Hansen, J.W., Greiner, D.L., Mordes, J.P. and Rossini, A.A. (1997) Dietary cow’s milk protein does not alter the frequency of diabetes in the BB rat. Diabetes, 46, 1133–1140. Markholst, H., Eastman, S., Wilson, D., Andreasen, B.E. and Lernmark, Å. (1991) Diabetes segregates as a single locus in crosses between inbred BB rats prone or resistant to diabetes. J. Exp. Med., 174, 297–300. Markholst, H., Klaff, L.J., Klöppel, G., Lernmark, Å., Mordes, J.P. and Palmer, J. (1990) Lack of systematically found insulin autoantibodies in spontaneously diabetic BB rats. Diabetes, 39, 720–727. Martin, A.-M., Blankenhorn, E.P., Maxson, M.N., Zhao, M., Leif, J., Mordes, J.P., et al. (1999) Non-MHC-linked diabetes susceptibility loci on chromosomes 4 and 13 in a backcross of the diabetes-prone BB/Wor rat to the Wistar Furth rat. Diabetes, 48, 50–58. McKeever, U., Mordes, J.P., Greiner, D.L., Appel, M.C., Rozing, J., Handler, E.S., et al. (1990) Adoptive transfer of autoimmune diabetes and thyroiditis to athymic rats. Proc. Natl. Acad. Set. USA, 87, 7718–7722. McVary, K.T., Rathnau, C.H. and McKenna, K.E. (1997) Sexual dysfunction in the diabetic BB/WOR rat: A role of central neuropathy. Am. J. Physiol. Regul. Integr. Comp. Physiol., 272, R259—R267. Meehan, C.J., Krajewski, A.S., Butcher, G.W., Smith, W. and Baird, J.D. (1993) Lymphoma in the BB/E rat: c-myc translocation identified. J. Pathol., 170, 87–93. Misra, M., Duguid, W.P. and Marliss, E.B. (1996) Prevention of diabetes in the spontaneously diabetic BB rat by the glutamine antimetabolite acivicin. Can. J. Physiol. Pharmacol., 74, 163–172. Mohseni, S. and Hildebrand, C. (1998) Neuropathy in diabetic BB/Wor rats treated with insulin implants. Acta Neuropathol., 96, 144–150.

JOHN P.MORDES ET AL.

33

Mooij, P., De Wit, H.J. and Drexhage, H.A. (1993) An excess of dietary iodine accelerates the development of a thyroid-associated lymphoid tissue in autoimmune prone BB rats. Clin. Immunol. Immunopathol., 69, 189–198. Mordes, J.P., Bortell, R., Doukas, J., Rigby, M.R., Whalen, B.J., Zipris, D., et al. (1996) The BB/Wor rat and the balance hypothesis of autoimmunity. Diabetes/Metab. Rev., 2, 103— 109 Mordes, J.P., Greiner, D.L. and Rossini, A.A. (1996) Animal models of autoimmune diabetes mellitus. In Diabetes mellitus. A fundamental and clinical text, edited by D.LeRoith, S.I.Taylor and J.M.Olefsky, pp. 349–360. Philadelphia: Lippincott-Raven. Mordes, J.P., Schirf, B., Roipko, D., Greiner, D.L., Weiner, H., Nelson, P., et al. (1996) Oral insulin does not prevent insulin-dependent diabetes mellitus in BB rats. Ann. NY Acad Sci., 778, 418–421. Mori, K., Mori, M., Stone, S., Braverman, L.E. and DeVito, W.J. (1998) Increased expression of tumor necrosis factorα and decreased expression of thyroglobulin and thyroid peroxidase mRNA levels in the thyroids of iodide-treated BB/Wor rats. Eur. J. Endocrinol., 139, 539–545. Murthy, V.K., Shipp, J.C., Hanson, C. and Shipp, D.M. (1992) Delayed onset and decreased incidence of diabetes in BB rats fed free radical scavengers. Diabetes Research & Clinical Practice, 18, 11–16. Myers, M.A., Laks, M.R., Feeney, S.J., Mandel, T.E., Koulmanda, M., Bone, A., et al. (1998) Antibodies to ICA512/IA-2 in rodent models of IDDM. J. Autoimmun., 11, 265–272. Nagata, M. and Yoon, J.-W. (1994) Prevention of autoimmune type I diabetes in BioBreeding (BB) rats by a newly established, autoreactive T cell line from acutely diabetic BB rats. J. Immunol., 153, 3775–3783. Nakhooda, A.F., Like, A.A., Chappel, C.I., Murray, F.T. and Marliss, E.B. (1977) The spontaneously diabetic Wistar rat. Metabolic and morphologic studies. Diabetes, 26, 100–112. National Research Council (1996) Laboratory Animal Management: Rodents, pp. 141–146. National Academy Press: Washington, D.C. Nepom, G.T. and Kwok, W.W. (1998) Molecular basis for HLA-DQ associations with IDDM. Diabetes, 47, 1177–1184. Nicoletti, F., Meroni, P.L. and Bendtzen, K. (1996) Fusidic acid and insulin-dependent diabetes mellitus. Autoimmunity, 24, 187–197. Nicoletti, F., Zaccone, P., Di Marco, R., Lunetta, M., Magro, G., Grasso, S., et al. (1997) Prevention of spontaneous autoimmune diabetes in diabetes-prone BB rats by prophylactic treatment with antirat interferon-gamma antibody. Endocrinology, 138, 281–288. Nicoletti, F., Zaccone, P., Di Marco, R., Magro, G., Grasso, S., Stivala, F., et al. (1998) Paradoxical antidiabetogenic effect of gamma-interferon in DP-BB rats. Diabetes, 47, 32–38. Oschilewski, U., Kiesel, U. and Kolb, H. (1985) Administration of silica prevents diabetes in BB-rats. Diabetes, 34, 197–199. Petersen, J.S., MacKay, P., Plesner, A., Karlsen, A., Gotfredsen, C., Verland, S., et al. (1997) Treatment with GAD65 or BSA does not protect against diabetes in BB rats. Autoimmunity, 25, 129–138. Pettersson, A., Wilson, D., Daniels, T., Tobin, S., Jacob, H.J., Lander, E.S., et al. (1995) Thyroiditis in the BB rat is associated with lymphopenia but occurs independently of diabetes. J. Autoimmun., 8, 493–505. Plamondon, C., Kottis, V., Brideau, C., Métroz-Dayer, M.-D. and Poussier, P. (1990) Abnormal thymocyte maturation in spontaneously diabetic BB rats involves the deletion of CD4− 8+ cells. J. Immunol., 144, 923–928. Posselt, A.M., Barker, C.F, Friedman, A.L. and Naji, A. (1992) Prevention of autoimmune diabetes in the BB rat by intrathymic islet transplantation at birth. Science, 256, 1321– 1324. Powrie, F. and Mason, D. (1990) OX-22high CD4+ T cells induce wasting disease with multiple organ pathology: prevention by the OX-22Low subset. J. Exp. Med., 172, 1701– 1708. Powrie, F. and Mason, D. (1990) Subsets of rat CD4+ T cells defined by their differential expression of variants of the CD45 antigen: Developmental relationships and in vitro and in vivo functions. Curr. Top. Microbiol. Immunol., 159, 79–96. Prins, J.-B., Herberg, L., Den Bieman, M. and Van Zutphen, B.F.M. (1991) Genetic characterization and interrelationship of inbred lines of diabetes-prone and not diabetes-prone BB rats. In Frontiers in diabetes research. Lessons from animal diabetes III, edited by E.Shafrir, pp. 19–24. London: Smith-Gordon.

34

AUTOIMMUNE DIABETES MELLITUS IN THE BB RAT

Rabinovitch, A., Suarez-Pinzon, W., El-Sheikh, A., Sorensen, O. and Power, R.F. (1996) Cytokine gene expression in pancreatic islet-infiltrating leukocytes of BB rats—Expression of Th1 cytokines correlates with β-cell destructive insulitis and IDDM. Diabetes, 45, 749– 754. Rabinovitch, A. and Suarez-Pinzon, W.L. (1998) Cytokines and their roles in pancreatic islet β-cell destruction and insulindependent diabetes mellitus. Biochem. Pharmacol. ,55, 1139— 1149. Racke, M.K., Bonomo, A., Scott, D.E., Cannella, B., Levine, A., Raine, C.S., et al. (1994) Cytokine-induced immune deviation as a therapy for inflammatory autoimmune disease. J. Exp. Med., 180, 1961–1966. Rajatanavin, R., Appel, M.C., Reinhardt, W., Alex, S., Yang, Y.-N. and Braverman, L.E. (1991) Variable prevalence of lymphocytic thyroiditis among diabetes-prone sublines of BB/Wor rats. Endocrinology, 128, 153–157. Ramanathan, S., Norwich, K. and Poussier, P. (1998) Antigen activation rescues recent thymic emigrants from programmed cell death in the BB rat. J. Immunol., 160, 5757–5764. Reimers, J.I., Rasmussen, Å.K., Karlsen, A.E., Bjerre, U., Liang, H., Morin, O., et al. (1996) Interleukin-1β inhibits rat thyroid cell function in vivo and in vitro by an NO-independent mechanism and induces hypothyroidism and accelerated thyroiditis in diabetes-prone BB rats. J. Endocrinol., 151, 147–157. Remmers, E.F., Longman, R.E., Du, Y., O’Hare, A., Cannon, G.W., Griffiths, M.M., et al. (1996) A genome scan localizes five non-MHC loci controlling collagen-induced arthritis in rats. Nature Genet., 14, 82–85. Renold, A.E., Porte, D., Jr. and Shafrir, E. (1988) Definitions for diabetes types: Use and abuse of the concept ‘animal models of diabetes mellitus’. In Frontiers in diabetes research. Lessons from animal diabetes II, edited by E.Shafrir and A.E.Renold, pp. 3–7. London: John Libbey & Co. Rigby, M.R., Bortell, R., Greiner, D.L., Czech, M.P., Klarlund, J.K., Mordes, J.P., et al. (1996) The rat T-cell surface protein RT6 is associated with src family tyrosine kinases and generates an activation signal. Diabetes, 45, 1419–1426. Rigby, M.R., Bortell, R., Stevens, L.A., Moss, J., Kanaitsuka, T., Shigeta, H., et al. (1996) Rat RT6.2 and mouse Rt6 locus 1 are NAD+: arginine ADP-ribosyltransferases with auto-ADP-ribosylation activity. J. Immunol, 156, 4259—4265. Ritter, M.A., Gordon, L.K. and Goldschneider, I. (1978) Distribution and identity of Thy-1-bearing cells during ontogeny in rat hemopoietic and lymphoid tissues. J. Immunol., 121, 2463–2471. Rossini, A.A., Mordes, J.P., Greiner, D.L., Nakano, K., Appel, M.C. and Handler, E.S. (1986) Spleen cell transfusion in the BB/W rat: Prevention of diabetes, MHC restriction and long term persistence of transfused cells. J. Clin. Invest., 77, 1399–1401. Rothe, H., Öngören, C., Martin, S., Rösen, P. and Kolb, H. (1994) Abnormal TNF-production in diabetes-prone BB rats: Enhanced TNF-α expression and defective PGE2 feedback inhibition. Immunology, 81, 407–413. Rothe, H., Schuller, I., Richter, G., Jongeneel, C.V., Kiesel, U., Diamantstein, T., et al. (1992) Abnormal TNF production in prediabetic BB rats is linked to defective CD45R expression. Immunology, 77, 1–6. Roza, A.M., Slakey, D.P., Pieper, G.M., Van Ye, T.M., Moore-Hilton, G., Komorowski, R.A., et al. (1994) Hydroxyethyl starch deferoxamine, a novel iron chelator, delays diabetes in BB rats. J. Lab. Clin. Meet., 123, 556–560. Rozing, J., Coolen, C., Tielen, F.J., Weegenaar, J., Schuurman, H.-J., Greiner, D.L., et al. (1989) Defects in the thymic epithelial stroma of diabetes prone BB rats. Thymus, 140, 125–135. Saoudi, A., Seddon, B., Fowell, D. and Mason, D. (1996) The thymus contains a high frequency of cells that prevent autoimmune diabetes on transfer into prediabetic recipients. J. Exp. Med., 184, 2393–2398. Sarkar, P., Crisá, L., McKeever, U., Haag, F., Koch, F., Thiele, H.-G., et al. (1992) Diabetes prone (DP) BB/Wor rats express functional mRNA for RT6 antigen but lack peripheral phenotypic RT6+ T cells due to shortened T cell life span. Diabetes, 4I (Suppl. 1), 96A– 96A. Satoh, J., Seino, H., Shintani, S., Tanaka, S.-I., Ohteki, T., Masuda, T., et al. (1990) Inhibition of type 1 diabetes in BB rats with recombinant human tumor necrosis factor-a. J. Immunol, 145, 1395–1399. Scott, F.W., Cloutier, H.E., Kleemann, R., Wörz-Pagenstert, U., Rowsell, P., Modler, H.W., et al (1997) Potential mechanisms by which certain foods promote or inhibit the development of spontaneous diabetes in BB rats— Dose, timing, early effect on islet area and switch in infiltrate from Th1 to Th2 cells. Diabetes, 46, 589–598.

JOHN P.MORDES ET AL.

35

Seddon, B. and Mason, D. (1997) Role of CD4+CD8− thymocytes in the prevention of autoimmune diabetes. Biochem. Soc. Trans., 25, 620–624. Seddon, B., Saoudi, A., Nicholson, M. and Mason, D. (1996) CD4+CD8−thymocytes that express L-selectin protect rats from diabetes upon adoptive transfer. Eur. J. Immunol., 26, 2702–2708. Sellins, K.S., Gold, D.P. and Bellgrau, D. (1996) Resistance to tolerance induction in the diabetes-prone Biobreeding rat as one manifestation of abnormal responses to superantigens. Diabetologia, 39, 28–36. Shirai, T., Hirose, S., Okada, T. and Nishimura, H. (1991) CD5+ B cells in autoimmune disease and lymphoid malignancy. Clin. Immunol Immunopathol, 59, 173–186. Shockett, P. and Woodland, R.T. (1988) A unique B cell subpopulation in diabetes-prone BB/ W rats, absent in diabetes resistant sublines. J. Cell Biochem., Suppl. 12B, 154–154. Shyp, S., Tishon, A. and Oldstone, M.B.A. (1990) Inhibition of diabetes in BB rats by virus infection. II. Effect of virus infection on the immune response to non-viral and viral antigens. Immunology, 69, 501–507. Sima, A.A.F. and Sugimoto, K. (1999) Experimental diabetic neuropathy: an update. Diabetologia, 42, 773–788. Simons, P.J., Delemarre, F.G., Jeucken, P.H. and Drexhage, H.A. (1998) Pre-autoimmune thyroid abnormalities in the biobreeding diabetes-prone (BB-DP) rat: a possible relation with the intrathyroid accumulation of dendritic cells and the initiation of the thyroid autoimmune response. J. Endocrinol., 157, 43–51. Sobel, D.O., Azumi, N., Creswell, K., Holterman, D., Blair, O.C., Bellanti, J.A., et al. (1995) The role of NK cell activity in the pathogenesis of poly I: C accelerated and spontaneous diabetes in the diabetes prone BB rat. J. Autoimmun., 8, 843–857. Sobel, D.O., Creswell, K., Yoon, J.W. and Holterman, D. (1998) Alpha interferon administration paradoxically inhibits the development of diabetes in BB rats. Life Set., 62, 1293– 1302. Sobel, D.O., Ewel, C.H., Zeligs, B., Abbassi, V., Rossio, J. and Bellanti, J.A. (1994) Poly I:C induction of α-interferon in the diabetes-prone BB and normal Wistar rats: Dose-response relationships. Diabetes, 43, 518–522. Sobel, D.O., Goyal, D., Ahvazi, B., Yoon, J.W., Chung, Y.H., Bagg, A, et al. (1998) Low dose poly I: C prevents diabetes in the diabetes prone BB rat. J. Autoimmun., 11, 343–352. Sobel, D.O. and Newsome, J. (1997) Gamma interferon prevents diabetes in the BB rat. Clinical & Diagnostic Laboratory Immunology, 4, 764–768. Sobel, D.O., Newsome, J., Ewel, C.H., Bellanti, J.A., Abbassi, V., Creswell, K., et al. (1992) Poly I: C induces development of diabetes mellitus in BB rat. Diabetes, 41, 515–520. Sun, S.H., Silver, P.B., Du, Y., Caspi, R.R., Wilder, R.L. and Remmers, E.F. (1997) Identification of a locus on rat chromosome 4 associated with EAU susceptibility. Invest. Ophthalmol.Vis.Sci., 38, 5436-. Tafuri, A., Bowers, W.E., Handler, E.S., Appel, M., Lew, R., Greiner, D., et al. (1993) High stimulatory activity of dendritic cells from diabetes-prone BioBreeding/Worcester rats exposed to macrophage-derived factors. J. Clin.Invest., 91, 2040–2048. Takada, T., lida, K. and Moss, J. (1994) Expression of NAD glycohydrolase activity by rat mammary adenocarcinoma cells transformed with rat T cell alloantigen RT6.2. J. Biol. Chem., 269, 9420–9423. Takahashi, K., Satoh, J., Seino, H., Zhu, X.P., Sagara, M., Masuda, T., et al. (1993) Prevention of type I diabetes with lymphotoxin in BB rats. Clin. Immunol. ImmunopathoL, 69, 318– 323. Teruya, M., Takei, S., Forrest, L.E., Grunewald, A., Chan, E.K. and Charles, M.A. (1993) Pancreatic islet function in nondiabetic and diabetic BB rats. Diabetes, 42, 1310–1317. Thiele, H.-G., Haag, F. and Nolte, F. (1993) Asymmetric expression of RT6.1 and RT6.2 alloantigens in (RT6a×RT6b)F1 rats is due to a pretranslational mechanism. Transplant. Proc., 25, 2786–2788. Thiele, H.-G., Koch, F. and Kashan, A. (1987) Postnatal distribution profiles of Thy-I+ and RT6+ cells in peripheral lymph nodes of DA rats. Transplant. Proc., 19, 3157–3160. Thliveris, J.A., Begleiter, A., Manchur, D. and Johnston, J.B. (1997) Immunotherapy for insulin-dependent diabetes mellitus in the “BB” rat. Life Sci., 61, 283–291. Thomas, V.A., Woda, B.A., Handler, E.S., Greiner, D.L., Mordes, J.P. and Rossini, A.A. (1991) Altered expression of diabetes in BB/Wor rats by exposure to viral pathogens. Diabetes, 40, 255–258.

36

AUTOIMMUNE DIABETES MELLITUS IN THE BB RAT

Timsit, J., Savino, W., Boitard, C. and Bach, J.F. (1989) The role of class II major histocompatibility complex antigens in autoimmune diabetes: animal models. J. Autoimmun., 2, 115–129. Todd, D., Singh, A.J., Greiner, D.L., Mordes, J.P., Rossini, A.A. and Bortell, R. (1999) A new isolation method for rat intraepithelial lymphocytes. J. Immunol. Methods, 224, 111– 127. Tullin, S., Farris, P., Petersen, J.S., Hornum, L., Jackerott, M. and Markholst, H. (1997) A pronounced thymic B cell deficiency in the spontaneously diabetic BB rat. J. Immunol., 158, 5554–5559. Verhaeghe, J., Van Herck, E., Visser, W.J., Suiker, A.M.H., Thomasset, M., Einhorn, T.A., et al. (1990) Bone and mineral metabolism in BB rats with long-term diabetes: Decreased bone turnover and osteoporosis. Diabetes, 39, 477–482. Vermeer, L.A., de Boer, N.K., Bucci, C., Bos, N.A., Kroese, F.G. and Alberti, S. (1994) MRC OX19 recognizes the rat CD5 surface glycoprotein, but does not provide evidence for a population of CD5bright B cells. Eur. J. Immunol., 24, 585–592. Vlahos, W.D., Seemayer, T.A. and Yale, J.-F. (1987) Diabetes prevention in BB rats by inhibition of endogenous insulin secretion. Metabolism, 40, 825–829. Waite, D.J., Appel, M.C., Handler, E.S., Mordes, J.P., Rossini, A.A. and Greiner, D.L. (1996) Ontogeny and immunohistochemical localization of thymus dependent and thymus independent RT6+ cells in the rat. Am. J. Pathol., 148, 2043–2056. Waite, D.J., Handler, E.S., Mordes, J.P., Rossini, A.A. and Greiner, D.L. (1993) The RT6 rat lymphocyte alloantigen circulates in soluble form. Cell. Immunol., 152, 82–95. Watson, W.C., Thompson, J.P., Terato, K., Cremer, M.A. and Kang, A.H. (1990) Human HLA-DRβ gene hypervariable region homology in the biobreeding BB rat: Selection of the diabetic-resistant subline as a rheumatoid arthritis research tool to characterize the immunopathologic response to human type II collagen. J. Exp. Med., 172, 1331–1339. Whalen, B.J., Greiner, D.L., Mordes, J.P. and Rossini, A.A. (1994) Adoptive transfer of autoimmune diabetes mellitus to athymic rats: Synergy of CD4+ and CD8+ T cells and prevention by RT6+ T cells. J. Autoimmun., 7, 819–831. Whalen, B.J., Mordes, J.P. and Rossini, A.A. (1996) The BB rat as a model of human insulin-dependent diabetes mellitus. In Animal models for autoimmune and inflammatory disease, edited by J.E.Coligan, A.M.Kruisbeck, D.H.Margulies, E.M.Shevach and W.Strober, pp. 15.3.1–15.3.15. New York: John Wiley & Sons. Whalen, B.J., Rossini, A.A., Mordes, J.P. and Greiner, D.L. (1995) DR-BB rat thymus contains thymocyte populations predisposed to autoreactivity. Diabetes, 44, 963–967. Whalen, B.J., Weiser, P., Marounek, J., Rossini, A.A., Mordes, J.P. and Greiner, D.L. (1999) Recapitulation of normal and abnormal BB rat T cell development in adult thymus organ culture. J. Immunol., In press. Wilkin, T., Kiesel, U., Diaz, J.-L., Burkart, V. and Kolb, H. (1986) Autoantibodies to insulin as serum markers for autoimmune insulitis. Diabetes Res., 3, 173–174. Woda, B.A., Handler, E.S., Greiner, D.L., Reynolds, C., Mordes, J.P. and Rossini, A.A. (1991) T lymphocyte requirement for diabetes in RT6-depleted diabetes resistant BB rats. Diabetes, 40, 423–428. Woda, B.A., Like, A.A., Padden, C. and McFadden, M. (1986) Deficiency of phenotypic cytotoxic-suppressor T lymphocytes in the BB/W rat. J. Immunol., 136, 856–859. Wu, G., Field, C.J. and Marliss, E.B. (1991) Elevated glutamine metabolism in splenocytes from spontaneously diabetic BB rats. Biochem. J., 274, 49–54. Wu, G., Field, C.J. and Marliss, E.B. (1991) Glucose and glutamine metabolism in rat macrophages: Enhanced glycolysis and unaltered glutaminolysis in spontaneously diabetic BB rats. Biochim. Biophys. Acta Gen. Subj., 1115, 166–173. Wu, G. and Marliss, E.B. (1991) Deficiency of purine nucleoside phosphorylase activity in thymocytes from the immunodeficient diabetic BB rat. Clin. Exp. Immunol., 86, 260–265. Wu, G. and Marliss, E.B. (1993) Enhanced glucose metabolism and respiratory burst in peritoneal macrophages from spontaneously diabetic BB rats. Diabetes, 42, 520–529. Yagihashi, S., Kamijo, M. and Nagai, K. (1990) Peripheral neuropathy in diabetic animals: heterogeneous expression of neuropathic patterns in different animal models. In Frontiers in diabetes research. Lessons from animal diabetes III, edited by E.Shafrir, pp. 459–463. London: Smith-Gordon.

JOHN P.MORDES ET AL.

37

Yoon, J.-W. (1990) The role of viruses and environmental factors in the induction of diabetes. Curr. Top. Microbiol. Immunol, 164, 95–124. Young, A.A., Gedulin, B., Vine, W., Percy, A. and Rink, T.J. (1995) Gastric emptying is accelerated in diabetic BB rats and is slowed by subcutaneous injections of amylin. Diabetologia, 38, 642–648. Zadeh, H.H., Greiner, D.L., Wu, D.Y, Tausche, F. and Goldschneider, I. (1996) Abnormalities in the export and fate of recent thymic emigrants in diabetes-prone BB/W rats. Autoimmunity, 24, 35–46. Ziegler, M., Schlosser, M., Hamann, J., Vieregge, P., Lühder, F., Klöting, I., et al. (1994) Autoantibodies to glutamate decarboxylase detected in diabetes-prone BB/OK rats do not distinguish onset of diabetes. Exp. Clin. Endocrinol., 102, 98–103. Zipris, D. (1996) Evidence that Th1 lymphocytes predominate in islet inflammation and thyroiditis in the BioBreeding (BB) rat. J. Autoimmun., 9, 315–319. Zipris, D., Greiner, D.L., Malkani, S., Whalen, B.J., Mordes, J.P. and Rossini, A.A. (1996) Cytokine gene expression in islets and thyroids of BB rats: Interferon gamma and IL-12 p40 mRNA increase with age in both diabetic and insulin treated nondiabetic BB rats. J. Immunol, 156, 1315–1321.

2. THE NOD MOUSE AND ITS RELATED STRAINS HIROSHI IKEGAMI1 and SUSUMU MAKINO2 1Department

of Geriatric Medicine, Osaka University Medical School 2AC

Center, Shionogi Laboratories

INTRODUCTION Type 1 (insulin-dependent) diabetes mellitus is caused by autoimmune destruction of insulin-producing βcells of the pancreas in genetically susceptible individuals. Understanding of the genetics and autoimmune mechanisms of the disease has been greatly facilitated by an animal model, the nonobese diabetic (NOD) mouse, which spontaneously develops Type 1 diabetes. The origins and characteristics of this strain and a closely related strain, the nonobese nondiabetic (NON) mouse, are summarized in this chapter. ESTABLISHMENT OF NOD AND ITS RELATED STRAINS Origin of the NOD Mouse The NOD mouse was established as an inbred strain of mice with spontaneous development of autoimmune Type 1 diabetes, by one of the authors, Makino (Makino et al., 1980). Discovery of a mouse with Type 1 diabetes, which eventually led to the development of the NOD mouse strain, was rather paradoxical. The origin of the NOD mouse traces back to a mouse with cataract identified in a closed colony of Jcl: ICR mice. Ohtori at the Toxicological Department of Shionogi Aburahi Laboratories found a mouse with cataract in Jcl: ICR mice in 1966, and an inbred strain of mice with cataract and small eyes, the CTS mouse, was established by brother-sister mating (Ohtori et al., 1968). The designation of “CTS” was originally for “cataract and small eyes”, but it was subsequently redesignated “cataract Shionogi”. The presence of cataract in CTS mice is an autosomal dominant trait, whereas small eyes are an autosomal recessive trait.

THE NOD MOUSE AND ITS RELATED STRAINS

39

Figure 1 Genealogy of NOD and its related strains.

Thus, mice homozygous for the CTS allele at cataract locus (Cs) show cataract and small eyes, whereas heterozygous mice show only cataract, but not small eyes. At the 4th generation of inbreeding of CTS, a control line with normal eyes was separated and an inbred NCT (non-cataract) strain was established (Figure 1) (Makino et al, 1988). At the 6th generation, selective breeding for euglycemic and hyperglycemic mice was performed, expecting to obtain hyperglycemic mice, prompted by the fact that cataract is a common complication in human diabetes. At this stage, it was not clear whether mating was strictly brother-sister mating, or only selective breeding was performed for elevated or normal glucose levels. After selective breeding for hyperglycemia and euglycemia for about 10 generations, these two sublines were transferred from the Toxicological Department to the Laboratory Animal Department, because of limitation of space for animals in the former department. Makino at the Laboratory Animal Department took over this project, and from this stage it is clear that strict brother-sister mating was started. Two sister strains, one with slightly higher fasting blood glucose levels (approximately 150 mg/dl) and the other with normal fasting glucose levels (approximately 100 mg/dl), both free of cataracts, were obtained after selective breeding for 13 generations (Makino et al., 1980). At the 20th generation in selective breeding, a mouse with polyuria, polydipsia and weight loss was found in the line with normal, but not increased, fasting blood glucose levels. This is a rather paradoxical and unexpected observation in that, contrary to the initial expectation of obtaining diabetic mice in the hyperglycemic line, a diabetic mouse was detected in the normoglycemic line. Inbreeding was continued with this mouse to establish an inbred strain with spontaneous development of diabetes, which culminated in a strain currently known as the NOD mouse. A slightly hyperglycemic subline, on the contrary, never developed overt diabetes, and this line was subsequently designated the nonobese nondiabetic (NON) mouse. As expected from the history described above, glucose tolerance in NON mice is not completely normal, but slightly impaired as compared with other control laboratory strains. Selection for hyperglycemia, however, in this line was discontinued after discovery of a diabetic mouse in the line subsequently established as NOD. NON mice, therefore, should not be regarded as a model for Type 2 diabetes or glucose intolerance because the phenotype for glucose intolerance is not fixed in this strain.

40

H.IKEGAMI AND S.MAKING

MAINTENANCE OF THE NOD MOUSE NOD mice are derived from Jcl: CR mice, which are a closed colony derived from so-called “Swiss” mice. The name ICR denotes the Institute of Cancer Research in Philadelphia, USA, where a closed colony of Swiss mice was maintained by random breeding. These mice were subsequently transferred to commercial breeders and are Table 1 Reproductive efficiency of NOD/Shi, NON/Shi, and C57BL/6J strains. Birth order Strain C57BL/6J

NOD/Shi

NON/Shi

No. of female mice Average litter size Weaning rate (%) No. of female mice Average litter size Weaning rate (%) No. of female mice Average litter size Weaning rate (%)

I

II

III

IV

V

25 6.3 82.6 43 8.4 70.7 37 6.5 85.3

25 7.0 91.9 22 10.4 81.7 31 8.4 88.7

24 6.5 87.5 5 11.0 27.0 28 8.2 82.9

22 6.4 89–6 1 2 0 15 5.9 70.0

18 6.6 98.6 0 – – 7 6.1 56.8

Reproductive Efficiency* 26.1

8.3

16.7

*: Total number of weanlings obtained from one female mouse

now commercially available as Jcl: ICR (CLEA, Japan) and CD-1: Crj (Charles River, Japan). Jcl: ICR (or CD-1) are known to breed well by producing large litters. It was therefore fortunate that NOD mice were derived from Jch: lCR mice in terms of maintenance of the strain. In fact, NOD mice produced an average of 9.1 offspring per litter as compared with 6.6 offspring per litter in C57BL/6J mice in Shionogi Aburahi Laboratories (Table 1), and therefore, it is relatively easy to maintain a colony. Most female NOD mice, however, raise at most two litters due to the development of diabetes. Diabetic females show very low reproductive capacity. Although most of the mothers before the onset of diabetes can nurse their pups well, those with overt diabetes cannot complete weaning of their pups. As a consequence, the total efficiency of breeding is much lower (about 30%) than that in usual laboratory strains, such as C57BL/6 mice (Table 1). If female NOD mice in a colony continuously breed well, delivering more than three litters, then there is a possibility that the incidence of diabetes in the colony will go down due to selection of mice less diabetes prone as mating pairs, as will be discussed in the next section. The NOD mouse is at present commercially available from the breeders listed in Table 2. At the time of writing, a NOD mouse is supplied by Clea Japan at a price of approximately 8,000 yen ($70). The cost of maintenance of non-diabetic NOD mice is similar to that of usual laboratory strains, except for the cost for test strips for screening of urinary glucose. For diabetic NOD mice, the cost is higher because frequent changes of cage bedding, almost everyday, are required due to polyuria in diabetic mice. In each regular cage, which is usually used for housing 3–5 mice, only one diabetic NOD mouse should be housed because the cages would become very dirty with multiple diabetic mice due to polyuria, and such an environment would lead to an unhealthy condition, and predispose to infectious diseases. Because of this and the low reproductive efficiency, approximately three times more space is required to maintain an NOD colony than that for usual laboratory strains.

THE NOD MOUSE AND ITS RELATED STRAINS

41

Table 2 Commercial breeders for NOD mice. Strain name

Breeder

NOD/ShiJic

Clea Japan, Inc.

Address

Dai 2 inari biru, 2–20–14 Aobadai, Meguroku 153–8533, Japan Tel: +81 3 5704 7123 Fax: +81 3 3791 2859 NOD/Bom Bomholtgard Breeding and Research Center Ltd Bomholtvej 10, PO Box 39, DK8680 Ry, Denmark Tel: +45 86 84 1211 Fax: +45 86 84 1699 NOD/RijHsd Harlan UK Ltd Shaw’s Farm, Blackthorn, Bicester, Oxon OX6 OTP, UK Tel: +44(0)1869 243 241 Fax: +44(0)1869 246 759 NOD/Orllo IFFA CREDO BP 0109–69592, 1’Arbresle Cedex, France Tel: +33 74 01 6969 Fax: +33 74 01 6999 NOD/LtSanIbm RCC Biotechnology and Breeding Division BRL Biological Research Laboratories Ltd, Wolferstrasse 4, CH-4414 Fullinsdorf, Switzerland Tel: +41 61 906 4242 Fax: +41 61 901 2565 NOD/MrkTac Taconic 273 Hover Avenue, Germantown, New York 12526, USA Tel: +1 518 537 6208 Fax: +1 518 537 7287

CHARACTERISTICS OF THE NOD MOUSE The NOD mouse spontaneously develops autoimmune Type 1 diabetes. Prior to the development of diabetes, infiltration of mononuclear cells into the pancreatic islets (insulitis) is observed. Insulitis is not observed before 3 weeks of age, but appears spontaneously at around 4 weeks of age. The frequency of insulitis reaches 70–90% by 9 weeks of age, and almost 100% of mice of both sexes develop insulitis by 20 weeks of age (Makino et al., 1985a). Mononuclear cells infiltrating the islets are mostly T cells (CD4+ and CD8+), but B cells, dendritic cells and macrophages are also observed. Despite massive infiltration of mononuclear cells into the islets, β-cells remain intact until 12–15 weeks of age, when destruction of P-cells becomes aggressive and overt diabetes develops. Although almost all NOD mice of both sexes develop insulitis, only some of them develop overt diabetes, suggesting that several factors modify the process from insulitis to βcell destruction and the development of overt diabetes. The molecular mechanisms involved in the initiation of insulitis and destruction of β-cells are not fully characterized and are now under extensive investigation (Tisch and McDevitt 1996) (Benoist and Mathis 1997). After the onset of overt diabetes, marked polyuria and polydipsia develop, and mice lose weight and die within one to two months unless treated with a daily injection of insulin (Makino et al., 1980), as in the case of human Type 1 diabetes.

42

H.IKEGAMI AND S.MAKING

Table 3 Cumulative incidence of diabetes at 30 weeks of age in NOD mice under SPF and GF conditions. Conditions

Female

Male

Germ free (GF) Specific pathogen free (SPF)

70.2% (47) 81.4% (43)

26.8% (41) 13.3% (45)

ENVIRONMENTAL EFFECT ON INCIDENCE OF DIABETES The cumulative incidence of Type 1 diabetes in the original colonies at Shionogi Aburahi Laboratories was approximately 80% in females and less than 20% in males at 30 weeks of age (Makino et al., 1980). The incidence in males subsequently increased and currently the cumulative incidence of Type 1 diabetes at 30 weeks of age is 90% in females and 50% in males. A sex-related difference in the frequency of diabetes has been observed in most NOD colonies. Orchiectomy in males has been reported to increase the frequency of diabetes, while oophorectomy in females has been reported to decrease the incidence of diabetes (Makino et al., 1981). Prevention of diabetes in female NOD mice by treatment with androgen has also been reported (Fox 1992), suggesting the modifying effect of sex steroids on the development of diabetes. Although the NOD mouse is an inbred strain with more than 70 generations of brother-sister mating, considerable variation in the incidence of diabetes has been reported among different colonies (Leiter et al., 1990) (Pozzili et al., 1993). Diet (Elliott et al., 1988) (Coleman et al., 1990), room temperature (Williams et al., 1990) and pathogens (Ohsugi and Kurosawa 1994) have been suggested as factors influencing the incidence of diabetes, but the exact reason is still unknown. The following are recommendations to maintain a constant incidence of diabetes in the NOD mouse: (1) The colony should be raised under specific pathogen-free (SPF) conditions, because a lower incidence of diabetes has been reported in NOD mice in conventional conditions (Ohsugi, Kurosawa 1994) and in those infected with viruses (Oldstone 1988) (Wilberz et al., 1991). (2) The same diet should be maintained, because changes in the incidence of diabetes have been reported in NOD mice fed a different diet. It was previously reported that 100% of female NOD mice maintained in germfree conditions developed Type 1 diabetes (Suzuki et al, 1987). This observation, however, was based on a relatively small number of animals (n=15 for germ-free conditions). In our experience with a larger number of animals at Shionogi Aburahi Laboratories, no such increase in the incidence of diabetes was observed under germfree conditions as compared with SPF condition (Table 3), indicating that SPF condition is sufficient for the maintenance of NOD mouse colonies. Even in the same colony maintained under the conditions described above, the incidence of diabetes readily goes down unless careful breeding is performed. Although the exact reason for the variation in the incidence of Type 1 diabetes in genetically homogeneous NOD mice is unknown, mating should be performed between NOD mice both or at least one of which had developed or subsequently developed Type 1 diabetes, in order to maintain a high incidence of diabetes. In practice, since the reproductive capacity of diabetic NOD females is low, the development of diabetes in mice used for mating should be followed-up and only pups from parents or a parent that subsequently developed diabetes should be used for breeding of the next generation. Deviation of the phenotypes by random mating is observed not only in NOD mice, but also in other inbred animal models with polygenic, multifactorial inheritance.

THE NOD MOUSE AND ITS RELATED STRAINS

43

It is therefore recommended that the incidence of diabetes in each NOD colony should be monitored and used as a reference incidence instead of referring to the incidence of diabetes in other colonies reported in the literature. This is particularly important in such experiments where the incidence of diabetes is used as the outcome. In addition, the incidence and degree of insulitis should also be monitored and used as a phenotype. Even in colonies with a low incidence of diabetes, insulitis is similarly observed as in the original NOD colony. GENETICS OF DIABETES The NOD Mouse as a Model for Multifactorial Diseases Inheritance of Type 1 diabetes in NOD mice is multifactorial, as in the case of human Type 1 diabetes, with both genetic and environmental factors contributing to disease development. The contribution of environmental factors is evident from the fact that not all NOD mice develop Type 1 diabetes despite their identical genetic background. The variation in the incidence of diabetes among colonies or even within the same colony as described above provides other evidence for an environmental effect on the development of diabetes. The identity of such environmental factors, however, is not fully characterized, with a few reports on viral infection (Oldstone 1988) (Wilberz et al., 1991), diet (Elliott et al., 1988) (Coleman et al., 1990) and room temperature (Williams et al., 1990). Elucidation of the environmental factors that affect the incidence of diabetes is important because such factors, if identified, would be beneficial for the prevention of Type 1 diabetes in humans (Todd, 1991). The contribution of multiple genes on different chromosomes to disease susceptibility has been directly demonstrated in NOD mice (Hattori et al., 1986) (Wicker et al., 1987) (Prochazka et al., 1987) (Todd et al., 1991) (Ghosh et al., 1993), which has greatly contributed to understanding of the genetics of Type 1 diabetes as well as that of multifactorial diseases in general in humans. To date, at least 18 susceptibility loci have been mapped to the mouse genome (Table 4) (Wicker et al., 1995) (Ikegami et al., 1996). Congenic Strains The contribution of multiple genes to disease predisposition in the NOD mouse has been clearly demonstrated using congenic strategy. Congenic strains are strains Table 4 Susceptibility loci for Type 1 diabetes mapped in the NOD mouse. Locus

Chromosome

Insulitis

Diabetes

Susceptibility to Type 1 diabetes*

References

Idd1

17

+**

+

1–4

Idd2 Idd3

9 3

– +

+ +

Idd4 Idd5 Idd6

11 1 6

– + –

+ + +

NOD (+); B6, BIO, C3H, NON (-) NOD > NON, B10 NOD > B6, B10, NON NOD > B10 NOD > B10 NOD > B10

2, 5, 6 5–8 5, 7 9, 10 5

44

H.IKEGAMI AND S.MAKING

Locus

Chromosome

Insulitis

Diabetes

Susceptibility to Type 1 diabetes*

References

+

?

11

+ + + + + + + + + + + + + +

NOD > Mus Spretus NON > NOD PWK > NOD B10, NON > NOD B10 > NOD NOD > B6, B10, NON NOD > B6, B10 NOD > B6+SJL NOD > B6+SJL NOD > NOR NOD > NON NOD > NON NOD > CTS NOD > B10 NOD > B6

Idd7 Idd8 Idd9

7 14 4

– – – – ?

Idd10 Idd11 Idd l 2 Idd 13 Idd l4 Idd 15 Idd 16 Idd 17 Idd 18

3 4 14 2 13 5 17 3 3

+ ? ? – – – ? ? ?

6 12 5, 6 5 5, 6, 13 5, 6, 8, 14–16 17 17 18 6 6 19 14 16

*With the exception of Idd1, none of the Idd loci is absolutely required for susceptibility to diabetes or insulitis. Strength of susceptibility to diabetes conferred by NOD alleles is therefore expressed relative to that of strains with which NOD mice are outcrossed. In most cases, of course, NOD alleles are more susceptible to insulitis and diabetes than alleles of diabetes-resistant control strains. In some strain combinations, however, alleles from diabetes-resistant strains are more susceptible than NOD alleles (e.g. Idd7 and Idd8). In case of Idd6, alleles of NON and PWK strains are more susceptible, but alleles of B10 and Mus Spretus are less susceptible to diabetes than NOD alleles. **(+): linkage with insulitis or diabetes; (—): lack of linkage with insulitis or diabetes; (?): data not available 1 Hattori M et al., Science 231:733–735, 1986 2 Prochazka M et al., Science 237:286–289, 1987 3 Wicker LS et al., J Exp Med 165:1639–1654, 1987 4 Ikegami H et al., Diabetologia 31:254–258, 1988 5 Ghosh S et al., Nat Genet 4:404–409, 1993 6 McAleer MA et al., Diabetes 44:1186–1195, 1995 7 Todd JA et al., Nature 351:542–547, 1991 8 Wicker LS et al., J Exp Med 180:1705–1713, 1994 9 Garchon H-J et al., Nature 353:260–262, 1991 10 Cornall RJ et al., Nature 353:262–265, 1991 11 de Gouyon B et al., Proc Natl Acad Sci USA 1877–1881, 1993 12 Melanitou E et al., Genome Res 8:608–620, 1998 13 Rodrigues NR et al., Mamm Genome 5:167–170, 1994 14 Prins JB et al., Science 260:695–698, 1993 15 Podolin PL et al., J Immunol 159:1835–1843, 1997 16 Podolin PL et al., Mamm Genome 9:283–286, 1998 17 Morahan G et al., Proc Natl Acad Sci USA 91:5898–5902, 1994 18 Serreze DV et al., J Exp Med 180:1553–1558, 1994 19 Ikegami H et al., J.Clin.Invest. 96:1936–0942, 1995

THE NOD MOUSE AND ITS RELATED STRAINS

45

Figure 2 Congenic strains. A genome of an NOD strain congenic for the MHC of the C57BL/6 mouse (NOD.B6-H-2) is shown in comparison with the NOD genome. Shaded rectangles: the B6 MHC introgressed onto the NOD genome.

differing at a locus or a chromosomal segment encompassing the locus, with all other background genes identical (Figure 2). For example, a NOD mouse strain congenic for the major histocompatibility complex (MHC) of the C57BL/6 (B6) mouse means that the MHC region on chromosome 17 is from B6 strain, but all other genetic background is identical to that of the NOD mouse. In practice, such congenic strains can be produced as follows (conventional congenic strategy, Figure 3A): NOD mice (recipient strain) are crossed with B6 mice (donor strain) to produce F1 mice (denoted N1 generation in congenic strain), and the F1 mice are backcrossed to NOD mice to produce the 1st backcross generation (denoted N2 generation). N2 mice with B6 alleles in the MHC are selected for backcrossing with NOD mice and N3 mice are produced. This process is repeated. Successive backcrossing with NOD mice gradually increases the proportion of NOD genetic background with the B6 MHC retained by selection. A sufficient number of backcrosses (to at least N8, usually N12) will eventually replace almost all background genes with NOD genome, but only the MHC from B6 strain is retained by selection. By this strategy, the effects of a single locus or a single chromosomal segment on disease predisposition can be assessed. Theoretically, half of the donor genome will be replaced by the recipient genome in each generation, and therefore 99.2% (1–[1/ 2]8–1) of the genome will be homozygous for the NOD genome at N8 generation, and 99–95% (1-[C1/2]12–1) at N12 generation in NOD MHC congenic strains. Classically, N12 generation, or at least N8 generation, was

46

H.IKEGAMI AND S.MAKING

Figure 3 A. Conventional congenic protocol. To introgress (genetically transfer) the MHC of a donor strain (e.g. B6) onto the genetic background of a recipient strain (e.g. NOD), B6 mice are crossed with NOD mice to produce N1 generation, and N1 mice are backcrossed to NOD mice to produce N2 generation. N2 mice are screened for the MHC, and those with B6 alleles are selected for backcrossing with NOD mice and N3 mice are produced. This process is repeated until N8–12 generations. On average, half of the heterozygous genome becomes homozygous for the NOD genome in each backcross, and therefore, 99–2% (1-[1/2]8–1) and 99.95% (I-U/2}12–1) of the genome is homozygous for the NOD genome at N8 and N12 generations, respectively. The introgressed B6 MHC is then made homozygous by intercrossing, resulting in a congenic strain, NOD.B6-H-2, which possesses the B6 MHC on the genetic background of the NOD mouse. It should be noted that male recipient mice have to be used instead of females at some stage of backcrossing, so that both X and Y chromosomes are of recipient origin. This is particularly important for polygenic traits, such as diabetes, in which the contribution of genes on sex chromosomes to disease susceptibility is not fully excluded. In this figure, for example, male N1 mice inherit Y chromosome from NOD father (YN) and X chromosome from B6 mother (XB6). Subsequent backcrossing to NOD females produces N2 males with YN from N1 father and XN from NOD mother, and thus both X and Y chromosomes are of NOD origin. If female mice heterozygous for X chromosomes, one from NOD (XN) and the other from B6 (XB6), are used for backcrossing to NOD recipient, recombination in X chromosome will occur and recombinant X chromosome with both B6 and NOD segments will be introduced to subsequent offspring, leading to incomplete replacement of background genes. Closed: homozygous for NOD genotype, Open: homozygous for B6 genotype, Shaded: heterozygous for NOD and B6.

proposed to be required for sufficient replacement of background genes in congenic strains. Although theoretical calculations indicate that more than 99% of the background genes are identical to the recipient strain after N8 generation, in the case of polygenic diseases it is possible that a certain locus critical for the

THE NOD MOUSE AND ITS RELATED STRAINS

47

Figure 3 B. “Speed” congenic protocol. To increase the speed and efficiency of developing congenic strains, background genes are genotyped with polymorphic markers throughout the genome at the N2 generation, and mice with the least amount of the undesired genome (i.e. the B6 genome) are selected for mating with NOD mice to produce the N3 generation. By selecting progeny at each generation that carry the B6 MHC (desired gene) and the least amount of the undesired genome of B6 origin, the number of generations necessary to eliminate the undesired genome and to construct a congenic strain can be greatly reduced (as early as N5 generation as compared with N8–12 generation with the conventional congenic protocol).

disease susceptibility is located in <1% of the genome of donor origin. Furthermore, the percentage of the genome of donor origin can be larger than the theoretical value in some of the congenic mice, making the possibility of involvement of the undesired genome in congenic mice higher. Thus, we should be careful in studying the effect of a locus on the susceptibility to a polygenic disease, such as Type 1 diabetes, by congenic strategy. It would, therefore, be desirable that certain loci critical for disease susceptibility, such as Idd1 in the MHC of the NOD mouse, are confirmed to be of recipient origin (Ikegami et al., 1995). Alternatively, intercrossing of congenic mice heterozygous for a certain locus should be performed and the incidence of diabetes should be compared relative to the inheritance of alleles at the locus of interest. Recent progress in genotyping with a microsatellite marker throughout the genome has made it possible to monitor background genes in each generation, and to use mice with the least undesired genome for mating to generate the next generation. By this method, termed “speed congenic” strategy (Figure 3B), it is now possible to establish congenic strains in a much shorter period compared with the classical congenic strategy. In our experience, background genes can completely be replaced with recipient strain by N5

48

H.IKEGAMI AND S.MAKING

generation. “Speed congenic” strategy is not only fast, but also accurate and reliable in that replacement of background genes with the recipient genome can clearly be monitored. This is particularly useful in polygenic diseases, such as Type 1 diabetes, in which susceptibility loci have been mapped to the genome, because markers linked to susceptibility loci can be used for monitoring of background genes. In this way, the possibility of involvement of the undesired genome in congenic mice with polygenic diseases, as mentioned above, can be eliminated. Congenic NOD strains have been established, either by classical or speed congenic strategies, in a limited number of laboratories, and have contributed greatly to understanding of the genetics of not only Type 1 diabetes, but also polygenic, multifactorial diseases in general. MHC Genes MHC gene is necessary, but not sufficient for disease development As summarized in Table 5, none of the NOD mouse strain congenic for the MHC from control strains was reported to develop Type 1 diabetes (Prochazka et al., 1989) Table 5 NOD strains congenic for the MHC. Strains

Donor

MHC (H-2) Recipient Generations Cumulative incidence % (n, age)

References

NOD.B10-H-2

C57BL/10

b

NOD

NOD.H-2i7 NOD.H-2i5 NOD.H-2h4 NOD.H-2h2 NOD.H-2k NOD.NON-H-2

B10.D2(R107) B10.A(5R) B10.A(4R) B10.A(2R) B10.BR NON

i7 i5 h4 h2 k nb1

NOD NOD NOD NOD NOD NOD

1 2 1 1 1 1 1 3 4

N6F1 N12F1 N6F1 N6F1 N6F1 N6F1 N6F1 N6F1 N15F2

0% (0/21, 5–13 Mo) 0% (0/6, 8–12 Mo) 0% (0/40, 5–13 Mo) 0% (0/39, 5–13 Mo) 0% (0/54, 5–13 Mo) 0% (0/38, 5–13 Mo) 0% (0/31, 5–13 Mo) 0% (0/6, 41 weeks) 0% (0/28, 30 weeks)

1. J Exp Med 178:793, 1993; 2. J Exp Med 176:67, 1992; 3. Diabetes 38:1446, 1989; 4. Makino et al. unpublished data

(Wicker et al, 1992) (Podolin et al, 1993) (Ikegami and Makino 1993) (Ikegami et al, 1996), indicating that the NOD MHC is essential for disease development. However, a control strain, B6, congenic for the NOD MHC, B6.NOD-H-2, does not develop diabetes despite the presence of the diabetogenic MHC from the NOD mouse, indicating that the NOD MHC alone is not in itself sufficient for disease development. A similar finding was also reported by Wicker et al., in B10.NOD-H-2 congenic mice (Wicker et al., 1995). This is a typical example of polygenic inheritance of a disease: any one of the genes that predispose to the disease is sufficient for disease susceptibility and combinations of multiple susceptibility genes are necessary for disease development (Mohan et al, 1999).

THE NOD MOUSE AND ITS RELATED STRAINS

49

Multiple components in MHC-linked susceptibility genes Although data on MHC congenic NOD strains indicate that the NOD MHC is diabetogenic and MHC from control strains confer resistance to diabetes, the data do not show identity of MHC-linked susceptibility gene (s) (Idd1). The strongest candidate for Idd1 is class II MHC of the NOD mouse: unique I-A molecules (IAg7) differing from any known I-A from control laboratory strains, and lack of I-E expression due to a deletion mutation in the promoter region of the gene encoding the α chain of the I-E molecule. Due to strong linkage disequilibrium within the MHC, however, the NOD MHC is transmitted to offspring as a set of genes, termed a haplotype, and therefore, the independent effect of class II MHC from other nearby genes within the MHC cannot be assessed in usual breeding studies. One way to overcome this problem and to directly demonstrate the effect of class II MHC on susceptibility to Type 1 diabetes is to use intra-MHC recombinants. Although such recombinants cannot be easily obtained in usual breeding studies between two inbred strains due to the low frequency of recombination within the MHC, recombination events have occurred during many historical meioses among ancestral haplotypes contained in an original outbred colony of the NOD mouse, giving rise to intra-MHC recombinants. By screening NOD-related strains, such intra-MHC recombinants were identified in a sister strain, the CTS mouse, and some mice in outbred Jcl: ICR mice, from which NOD mice were derived. In particular, the CTS mouse was found to have the same class II MHC as the NOD mouse, but different class I MHC (Ikegami et al., 1989a) (Ikegami et al., 1990). With this recombinant MHC, it became possible to dissect out the effect of class II MHC on disease susceptibility from that of other nearby genes within the MHC. Unlike previously reported NOD strains congenic for the MHC, which were completely resistant to diabetes, the NOD strain congenic for the CTS MHC, NOD.CTS-H-2, developed diabetes, indicating that the CTS MHC contained a MHC-linked susceptibility gene (Idd1) for Type 1 diabetes (Ikegami et al., 1995). To our surprise, however, the incidence of diabetes in NOD.CTS-H-2 mice was much lower than that in the NOD parental strain (Figure 4) (Ikegami et al, 1995), indicating that MHC-linked susceptibility consists of multiple components and that the CTS MHC contains part, but not all, of the components. Since class II MHC is identical in the CTS and the NOD mouse, the data indicate that class II MHC is diabetogenic, but other genes adjacent to, but distinct from, class II MHC also contribute to disease susceptibility. A combination of the former (Idd1) and latter (Idd16) components is necessary for full expression of the MHC-linked susceptibility to Type 1 diabetes in the NOD mouse (Figure 5A). Non-MHC Genes Multiple susceptibility genes on chromosome 3 By genome scan in crosses of NOD with B10.H-2g7 mice, strong linkage of Type 1 diabetes was identified in the central part of chromosome 3 near the marker D3Nds1 (Figure 5) (Todd et al., 1991). Subsequent studies in congenic strains, however, revealed that this localization was caused by the combined effect of two independent loci, Idd3 and Idd10, flanking D3Nds1 (Figure 5B) (Wicker et al., 1994). In fact, NOD mice congenic for B10 or B6 alleles at D3Nds1 were not protected against diabetes, indicating lack of a susceptibility gene in the D3Nds1 region, where strong linkage was initially detected. Fine mapping in additional congenic strains has now revealed the contribution of at least four independent loci (Idd3, 10, 17, 18) to disease susceptibility, with no susceptibility gene in the initial location of peak linkage (Figure 5B) (Podorin et al., 1997) (Podorin et al., 1998). These data, together with identification of multiple susceptibility genes in the MHC as mentioned in the previous section, emphasize the power and importance

50

H.IKEGAMI AND S.MAKING

Figure 4 Cumulative incidence of Type 1 diabetes in NOD.CTS-H-2 congenic mice. MHC heterozygotes (N13–19) were intercrossed and the incidence of diabetes was monitored relative to the genotypes of the MHC. A: females, B: males. Closed circles: NOD MHC homozygotes Open circles: CTS MHC homozygotes

of the congenic strategy in fine mapping and characterization of susceptibility genes for polygenic diseases, such as Type 1 diabetes.

THE NOD MOUSE AND ITS RELATED STRAINS

51

Figure 5 Multiple susceptibility genes on chromosome 17 (A) and chromosome 3 (B). Congenic mapping revealed the existence of at least two (Idd1 and Idd16), and probably more, susceptibility genes for Type 1 diabetes on chromosome 17. Similarly, at least four (Idd3, 10, 17, 18) susceptibility genes have been mapped on chromosome 3 by congenic mapping.

THE NON MOUSE Origin As mentioned above, two sublines were separated at the 6th generation during the establishment of the CTS strain to produce euglycemic and hyperglycemic lines (Figure 1). The NOD mouse was established from one diabetic female mouse found in the former line, an unexpected finding as mentioned previously. Contrary to the initial expectation, the latter line never developed overt diabetes and was subsequently designated as the nonobese nondiabetic (NON) mouse. Due to the selection for hyperglycemia in the initial part of inbreeding, glucose tolerance in the NON mouse is not completely normal, but slightly impaired. Selection for hyperglycemia, however, was discontinued when a diabetic mouse was found in the former line. The NON strain, therefore, should not be regarded as a model for Type 2 diabetes or glucose intolerance because the phenotype for glucose intolerance is not fixed in this strain, nor is this being maintained by selective breeding. This is in clear contrast to the NSY mouse (Ueda et al., 1995), which was established as an inbred animal model for Type 2 diabetes by selective breeding for glucose intolerance from Jcl: CR mice, and is being maintained by selection for glucose intolerance (see chapter XX). NON mice are not available from commercial breeders, but they are maintained at Shionogi Aburahi Laboratories in Japan and Jackson Laboratory in USA.

52

H.IKEGAMI AND S.MAKING

Phenotypes Glucose tolerance Although the phenotype for glucose tolerance has not been fixed in the NON mouse, glucose tolerance in this strain as a group is reported to be slightly impaired (Hasegawa et al., 1992). Mild glucose intolerance as reported in the NON mouse, however, has also been reported in some inbred laboratory strains, such as C57BL/ 6 mice (Kaku et al., 1988). Glucose intolerance in NON mice is observed as early as 8–10 weeks of age. Unlike human Type 2 diabetes, however, glucose intolerance in NON mice does not deteriorate with age, but rather improved in both males and females (Wainai et al., 1992). This is in clear contrast to the NSY mouse, an animal model of Type 2 diabetes, in which age-dependent deterioration of glucose tolerance is observed, as in the case of Type 2 diabetes in humans (Ueda et al., 1995). In addition, marked interindividual variation in glucose tolerance is reported in NON mice (Wainai et al., 1992), probably because selective breeding for hyperglycemia was discontinued a long time ago and therefore, glucose intolerance was not fixed as a phenotype in this strain. Similar to other strains with impaired glucose tolerance (Ueda et al., 1995), glucose tolerance is more markedly impaired in male NON mice than in female mice (Hasegawa et al, 1992) (Wainai et al, 1992). Infiltration of mononuclear cells into islets (insulitis) as observed in the NOD mouse is not observed in the NON mouse. To evaluate glucose tolerance, intraperitoneal glucose tolerance test (ipGTT) is usually performed. Glucose (2 g/kg) is injected i.p. in overnight-fasted mice, and blood glucose levels are monitored at 0, 30, 60, (90) and 120 min after glucose administration (Ueda et al., 1995). The easiest way to monitor blood glucose is to use a glucose meter developed for self-monitoring of blood glucose in diabetic patients. At each time point for measuring blood glucose, a small cut is made in the tail with a razor blade, and a small amount of blood is aspirated with a test strip attached to the glucose monitoring device. Most sensors with a capillary aspiration system can measure blood glucose level with a small amount of blood, as little as 3–5 µl, and the results can be obtained in 10–30 seconds. Diabetes is usually diagnosed when blood glucose level at 120 min after glucose load is equal to or above 200 mg/dl (11.1 mM). Renal lesions PAS-positive deposits in glomerular capillaries have been reported in the NON mouse (Muraoka et al., 1992). The deposits were reported to be positive for IgM as well as PAS, and to consist of fine lipid-like droplets on electron microscopic observation. These characteristics are similar to those observed in lipoprotein glomerulopathy in humans (Saito et al., 1989). The frequency of the lesion is 37% in females and 31% in males at 16–24 weeks of age, and 78% in females and 53% in males at 36–38 weeks of age. Since this lesion is observed in NON mice with normal glucose tolerance, it is not related to diabetes or glucose intolerance, and therefore, is different from diabetic nephropathy. Genetic backgrounds The recent identification of highly polymorphic microsatellite markers throughout the genome has made it possible to compare genetic identity and differences among different inbred strains (Dietrich et al., 1994). Despite the close relationship between the NOD and NON strains as sister strains, allelic variation in microsatellite markers between NOD and NON strains is as large as that between NOD and inbred laboratory strains, or even larger than that between some of the inbred laboratory strains, such as BALB/c

THE NOD MOUSE AND ITS RELATED STRAINS

53

and C3H/He (Dietrich et al., 1994). This may be partly due to the non-strict process of the initial part of inbreeding as described above. If strict brother-sister mating had been performed from the initiation of inbreeding of these strains, there would have been much smaller differences. These differences made it possible to use NON mice as an outcross partner with NOD mice for mapping of susceptibility genes for Type 1 diabetes (Ikegami et al., 1989a) (Ikegami et al., 1989b) (Prochazka et al., 1987) (McAleer et al., 1995), as well as a partner for congenic strains (Prochazka et al., 1989) (Ikegami et al., 1996). Despite the differences in genetic markers, however, sharing of some of the susceptibility genes for insulitis and diabetes between NOD and NON mice was suggested by breeding studies (Makino et al., 1985b) (McAleer et al., 1995) and examination of congenic strains (Ikegami, Makino 1993) (McAleer et al., 1995). APPLICATION OF NOD AND RELATED STRAINS FOR BIOMEDICAL RESEARCH The NOD mouse has contributed to our understanding of the genetics and pathogenesis of Type 1 diabetes and to the development of effective methods of disease prediction, prevention and intervention. Pioneering studies using genome scanning and congenic and transgenic strategies were all performed with NOD mice. In addition, information generated with this strain has contributed and will continue greatly to elucidation of the genetics of multifactorial diseases, and of the pathogenesis and molecular mechanisms of autoimmune diseases in general. The NON mouse can potentially be used for studies related to glucose intolerance, such as examination of the mechanisms, prevention and treatment of glucose intolerance. Glucose intolerance in NON mice, however, is not fixed as a phenotype as described above. Glucose tolerance in each mouse should, therefore, be monitored and only mice with established glucose intolerance should be used for experiments. REFERENCES Benoist, C. and Mathis, D. (1997) Cell death mediators in autoimmune diabetes-no shortage of suspects. Cell, 89, 1–3. Coleman, D.L., Kuzuva, J.E. and Leiter, E.H. (1990) Effect of diet on incidence of diabetes in nonobese diabetic mice. Diabetes, 39, 432–436. Dietrich, W., Miller, J., Steen, R., Merchant, M., Damron, D., Nahf, R., et al. (1994) A genetic map of the mouse with 4006 simple sequence length polymorphisms. Nature Genetics, 7, 220–245. Elliott, R., Reddy, S., Bibby, N. and Kida, K. (1988) Dietary prevention of diabetes in the non-obese diabetic mouse. Diabetologia, 31, 62–64. Fox, H. (1992) Androgen treatment prevents diabetes in nonobese diabetic mice. J Exp Med, 175, 1409–1412. Ghosh, S., Palmer, S.M., Rodrigues, N.R., Cordell, H.J., Hearne, C.M., Cornall, R.J., et al. (1993) Polygenic control of autoimmune diabetes in nonobese diabetic mice. Nat. Genet., 4, 404–9. Hasegawa, G., Hata, M., Nakano, K., Kondo, M. and Kanatsuna, T. (1992) Diabetic syndrome in the NON mouse. In Current concepts of a new animal model: the NON mouse, edited by N.Sakamoto, N.Hotta, and K.Uchida, pp. 41–50. Amsterdam: Elsevier Science Publishers, BV. Hattori, M., Buse, J., Jackson, R., Glimcher, L., Dorf, M., Minami, M., et al. (1986) The NOD mouse: recessive diabetogenic gene in the major histocompatibility complex. Science, 231, 733–735. Ikegami, H., Eisenbarth, G.S. and Hattori, M. (1990) Major histocompatibility complex-linked diabetogenic gene of the nonobese diabetic mouse. Analysis of genomic DNA amplified by the polymerase chain reaction. J. Clin. Invest., 85, 18–24. Ikegami, H. and Makino, S. (1993) Genetic susceptibility to insulin-dependent diabetes mellitus: from the NOD mouse to man. In Frontiers in diabetes research. Lessons from animal diabetes, edited by E.Shafrir, pp. 39–50. London: Smith-Gordon.

54

H.IKEGAMI AND S.MAKING

Ikegami, H., Makino, S., Harada, M., Eisenbarth, G. and Hattori, M. (1989a) The cataract Shionogi mouse, a sister strain of the non-obese diabetic mouse: similar class II bur different class I gene products. Diabetologia, 31, 254–258. Ikegami, H., Makino, S. and Ogihara, T. (1996) Molecular genetics of insulin-dependent diabetes mellitus: analysis of congenic strains. In Lessons from Animal Diabetes VI, edited by E.Shafrir, pp. 33–46. Boston: Birkhauser. Ikegami, H., Makino, S., Yamato, E., Kawaguchi, Y., Ueda, H., Sakamoto, T., et al. (1995) Identification of a new susceptibility locus for insulin-dependent diabetes mellitus by ancestral haplotype congenic mapping. J. Clin. Invest., 96, 1936–1942. Ikegami, H., Yano, N., Sato, T. and Hattori, M. (1989b) Immunogenetics and immunopathogenesis of the NOD mouse. In Immunotherapy of Diabetes and Selected Autoimmune Diseases, edited by E.GS, pp. Boca Raton: CRC Press. Kaku, K., Fiedorek Jr, F.T., Province, M. and Permutt, M.A. (1988) Genetic analysis of glucose tolerance in inbred mouse strains: evidence for polygenic control. Diabetes, 37, 707–713. Leiter, E.H., Serreze, D.V. and Prochazka, M. (1990) The genetics and epidemiology of diabetes in NOD mice. Immunology Today, 11, 147–149. Makino, S., Hayashi, Y., Muraoka, Y. and Tochino, Y. (1985a) Establishment of the nonobesediabetic (NOD) mouse. In Current topics in clinical and experimental aspects of diabetes mellitus, edited by N.Sakamoto, H.K.Min, and S.Baba, pp. 25–32. Amsterdam: Elsevier Science Publisher, B.V. Makino, S., Kunimoto, K., Muraoka, Y. and Katagiri, K. (1981) Effect of castration on the appearance of diabetes in NOD mouse. Exp. Anim., 30, 137–140. Makino, S., Kunimoto, K., Muraoka, Y., Mizushima, Y., Katagiri, K. and Tochino, Y. (1980) Breeding of non-obese, diabetic strain of mice. Exp. Anim., 29, 1–13. Makino, S., Muraoka, Y., Harada, M., Kishimoto, Y. and Konishi, T. (1988) Characteristics of the NOD mouse and its relatives. In New lessons from diabetes in animals, edited by R. Larkins, P.Zimmet, and D.Chisholm, pp. 747–750. Amsterdam: Elsevier Science Publishers B.V. Makino, S., Muraoka, Y., Kishimoto, Y. and Hayashi, Y. (1985b) Genetic analysis for insulitis in NOD mice. Exp. Anim., 34, 425–432. McAleer, M.A., Reifsnyder, P., Palmer, S.M., Prochazka, M., Love, J.M., Copeman, J.B., et al (1995b) Crosses of NOD mice with the related NON strain: a polygenic model for IDDM. Diabetes, 44, 1186–1195. Mohan, C., Morel, L., Yang, P., Watanabe, H., Croker, B., Gilkeson, G., et al. (1999) Genetic dissection of lupus pathogenesis: a recipe for nephrophilic autoantibodies. J. Clin. Invest., 103, 1685–1695. Muraoka, Y., Matsui, S., Watanabe, H. and Makino, S. (1992) Histopathological observation of the development of glomerular intracapillary deposits in the NON mouse. In Current concepts of a new animal model: the NON mouse, edited by N.Sakamoto, N.Hotta, and K. Uchida, pp. 107–120. Amsterdam: Elsevier Science Publishers, BV. Ohsugi, T. and Kurosawa, T. (1994) Increased incidence of diabetes mellitus in specific pathogen-eliminated offspring produced by embryo transfer in NOD mice with low incidence of the disease . Laboratory Animal Science, 44, 386–388. Ohtori, H., Yoshida, T. and Inuma, T. (1968) Small eye and catarct, a new dominant mutation in the mouse. Exp. Animals, 17, 91–96. Oldstone, M.B. (1988) Prevention of Type 1 diabetes in nonobese diabetic mice by virus infection. Science, 239, 500–502. Podolin, P.L., Pressey, A., DeLarato, N.H., Fischer, PA., Peterson, L.B. and Wicker, L.S. (1993) I-E+nonobese diabetic mice develop insulitis and diabetes. J. Exp. Med., 178, 793–803. Podorin, P., Denny, P., Armitage, N., Lord, C., Hill, N., Levy, E., et al. (1998) Localization of two insulin-dependent diabetes (Idd) genes to the Idd10 region on mouse chromosome 3. Mamm. Genome, 9, 283–286. Podorin, P., Denny, P., Lord, C., Hill, N., Todd, J., Peterson, L., et al. (1997) Congenic mapping of the insulindependent diabetes (Idd) gene, Idd10, localizes two genes mediating the Idd10 effect and eliminates the candidate Fcgr1. J. ImmunoL, 159, 1835–1843. Pozzili, P., Signore, A., Williams, A. and Beales, P. (1993) NOD mouse colonies around the world—recent facts and figures. Immunology Today, 14, 193–196.

THE NOD MOUSE AND ITS RELATED STRAINS

55

Prochazka, M., Leiter, E.H., Serreze, D.V. and Coleman, D.L. (1987) Three recessive loci required for insulindependent diabetes in nonobese diabetic mice. Science, 237, 286–289. Prochazka, M., Serreze, D., Worthen, S. and Leiter, E. (1989) Genetic control of diabetogenesis in NOD/Lt mice. Development and analysis of congenic stocks. Diabetes, 38, 1446–1455. Saito, T., Sato, H., Kudo, K., Oikawa, S., Shibata, T., Hara, Y., et al. (1989) Lipoprotein glomerulopathy: glomerular lipoprotein thorombi in a patient with hyperlipoproteinemia. Am. J. Kidney Dis., 13, 148–153. Suzuki, T., Yamada, T., Fujimura, T., Kawamura, E., Shimizu, M., Yamashita, R., et al. (1987) Diabetogenic effects of lymphocyte transfusion on the NOD or NOD nude mouse. In Immune-deficient animals in biochemical research, edited by Rygaard, Brunner, and Graem-Thomsen, pp. 112–116. Basel: Karger. Tisch, R. and McDevitt, H. (1996) Insulin-dependent diabetes mellitus. Cell, 85, 291–297. Todd, J., Aitman, T., Cornall, R., Ghosh, S., Hall, J., Hearne, C., et al. (199D Genetic analysis of autoimmune Type 1 diabetes mellitus in mice. Nature, 351, 542–547. Todd, J.A. (1991) A protective role of the environment in the development of Type 1 diabetes? Diabetic Medicine, 8, 906–910. Ueda, H., Ikegami, H., Yamato, E., Fu, J., Fukuda, M., Shen, G., et al. (1995) The NSY mouse, a new animal model of spontaneous NIDDM with moderate obesity. Diabetologia, 38, 503– 508. Wainai, H., Maruyama, T., Takei, I., Kataoka, K., Saruta, T. and Ogata, K. (1992) Renal pathological findings and abnormal GTT. In Current concepts of a new animal model: the NON mouse, edited by N.Sakamoto, N.Hotta, and K.Uchida, pp. 149–157. Amsterdam: Elsevier Science Publishers, BV. Wicker, L., Todd, J. and Peterson, L. (1995) Genetic control of autoimmune diabetes in the NOD mouse. Annu. Rev. Immunol., 13, 179–200. Wicker, L.S., Appel, M.C., Dotta, F., Pressey, A., Miller, B.J., DeLarato, N.H., et al. (1992) 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., 176, 67–77. Wicker, L.S., Miller, B.J., Coker, L.Z., McNally, S.E., Scott, S., Mullen, Y., et al. (1987) Genetic control of diabetes and insulitis in the nonobese diabetic (NOD) mouse. J. Exp. Med., 165, 1639–1654. Wicker, L.S., Todd, J.A., Prins, J.-B., Podolin, P.L., Renjilian, R.J. and Peterson, L.B. (1994) Resistance alleles at two non-major histocompatibility complex-linked insulin-dependent diabetes loci on chromosome 3, Idd3 and Idd10, protect nonobese diabetic mice from diabetes. J. Exp. Med., 180, 1705–1713. Wilberz, S., Partke, H., Dagnaes-Hansen, F. and Herberg, L. (1991) Persistent MHV (mouse hepatitis virus) infection reduces the incidence of diabetes mellitus in non-obese diabetic mice. Diabetologia, 34, 2–5. Williams, A., Krug, J., Lampeter, E., Mansfield, K., Beales, P., Signore, A., et al. (1990) Raised temperature reduces the incidence of diabetes in the NOD mouse. Diabetologia, 33, 635– 637.

3. OBESITY/DIABETES IN MICE WITH MUTATIONS IN THE LEPTIN OR LEPTIN RECEPTOR GENES LIESELOTTE HERBERG and EDWARD H.LEITER Diabetes Research Institute at the Heinrich-Heine-University of Düsseldorf, Germany and The Jackson Laboratory, Bar Harbor, Maine 04609, USA

INTRODUCTION This chapter is meant to serve as a “primer” for use of two of the most intensively-studied monogenic obesity mutations in the mouse, obese (ob) and diabetes (db). The previous observations that both mutations, although mapping to separate chromosomes, produced nearly identical obesity/diabetes syndromes when studied on a common inbred background, coupled with the results of parabiosis studies (Coleman, 1973, 1978), suggested that the two mutations affected a common pathway. This has been confirmed by recent discoveries showing the ob mutation to be a defect in the gene encoding leptin, and the db mutation to be a defective leptin receptor. Several recent reviews contrast these two mutations with other monogenic obesity-producing genes in the mouse (Kim et al., 1998; McIntosh and Pederson, 1999). In the rat, the fatty (fa) mutation on Chromosome 5 (and its allele, corpulent, cp) are homologs of the mouse db mutation (Chua et al., 1996a, 1996b). This chapter will help to integrate the extensive early literature describing the physiologic, biochemical, and behavioral effects of the mouse mutations with the more current information gained from the molecular information and the availability of recombinant leptin protein. CURRENT NOMENCLATURE FOR THE MOUSE MUTATIONS Since the discovery that ob is a mutation in the leptin structural gene, and db is a mutation in the leptin receptor gene, the nomenclature has been changed to reflect

OB/DB IN MICE WITH MUTATIONS

57

Table 1 Current genetic nomenclature for mouse ob and db mutations. Common Name Gene Symbol Formal Designation

Earlier Symbol Gene Product Current Name Synonyms

obese

Lepab−J

ob

diabetes

Leprdb-J

db

leptin leptin mRNA leptin receptor leptin receptor mRNA

Ob-protein Ob-mRNA Ob-receptor Ob-receptor mRNA

their molecular bases as noted in Table 1 below. This new nomenclature will henceforth be used in preference to the inaccurate descriptors “Ob-protein” and “Ob-receptor”. Gene symbols are italicized whereas gene products (leptin, leptin receptor) are not. It should be noted that the gene symbols for the rat homolog of the leptin receptor mutation have also been changed to reflect this new nomenclature. Hence, the fatty mutation is now referred to as Leprfa and the corpulent mutation as Leprfacp. As noted in Table 2 below, spontaneous mutations at the Leprdb locus are relatively frequent, whereas only two mutations at the Lepob locus have been described. Throughout this chapter, Lepab−J and Leprdb−J will henceforth be denoted as Lepab and Leprdb. The symbols Lepab and Leprdb will denote homozygous mutants in place of Lepob/Lepob and Leprdb/Leprdb. The wild-type alleles are denoted simply as Lep and Lepr respectively. However, in descriptions of specific crosses entailing use of linked coat color mutations, any wild-type allele will be noted by “+”. Origin of the C57BL/6J (B6) and the C57BLKS/J (BKS) strains Both the B6 and BKS strain can be traced to the C57BL/J strain founded by Little in 1921. In the 1930s, some substrains were separated; one of them was called C57BL/6J (B6). B6 mice sent to other places at F22 came back to The Jackson Laboratory, Bar Harbor, ME, USA at F24 in 1948 after a devastating fire had destroyed the mouse colonies there in 1947. C57BL/6J mice sent to Dr. Nathan Kaliss at Sloan-Kettering prior to the fire in 1947 were taken back by him to The Jackson Laboratory in 1948. Tissue transplantation experiments revealed that Dr. Kaliss’s mice were not histocompatible with the B6 stock reimported to The Jackson Laboratory. The Kaliss stock was found to express a different H2 haplotype (H2d instead of H2b). At that time, it was erroneously thought that H2 comprised a single gene, so the Kaliss stock was initially assumed to be a point mutation at the H2b locus of B6 mice. Hence, a substrain designation, C57BL/Ks (the Ks devoting Kaliss) was assigned. However, subsequent genomic analysis of this strain revealed that the BKS genome contains about 84% B6-like and 16% DBA-like alleles indicating a genetic contamination early in the strain’s history (Naggert et al., Table 2 Additional alleles of Lepob and Leprdb. Gene Symbol

Spontaneous Occurrence of the Alleles:

Lepab−2J

in the SM/J inbred strain at The Jackson Laboratory;

Leprdb-2J

in a stock that has been discontinued; in the 129/J strain at The Jackson Laboratory;

Leprdb−3J

Address correspondence to: Lieselotte Herberg, Diabetes Research Institute at the Heinrich-HeineUniversity Düsseldorf, Aufm Hennekamp 65, 40225 Düsseldorf, Germany. Fax: 49–211–338–2603

58

L.HERBERG AND E.H.LEITER

Gene Symbol

Spontaneous Occurrence of the Alleles:

Lepab−2J

in the SM/J inbred strain at The Jackson Laboratory;

Leprdb-4J

in the BXD-16 stock of Dr. Ben Taylor at The Jackson Laboratory†; in the DW inbred strain at the Institut Pasteur in Paris; in the DW inbred strain at the Institut Pasteur in Paris; in a heterogenous strain at the Institute of Genetics in Edinburgh; in the B10.D2-H8b(57N)/SnJ congenic stock at The Jackson Laboratory*; in the CBA/J strain at The Jackson Laboratory*.

Leprdb-Pas1 Leprbd−Pas2 Leprdb−ad Leprdb-dmpg (dumpling) Leprdb-nnd (rotund)

† Dr. Ben Taylor, The Jackson Laboratory, personal communication (mutation no longer maintained) * Dr. Jung-Han Kim, The Jackson Laboratory, personal communication

1995). Hence, BKS is not a mutation bearing substrain of B6, but rather is a recombinant congenic strain. To indicate this distinction, the Mouse Genome Nomenclature Committee has changed the string descriptor for the BKS substrain from C57BL/KsJ to C57BLKS/J. Influence of B6 and BKS background modifiers on the expression of Lepob and Leprdb The major effects of inbred strain background on development of the obesity/ diabetes syndrome produced by mutations at the Lepob and Leprdb loci have been well-reviewed (Coleman and Hummel, 1973, 1975; Coleman, 1978, 1981, 1982a; Leiter and Herberg, 1997; McIntosh and Pederson, 1999). The Lepob mutation on Chromosome 6 was discovered at The Jackson Laboratory in a multiple recessive (“V”) stock in 1949 (Ingalls et al., 1950). It was immediately recognized because of the marked obesity and hyperphagia exhibited by homozygous mutant mice. As was customary with most mutations developed at The Jackson Laboratory, the Lepob mutation was subsequently transferred to the B6 inbred strain background, initially by standard backcrossing (Drasher et al., 1955). The Lepob mutation on the B6 background produces a juvenile onset obesity, hyperinsulinemia with increasing insulin resistance, but a hyperglycemia that was relatively mild and transient. This remission from chronic hyperglycemia was correlated with a sustained hypertrophy of pancreatic islets primarily contributed by hyperplasia of the B-cell mass. The Leprdb mutation is a recessive mutation on Chromosome 4 that occurred spontaneously in the BKS inbred strain in 1966 (Hummel et al., 1966). This mutation, as well as a second mutation (Leprdb–2J) arising in a mixed background stock (Hummel et al., 1972) and backcrossed to BKS was named “diabetes” because the obesity/diabetes syndrome did not remit as it did in B6-Lepob mice. Instead, an early hyperinsulinemia was not sustained; a progressively more severe hyperglycemia was correlated at the morphologic level with pancreatic B-cell necrosis and islet atrophy. During this “end stage” period, a precipitous drop in insulin occurs concomitant with islet atrophy and B-cell degeneration as well as a loss of body weight. Lifespan of the mutants is shortened compared to lean littermate controls. Although B6 and BKS represent two very closely-related inbred strains, these two backgrounds interact with either the Lepob or the Leprdb mutation in strikingly different ways. The inbred strain background effect on the severity of the obesity/diabetes syndromes was first elucidated in the course of transferring the Leprdb−2J allele through 5 backcross cycles (designated as an N6 generation) to each inbred strain background (Hummel et al., 1972). The Leprdb−2J allele produced the same diabetogenic effects on the BKS background as the original Leprdb allele. In contrast, the same mutation appeared to produce an identical phenotype as the Lepob mutation on the B6 background. An

OB/DB IN MICE WITH MUTATIONS

59

identical obesity syndrome uncomplicated by permanent diabetes due to a similar massive increase in islet size and numbers was observed when the Leprdb mutation was studied on the B6 inbred strain background (Coleman, 1978; Gapp et al., 1983). The exact nature of the background modifiers contributing to the more severe diabetes syndrome on the BKS background has not yet been established. Although the two strains differ at the major histocompatibility complex (MHC), segregation analysis has eliminated this complex locus as a diabetogenic determinant (Leiter et al., 1987a). Rather, gender-associated factors were implicated, especially those associated with aberrant regulation of estrogen and dehydroepiandrosterone (DHEA) sulfotransferases (Leiter and Chapman, 1994). Initial islet hypertrophy/hyperplasia developing in BKSLeprbd and Lepob mutants is limited in duration, whereas this compensatory response at the islet level is sustained in B6-Leprdb and Lepob mutant mice. Because B6-+/+ islet proliferative rate in vitro was twice that observed in BKS-+/+ islets (Swenne and Andersson, 1984), the differential background effect was suggested to reflect intrinsic differences in B-cell proliferative capacity. However, an intrinsic BKS background-mandated limitation in B-cell proliferative activity has been questioned by the finding that another recessive mutation on the BKS background, the fat allele at the carboxypeptidase E (Cpe) locus is not associated with β-cell atrophy, but rather with hypertrophied/hyperplastic islets (Leiter et al., 1999). The difference in B-cell replicative ability may be a secondary response of the BKS genome to chronically elevated glucose in Leprdb or Lepob mutations on the BKS background since studies in vitro using wild-type BKS versus B6 islets exposed to high glucose show differential BKS induction of endogenous retroviral genomes (Leiter et al., 1986). If hyperglycemia is controlled, as can be done with estrogen therapy (Prochazka et al., 1986), then continued islet hypertrophy/hyperplasia can be observed in BKS-Leprdb mutants. In a F2 cross, higher hepatic malic enzyme activity level characteristic of grandparental B6-Leprdb mice was associated with less severe hyperglycemia while the lower activity characteristic of BKS-Leprdb grandparental mice was associated with greater diabetes severity (Coleman, 1992). A gene, malic enzyme regulator (Mod1r) on Chromosome 12, represented a possible candidate gene. In a repetition of this F2 cross in which atherogenic responses to a high fat diet was tested, an atherosclerosis susceptibility locus on Chromsome 12 conferring higher susceptibility to BKS was indicated (Mu et al., 1999). However, this locus did not appear to control severity of diabetes under the dietary conditions used. It should be noted that when B6-+/+ and BKS-+/+ lean male mice were tested for diabetogenic responses elicited by a “Western-style” high (40%) fat diet, B6 males rather than BKS males proved to be more susceptible to induction of obesity accompanied by hyperglycemia and hyperinsulinemia (Surwit et al., 1994). Hence, genes conferring higher atherogenic susceptibility to BKS mice do not appear to be identical to those conferring increased diabetes susceptibility (Mu et al., 1999). The expression of the Leprdb mutation maintained in strains other than BKS and B6 has been discussed recently (Leiter and Herberg, 1996) and are summarized in Table 3 below. Table 3 Diabetes-susceptible/resistant strains. Susceptible

Gender

Resistant

Gender

(C57BLKS/J) BKS CBA/Lt DBA/2J SM/J* C3HeB/FeJ

both sexes males males both sexes males

(C57BL/6J) B6 V stock 129/J MA/J

both sexes both sexes both sexes both sexes

*(Leiter, unpublished)

60

L.HERBERG AND E.H.LEITER

METABOLIC DEFECTS AND THE SEARCH FOR THE PRIMARY LESIONS Mice with inappropriate hyperglycemia and obesity exhibit metabolic abnormalities which, albeit in a less pronounced manner, are found in rats with hypothalamic lesions (Bray and York, 1979). Reproductive failure coupled with inability to control appetite clearly pointed to neuroendocrine defects in both Lepob and Leprdb mice. With regard to the etiology of the obese-hyperglycemic syndrome, a primary involvement of structures of the central nervous system, of the hypothalamic-pituitary-adrenal axis, of brain neurotranstransmitters, and of the autonomous nervous system have been discussed in various reviews (Bray and York, 1971; Bray, 1984; Jeanrenaud, 1978; Jeanrenaud, 1985; Jeanrenaud, 1995; Shafrir, 1992; Proietto and Thorburn, 1994). Adrenal and thyroidal function have been studied in Lepob and Leprdb mice by numerous groups. Whereas hypercorticism usually present in the mutants exacerbates the hyperglycemic syndrome, the role of thyroid function is still unclear and controversial (Dubuc, 1989). Dysfunction of both systems, however, seems to be secondary and can modulate the clinical picture. Inherent defects in the B-cell itself have been discussed with special regard to the dynamics and magnitude of glucose-induced insulin secretion and the deranged regulation of β-cell membrane potential (Flatt et al., 1992). However, the -cell’s candidacy of being the primary target of the Leprdb mutation (on the BKS background) was excluded by the observation that treatment with estrogen (estrone or 17B-estradiol) blocked both hyperglycemia and subsequent B-cell necrosis while the development of obesity remained unaffected (Prochazka et al., 1986; Leiter et al., 1987b). Feeding experiments on rats with or without hypothalamic lesions revealed that “the young rat adjusts its food intake so precisely to its energy needs that its fat stores remain almost constant”. This observation led Kennedy (1950) to state “Reasons are advanced for regarding the hypothalamic mechanisms as being sensitive to chemical changes in the blood”. The hypothesis of a humoral factor linking the fat cell with hypothalamic centres was in agreement with the observation that parabiosis of Lepob with lean mice was followed by a suppressed or completely prevented weight gain in the Lepob partner of the pair. However, after separation with skin flaps from each mouse left on the partner Lepob mice started again gaining weight (Hausberger, 1958). Moreover, transplants of gonadal adipose tissue from lean B6-+/+ donors under the kidney capsule of B6-Lepob recipients failed to prevent the latter from gaining weight as well (Ashwell et al., 1977; Meade et al., 1979). In either study, the lack of an adequate innervation seems to be the cause of the failure rather than Hausberger’s conclusion “that this type of obesity is caused not by the qualities of adipose tissue itself but by the lack of a factor which can be transmitted by successful parabiosis.” These studies failed to demonstrate the link between fat cell and hypothalamus. However, the observation that donor fat cells took on the morphological characteristics of the host anticipated the hypothesis that a humoral factor present in only one partner of a lean/obese parabiosed pair would be able to regulate adipose tissue mass in the factor-deficient partner (Ashwell et al., 1977). Later, in studying whether circulating factors might be responsible for the increased insulin secretion, an early metabolic abnormality in both Lepob and Leprdb mice on either the B6 or BKS background, Coleman and Hummel (1969a) observed that food intake was drastically reduced in the lean (+/+) partner parabiosed with a mutant (Leprdb) mouse. Whereas, in the lean parabiont, blood glucose and body weight dropped until it died of starvation the Leprdb partner of the pair remained hyperglycemic and gained weight. In a second set of experiments Coleman (1973) compared the effect of Lepob and Leprdb on a congenic (B6) background. Parabiosis of Lepob with Leprdb mice led to the dramatic following of events in the Lepob partner as had been observed in the foregoing experiment in the lean partner (Coleman and Hummel, 1969a), namely, the Lepob parabiont stopped eating, developed hypoglycemia, and died of starvation. Coleman (1973) concluded that

OB/DB IN MICE WITH MUTATIONS

61

the Leprdb mouse produces a satiety factor which is missing in the Lepob mouse. Whereas the satiety centre of the Leprdb mouse is insensitive to its own satiety factor the satiety centre of the Lepob mouse responds to the factor if it is provided by either the Leprdb or the lean parabiont. This hypothesis was proven correct 21 years later. IDENTIFICATION OF THE PRIMARY MOLECULAR DEFECTS The Leptin gene. Confirmation of the Coleman hypothesis concerning the relationship between the Lepob and Leprdb mutations came two decades later with the positional cloning of the obese gene and identification of its product, leptin, as a new protein hormone (Zhang et al., 1994). Shortly thereafter, the diabetes mutation was identified as a defect in the unlinked leptin receptor (Tartaglia et al., 1995). These molecular discoveries revolutionized our understanding of the role of the adipocyte in the neuroendocrine control of body fat mass. Furthermore, the endocrine organ fat cell, known to synthesize estrogen (Simpson et al., 1989) and angiotensinogen (Frederich et al., 1992) was now found to synthesize leptin as well which orpened a fascinating area of research. The positional cloning of the leptin and leptin receptor genes was preceded by the generation of high resolution molecular maps localizing the genes relative to chromosomal markers (Bahary et al, 1990; Friedman et al, 1991; Bahary et al, 1991; Bahary et al., 1993). The wild-type mouse Leptin gene contains 3 exons and 2 introns, with the coding sequence in exons 2 and 3 producing a 4.5 kb mRNA. The 167-amino acid secreted protein made from this message was named leptin (Greek leptós=thin). Leptin is synthesized in white adipose tissue with different mRNA levels in different localizations. Other sites of leptin synthesis are brown adipose tissue, the placenta (Trayhurn et al., 1998), the stomach (Bado et al., 1998), and the ovary (Chebab et al, 1996). Leptin secreted into the blood stream after ingestion of meals, binds to leptin receptors in the hypothalamic area and the chorioid plexus and leads to a suppression of food intake. The anorectic effect of leptin is supported by an increased secretion of cholecystokinin (CCK) as well as a decreased synthesis and secretion of neuropeptide Y (Friedman and Halaas, 1998). Leptin binds also to leptin receptors of pancreatic islets [β-, α-, and δ-cells express leptin receptor (Fehmann et al., 1997; Kieffer et al., 1996)], and inhibits insulin secretion in vivo (Pelleymounter et al., 1995) and in vitro (Emilsson et al., 1997; Poitout et al., 1998). In Lepob mice, a nonsense mutation (CØT changing an arginine at position 105 to a stop codon) results in the synthesis of a truncated, inactive protein that probably is rapidly degraded in the fat cell (Zhang et al., 1994). The manifold increase of leptin mRNA and protein in adipocytes and serum of Leprdb mice reflects a dys-functional feedback loop between the hypothalamic centre that cannot bind leptin and the periphery as proposed by Coleman (1973). A second mutation, named ob2J (now Lepob–2J) has been shown to be unable to synthesize mature leptin mRNA due to an insertion of a retroviral-like transposon into the first intron of the leptin gene (Moon and Friedman, 1997). Thus, the serum of mice carrying this allele (SW/Ckc-Lepob−2JLepob–2J) does not contain either mature leptin mRNA or leptin protein. Accordingly, the phenotypes of mice presenting either mutation are identical. Leprdb encodes the leptin receptor The leptin receptor, often referred to in the literature as “Ob-R”, is a member of the IL-6 cytokine family of receptors. It was identified by means of cell lines and tissue binding survey with leptin-alkaline phosphatase

62

L.HERBERG AND E.H.LEITER

fusion proteins. The receptor was localized in the mouse chorioid plexus and the hypothalamus as well as in other tissues (Tartaglia et al., 1995). The leptin receptor gene encodes five alternatively spliced forms of the leptin receptor (denoted as Ra—Re) with an identical extracellular ligand-binding domain and a long intracellular domain characteristic of the Rb form only. This long form is required for signal transduction processes (Lee et al, 1996; Chen et al, 1996; Ghilardi et al, 1996). The spliced receptor variant identified in the mouse carrying the Leprdb mutation shows a 106 bp insertion at the splice junction predicting a premature stop codon and changing Rb to Ra (Lee et al., 1996). Since the expression of other “Ob-R” forms is unaffected in mice carrying Leprdb, the weight-reducing effect of leptin depends on the Rb form. The Re form appears to be a soluble leptin binding protein (Gavrilova et al, 1997). Three mutant alleles of the leptin receptor gene have been characterized at the molecular level, Leprdb-J, Leprdb-3J and Leprdb-Pas1. Whereas the Leprdb-J mutation contains a 106-bp insert with a premature stop codon due to an abnormal splicing (Lee et al., 1996), the Leprdb−3J mutation results in a leptin receptor truncated at amino acid 625 (Lee et al, 1997) and the Leprdb-pas1 mutation results in a stop codon one residue after amino acid 281 (Li et al, 1998). Thus, as a result of the splicing mutation in Leprdb-J mice, the cytoplasmic region involved into signal transduction is missing whereas Leprdb-3J and Leprdb-Pas1 mice exhibit a truncated receptor in the extracellular region. Leptin mRNA in tissues and relative or absolute plasma leptin levels in mice Only fully developed white adipocytes express leptin mRNA (MacDougald et al, 1995). Leptin mRNA is markedly higher in fat depots from B6-Lepob and BKS-Leprdb mice when compared with the respective lean littermates. Leptin mRNA expression correlates well with fat cell size rather than fat cell number: in Lepob mice of the Aston, U.K., substrain (see below) and in BKS-Leprdb mice leptin mRNA expression was found to be higher in intraabdominal fat pads with large fat cells when compared with subcutaneous fat pads with small cells (Trayhurn et al., 1995; Maffei et al., 1995b). In brown adipose tissue leptin mRNA is markedly lower when compared with white adipose tissue (Moinat et al., 1995; Maffei et al., 1995a) and differs according to ambient temperature and functional stage of brown adipose tissue (Cancello et al., 1998). Fully differentiated uncoupling protein-1 (UCP-1)-positive multilocular brown adipocytes are leptin-negative whereas unilocular, UCP-negative brown adipocytes localized mainly at the periphery of the intrascapular brown adipose tissue depot express leptin mRNA (Cinti et al., 1997). Sensitive radioimmunoassay and ELISA kits for measurement of mouse leptin are available from several suppliers. Serum levels vary depending upon inbred strain, gender (females higher than males), and age. In lean mice, serum leptin levels are usually found in a range between 5–20 ng/ml. Leptin is missing in plasma from Lepob mice due to a nonsense mutation in codon 105. It is about 9 times higher in plasma from BKSLeprdb mice when compared with lean littermates (Frederich et al., 1995). Plasma leptin levels correspond with BMI (body mass index=weight in grams/nose-to-anus length in centimeters2): B6-tub/tub (“tubby”) mice and B6-Avy/+ (yellow agouti) mice exhibit plasma leptin levels twice and 10 times the levels seen in B6-+/+ mice. In BKS-Cpefat/Cpefat and BKS-Leprdb/Leprdb mice, plasma leptin levels are 5 times and 10 times higher than in lean littermates, respectively (Maffei et al., 1995b). Plasma leptin levels are affected by heterozygosity of the Leprdb mutation. Adjusted for fat mass, plasma leptin levels are 9.6, 11.5, and 6.5 ng/ml in B6-+/+, B6-Leprdb/+, and B6-Lepob/+ mice, respectively. In B6 mice heterozygous at both Lepob and Leprdb (compound heterozygotes), leptin levels adjusted for fat mass are reported to be 6.2 ng/ml. The equivalence of plasma leptin levels in B6-Lepob/+ and B6-Lepob/+, Leprdb/ + mice (6.5 vs 6.2 ng/ml) as well as the dissimilarity in plasma leptin levels in B6-Lepob/+ and B6-+/+ mice (6.5 vs 9.6 ng/ml) led the authors to suggest that the single wild-type allele in the B6-Lepob/+ mouse

OB/DB IN MICE WITH MUTATIONS

63

maximally produces leptin which, however, is not sufficient to compensate for the defective (Lepob) allele (Chung et al, 1998). Effect of leptin administration to Lepob and Leprdb mice Leptin not only plays a major role in food intake but also is involved in various regulatory mechanisms of metabolism in the neuroendocrine, the endocrine, and the immune system (Friedman and Halaas, 1998). The close link between central neuronal networks controlling food intake and the fat cell sending its feedinginhibitory signal was demonstrated in B6 mice by various groups at the same time (Campfield et al., 1995; Halaas et al., 1995; Pelleymounter et al., 1995; Stephens et al., 1995; Weigle et al., 1995). The lack of endogenously-produced leptin was compensated by the intraperitoneal or intracranial injection of recombinant leptin. In these experiments, recombinant leptin injected into Lepob mice, was followed by a dosedependent decrease in food intake and a loss of body weight. Furthermore, a recent study has shown that in B6-Lepob mice, a 5- to 6-hr i.v. infusion of leptin resulted in a 60% increased glucose turnover when compared with PBS-treated Lepob. Hepatic glucose output as well as glucose-6-phosphatase increased whereas hepatic phosphoenol pyruvate carboxykinase (PEPCK) decreased, suggesting that insulin-sensitivity was at least partly restored by leptin. Moreover, leptin-induced glucose uptake was markedly increased in brown adipose tissue (Burcelin et al, 1999). Therefore, one may conclude that the normalization of the increased metabolic efficiency after leptin administration (Pelleymounter et al., 1995) came about by a more complete oxidation of free fatty acids in the presence of extra endogenous carbohydrate. In Leprdb mice on either background, administration of recombinant leptin remains ineffective due to the defective leptin receptor. BEHAVIORAL CHARACTERISTICS OF B6-Lepab AND BKS-Leprdb MICE Physical activity and demand for high ambient temperature Male Lepob and Leprdb mice do not exhibit any intra-strain aggressiveness as do the lean controls. B6-Lepob and BKS-Leprdb mice exhibit a considerably reduced physical activity secondary to obesity (Yen and Acton, 1972). Due to a defective brown adipose tissue thermogenesis (Himms-Hagen, 1985) B6-Lepob mice prefer higher (25–35°C) ambient temperature (Wilson and Sinha, 1985) facilitating a mean core temperature of about 36.5°C (Carlisle and Dubuc, 1984). Lepob and Leprdb mice exhibit a higher rate of heat loss when compared with lean controls (Bellward and Dauncy, 1988). Therefore, during resting periods Lepob and Leprdb mice cling closely to each other in order not to lose heat. Food intake Hyperphagia, one of the main characteristics of Lepob and Leprdb mice, is not a prerequisite of obesity since adipose tissue mass of mutants pair-fed with lean mice remains still higher and even increases due to an increased metabolic efficiency in mutants (Coleman, 1985). The increase in meal size and the proportionally higher food intake in the light period of the light-dark cycle in male B6-Lepob mice when compared with controls is easily to understand in the light of the lack of leptin in Lepob mice. The equal number and periodicity of meals in Lepob and lean mice in either 12-hour period in the light-dark cycle points to a normal circadian feeding pattern in Lepob mice (Ho and Chin, 1988). When either gender of Lepob mice was studied, it became obvious that female Lepob ate fewer meals in the dark when compared with male Lepob

64

L.HERBERG AND E.H.LEITER

and that the number of meals ingested in the dark were higher in lean when compared with Lepob mice, respectively. The number of meals consumed during the light period was similar in Lepob and lean mice of either gender (Strohmayer and Smith, 1987). Both B6-Lepob and BKS-Leprdb mice show a clear preference for sweet liquid diets (Sprott, 1972). On a free-choice feeding B6-Lepob mice prefer a high-carbohydrate diet (Mayer et al., 1951). In this study the diets offered to mice consisted of about 90% protein, or 90% carbohydrate, or 90% fat. Therefore, the lesser intake (g) of the fat-enriched diet when compared with the slightly higher intake (g) of the highcarbohydrate diet resulted in a nearly 100% higher intake of calories from fat and reflects the poor adjustment to dietary diluted diets rather than a preference for high-fat diets as concluded by the authors and frequently cited in the literature. Autonomous nervous system Enhanced sensitivity to various forms of stress is also a characteristic of Lepob and Leprdb mice. The observation that food-deprived and physically restrained B6-Lepob mice developed gastric stress ulcers and became hypothermic (Greenberg and Ackerman, 1984) points to a possible relationship between the autonomous nervous system and the leptin-leptin receptor loop. Various hypothalamic nuclei containing leptin receptors are known to modulate the activity of both the sympathetic and the parasympathetic nervous system (Friedman and Halaas, 1998). Usually, sympathetic activity is reduced in genetic obesity (Bray, 1991). Thus, blood glucose was increased either by systemic administration of epinephrine to B6Lepob mice or by social stress evoked by different grouping conditions (Surwit and Williams, 1996). An exaggerated peripheral response to catecholamines is suggested to contribute to stress-induced hyperglycemia in B6-Lepob mice (Kuhn et al., 1987). Central injection of epinephrine or the α-2 agonist clonidine into B6-Lepoh mice was followed by an enhanced feeding response at the beginning of the dark period of the light-dark cycle (Currie and Wilson, 1993). The hyperphagic effect of norepinephrine and clonidine as well as the anorectic effect of 5-hydroxytryptamine (serotonin) was shown to be dose-dependent and nutrient selective in that B6-Lepob mice showed an increased preference for carbohydrate (Currie, 1993). Central effects The hypothalamic-pituitary-adrenal axis has been discussed with regard to its role in the development of the obese-hyperglycemic syndrome (Bray, 1991). Since leptin is known to stimulate corticotropin-releasing hormone (CRH) mRNA in the paraventricular nucleus (Friedman and Halaas, 1998), the observation that CRH decreases feeding, oxygen consumption, and grooming activity in B6-Lepob mice suggested a missing leptin stimulation of CRH release “rather than abnormal CRH action mechanisms” (Drescher et al., 1994). A direct leptin-corticosterone interaction seems to be unlikely since leptin affects energy and fat metabolism in the absence of an intact hypothalamic-pituitary-adrenal axis in B6-+/+ mice (Arvaniti et al., 1998a). A poor adjustment to diluted diets explained by “a defect in the satiety mechanism” (Parson et al., 1954) can be easily understood with regard to the missing food intake regulating leptin levels in serum of B6-Lepob mice. Opioids modulate food intake as well as behavioral activation when administered to BKS-Leprdb (Levine et al., 1982) and B6-Lepob mice (Shimomura et al., 1982; Calcagnetti et al., 1987). Experiments on the effect of the opiate antagonist, naxolone, on grooming, rearing, and jumping following an immobilization or heat stress or a combination of both revealed possible links between endogenous opioids (endorphins) and

OB/DB IN MICE WITH MUTATIONS

65

behavioral responses in both B6-Lepob and lean controls. Under the conditions tested, the naxolone effect was clearly stronger in Lepob when compared with the lean controls (Amir, 1981). PHYSIOLOGICAL AND BIOCHEMICAL CHARACTERISTICS IN B6-Lepob MICE B6-Lepob mice are characterized by hyperphagia, massive obesity, pronounced hyperinsulinemia, and transient hyperglycemia as well as abnormalities of the neuroendocrine, endocrine, and immune system. Whereas earlier adipose tissue transplantation studies in the 1950s and 1970s (Hausberger, 1958; Ashwell et al., 1977; Meade et al., 1979) had failed to demonstrate the link between the fat cell and the hypothalamus, the molecular identification of a defect in the leptin gene as the basis for the obese mutation, the finding that leptin is produced by normal fat cells, and the fact that the above-mentioned abnormalities in B6-Lepob mice are correctable by leptin administration (Friedman and Halaas, 1998) clearly demonstrate the existence of an adipocyte-neuroendocrine axis. Adipose tissue size The abnormal adipose tissue enlargement in B6-Lepob mice is due to hyperphagia, increased food utilization, increased lipogenesis and depressed lipolysis. Milk intake is identical in suckling Aston-Lepob and Aston-+/ + pups (Contaldo et al., 1981). Enlargement of adipose tissue, however, starts already before weaning since carcass energy (kJ) of Aston-Lepob mice is recorded to be 12% and 40% higher at day 10 and 17 of life, respectively, when compared with lean mice (Thurlby and Trayhurn, 1978). Total body fat content being 15% in 6- to 16-d-old B6-Lepob, increases to 60–75% in 1-yr-old Lepob mice whereas in controls body fat content remains constantly around 15% (L.H., unpublished observation). Adipose tissue localization In young B6-Lepob mice, the main portion of white adipose tissue is located subcutaneously in the inguinal and axillar region. In 1-mo-old mutants the ratio of total/intraperitoneal body fat is 7.1/1 and decreases to 4. 1/1 in 9-mo-old B6-Lepob mice. The increase in intraperitoneal adipose tissue in older mice is mainly due to an enlargement of visceral adipose tissue (L.H., unpublished observation). In older B6-Lepob mice, epididymal adipose tissue becomes less resilent, less pliable, and assumes a yellowish colour (Herberg and Coleman, 1977). Cellularity of adipose tissue B6-Lepob mice exhibit the hypertrophic-hyperplastic type of obesity. In 6- to 16-d-old mutants mean fat cell volume is about 6 times higher than in the controls; mean fat cell volume increases with aging. In 1-yr-old B6-Lepob and lean controls, fat cell volume enlarges by 5.5-fold and 3-fold, respectively, when compared with 6- to 16-d-old pups. In 1-yr-old B6-Lepob mice, mean fat cell volume is about 10 times that observed in the controls (Table 4). It is smallest in the retroperitoneal and largest in the subcutaneous pad (L.H., unpublished observation).

66

L.HERBERG AND E.H.LEITER

Table 4 Fat cell volume (×103µ3). Mice

day 6–16

>1yr

B6-+/+ B6-Lepob

33.73 198.70

111.49 1106.32

On testing the frequency of adipose cell size distribution in 3- to 18-wk-old B6-Lepob and B6-+/+ mice, Kaplan et al. (1976) observed an increase in the frequency of small cells in Lepob mice only, suggesting that, in mutants, the number of adipocytes is not prefixed. Accordingly, preadipocytes isolated from both the epididymal and retroperitoneal fat pad of B6-Lepob mice show a greater ability to proliferate than stromal-vascular cells from lean controls (Black and Bégin-Heick, 1995). Total fat cell number is highest in the subcutaneous and lowest in the gonadal fat pad (Johnson and Hirsch, 1972). Thermogenesis A defective dietary and cold-induced thermogenesis contributes to adipose tissue enlargement. Thus, overfeeding Aston-Lepob and lean mice with a cafeteria type diet revealed that Lepob mice deposit most of the extra ingested energy into fat, in contrast to lean controls (Trayhurn et al., 1982). About 70% of the heat production in non-shivering thermogenesis is produced in brown adipose tissue mitochondria. In B6-Lepob mice the heat-producing proton conductance system is less active on cold exposure when compared with controls (Hogan and Himms-Hagen, 1980). Both a diminished Na+/K+-ATPase (Bray and York, 1971) as well as a masking of nucleotide binding in mutants (Hogan and Himms-Hagen, 1980) contribute to the defective thermogenesis. A reduced binding of purine nucleotides (GDP, ADP) to the mitochondrial membrane of brown adipose tissue as well as an attenuation of the cold-induced increase in uncoupling protein-1 (UCP-1) mRNA are followed by reduced heat production and increased energy storage in B6Lepob when compared with controls (Reichling et al., 1988). The decreased GDP binding to brown adipose tissue mitochondria of 14-d-old Aston-Lepob (Goodbody and Trayhurn, 1982) confirms the observation that adipose tissue enlargement precedes hyperphagia in mutants. Measurements of oxygen consumption at different environmental temperature0s showed that Aston-Lepob mice spend less energy on thermoregulatory thermogenesis when compared with controls (Trayhurn and James, 1978). With increasing environmental temperature from 17°C to thermoneutrality at 33°C, lean mice accumulated more energy whereas Aston-Lepob showed little change in energy gain, reflecting their reduced energy expenditure (Thurlby and Trayhurn, 1979). B6-Lepob mice kept at 4°C become hypothermie and die. However, if a 24 hr period at 10°C precedes the 4°C period, the mice are capable to adapt (Coleman, 1982b). The defect in thermoregulation, therefore, is only partial and, as shown, not severe enough to explain the increased metabolic efficiency characteristic of B6-Lepob mice (Coleman, 1985). Lipogenesis De novo lipogenesis is markedly enhanced in B6-Lepob when compared with control mice. This is true especially in young animals. In Lepob mice, hepatic fatty acid synthesis is increased 6 fold per total liver and 2.2 fold per total small intestine (Memon et al., 1994). Recently, fatty acid translocase mRNA and plasma membrane fatty acid binding protein mRNA were reported to be increased in both hepatic and adipose tissue in B6-Lepob when compared with lean mice. The increase in hepatic microsomal acyl-CoA synthase activity suggests an increased esterification of free fatty acids in Lepob hepatic tissue (Memon et al., 1999).

OB/DB IN MICE WITH MUTATIONS

67

Hepatic lipogenesis accounts for about 50% of total lipogenesis in B6-Lepob and for 20% in lean control mice. Both hepatic and adipose tissue lipogenesis remain clearly enhanced under fasting conditions as shown in pair-feeding experiments. Although the high serum insulin levels favor adipose tissue enlargement, the rate of liver and adipose tissue lipogenesis does not show a strict proportionality to circulating serum insulin levels (Herberg and Coleman, 1977). In B6-Lepob mice reesterification is also enhanced due to an increased glycerol kinase activity (Stern et al., 1983) which, theoretically, is capable of phosphorylating about 30% of glyceride-glycerol. Lipolysis The role of lipolysis in relation to adipose tissue enlargement has been discussed previously (Herberg and Coleman, 1977). The marked increase in free fatty acids and glycerol in serum from starved B6-Lepob mice seems to reflect a high lipolytic rate as does the high basal and stimulated lipolysis per single fat cell. However, in white adipose tissue from B6-Lepob mice β3-adrenergic receptor function markedly decreased (Bégin-Heick, 1996). The lipolytic defect in adipose tissue, therefore, might be due to the reduced β3receptor function. Plasma insulin The developmental pattern of non-fasted serum insulin and blood glucose related to body weight (main group) and age (subgroup) of B6-Lepob mice of the former Düsseldorf colony is given in Figure 1. In adult lean controls, mean serum insulin level remained nearly constant throughout lifetime (<50 µU/ml). In B6Lepob mice, hyperinsulinemia was present already before weaning since serum insulin levels were twice as high in 6- to 16-d-old Lepob mice when compared with controls (data not shown). Serum insulin levels markedly increased with age. In 4-mo-old and 7-moold Lepob mice, serum insulin levels were 2–3 times and 20–30 times respectively the level observed in controls. The highest serum insulin levels were observed in 7-mo-old Lepob mice weighing about 75 g. Thereafter, serum insulin levels spontaneously declined and showed considerable variation among individual mutant mice. In general, highest insulin levels were observed in B6-Lepob males, indicating a more severe insulin resistance in mutant males as compared with mutant females. As been reviewed before (Friedman and Halaas, 1998), high serum insulin levels are correctable by leptin in a dose-dependent manner. The effects of fasting or different composition of diets on serum or plasma insulin concentrations have been discussed earlier (Herberg and Coleman, 1977). Insulin secretion and the adipo-insular axis in Lep ob mutants Both basal and stimulated insulin secretion is increased in B6-Lepob mice, under in vivo conditions as well as in vitro. Arginine stimulation is followed by a prompt and highly pronounced increase in insulin levels. After a glucose challenge, however, serum insulin initially drops and the subsequent increase is delayed. A preceeding 16-hr-fasting period abolishes the delayed glucose-induced insulin response, whereas argininestimulated insulin release remains unaffected by fasting. Likewise, long-term food-restriction keeping body weight of B6-Lepob mice within the normal range leads to a reduced glucose-induced insulin secretion, known as “starvation diabetes” (Herberg and Coleman, 1977). Recently, various factors known to affect insulin secretion by different mechanism have been reviewed (Chan, 1995).

68

L.HERBERG AND E.H.LEITER

Figure 1 Blood glucose and serum insulin in relation to body weight and age in B6-Lepob mice. Body weight (g), blood glucose (mg/dl) and serum insulin (µU/ml) were determined once per month. Blood was harvested at 8 a.m. from non-fasted mice. The mice were grouped according to body weight (main groups) and age (mo) (subgroups). Blood glucose and serum insulin values which were identical within a weight and/or an age group were pooled. Numbers of mice are given in the respective columns. Gender-specific insulin values are indicated. MEANS ± SEM.

The link between the adipocyte and the pancreatic β-cell was demonstrated by various groups. The findings of leptin receptor mRNA expression on pancreatic β-cells from normal rats (Kieffer et al., 1996) and of an insulin-induced up-regulation of leptin gene expression in adipose tissue (Mizuno et al., 1996) suggested an adipoinsular feedback loop or an adipo-insular axis. Thus, in normal and Aston-Lepob mice, leptin administration led to a marked decrease in plasma insulin levels whereas blood glucose increased (Kulkarni et al., 1997). In isolated pancreatic islets from normal mice and rats, leptin failed to depress basal insulin secretion whereas insulin secretion stimulated by glucose plus isobutylmethylxanthine (IBMX) was significantly inhibited (Poitout et al., 1998). However, in isolated pancreatic islets from Aston-Lep0b mice (islets from Lepob mice are very sensitive to leptin), recombinant leptin acutely inhibited both basal and glucose-stimulated insulin secretion (Emilsson et al., 1997). That leptin directly acts at the B-cell was demonstrated by both the leptin-induced activation of ATP-sensitive K+ channels which was reversible by the specific KATP inhibitor tolbutamide (Kieffer et al., 1997) and the concomitant leptin-induced decrease in glucose-stimulated intracellular Ca2+ concentration in islets from both Lepob mice and normal mice and rats (Kieffer et al., 1997; Fehmann et al., 1997). Furthermore, a leptin-induced decrease in intracellular Ca2+ concentration was shown to be reversible by glucose plus GLP-1 and leptin suppression of insulin release was overcome by glucose plus GLP-1 (Kieffer et al., 1997). The maximal insulin secretion inhibiting effect of

OB/DB IN MICE WITH MUTATIONS

69

leptin which occurred in normal rat islets after a prolonged exposure to leptin suggested the leptin effect to be mediated through a change in proinsulin gene expression (Roduit and Thorens, 1997). Recently, it was shown in isolated human pancreatic islets that leptin directly inhibits basal and stimulated insulin secretion by both interfering with the secretory process and suppressing the GLP-1-stimulated expression of proinsulin mRNA (Seufert et at., 1999). Pancreas The pancreas of B6-Lepob mice is character ized by islet cell hypertrophy and hyperplasia. Islet volume in older Lepob mice can be enlarged up to 10 times compared to controls. Islet enlargement is due mainly to Bcells. Histological and histochemical studies revealed a markedly higher B-cell activity in islets from Lepob when compared with control mice. In young Lepob mice, B-cells may be found to be moderately to severely degranulated suggesting a high secretory activity. In older mutants, exhibiting decreasing serum insulin levels, β-cells are well granulated (Herberg and Coleman, 1977). Insulin resistance Insulin resistance has been described in B6-Lepob mice by various groups. In early studies on insulin resistance in Lepob mice, a decrease in insulin binding membranes of liver, fat, and kidney cells as well as of thymic lymphocytes was concluded to be due to a reduced receptor number. A direct relationship between the degree of hyperinsulinemia and the receptor loss was suggested to be present in B6-Lepob mice (Herberg and Coleman, 1977). Identification of the insulin receptor gene revealed the decreased insulin receptor binding to be due to a defect at the post-transcriptional level (Ludwig et al., 1988). Recently, in liver and muscle cells from Lepob mice it was shown that multiple alterations in the early steps of intracellular insulin signalling are responsible for the insulin insensitivity of the target organs (Kerouz et al., 1997). Blood glucose In B6-Lepob mice, hyperglycemia is transient (Figure 1). In 1-mo-old mutants of the Düsseldorf colony nonfasted blood glucose was slightly lower than in controls (data not shown). Highest blood glucose levels occurred in 4-mo-old B6-Lepob mice at which time serum insulin was still increasing. In B6-Lepob older than 5 mo of age, blood glucose significantly dropped by 40% in either gender. From the 5th– 6th month of life until the 13th month, blood glucose remained nearly constant. The high circulating plasma glucagon levels as measured in B6-Lepob of the Düsseldorf colony (Buchanan and Herberg, unpublished) and by Flatt et al. (1980) in Aston-Lepob mice may contribute to the pronounced hyperglycemia in Lepob mice. The possible relationship between blood glucose levels and the activity of the gluconeogenic and glycolytic pathway has been discussed in detail (Herberg and Coleman, 1977). Numerous studies on the relationship between blood glucose and the adrenal cortex revealed hypercorticosteronemia to be secondary in the development of the obese-hyperglycemic syndrome in Lepob mice (Herberg and Coleman, 1977). PHYSIOLOGIC AND BIOCHEMICAL CHARACTERISTICS OF MICE BKS-Leprbd mice are phenotypically similar to B6-Lepob mice. However, the degree of severity of the obese-hyperglycemic syndrome is considerably more pronounced and life span in most colonies is

70

L.HERBERG AND E.H.LEITER

considerably shortened. Leprdb mice are leptin-resistant due to the defective leptin receptor (Friedman and Halaas, 1998). Body weight development mice, like B6-Lepob mice, are hyperphagic and 3- to 4-wk-old mutants can be identified by their plump appearance. In BKS-Leprdb mice, the pattern of body weight development can differ from colony to colony. In principle, two types of developmental pattern of body weight have been observed. The one type described by Coleman and Hummel (1967) is characterized by a continuous increase in body weight up to 50–60 g in 3- to 4-mo-old mice. Afterwards, the mutants lose weight and spontaneous death occurs at 5–8 months of life. This pattern of body weight development was called “representative” by Coleman and Hummel (1967). In contrast, BKS-Leprdb mice of the Düsseldorf colony exhibited either a continuous weight gain up to 75–80 g at 8–12 months or a moderate to severe decrease in body weight in 6to 7-mo-old mice. Spontaneous death occurred at 10–12 and 8–10 months in mutants maintaining high body weight versus those showing body weight reductions, respectively. Currently, the BKS-Leprdb mice distributed by The Jackson Laboratory show less B-cell compensation during the early phase of syndrome progression, such that mutant mice are severely hyperglycemic by 16 weeks of age (plasma glucose >500 mg/ dl). These mice seldom attain weights above 45 g, becoming unthrifty and losing weight after this maximum weight is attained. Adipose tissue In young BKS-Leprdb mice, white adipose tissue is mainly localized in the inguinal and axillar subcutaneous region. Increasing body fat content in older BKS-Leprdb mice is mainly due to the visceral fat portion. Total body fat content was found to be 16% in 6- to 16-d-old mice, 33% in 2-mo-old and 53% in 1yr-old BKS-Leprdb mice (L.H. unpublished observation). During the first, dynamic phase of the syndrome body fat mass increases mainly by hypertrophy (Herberg and Coleman, 1977). Mean fat cell volume is smaller until weaning in BKS-Leprdb when compared with B6-Lepob mice; i.e., 20% the value recorded in 6- to 16-d-old Lepob and 80% the value recorded in 18- to 22-d-old Lepob mice. In 26-wk-old male BKSLeprdb mice, wet weight, fat cell volume, and fat cell number of the retroperitoneal, gonadal, and subcutaneous adipose tissue were found to be lower when compared with B6-Lepob mice (Johnson and Hirsch, 1972). However, BMI is identical in B6-Lepob and BKS-Leprdb mice. Thermogenesis Energy expenditure on thermoregulatory thermogenesis is reduced in mice (Trayhurn, 1979). However, from studies on thermoregulation, Coleman (1985) concluded that, in mutants (Leprdb and Lepob on either the BKS or the B6 background) and lean controls, equivalent amounts of energy were used for heat production. In pair-fed mutants, the energy used by controls for thermoregulation (33%) was not sufficient to account for the increased metabolic efficiency in the mutants. Lipogenesis BKS-Leprdb mice are characterized by an increased lipogenesis, high plasma triglyceride and free fatty acid levels. Compared with lean controls Leprdb mice showed a 9 fold higher hepatic and a 10.5 fold higher

OB/DB IN MICE WITH MUTATIONS

71

Figure 2 Blood glucose and serum insulin in relation to body weight and age in Body weight (g), blood glucose (mg/dl) and serum insulin (µU/ml) were determined once per month. Blood was harvested at 8 a.m. from non-fasted mice. The mice were grouped according to body weight (main groups) and age (mo) (subgroups). Blood glucose and serum insulin values which were identical within a weight and/or an age group were pooled. Numbers of mice are given in the respective columns. MEANS ± SEM.

small intestinal fatty acid synthesis when expressed on a per organ basis (Memon et al., 1994). In both BKSLeprdb and B6-Lepob mice, the hypothesis that adipsin, a serine protease homologue of complement factor D in humans, acts as a fat cell-derived regulatory molecule (Flier et al., 1987) could not be confirmed. The age- and obesity-related pattern of adipsin revealed that the impairment of adipsin expression is secondary to obesity (Dugail et al., 1990). Blood glucose Chronic and increasingly more severe hyperglycemia is the most prominent feature distinguishing BKSLeprbd and BKS-Lepob mice from the same mutations on the B6 inbred background. In 4-wk-old BKSLeprbd and BKS-Lepob mutants and controls, blood glucose levels are similar. In a research colony at The Jackson Laboratory, significant elevations in blood glucose can be detected in BKS-Leprdb and BKS-Lepob mice within a week post-weaning. By 6–8 weeks of age, most individuals exhibit plasma glucose levels above 400 mg/dl, and by 12 weeks of age, values range between 500–800 mg/dl, with males generally

72

L.HERBERG AND E.H.LEITER

exhibiting higher values than females. An occasional mutant female will exhibit a B6-like pattern (e.g., mild if any hyperglycemia with compensatory sustained hyperinsulinemia). It is interesting to note that since an earlier description of a more protracted development of hyperglycemia in BKS-Leprdb mice (Herberg and Coleman, 1977), the mutant stock currently distributed by The Jackson Laboratory exhibits an earlier onset of severe hyperglycemia associated with reduced histologic evidence of ß-cell neogenesis. A less severe syndrome was observed in BKS-Leprdb mice of the former Düsseldorf colony (Figure 2). The less severe syndrome and the longer lifespan of BKS-Leprdb mice of the Düsseldorf colony when compared with BKS-Leprdb mice at The Jackson Laboratory (Coleman and Hummel, 1967) might be due to different husbandry conditions rather than genetic drift. The maintenance conditions of the Düsseldorf animal unit are described in the section below on “Husbandry Considerations”. Plasma and pancreas insulin Normal plasma insulin levels in BKS-+/+ mice is approximately 2 ng/ml. BKS-Leprdb mice exhibit increased plasma insulin levels at 2 weeks of age when compared with lean controls. Increased plasma insulin levels persist until the end of the 3rd mo of age and then decrease. Maximum plasma insulin levels in mutant males in a research colony at The Jackson Laboratory rarely exceeded 10 ng/ml. When mutants become unthrifty and body weights decline, plasma insulin drops to approximately 2 ng/ml. Already in 1mo-old mutants, pancreatic insulin content is lower when compared with controls. Whereas pancreatic insulin remains similar in 1- to 5-mo-old BKS-+/+ mice, in BKS-Leprdb mice, it declines progressively. In 5-mo-old mutants, pancreatic insulin is about 20% that measured in controls, thus reflecting increasing islet atrophy and B-cell degeneration (Coleman and Hummel, 1969b) A more detailed description of the insulin pattern is given in the following paragraph on the seventy of the syndrome in BKS-Leprdb mice. Insulin resistance In BKS-Leprdb mice, insulin administration up to 100 U/100 g body weight failed to normalize blood glucose concentrations of about 250 mg/dl (Coleman and Hummel, 1967). The persistence of reduced numbers of hepatocyte plasma membrane insulin receptors in BKS-Leprdb mice in which hyperinsulinemia was abolished after estrone treatment indicated that the reduced insulin binding capacity resulting from reduced numbers of insulin receptors was in some way closely marking the primary genetic defect rather than reflecting secondary ligand-mediated downregulation (Prochazka et al., 1986). SECONDARY COMPLICATIONS Vasculopathy Both microvascular (Bohlen and Niggl, 1979) and macrovascular (Kamata and Kojima, 1997) lesions are present in BKS-Leprdb mice. Increased protein glycation in BKS-Leprdb mice is suggested to lead to a generalized endothelial damage (Cohen et al., 1996). Neuropathy In contrast to B6-Lepob or B6-Leprdb mice, BKS-Leprdb develop neuropathy (Hanker et al., 1980). Both motor and sensory velocities are reduced (Moore et al., 1980). Functional impairment precedes

OB/DB IN MICE WITH MUTATIONS

73

degenerative changes. Thus, a decrease in MNCV (motor nerve conduction velocity) was observed in 5-wkold BKS-Leprdb mice. In 7-wk-old Leprdb MNCV was reported to be significantly lower when compared with BKS-Leprdb +/+ m mice. MNCV was significantly improved by insulin treatment of young mice (Robertson and Sima, 1980) or ganglioside treatment of older BKS-Leprdb mice (Norido et al., 1984). Therefore, the first phase of neuropathy was labeled “metabolic” and the second phase “neuronal”. Distal hind limb nerves are most affected by functional defects. This can be verified by a simple test: when lifted by the tail BKS-Leprdb mice adduct their hind limbs and keep them near the belly with the digits clenched. With increasing age and severity of the syndrome the frequency of adduction increases. In contrast, BKS-+/+ and B6-Lepob extend their legs when held in this position (Carson et al., 1980; Hanker et al., 1980). Before any degenerative morphological changes become visible, axonal transport of proteins is markedly decreased (Vitadello et al, 1983, 1985; Calcutt et al., 1988). In both myelinated and unmyelinated fibers, axons are swollen and contain conglomerates of membranous profiles. Honeycombed Schwann cellaxon networks develop and are followed by axonal atrophy (Sima and Robertson, 1979). In Schwann cells RER (rough endoplasmatic reticulum) is dilated, mitochondria are swollen and contain electron-lucent vacuoles (Carson et al., 1980). Loss and shrinkage of myelinated fibers occur and diameters become smaller in both myelinated and unmyelinated fibers (Robertson and Sima, 1980). The sequence of events suggests metabolic abnormalities the primary cause of functional nerve defects rather than degenerative processes (Sima and Robertson, 1979). Nephropatby An early symptom of nephropathy is an increased GFR (glomerular filtration rate) determined by total clearance of 51Cr-EDTA. GFR is increased in both BKS-Leprdb with extremely elevated bloood glucose and B6-Leprdb with moderately elevated blood glucose. In older mutants, GFR decreases slowly and approaches the levels of controls, presumedly because of an advanced glomerular damage (Gärtner, 1978). Glomerular damage becomes evident in 1-mo-old BKS-Leprdb mice by the deposition of immunoglobulins in the glomerular mesangium (Lee and Graham, 1980) which leads to marked thickening of the mesangium. Immunoglobulins are deposited mainly at the glomerular hilus and in the juxtaglomerular mesangium of the juxtaglomerular apparatus (Bower et al., 1980). Immunoglobulin deposits in the distal tubuli are seen in BKSLeprdb of all ages and in hyperglycemic B6-Lepob mice (Meade et al, 1981) and suggested to be derived from the mesangium of associated glomeruli. Albuminuria is present only in BKS-Leprdb mice older than 6 months of age (Meade et al., 1981). In BKS-Leprdb mice, immunoglobulin deposition can be prevented by normalization of blood glucose by dietary means (Lee and Bressler, 1981) or pharmacological intervention (Lee et al., 1982). Arginine and its metabolites reduce the accumulation of collagen in the glomerular basement membrane of BKS-Leprdb mice by their inhibitory effect on advanced stage nonenzymatic glycolation end products (AGEs) (Weniger et al, 1992; Lubec et al, 1994; Marx et al, 1995). Recently, treatment of BKS-Leprdb mice with monoclonal antibodies specific for Amadori-modified glycated albumin prevented the formation of AGEs reduced albuminuria and decreased the elevated type IV collagen as well as the elevated fibronectin gene expression by about 50%. Concomittantly, mesangial enlargement of glomeruli was markedly reduced and the increase in serum creatinine as well as the decrease in creatinine clearance were prevented while the rise in blood urea nitrogen was attenuated (Cohen et al., 1996). The absence of receptors for advanced glycation endproducts AGE (RAGE) in renal tissue from BKS-Leprdb mice suggests that factors other than hyperglycemia affect RAGE expression (Ziyadeh et al, 1997). However, hyperglycemia may affect cell cycle arrest and hypertrophy of mesangial cells as demonstrated by the expression of p27Kip1 in nuclei of mesangial cells (Wolf et al., 1998).

74

L.HERBERG AND E.H.LEITER

Myocardial disease Lipid droplets are recorded to be scattered in myocytes from both BKS-Leprdb and B6-Lepob mice. Atrophy of cardiomyocytes increases with age in BKS-Leprdb mice only (Giacomelli and Wiener, 1979). Degenerative changes of mitochondria and abnormalities in various mitochondrial enzyme activities are also observed in BKS-Leprdb mice only (Skoza et al., 1980; Kuo et al., 1983, 1985). Heart collagen accumulation can be reduced by arginine (Khaidar et al, 1994), presumably by inhibiting AGE formation. Retinopathy Studies on retinopathy in mutants are scarce. In BKS-Leprdb mice, a significant decrease in retinal, intramural pericytes has been recorded (Midena et al, 1989). In lens tissue sorbitol levels were found to be similar in B6-Lepob mice and lean littermates and to be higher in BKS-Leprdb when compared with lean littermates. Feeding of an aldose reductase inhibitor resulted in a 70% decrease in lens sorbitol levels in BKS-Leprdb mice (Vicario et al., 1989). REPRODUCTIVE CHARACTERISTICS IN B6-Lepob MICE Recently, it was shown, that administration of recombinant leptin was followed by successful pregnancies in B6-Lepob mice (Chebab et al., 1996). Hence, leptin treatment of mutants provides an effective new method to directly propagate mutants. Depletion of adipose depots by food reduction did not restore the greatly reduced reproductive capacity of female Lepob mice (Drasher et al, 1955). However, elicitation of fertility by pituitary extracts (Runner, 1954), hypothalamic extracts (Batt, 1972), or gonadotropic hormones (Runner and Gates, 1954) led to the suggestion of an alteration of the hypothalamic-pituitary system in female Lepob mice. In male Lepob mice, an impaired response to LH-RH (Swerdloff et al., 1978) and an inadequate release of LH-RH (Batt et al., 1982) led to the suggestion of a defect in the hypothalamic-pituitary axis. In male Lepob mice the descent of testes into the poorly developed scrotum is usually incomplete. A poor vascularization of the testes and the accessory glands as well as reduced weights of testes, epididymides, and seminal vesicles is common in Lepob mice (Jones and Harrison, 1958). Leydig cell mass is reduced as is nuclear size of the frequently atrophic Leydig cells. In B6-Lepob males, an incomplete spermatogenesis has been described (Jones and Harrison, 1958), whereas in V stock-Lepob mice, an intact spermatogenesis was reported (Hellman et al., 1963). Reproductive capability may be intact in both food-restricted and foodnonrestricted males (Lane and Dickie, 1954; Lane, 1959). That hypogonadism is secondary was shown by the prompt LH-stimulated testicular production of testosterone in vivo (Swerdloff et al., 1976) and in vitro (Wilkinson and Moger, 1981). Accordingly, administration of testosterone propionate was followed by a near normal enlargement of the accessory glands (Jones and Harrison, 1958). In female B6-Lepob mice, ovaries contain a few Graffian follicles only whereas corpora lutea are totally lacking. The uteri contain only 50% of the number of cells when compared with lean controls (Drasher et al., 1955). Female Lepob never come into oestrus (Jones and Harrison, 1958). The reproductive organs remain immature. Uterine tissue promptly responses to cessation of estrogen therapy by regression (Drasher et al., 1955). However, after induction of ovulation by gonadotropin ova and ovaries transplanted into lean mice can produce viable offspring (Runner and Gates, 1954). Implantation of fertilized ova from gonadotropin-stimulated obese females can be protected by daily injections of progesterone from the day of copulation until day 18 p.c. (Smithberg and Runner, 1957).

OB/DB IN MICE WITH MUTATIONS

75

REPRODUCTIVE CHARACTERISTICS IN BKS-Leprdb MICE In BKS-Leprdb and Leprdb-ad mice kept on a heterogenous background (Falconer and Isaacson, 1959), testes were described to be normal or slightly reduced in size. Numbers of Leydig cells, spermatozoa or tailed spermatides are also reduced (Johnson and Sidman, 1979; Batt and Harrison, 1960). Ovaries are filled with Graffian follicles at all stages of development and are lacking any corpora lutea and the uteri are small and immature (Batt and Harrison, 1963). Ovary transplantation into lean controls resulted in the successful production of litters (Johnson and Sidman, 1979). Estrogen treatment resulted in a prompt disappearance of the numerous intracellular uterine lipid vacuoles (Garris, 1989). Measurements of gonadotropins before and after stimulation by releasing hormones in both male and female Leprdb mice revealed the sterility to be of central origin (Johnson and Sidman, 1979). GESTATIONAL DIABETES IN BKS-Leprdb MICE Gestastional diabetes and macrosomia in pups have been described in heterozygous Leprdb/+ mice (Kaufmann et al., 1981). A recent study revealed hepatic glucose production as well as plasma leptin levels to be markedly elevated in pregnant hyperglycemic Leprdb/+ mice. Since down-regulated insulin receptor β (IR-β), insulin receptor substrate 1 (IRS-1), and phosphoinositol (PI) 3-kinase could be stimulated in heterozygous, pregnant, transgenic mice overexpressing the human GLUT 4 gene the authors suggested abnormalities in insulin receptor signalling in maternal muscle and a defect in insulin secretion as well as an enhanced nutrient availability to be the cause of gestational diabetes. A direct relationship between maternal hyperglycemia and fetal macrosomia was excluded (Ishizuka et al., 1999). OTHER FEATURES OF BKS-Leprdb AND/OR B6-Lepob MICE Kidney pathology Polycystic kidneys and hydronephrosis occur in BKS-+/+ mice. The frequency increases with aging and can be as high as 63% in 11- to 15-wk-old BKS-+/+ (Weide and Lacy, 1991). In the former Düsseldorf colony hydronephrosis or polycystic kidneys occurred in 25/57 and 20/36 BKS-+/+ at wk 15–25 and 30–54, respectively. BKS-Leprdb mice were found to be free from kidney anomalies. However, 4/30 BKS-Lepob mice had developed polycystic kidneys at wk 25. Ability to metabolize ketone bodies Increased metabolic efficiency of mutants extends to heterozygous mice. The observation that not only homozygous B6-Lepob mice, but also heterozygous, phenotypically normal, (Lepobl+) mice are able to withstand a total fast longer than homozygous lean (+/+) controls led Coleman (1980) to study the ability of metabolizing ketone bodies in heterozygous lean (BKS-Leprdb/+ and B6-Lepob/+) and homozygous lean (BKS-+/+ and B6-+/+) mice. Both in vivo and in vitro studies revealed a more effective conversion of (2– 14C)-acetone provided by an incomplete fatty oxidation to lactate. This ability was more pronounced in heterozygous when compared with homozygous lean mice. The acetone pathway provides the mice with additional energy and partly refills the limited stores of endogenous carbohydrate necessary for a more complete oxidation of free fatty acids. The ability to totally utilize endogenous fat stores by inducing the acetone-conversion pathway was found to be present in both heterozygous B6-Lepob/+ and BKS-Leprdbl+

76

L.HERBERG AND E.H.LEITER

mice. The increased energy efficiency in both homozygous and heterozygous lean mice conforms to the definition of a thrifty genotype (Neel, 1962; Neel et al., 1998): The advance to store energy when periods of plenty and shortage follow one another becomes a disadvantage in times of continuous abundance. Increased metabolic efficiency is correctable in Lepob mice by leptin administration (Friedman and Halaas, 1998). In B6-Lepob mice, the decreased corticosterone-induced expression of intrascapular brown adipose tissue UCP-1 mRNA, however, seems to be unaffected by leptin (Arvaniti et al., 1998b). Resistance to atherosclerosis Plasma lipid and cholesterol levels are elevated in B6-Lepob and BKS-Leprdb mice (Herberg and Coleman, 1977). By fractionating total cholesterol into its constituents HDL-C and combined VLDL+LDL-C, Nishina et al. (1994a) found the elevation of cholesterol levels to be due to an increase in the HDL subfraction. Lepob and Leprdb mice on either the B6 or BKS background exhibited higher HDL-C levels than homozygous or heterozygous lean controls. In B6-Leprdb mice between the 2nd and 4th wk of life, HDL-C and blood glucose levels transiently increased. Since insulin precedes the early blood glucose increase in mutants, the authors determined plasma HDL-C levels in 14-wk-old and 14-mo-old female Leprdb mice. B6Leprdb mice in which plasma insulin levels remain high throughout life showed similar HDL-C levels at either time. In BKS-Leprdb in which plasma insulin usually drops with aging, HDL-C levels were clearly higher in the 14-wk-old when compared with the 14-mo-old group. However, HDL-C levels were lower in BKS-Leprdb when compared with B6-Leprdb mice. VLDL+LDL-C levels were remarkably low in either age group. These results led the authors to speculate that cholesterogenesis is stimulated by insulin. In a second set of experiments Nishina et al. (1994b) evaluated aortic lesions in Lepob Leprdb females on both the B6 and the BKS backgrounds. Both wild-type and mutant males are resistant to atherosclerosis under the dietary conditions used. After a long-term (14 weeks) feeding of a high-fat, high-cholesterol diet, both Lepob and Leprdb mutant females exhibited reduced lesion area in comparison to normal females. Presumably, the greater resistance of the mutants compared to the controls to diet-induced atherosclerosis was due to higher levels of HDLC in the former. In BKS controls, the lesions were twice as large when compared with B6 controls. In both Lepob and Leprdb mutants, aortic lesions were significantly larger if the mutations were carried in the BKS background. A BKS allele on proximal Chromosome 12, designated Ath6, was found to be a significant contributor of the increased atherogenic susceptibility of this strain in a segregating F2 population produced by outcross of BKS-Leprdb and B6-Leprdb heterozygotes (Mu et al., 1999). Interestingly, a malic enzyme regulator gene (Mod1r) had been mapped into this same region previously in the same type of F2 cross, with the allele conferring low malic enzyme activity (Mod1rb) carried by BKS associated with more severe diabetes (Coleman and Kuzava, 1991). However, the BKS-derived Ath6 atherogenic allele did not correlate with glycemic state (Mu et al., 1999). Antihyperglycemic effects of DHEA and its metabolites Dehydroepiandrosterone (DHEA) depresses gluconeogenesis by reducing lactate conversion and PEPCK activity and is suggested to reduce metabolic efficiency by inducing one or more futile cycles (Coleman, 1988). In BKS-Leprdb mice, orally administered DHEA and the 5β stereoisomers of its metabolites Cthydroxyetiocholanolone (α-ET) and β-hydroxyetiocholanolone (β-ET) exert hypoglycemie effects, prevent the decrease in plasma insulin as well as islet atrophy and B-cell necrosis while food intake and body weight remain unaffected. Suboptimal doses of estradiol in combination with either α-ET or β-ET markedly potentiate the α-ET or β-ET effect (Coleman et al., 1984a). When DHEA therapy was started at 2 wk of age

OB/DB IN MICE WITH MUTATIONS

77

the development of the diabetes symptoms was partly prevented in Leprdb pups of either the B6 or BKS strain (Coleman et al., 1984b). In B6-Leprdb mice, DHEA and its metabolites not only reduce body weight gain but prevent any increase in blood glucose in spite of a 50% decrease in plasma insulin suggesting an increased insulin sensitivity (Coleman, 1990). In order to clarify the mechanism of the blood glucose lowering effect of DHEA in BKS-Leprdb mice, plasma sex steroid pattern after DHEA administration was analyzed. DHEA is metabolized to both androgens and estrogens, irrespective of the presence of intact gonads or adrenals. An i.v. injection of labelled 17β-estradiol was followed by a 6 fold accumulation of radioactivity in epididymal adipose tissue from BKS-Leprdb when compared with controls. A comparison of the blood glucose lowering effect of DHEA and its metabolites revealed estrone (E1), 17β estradiol (E2) and androstendiol (ADIOL) which bind to estrogen receptors the most potent hypoglycemic sex steroids. Hydroxypregnenolone and hydroxyprogesterone exhibited the weakest effects. DHEA, DHEA-S, androsterone, epiandrosterone, and α-ET or (β-ET showed an intermediate potency (Leiter et al. 1987b). Possibly the low endogenous levels of DHEA and DHEA-S in rodents and the lack of specific sex hormonebinding globulins in this species explains the effectiveness of high levels of exogeneously-administered DHEA and its metabolites on the obese-hyperglycemic syndrome. Recently, the role of sex steroid sulfotransferases involved into the regulation of androgen/estrogen balance has been discussed in detail (Leiter and Herberg, 1997). High sensitivity to refined carbohydrates BKS-Leprdb mice are highly sensitive to dietary carbohydrate. In weaning BKS-Leprdb a carbohydrate-free diet retarded or even prevented the development of the extreme hyperglycemia characteristic of chow-fed BKS-Leprdb mice (Coleman and Hummel, 1967). In order to test BKS-Leprdb and control mice for carbohydrate sensitivity, Leiter et al. (1983) offered isocaloric diets either totally devoid of carbohydrate or containing complex (dextrin starch) or refined carbohydrate (glucose, fructose, sucrose). Survival time was longest in BKS-Leprdb mice fed the carbohydrate-free or low-sucrose (8%) diet. Survival time was shortest in BKS-Leprdb on the 60% glucose, fructose, or sucrose diet. Mutants on the carbohydrate-free diet showed highest weight gain and lowest increase in blood glucose. Mutants fed diets containing 60% refined sugars exhibited the highest blood glucose levels whereas body weight increase was retarded. Histological examination of pancreatic and hepatic tissue revealed that structurally intact islets and granulated B-cells were present only in mutants fed on a carbohydrate-free diet. In mutants on highcarbohydrate diets, a fatty infiltration of the liver was the predominant feature. This was true especially in high-fructose fed BKS-Leprdb mice in which a considerable part of liver parenchyma seemed to be replaced by lipid deposits. Analysis of variance revealed significant effects of genotype and diet on body weight and blood glucose levels in both female and male mice. This study as well as the foregoing one (Leiter et al., 1981b) clearly show that dietary composition and not caloric intake is primarily involved in the β-cell’s function and its metabolic consequences in diabetes-susceptible strains. Immunodeficiency Both B6-Lepob and BKS-Leprdb mice are immunodeficient. In B6-Lepob mice a reduced weight of the spleen and the thymus (Bray and York, 1979), a decreased number of IgG-producing lymphocytes (Chandra, 1980), and an impaired ability to reject skin grafts (Sheena and Meade, 1978) suggest a reduced immune function. In BKS-Leprdb mice precocious and severe thymic involution due to a lymphocytic depletion, reduced levels of serum thymic factor (FST) (Dardenne et al., 1983), as well as T-cell lymphopenia and T-

78

L.HERBERG AND E.H.LEITER

cell imbalance (Boillot et al., 1986) indicate immunodeficiency. That the anti-pancreatic B-cell immunity is secondary in the development of the hyperglycemic syndrome has been shown in a study using double mutant mice homozygous for Leprdb in combination with one of the immune deficiency mutations (xid, scid, or nu) (Leiter et al, 1987e). On comparing organ-reactive autoantibody (ICA) production in diabetessusceptible (BKS, DBA/2J) and diabetes-resistant (B6, 129/J) strains Yoon et al. (1988) observed genetic and gender-dependent differences. Antibodies against pancreatic islets, gastric mucosa, and testes were present in male BKS-+/+ mice only. However, BKS-Leprdb B6-Leprdb mice of either gender expressed antibodies in the 3 tested organs. The induction of autoantibodies was highest in BKS-Leprdb (H2d haplotype) and weakest in B6-Leprdb (H2b haplotype). After transfer of the H2h haplotype on the BKS background in hyperglycemic BKS.B6 H2b mice, autoantibodies were absent. The authors concluded that the immunresponsiveness to the effects of the Leprdb and Lepob mutation is H2-linked and that the organspecific reactivities are under polygenic control. However, H2 haplotype did not prove to be a significant determinant of diabetes susceptibility in segregation analysis (Leiter et al., 1987a), and introduction of immunodeficiency genes that precluded autoantibody production also did not prevent diabetogenesis (Leiter et al., 1987c). APPROPRIATE CONTROL MICE B6-Lepob mice The most appropriate controls of B6-Lepob mice are B6-+/+ mice. When heterozygotes are mated, the socalled lean littermates consist of unknown numbers of +/+ and Lepob/+ mice. However, heterozygous lean (Lepob/+) mice of the B6 strain differ from homozygous lean (+/+) mice with regard to epididymal fat cell size (Joosten and van der Kroon, 1974), glucose oxidation in fat pads (Yen et al., 1968), survival time under fasting conditions (Coleman, 1979), conversion of acetone to lactate (Coleman, 1980), plasma corticosterone (Herberg and Kley, 1975) as well as plasma leptin levels and percentage body fat (Chung et al., 1 998). Moreover, basal blood glucose as well as glucose and insulin levels during a glucose tolerance test were found to be higher in the Aston stock of Lepob/+ mice when compared with Aston-+/+ mice (Flatt and Bailey, 1981). BKS-Leprdb mice The most appropriate controls of BKS-Leprdb mice are BKS-+/+ mice. Lean littermates are not suitable as controls since in mice of the BKS strain heterozygosity for Leprdb favors the development of obesity (Ishizuka et al., 1 999) and gestational diabetes (Kaufmann et al., 1987) with increased plasma leptin levels, abnormalities in insulin signalling, and a defective insulin secretion (Ishizuka et al., 1999). As for Lepob heterozygosity for Leprdb also affects plasma leptin levels and favors the development of body fat mass (Chung et al, 1998), affects the survival time under fasting conditions (Coleman, 1979) as well as the conversion of acetone to lactate (Coleman, 1980). SOURCE COLONIES OF Lepob MICE In the 1950s, the Lepob mutation was placed on the B6 genetic background at The Jackson Laboratory where it is still maintained. The consistent phenotype of obesity associated with transient hyperglycemia, marked hyperinsulinemia, and hypertrophy of the pancreatic islets exhibited by B6-Lepob mice originating from The

OB/DB IN MICE WITH MUTATIONS

79

Jackson Laboratory make this stock the most frequently described in the literature. Neuroendocrine defects in Lepob homozygotes preclude natural matings (see below). At The Jackson Laboratory, the B6-Lepob stock is maintained by transplantation of Lepob ovaries into +/+ hosts. If the hosts are mated with Lepob/+ males, 50% of the progeny are mutants (Lepob/Lepob) and 50% heterozygotes (Lepob/+). The heterozygotes can be used to produce the next generation. Although the heterozygotes are supplied as controls by The Jackson Laboratory, the user is advised to purchase wild-type B6-+/+ as controls for the reasons described above. If the ovarian transplant recipient females are mated with +/+males, the offspring are 100% heterozygotes which can be intercrossed to produce 25% mutants (Lepob), 25% +/+ and 50% Lepob/+ mice. Genotyping of nonobese black offspring is performed by PCR (Chung et al., 1997; Namae et al., 1998). The heterozygotes are then used both to continue the strain and to make up experimental pairs (Handbook on Genetically Standardized JAXR mice, 1997). Since each cycle of ovarian transplantation represents another backcross to B6, Lepab is truly congenic on the B6 inbred background. This is not necessarily the case for either BKSLepab, BKS-Leprdb or B6-Leprdb stocks which have been maintained as incipient congenic stocks after only 5 cycles of backcrossing to BKS. Although most male and female mutants in the Animal Resources Unit of The Jackson Laboratory exhibit spontaneous remission from hyperglycemia as they age beyond 12 weeks of age, recent quality control analysis showed occasional female mutants that rapidly transited into severe hyperglycemia shortly after weaning that did not remit with age (E.H.Leiter, unpublished observations). The BKS-Lepob incipient congenic stock is maintained as a cyropreserved stock at The Jackson Laboratory. A second mutation, called Lepob-2J arose spontaneously on the SM/J background. This mutant stock was sent to Dr. Jeff Friedman (Rockefeller University, NY) and is not maintained at The Jackson Laboratory. Colonies of B6-Lepob mice have been maintained at various places. The colony kept for many years by one of the authors (L.H.) at the Diabetes Research Institute in Düsseldorf/Germany originated from heterozygous breeding pairs (B6-Lepob/+) from The Jackson Laboratory. This colony has recently been discontinued. Around 1960 Hellerström and Hellman established a non-inbred colony of V stock obese mice at the Department of Histology, University of Uppsala/Sweden. In 1998, when the Uppsala colony was closed down, mice were transferred to the Department of Histology and Biology, University of Umea/Sweden as well as to the Rolf Luft Center for Diabetes Research, Karolinska Institute in Stockholm/ Sweden in order to build up two new colonies (C.Hellerström, Biomedical Center/ Uppsala, personal communication). V-stockLepob mice of the Swedish colonies differ from B6-Lepob by exhibiting even greater islet hypertrophy/ hyperplasia and are especially useful for in vitro studies on insulin secretion mechanism (Hellerström and Gunnarsson, 1970; Hellman, 1978). A further, non-inbred colony of V-stock-Lepob mice had been kept at the Centre de Sélection et d’Elévage des Animeaux de Laboratoire du Centre National de la Récherche Scientifique in Orléans, France (Lemonnier et al., 1971). This colony has been discontinued. At the Institute of Animal Genetics of the University of Edinburgh/UK B6-Lepob/+ mice originating from The Jackson Laboratory were outcrossed to two non-inbred local strains: JH selected for larger litter size and CRL selected for faster growth rate. Heterozygous mice from the CRL-derived stock were outcrossed to two further non-inbred local strains in 1966. In the same year, from the resultant stock heterozygous (Lepob/ +) breeding pairs were transferred to the University of Aston in Birmingham/UK where they were maintained in a closed, non-inbred colony for more than 30 generations (Flatt and Bailey, 1981). From the Aston colony mice were shipped to the Department of Biochemistry at the Imperial College in London/UK where they were maintained as a random-bred closed colony (Beloff-Chain et al., 1975). Lepob mice of the Aston University strain are still maintained at Aston University (Fraser et al., 1998) and at the Rowett Research Institute in Aberdeen/UK (P.Trayhurn, Rowett Research Institute, personal communication).

80

L.HERBERG AND E.H.LEITER

Aston-Lepob mice are reported to “display a unique combination of different features shown by Lepob on the B6, BKS and other backgrounds.” (Bailey et al., 1982). SOURCE COLONIES OF Leprdb MICE The Jackson Laboratory did not retain the original Leprdb mutation coisogenic on the BKS inbred background, but rather, as described below, maintains a large standing colony of this mutation on a BKS incipient congenic background in which the coat color gene has been introduced as a breeding aid. A smaller production colony of this mutation as an incipient congenic stock on the B6 background is maintained at The Jackson Laboratory. The incipient congenic BKS and B6 stocks carrying the Leprdb-2J allele (Hummel et al., 1972) no longer exist at The Jackson Laboratory. The Jackson Laboratory has conserved the Leprdb-3J mutation that occurred spontaneously in the 129/J strain (Leiter et al., 1980a; Leiter et al., 1980b). 129/J mice are resistant to Leprdb-gene induced diabetes, but are massively obese and hyperinsulinemic (Leiter et al., 1981a). 129/J-Leprdb-3J/+ is preserved in the embryo repository of The Jackson Laboratory (Handbook on Genetically Standardized JAXR Mice, 1997). Two further alleles are maintained in the inbred DW strain at the Institute Pasteur in Paris/France. The first one which appeared in the 1980s was called Leprdb-Pas1 and the second more recent one Leprdb-pas2 (J.L.Guenet, Institut Pasteur, personal communication). DW-Leprdb-Pas mice in which blood glucose is well controlled by insulin develop marked obesity in spite of an identical thermoregulatory capacity in mutant and control mice (Aubert et al, 1985). Another allele which had occurred in a heterogeneous stock at the Institute of Genetics in Edinburgh/UK (Falconer and Isaacson, 1959), gene symbol Leprdb-ad (dhad), had been maintained in the BKS strain for a limited period of time only. BREEDING SCHEMES FOR EARLY PHENOTYPIC IDENTIFICATION OF THE Leprdb MUTATION Neuroendocrine defects in Leprdb homozygotes preclude natural matings (see below) so that heterozygous matings must be utilized to propagate the mutant stock. Initially, identification of heterozygotes required progeny testing lean siblings of obese mice (for the presence of obesity in their offspring). A useful breeding tool was subsequently developed which utilized a linked coat color mutation, misty, (m) mapping within a centiMorgan of the Leprdb locus. The misty mutation arose spontaneously in the DBA/2J strain, and then was transferred onto the B6 inbred strain background. This congenic stock of B6-m/m mice were subsequently mated with BKS-Leprdb/+ mice and recombinants were selected that carried m and Leprdb in linkage (Coleman and Hummel, 1974). A separate stock was coselected that carried the two mutant genes in repulsion (i.e., on separate homologous chromosomes) (Coleman, 1978). Both the linkage and repulsion stocks represent 5 backcrosses to both the BKS and B6 backgrounds respectively following the initial outcross of BKS-Leprdb/+ to B6-m/m mice. The linkage stock permits early identification of Leprdb mice prior to weaning. Identificatin of postnatal mutants is facilitated by lighter pigmentation and white tips of the feet and the tail around day 4–5 of life. By two weeks of age, the grey pelage associated with the homozygous m mutation distinguishes the mutants (25% of progeny) from the black-furred controls [genotypes are +/+ (25%) and +/Leprdb (5Q%), the combination often denoted as +/? or lean littermates in publications]. Although progeny testing used to be the means of propagating this stock, the use of allelespecific PCR primers for the mutation would now represent the method of choice for genotyping the black lean mice. B6-m Leprdb/++ and BKS-m Leprdb/++ linkage stocks are preserved in the embryo repository at

OB/DB IN MICE WITH MUTATIONS

81

The Jackson Laboratory (The Jackson Laboratory catalog:http://jaxmice.jax.org). The repulsion stock permits easy identification of the heterozygous (m +/+ Leprdb) mutants (50% of weanlings) that are black pigmented and can be used for propagation of the mutation. Approximately 25% of weanlings are homozygous mutant mice (+ Leprdb/+ Leprdb). These black-pigmented mutants are identifiable by their plump body shape towards the end of the suckling period. The 25% of progeny that are lean and have grey (misty, m+/m+) pelage are usually discarded. The BKS-Leprdb and B6-Leprdb mice standardly available from The Jackson Laboratory are in repulsion (Handbook on Genetically Standardized JAXR Mice 1997). Since misty (m) in homozygosity is reported to negatively affect growth traits (Truett et al., 1998), homozygous BKS-+/+ mice rather than homozygous (m +/+ m) or heterozygous (m+/+Leprdb) misty mice should be taken as controls. Two stocks of “diabetes “(db) mice, also derived from matings between BKS-Leprdb/+ and B6-m/m but randomly bred and kept for a short period of time only, were called DBM (Chick and Like, 1970) and “db m” (Gunnarsson, 1975). These stocks should not be confused with the BKS-m+/+Leprdb and the B6-m +/ +Leprdb stocks currently distributed by The Jackson Laboratory. HUSBANDRY CONSIDERATIONS As noted above, mutants are quite stress sensitive. Hence, access to the mouseroom should be limited to the personnel involved in care of the mice and mice should not be disturbed during the dark period. Mouserooms are customarily set for either 14 hr/10 hr or 12 hr/12 hr light dark cycles. Procedures for entry will depend upon the barrier level. A complete barrier facility is not necessary for maintaining a high frequency of diabetes in BKS-Leprdb and Lepob mice. However, because the mutant mice, particularly on the BKS background, become increasingly immunodeficient as they age, the facility should be free of the standardly-encountered major murine pathogenic agents. This requires constant monitoring by veterinary care staff, and diligence on the part of the investigator to prohibit introduction of other research animals or biologicals into the same room that have not been prescreened to be specific pathogen-free (SPF). To insure a clean husbandry environment, caretakers should wear body covering (clean lab gown is a minimum requirement; donning of clean booties over shoes, and a hair net are recommended). Caretakers should wear gloves and handle mice by means of stainless steel forceps dipped in an iodine disinfectant solution. The room and associated clean supply area should be positively pressurized by HEPA-filtered, humidified air to remove particles of 0.3 microns or larger. Mouse cages may be kept on open shelves if covered by sterilizable filter bonnets, or they may be held in pressurized, individually ventilated (PIV) caging systems or in microisolator cages. Autoclaved or otherwise processed clean materials should be kept in a clean supply area and dirty materials removed from the mouseroom into a separate dirty materials corridor. HEPA-filtered cage changing stations are recommended. Because of the rapid post-weaning weight gains exhibited by mutants on both BKS and B6 backgrounds, a diet with a lower fat content (4–6%) should be selected. Protein is generally between 18–21% of constituents. Diets should be autoclavable. Drinking water should be acidified (with hydrochloric acid to attain a pH of 2.8– 3.2) to prevent growth of Pseudomonas sp. Alternatively, hyperchlorinated water (10 ppm sodium hypochlorite) may be used. Water bottles should be checked frequently since the mutant mice are polydipsic, their water consumption is higher than for cages containing lean controls. Bottles should be changed twice per week. Bedding material can either be autoclaved pine shavings of 2 mm length or other commercially-available absorptive material. BKS-Leprdb and Lepob mice become increasingly polyuric as the severity of diabetes increases. Although the bedding material used absorbs the moisture quite well, the quantity of urine produced and the peculiarity of the mice to urinate into a corner only of the cage requires a

82

L.HERBERG AND E.H.LEITER

minimum of two cage changes per week. Daily changing of the cages to minimize exposure of the mice to a wet substratum greatly extends lifespan of BKS-Leprdb mice and permits some of the longer-lived mice to go into remission from hyperglycemia (see Figure 2). Cellulose paper bedding is quite hygroscopic, and thus is useful for maintaining adult diabetic mice. However, such material may be excessively dry for litters in breeding cages. Breeding pairs can be provided with nesting material (nestlets made from cellulose paper). Cages containing newborn litters are not changed until the 7th day post partum; at that time, the nestlets containing pups can be transferred into the clean cages. Gloves should be worn when handling litters; if mating females, when disturbed, show a tendency to eat their pups, the pups can be sprinkled with urine from the dam, or alternatively, a small amount of an camphor containing petroleum jelly like Vick’s Vapor rub can be deposited on the dam’s nose. Because mutant mice gain weight rapidly after weaning and are polyuric, fewer mutant mice can be maintained in the same cage as compared to lean controls, with more frequent cage changes required to prevent morbidity. This is an important consideration in studies designed to study development of diabetic complications or to implement long-term therapies. At The Jackson Laboratory, cages for weaned mice are transparent polycarbonate double pens measuring 28 (1)×28 (w)×13 (h) cm. One breeding pair can be maintained in each pen. For weaned mice being aged for experimental purposes, not more than 3 mutants would be caged per side. In the Dusseldorf colony, Leprdb and Lepob offspring were grouped by sex at weaning with 4 mice housed in plastic cages measuring 26×20 cm and 5–10 housed in larger cages of 42×26 cm, respectively. Lean mice are usually grouped by sex and housed separately. As mentioned above, mutant males are quite docile so that mice from different litters can be pooled in aging studies. However, lean control male littermates should be caged together from weaning to reduce fighting behavior. EXPERIMENTAL TECHNIQUES Typical blood glucose and insulin curves for Leprdb and Lepob mutants have been published before (Herberg et al., 1970b). Because the mutants are so stress sensitive, it is recommended that they be conditioned (by handling) in advance of blood drawing. Small quantities of blood (20–70 µl) are typically withdrawn by heparanized capillary tube from either the retroorbital sinus of the eye, or from an incision into the tail. Regardless of inbred background, Leprdb and Lepob mutants exhibit higher than normal non-fasting blood glucose and exhibit impaired glucose tolerance following glucose loading. For this reason, if glucose tolerance testing (GTT) is to be done, a prior overnight fasting is recommended. Even under these conditions, BKS-Leprdb mice will exhibit fasting levels well above normal. Regardless of whether mice are fasted or not, GTT is performed in unanaesthetized mice. A 10% glucose solution is given intraperitoneally (i.p.). Lean mice and B6-Lepob can be compared at a loading dose of 2.5 mg glucose/g body weight. However, because of the elevations in fasting blood glucose, BKS-Leprdb mice should be tested with a lower dose of 1 mg glucose/g body weight in order to avoid extremely high blood glucose values. Blood samples are typically taken at various intervals (usually 0, 30, 60, and 120 min) after a glucose load. Oral GTT is performed only if the “incretin” effect of gastrointestinal hormones on glucose homeostasis is of interest. Glucose doses and concentrations are identical in both the i.p. and the oral GTT. Although a glucose clearance rate (k) can be extrapolated from the changes in glucose level between 30 and 60 or 120 min, the area under the curve is often used to express glucose clearance rates. The formula provided above (Table 5), with time units in hours, is based upon the “trapezoid rule” and subtracts the zero time value from the subsequent readings. Serum insulin determinations are performed on 50 µl serum (undiluted) or 25 µ serum (diluted 1:1, during a GTT when insulin is determined at various times) using a RIA Kit (Pharmacia GmbH Diagnostics, Munzinger Str. 9, 79111 Freiburg/ Germany) or using a rat insulin RIA kit from Linco (St. Louis, MO.,

OB/DB IN MICE WITH MUTATIONS

83

Table 5 Calculation of area under curves in a glucose tolerance test.

USA). Purified rat insulin (Novo Research Institute, Bagsvaerd/Denmark or Linco, USA) is used as standard. Preliminary screening of BKS-Leprdb and BK -Lepob serum samples for insulin are diluted at least 1:2. B6-Lepob and B6-Leprdb samples are diluted at 1:5 and higher depending upon age. Plasma can be stored frozen for insulin RIA without significant loss of insulin; 10% w/w aprotenin (Sigma, St. Louis, MO) may be added to prevent proteolysis. Insulin sensitivity is tested in nonfasted mice. 2 IU undiluted insulin (Actrapid, Insulin Novo)/40 g body weight is injected i.p. Blood samples are taken before and 120 min post insulin administration. Blood glucose drops to about 15% of basal in lean mice. In 15-wk-old and 32-wk-old BKS-Leprdb mice, blood glucose decreases to 30% and 20% of basal, respectively. Serum leptin can be assayed by RIA or ELISA. using Quantikine ™ M, Mouse Leptin Immunoassay (solid phase ELISA).Quantikine TM M is available from R&D Systems Oxford, UK or Minneapolis, MN (USA). Rodent leptin RIA kits are available from LINCO (St. Louis, MO.). Leptin challenge is performed with recombinant murine leptin (PREPROTECH INC., Princeton Business Park, G2, P.O. Box 275, Rockey Hill, NJ 08553, USA). On each day of a 3 day treatment, food is withdrawn at 8:00 a.m. and returned 1 hr after the subcutaneous injection of recombinant leptin at 2:00 p.m. Food consumption is measured 6 and 18 hr after leptin administration. In B6-Lepob mice, food intake decreases in a dose-dependent manner (0.8, 2.4, and 7.2 mg leptin/g body weight). A method for determining adipose tissue cellularity has been published before (Herberg et al., 1970a). Determination of total body fat content: Subcutaneous and intraabdominal fat pads are dissected and weighed. Intrascapular brown adipose tissue is dissected separately. All fat pads and the carcass are weighed, roughly minced and cooled in liquid nitrogen. Then the material is thoroughly minced in a grinding mill (F.Kurt Retsch GmbH, Postfach 1554, 42759 Haan/Germany). 22.5 g of the minced material +120 ml perchloethylene+60 g plaster of Paris are transferred into a homogenizer (Foss Electric Denmark; Foss Deutschland GmbH, Waidmannstr. 12b, 22769 Hamburg). After filtration, the solution is transferred into a measuring chamber (Foss-Let, A/S N.Foss Electric, Denmark) and the fat content measured, related to weight of fat pads+carcass and expressed as % body fat content. CONCLUDING REMARKS B6-Lepob

BKS-Leprdb

and mice are often labeled “Animal models of human NIDDM”. In fact, the obesehyperglycemic syndrome displayed by those mice shows various similarities to metabolic abnormalities present in both syndrome X and NIDDM. Insulin resistance, inappropriate hyperglycemia, impared glucose tolerance as well as increased insulin secretion finally leading to B-cell exhaustion are seen in both humans and mice. However, age, gender, and maintenance conditions may affect the phenotype of the mice. Thus, metabolic abnormalities such as high plasma insulin levels, increased volume of adipocytes and increased adipose tissue deposition occur already before weaning. Males usually display a more severe syndrome when compared with females. Avoidance of stress-conditions results in a less severe syndrome in mutants

84

L.HERBERG AND E.H.LEITER

of the BKS strain. Mutation, background genome as well as the interaction of the mutation with the genetic background may cause considerable variation in the phenotype of the mice. Thus, the syndromes displayed by Lepob and Leprdb mutants are quite similar when either mutation is maintained on the same genetic background: Lepob or Leprdb on the B6 background develop a less severe syndrome than Lepob or Leprdb on the BKS background. Secondary complications develop predominantly in BKS mutants. Genus-unique features limited to mice include the ability to metabolize ketone bodies, high HDL levels in serum and male resistance to atherosclerosis-inducing diet, the marked anti-hyperglycemic effects of DHEA and its down-stream steroid metabolites, and unusually high sensitivity to the diabetogenic effects of refined sugars. Nevertheless, research on leptin, the leptin-leptin receptor loop in both Lepob and Leprdb mice, and on the adipo-insular axis in Lepob mice has opened a new field of leptin studies in humans. As best summarized by others with regard to the utility of any single animal model of “diabesity” in terms of the complexity of obesity-associated diabetes in humans“None is identical to any human syndrome; all establish that the mechanism found in one animal is a mechanism in other mammals and thus worth testing for.” (Renold, Porte, and Shafrir, 1988). ACKNOWLEDGEMENTS This chapter is dedicated to our friend and colleague, Dr. Douglas L.Coleman, whose pioneering studies at The Jackson Laboratory pointed the way for molecular analyses to come. We thank Drs. Jung-Han Kim and Clay ton Mathews for their critical reviews. This work was supported by the Ministerium für Wissenschaft und Forschungs des Landes Nordrhein-Westfalen, the Deutsche Forschungsgemeinschaft (SFB 351), and NIH-RR08911. NOTE: A recent critical review of the significant differences in leptin endocrinology in humans versus mice has been published (Himms-Hagen, J. (1999)). Physiological roles of the leptin endocrine system: differences between mice and humans. Crit. Rev. Clin. Lab. Sci., 36, 575–655. REFERENCES Amir, S. (1981) Behavioral response of the genetically obese (ob/ob) mouse to heat stress: effects of naxolone and prior exposure to immobilization stress. Physiol. Behav., 27, 249–253. Arvaniti, K., Deshaies, Y. and Richard, D. (1998a) Effect of leptin on energy balance does not require the presence of intact adrenals. Am. J. Physiol., 275, R105-R111. Arvaniti, K., Ricquier, D., Champignym O. and Richard, D. (1998b). Leptin and corticosterone have opposite effects on food intake and the expression of UCP1 mRNA in brown adipose tissue of Lepob/Lepob mice. Endocrinology, 139, 4000–4003. Ashwell, M., Meade, C.J., Medawar, P. and Sowter, C. (1977) Adipose tissue: contributions of nature and nurture to the obesity of an obese mutant mouse (ob/ob). Proc. R.Soc. Lond. B, 195, 343–353. Aubert, R., Herzog, J., Camus, M.C., Guenet, J.L. and Lemonnier, D. (1985) Description of a new model of genetic obesity: the dbpas mouse. J. Nutr., 115, 327–333.

OB/DB IN MICE WITH MUTATIONS

85

Bado, A., Levasseur, S., Attoub, S., Kermorgant, S., Laigneau, J.P., Bortoluzzi, M.N., Moizo, L., Lehy, T., GuerreMillo, M., Le Marchand-Brustel, Y. and Lewin, M.J.M. (1998) The stomach is a source of leptin. Nature, 394, 790–793. Bahary, N., Leibel, R.L., Joseph, L. and Friedman, J.M. (1990) Molecular mapping of the mouse db mutation. Proc. Natl. Acad. Set., USA, 87, 8642–8646. Bahary, N., Zorich, G., Pachter, J.E., Leibel, R.L. and Friedman, J.M. (1991) Molecular genetic linkage maps of mouse chromosomes 4 and 6. Genomics, 11, 33–47. Bahary, N., Siegel, DA., Walsh, J., Zhang, Y., Leopold, L., Leibel, R., et al. (1993) Micro-dissection of proximal mouse chromosome 6: identification of RFLPs tightly linked to the ob mutation. Mamm Genome, 4, 511–515. Bailey, C.J., Flatt, P.R. and Atkins, T.W. (1982) Influence of genetic background and age on the expression of the obese hyperglycaemic syndrome in Aston ob/ob mice. Int. J. Obes., 6, 11–21. Batt, R.A.L. and Harrison, G.A. (1960) Features of the “adipose” mouse. Heredity, 15, 335– 337. Batt, R.A.L. and Harrison, G.A.(1963) The reproductive system of the adipose mouse. J. Hered., 54, 125–138. Batt, R.A.L. (1972) The response of the reproductive system in the female mutant mouse, obese (genotype obob) to gonadotrophin-releasing hormones. J. Reprod. Fert., 31, 496— 497. Batt, R.A.L., Everard, D.M., Gillies, G., Wilkinson, M., Wilson, C.A. and Yeo, TA. (1982) Investigation into the hypogonadism of the obese mouse (genotype ob/ob) J. Reprod. Fert., 64, 363–371. Bégin-Heick, N. (1996) β-adrenergic receptors and G proteins in the ob/ob mouse. Int.J. Obes., 20, Suppl. 2, S32—S35. Bellward, K. and Dauncey; M.J. (1988) Behavioural energy regulation in lean and genetically obese (ob/ob) mice. Physiol. Behav., 42, 433–438. Beloff-Chain, A., Hawthorn, J. and Green, D. (1975) Influence of the pituitary gland from the homozygote (+/+) and heterozygote (ob/+) lean mouse on insulin secretion in vitro. FEBS Letters 55, 72–74. Black, M.A. and Bégin-Heick, N. (1995) Growth and maturation of primary-cultured adipocytes from lean and ob/ob mice. J. Cell. Biochem., 58, 455–463. Bohlen, H.G. and Niggl, B.A. (1979) Arteriolar anatomical and functional abnormalities in juvenile mice with genetic or Streptozotocin-induced diabetes mellitus. Circ. Res., 45, 390–396. Bower, G., Brown, D.M., Steffes, M.W., Vernier, R.L. and Mauer, S.M. (1980) Studies of the glomerular mesangium and the juxtaglomerular apparatus in the genetically diabetic mouse. Lab. Invest., 43, 333–341. Boillot, D., Assan, R., Dardenne, M., Debray-Sachs, and J.F.Bach (1986) T-lymphopenia and T-cell imbalance in diabetic db/db mice. Diabetes, 35, 198–203. Bray, G.A. and York, D.A. (1971) Genetically transmitted obesity in rodents. Physiol.Rev., 51, 598–646. Bray, G.A. and York, D.A. (1979) Hypothalamic and genetic obesity in experimental animals: an autonomic and endocrine hypothesis. Physiol. Rev., 59, 719–809. Bray, G.A. (1984) Hypothalamic and genetic obesity: an appraisal of the autonomic hypothesis and the endocrine hypothesis. Int. J. Obes., 8, Suppl. 1, 119–137. Bray, G.A. (1991) Obesity, a disorder of nutrient partitioning: the MONA LISA hypothesis. J. Nutr., 121, 1146–1162. Burcelin, R., Kamohara, S., Li, J., Tannenbaum, G.S., Charron, M.J. and Friedman, J.M. (1999) Acute intravenous leptin infusion increases glucose turnover but not skeletal muscle glucose uptake in ob/ob mice. Diabetes, 48, 1264–1269. Calcagnetti, D.J., Flynn, J.J. and Margules, D.L. (1987) Opioid-induced linear running in obese (ob/ob) and lean mice. Pharmacol. Biochem. Behav., 26, 743–747. Calcutt, N.A., Willars, G.B. and Tomlinson D.R. (1988) Axonal transport of choline acetyltransferase and 6phosphofructokinase activities in genetically diabetic mice. Muscle and Nerve, 11, 1206–1210. Campfield, L.A., Smith, F.J., Guisez, Y., Devos, R. and Burn, P. (1995) Recombinant mouse OB protein: evidence for a peripheral signal linking adiposity and central neural networks. Science, 269, 546–549. Cancello, R., Zingaretti, M.C., Sarzani, R., Ricquier, D. and Cinti, S. (1998) Leptin and UCP1 genes are reciprocally regulated in brown adipose tissue. Endocrinology, 139, 4747–4750. Carlisle, H.J. and Dubuc, P.U. (1984) Temperature preference of genetically obese (ob/ob) mice. Physiol. Behav., 33, 899–902.

86

L.HERBERG AND E.H.LEITER

Carson, A., Bossen, E.H. and Hanker, J.S. (1980) Peripheral neuropathy in mouse hereditary diabetes mellitus. II. Ultrastructural correlates of degenerative and regenerative changes. Neuropathol. Appl. Neurobiol., 6, 361–374. Chan, C. (1995). β-cell stimulus—secretion coupling defects in rodent models of obesity. Can. J. Physiol. Pharmacol., 73, 1414–1424. Chandra, R.K. (1980) Cell-mediated immunity in genetically obese (C57BL/6J ob/ob) mice. Am. J. Clin.Nutr., 33, 13–16. Chebab, F.F., Lim, M.E. and Lu, R. (1996) Correction of the sterility defect in homozygous obese female mice by treatment with the human recombinant leptin. Nat. Genet., 12,318– 320. Chen, H., Charlat, O., Tartaglia, L.A., Woolf, E.A., Weng, X., Ellis, S.J., et al. (1996) Evidence that the diabetes gene encodes the leptin receptor: identification of an mutation in the leptin receptor gene in db/db mice. Cell, 84, 491–495. Chick, W.L. and Like, A.A. (1970) Studies in the diabetic mutant mouse. IV: DBM, a modified diabetic mutant produced by outcrossing of the original strain. Diabetologia, 6, 252–256. Chua, S.C., Jr., White, D.W., Wu-Peng, X.S., Liu, S.M., Okada, N., Kershaw, E.E., et al. (1996b) Phenotype of fatty due to Gln269Pro mutation in the leptin receptor (Lepr). Diabetes, 45, 1141–1143. Chua, S.C., Jr., Chung, W., Wu-Peng, X., Zhang, Y, Liu, S., Tartaglia, L., et al. (1996) Phenotypes of mouse diabetes and rat fatty due to mutations in the OB (leptin) receptor. (1996) Science, 271, 994–996. Chung, W.K., Chua, S.C., Lee, G.H. and Leibel, R.L. (1997) Polymerase chain reaction-restriction fragment length polymorphisms (PCR-RFLP) and electrophoretic assays for the mouse obese (Lepob) mutation. Obes. Res., 5, 183–185. Chung, W.K., Belfi, K., Chua, M., Wiley, J., Mackintosh, R., Nicolson, M., et al. (1998) Heterozygosity for Lepob or Leprdb affects body composition and leptin homeostasis in adult mice . Am. J. Physiol., 274, R985-R990. Cinti, S., Frederich, R.C., Zingaretti, M.C., De-Matteis, R., Flier, J.S. and Lowell, B.B. (1997) Immunohistochemical localization of leptin and uncoupling protein in white and brown adipose tissue. Endocrinology, 138, 797–804. Cohen, M.P, Clements, R.S., Cohen, J.A. and Shearman, C.W. (1996) Glycated albumin promotes a generalized vasculopathy in the db/db mouse. Biochem. Biophys. Res. Commun., 218, 72–75. Coleman, D.L. and Hummel, K.P (1967) Studies with the mutation, diabetes, in the mouse. Diabetologia, 3, 238–248. Coleman, D.L. and Hummel, K.P. (1969a) Effects of parabiosis of normal with genetically diabetic mice. Am. J. Physiol., 217, 1298–1304. Coleman, D.L. and Hummel, K.P. (1969b) The mutation, diabetes, in the mouse. In Diabetes, Proc. VI. Congr. Internat. Diab. Fed. Stockholm, edited by J.Östman and R.D.G.Milner, pp. 813–820, 1967. Internat. Congr. Ser. 172, Excerpta med. Amstrdam, 1969b. Coleman, D.L. (1973) Effects of parabiosis of obese with diabetes and normal mice. Diabetologia, 9, 294–298. Coleman, D.L. and Hummel, K.P. (1973) The influence of genetic background on the expression of the obese (ob) gene in the mouse. Diabetologia, 9, 287–293. Coleman, D.L. and Hummel, K.P. (1974) Hyperinsulinemia in pre-weaning diabetes (db) mice. Diabetologia, 10, 607–610. Coleman, D.L. and K.P.Hummel (1975) Influence of genetic background on the expression of mutations at the diabetes locus in the mouse. II. Studies on background modifiers. Israel. J. Med. Sci., 11, 708–713. Coleman, D.L. (1978) Obese and diabetes: two mutant genes causing diabetes-obesity syndromes. Diabetologia, 14, 141–148. Coleman, D.L. (1979) Obesity genes: beneficial effects in heterozygous mice. Science, 203, 663– 665. Coleman, D.L. (1980) Acetone metabolism in mice: increased activity in mice heterozygous for obesity genes. Proc. Natl. Acad. Sci., 11, 290–293. Coleman, D.L. (1981) Inherited obesity-diabetes syndromes in the mouse. In Mammalian genetics and cancer, pp. 145–158. The Jackson Laboratory Fiftieth Anniversary Symposium, Alan R.Liss, Inc., 150 Fifth Avenue New York, N.Y 10011. Coleman, D.L. (1982a) Diabetes-obesity syndromes in mice. Diabetes, 31, Suppl. 1, 1–6. Coleman, D.L. (1982b) Thermogenesis in diabetes-obesity syndromes in mice. Diabetologia, 22, 205–211.

OB/DB IN MICE WITH MUTATIONS

87

Coleman, D.L., Leiter E.H. and Applezweig, N. (1984a) Therapeutic effects of dehydroepiandro-sterone metabolites in diabetes mutant mice (C57BLKsJ-db/db). Endocrinology 115, 239– 243. Coleman, D.L., Schwizer, R.W. and Leiter, E.H. (1984b) Effect of genetic background on the therapeutic effects of dehydroepiandrosterone (DHEA) in diabetes-obesity mutants and in aged normal mice. Diabetes, 33, 26–32. Coleman, D.L. (1985) Increased metabolic efficiency in obese mutant mice. Int. J. Obes., 9, Suppl. 2, 69–73. Coleman, D.L. (1988) Therapeutic effects of dehydroepiandrosterone (DHEA) and its metabolites in obesehyperglycemic mutant mice . In Pathogenesis and new approaches to the study of noninsulin-dependent diabetes, edited by A.Y.Chang and A.R.Diani, pp. 161–175. New York: Alan R.Liss, Inc. Coleman, D.L. (1990) Hypoglycaemic action of the aetiocholanolones in mice. In New anti-diabetic drugs, edited by C.J.Bailey and P.R.Flatt, pp. 191–196. Smith-Gordon. Coleman, D.L. and Kuzava, J.E. (1991) Genetic regulation of malic enzyme activity in the mouse. J. Biol. Chem., 266, 21997–22002. Coleman, D.L. (1992) The influence of genetic background on the expression of mutations at the diabetes (db) locus in the mouse. VI. Hepatic malic enzyme activity is associated with diabetes severity. Metabolism, 41, 1134–1136. Contaldo, F., Gerber, H., Coward, W.A. and Trayhurn, P. (1981) Milk intake in pre-weanling genetically obese (oblob) mice. In Obesity: Pathogenesis and Treatment, edited by G. Enzi, G.Crepaldi, G.Pozza, and A.E.Renold, pp. 319–322. London and New York: Academic Press. Currie, P.J. (1993) Differential effects of NE, CLON, and 5-HT on feeding and macronutrient selection in genetically obese (ob/ob) and lean mice. Brain Res. Bull., 32, 133–142. Currie, P.J. and Wilson, L.M. (1993) Potentiation of dark onset feeding in obese mice (genotype ob/ob) following central injection of norepinephrine and clonidine. Eur. J. Pharmacol., 232, 227–234. Dardenne, M., Savino, W., Gastinel, L.N., Nabarra, B. and Back, J.F. (1983) Thymic dysfunction in the mutant diabetic (db/db) mouse. J. Immunol., 139, 1195–1199. Drasher, M.L., Dickie, M.M. and Lane, W.D. (1955) Physiological differences in uteri of obese stock mice. A comparison between obese mice and their thin sibs. J. Hered., 46, 209– 212. Drescher, V.S., Chen, H.L. and Romsos, D.R. (1994) Corticotropin-releasing hormone descreases feeding, oxygen consumption and activity of genetically obese (ob/ob) and lean mice. J. Nutr., 124, 524–530. Dubuc, P.O. (1989) Thyroid hormone responses to feeding in ob/ob mice. Norm. Metab. Res., 21, 546–549. Dugail, I., Quignard-Boulangé, A., Le Liepvre, X. and Lavau, M. (1990) Impairment of adipsin expression is secondary to onset of obesity in db/db mice. J. Biol. Chem., 265, 1831–1833. Emilsson, V., Liu, Y.L., Cawthorne, M.A., Morton, N.M. and Davenport, M. (1997) Expression of the functional leptin receptor mRNA in pancreatic islets and direct inhibitory action of leptin on insulin secretion. Diabetes, 46, 313–316. Falconer, D.S. and Isaacson, J.H. (1959) Adipose, a new inherited obesity of the mouse. J. Hered., 50, 290–292. Fehmann, H.C., Peiser, C., Bode, H.P., Stamm, M., Staats; P., Hedetoft, C., et al. (1997) Leptin: A potent inhibitor of insulin secretion. Peptides, 18, 1267–1273. Flatt, P.R., Bailey, C.J. and Buchanan, K.D. (1980) Circulating immunoreactive glucagon concentrations in aging obesehyperglycaemic (ob/ob) mice. Biochem. Soc. Trans., 8, 58–59. Flatt, P.R. and Bailey, C.J. (1981) Abnormal plasma glucose and insulin responses in heterozygous lean (ob/+) mice. Diabetologia, 20, 573–577. Flatt, P.R., Bailey, C.J., Berggren, P.O., Herberg, L. and Swanston-Flatt, S.K. (1992) Defective insulin secretion in diabetes and insulinoma. In Nutrient regulation of insulin secretion, edited by P.R.Flatt, pp. 341–386. Colchester: Protland Press. Flier, J.S., Cook, K.S., Usher, P. and Spiegelman, B.M. (1987) Severely impaired adipsin expression in genetic and acquired obesity. Science, 237, 405–408. Fraser, D.C., Srivastava, J., Woodward, C.J., Simpson, A.E.C.M. and Jones, R.B. (1998) A molecular investigation of the obese phenotype in the Aston University strain of ob/ob mice and the Japanese Kuo Kondo mice. Int. J. Obes., 22, 193–194.

88

L.HERBERG AND E.H.LEITER

Frederich, R.C., Kahn, B.B., Peach, M.J. and Flier, J.S. (1992) Tissue specific nutritional regulation of angiotensinogen in adipose tissue. Hypertension, 19, 339–344. Frederich, R.C., Löllmann, B., Hamann, A., Napolitano-Rosen, A., Kahn, B.B., Lowell, B.B., et al. (1995) Expression of ob mRNA and ist encoded protein in rodents. Impact of nutrition and obesity. J. Clin. Invest., 96, 1658–1663. Friedman, J.M., Leibel, R.L., Siegel, D.S., Walsh, J. and Bahary, N. (1991) Molecular mapping of the mouse ob mutation. Genomics, 11, 1054–1062. Friedman, J.M and Halaas, J.L. (1998) Leptin and the regulation of body weight in mammals. Nature, 392, 763–770. Gapp, D.A., Leiter, E.H., Coleman, D.L. and Schwizer, R.W. (1983) Temporal changes in pancreatic islet composition in C57BL/6J-db/db (diabetes) mice. Diabetologia, 25, 439– 443. Gärtner, K. (1978) Glomerular hyperfiltration during the onset of diabetes mellitus in two strains of diabetic mice (C57BL/6J db/db and C57BL/KsJ db/db). Diabetologia, 15, 59–63. Garris, D.R. (1989) Effects of estradiol and progesterone on diabetes-associated utero-ovarian atrophy in C57BL/KsJ (db/db) mutant mice. Anat. Rec,, 225, 310–317. Gavrilova, O., Barr, V., Marcus-Samuels, B. and Reitman, M. (1997) Hyperleptinemia of pregnancy associated with the appearance of a circulating form of the leptin receptor. J. Biol. Chem., 272, 30546–30551. Ghilardi, N., Ziegler, S., Wiestner, A., Stoffel, R., Heim, M.H. and Skoda, R.C. (1996) Defective STAT signaling by the leptin receptor in diabetic mice. Proc. Natl. Acad. Sci., USA, 93, 6231–6235. Giacomelli, F. and Wiener, J. (1979) Primary myocardial disease in the diabetic mouse. An ultrastructural study. Lab. Invest., 40, 460473. Goodbody, A.E. and Trayhurn, P. (1982) Studies on the activity of brown adipose tissue in suckling, pre-obese ob/ob mice. Biochim. Biophys. Acta., 680, 119–126. Greenberg, D. and Ackerman, S.H. (1984) Genetically obese (ob/ob) mice are predisposed to gastric stress ulcers. Bebav. Neurosci., 98, 435–440. Gunnarsson, R. (1975). Function of the pancreatic B-cell during the development of hyperglycaemia in mice homozygous for the mutations “diabetes” (db) and “misty” (m). Diabetologia, 11, 431–438. Halaas, J.L., Gajiwala, K.S., Maffei, M., Cohen, S.L., Chait, B.T., Rabinowitz, D., et al. (1995) Weight-reducing effects of the plasma protein encoded by the obese gene. Science, 269, 543– 546. Handbook on Genetically Standardized JAXR Mice, edited by R.R.Fox and B.A.Witham. Fifth Edition, The Jackson Laboratory, 1997. Hanker, J.S., Ambrose, W.W., Yates, P.E., Koch, G.G. and Carson, K.A. (1980) Peripheral neuropathy in mouse hereditary diabetes mellitus. I. Comparison of neurologic, histologic, and morphometric parameters with dystonic mice. Acta Neuropathol. (Berl), 51, 145–153. Hausberger, F.X. (1958) Parabiosis and transplantation experiments in hereditarily obese mice. Anat. Rec., 130, 313. Hellerström, C. and Gunnarsson, R. (1970) Bioenergetics of islet function: oxygen utilization and oxidative metabolism in the β-cells. Acta diabet. lat., 7, (Suppl. 1) 127–151. Hellman, B., Jacobsson, L. and Täljedal, I.E. (1963) Endocrine activity of the testis in obese hyperglycaemic mice. Acta Endocrinol., 44, 20–26. Hellman, B. (1978) Mechanism of insulin secretion. In Regulatory mechanisms of carbohydrate metabolism, edited by V.Esmann, pp 41–50. Oxford: Pergamon. Herberg, L., Gries, F.A. and Hesse-Wortmann, Ch. (1970a) Effect of weight and cell size on hormone-induced lipolysis n New Zealand obese mice and American obese hyperglycemic mice. Diabetologia, 6, 300–305. Herberg, L., Major, E., Hennigs, U., Grüneklee, D., Freytag, G. and Gries, F.A. (1970b) Differences in the development of the obese-hyperglycemic syndrome in obob and NZO mice. Diabetologia, 6, 292–299. Herberg, L and Kley, H.K. (1975) Adrenal function and the effect of a high-fat diet on C57BL/ 6J and C57BL/6J-ob/ob mice. Norm. Metab. Res., 7, 410–415. Herberg, L. and Coleman, D.L. (1977) Laboratory animals exhibiting obesity and diabetes syndromes. Metabolism, 26, 59–99. Himms-Hagen, J. (1985) Defective brown adipose tissue thermogenesis in obese mice. Int. J. Obes., 9, Suppl. 2, 17–24.

OB/DB IN MICE WITH MUTATIONS

89

Ho, A. and Chin, A. (1988) Circadian feeding and drinking patterns of genetically obese mice fed solid chow diet. Physiol. Behav., 43, 651–656. Hogan, S. and Himms-Hagen, J. (1980) Abnormal brown adipose tissue in obese (ob/ob) mice: response to cold. Am. J. Physiol., 239, E301-E309. Hummel, K.P., Dickie, M.M. and Coleman, D.L. (1966) Diabetes, a new mutation in the mouse. Science, 153, 1127–1128. Hummel, K.P., Coleman, D.L. and Lane, P.W. (1972) The influence of the genetic background on expression of mutations at the diabetes locus in the mouse. I. C57BL/KsJ and C57BL/ 6J strains. Biochem. Genet., 7, 1–13. Ingalls, A.M., Dickie, M.M. and Snell, G. (1950). Obese, a new mutation in the house mouse. J. Hered., 41, 317–318. Ishizuka, T., Klepcyk, P., Liu, S., Panko, L., Liu, S. and Gibbs, E.M. (1999) Effects of overexpression of humanGLUT4 gene on maternal diabetes and fetal growth in spontaneous gestational diabetic C57BLKS/J Leprdb/+ mice . Diabetes, 48, 1061–1069. Jeanrenaud, B. (1978) Hyperinsulinemia in obesity syndromes: its metabolic consequences and possible etiology. Metabolism, 27, 1881–1891. Jeanrenaud, B. (1985) An hypothesis on the aetiology of obesity: dysfunction of the central nervous system as a primary cause. Diabetologia, 28, 502–513. Jeanrenaud, B. (1995) Insulin, corticosterone and the autonomic nervous system in animal obesities: a viewpoint. Diabetologia, 38, 998–1002. Johnson, L.M.M. and Sidman, R.L. (1979) A reproductive endocrine profile in the diabetes (db) mutant mouse. Biol. Reprod., 20, 552–559. Johnson, P.R. and Hirsch, J. (1972) Cellularity of adipose depots in six strains of genetically obese mice. J. Lipid. Res., 13, 2–11. Jones, N. and Harrison, G.A. (1958) Genetically determined obesity and sterility in the mouse. Stud. Pert., 9, 51–61. Joosten, H.F.P. and van der Kroon, P.H.W. (1974) Enlargement of epididymal adipocytes in relation to hyperinsulinemia in obese hyperglycemic mice (ob/ob). Metabolism, 23, 59– 66. Kamata, K. and Kojima, S. (1997) Characteristics of contractile responses of aorta to norepinephrine in db/db mice. Res. Commun. Mol. Pathol. PharmacoL, 96, 319–328. Kaplan, M.L., Trout, J.R. and Leveille, G.A. (1976) Adipocyte size distribution in ob/ob mice during preobese and obese phases of development. Proc. Soc. Exp. Biol. Med., 153, 476— 482. Kaufmann, R.C., Amankwah, K.S., Dunaway, G., Maroun, L., Arbuthnot, J. and Roddick, J.W. (1981) An animal model of gestastional diabetes. Am. J. Obstet. Gynecol, 141, 479– 482. Kaufmann, R.C., Amankwah, K.S., Colliver, J.A. and Arbuthnot, J. (1987) The effect of genetic susceptibility for diabetes on fetal weight. Am J Perinatol, 4, 72–74. Kennedy, G.C. (1950) The hypothalamic control of food intake in rats. Proc. Roy. Soc. Lond. Ser. B 137, 535–549. Kerouz, N.J., Hörsch, D., Pons, S. and Kahn, C.R. (1997) Differential regulation of insulin receptor substrates-1 and -2 (IRS-1 and IRS-2) and phosphatidylinositol 3-kinase in liver and muscle from the obese diabetic (ob/ob) mouse. J. Clin. Invest., 100, 3164.3172. Khaidar, A., Marx, M., Lubec, B. and Lubec, G. (1994) A-arginine reduces heart collagen accumulation in the diabetic db/db mouse. Circulation, 90, 479–483. Kieffer, T.J., Heller, R.S. and Habener, J.F. (1996) Leptin receptors expressed on pancreatic β-cells. Biochem. Biophys. Res. Commun., 224, 522–527. Kieffer, T.J., Heller, R.S., Leech, C.A., Holz, G.G. and Habener, J.F. (1997) Leptin suppression of insulin secretion by the activation of ATP-sensitive K+ channels in pancreatic B-cells. Diabetes, 46, 1087–1093. Kim, J.H., Nishina, P.M. and Naggert, J.K. (1998) Genetic models for non insulin dependent diabetes mellitus in rodents. J. Basic. Clin. Physiol. PharmacoL, 9, 325–345. Kuhn, C.M., Cochrane, C., Feinglos, M.N. and Surwit, R.S. (1987) Exaggerated peripheral responses to catecholamines contributes to stress-induced hyperglycemia in the ob/ob mouse. Pharmacol.Biochem. Behau, 26, 491495.

90

L.HERBERG AND E.H.LEITER

Kulkarni, R.N., Wang, Z.L., Wang, R.M., Hurley, J.D., Smith, D.M., Ghatei, M.A., et al. (1997) Leptin rapidly suppresses insulin release from insulinoma cells, rat and human islets and, in vivo, in mice. J. Clin. Invest., 100: 2729–2736. Kuo, T.H., Giacomelli, F., Wiener, J. and Lapanowski-Netzel, K. (1985) Pyruvate dehydrogenase activity in cardiac mitochondria from genetically diabetic mice. Diabetes, 34,1075– 1081. Kuo, T.H., Moore, K.H., Giocomelli F. and Wiener, J. (1983) Defective oxidative metabolism of heart mitochondria from genetically diabetic mice. Diabetes, 32, 781–787. Lane, P.W. and Dickie, M.M. (1954) Fertile, obese male mice. Relative sterility on obese males corrected by dietary restriction. J. Hered., 45, 56–58. Lane, P.W. (1959) The pituitary-gonad response in genetically obese mice in parabiosis with thin and obese siblings. Endocrinology, 65, 863–868. Lee, G.H., Proenca, R., Montez, J.M., Carroll, K.M., Darvishzadeh, J.G., Lee, J.I. and Friedman, J.M. (1996) Abnormal splicing of the leptin receptor in diabetic mice. Nature, 379, 632–635. Lee, G.H., Li, C., Montez, J., Halaas, J., Darvishzadeh, J. and Friedman, J.M. (1997) Leptin receptor mutations in 129 db3J/db3J mice and NIH facp/fa/cp rats. Mamm. Genome., 8, 445– 447. Lee, S.M. and Graham, A. (1980) Early immunopathologic events in experimental diabetic nephropathy: a study in db/ db mice. Exp. Mol. Pathol., 33, 323–332. Lee, S.M. and Dressier, R. (1981) Prevention of diabetic nephropathy by diet control in the db/db mouse. Diabetes, 30, 106–111. Lee, S.M., Tutwiler, G., Bressler, R. and Kircher, C.H. (1982) Metabolic control and prevention of nephropathy by 2tetradecylglycidate in the diabetic mouse (db/db). Diabetes, 31, 12– 18 . Leiter, E.H., Coleman, D.L., Eisenstein, A.B. and Strack, I. (1980a) A new mutation (db3J) at the diabetes locus in strain 129/J mice. I. Physiological and histological characterization. Diabetologia, 19, 58–65. Leiter, E.H., Strack, I. and Eisenstein, A.B. (1980b) A new mutation (db3J) at the diabetes locus in strain 129/J mice. II. Studies of pancreatic alpha cell function in culture. Diabetologia, 19, 66–73. Leiter, E.H., Coleman, D.L. and Hummel, K.P. (198 1a) The influence of genetic background on the expression of mutations at the diabetes locus in the mouse. III. Effect of H-2 haplotype and sex. Diabetes, 30, 1029–1034. Leiter, E.H., Coleman, D.L., Eisenstein, A.E. and Strack, I. (1981b) Dietary control of pathogenesis in C57BL/KsJ db/ db diabetes mice. Metabolism, 30, 554–562. Leiter, E.H., Coleman, D.L., Ingram, D.K. and Reynolds, M.A. (1983) Influence of dietary carbohydrate on the induction of diabetes in C57BL/KsJ-J-db/db diabetes mice. J. Nutr., 113, 184–195. Leiter. E.H., Fewell, I.W. and Kuff, E.L. (1986) Glucose induces intracisternal type A retroviral gene trannscription and translation in pancreatic beta cells. J Exp Med, 163, 87–100. Leiter, E.H., Le, P.H. and Coleman, D.L. (1987a) Susceptibility to db gene and streptozotocin-induced diabetes in C57BL mice: control by gender-associated, MHC-unlinked traits. Immunogenetics, 26, 6–13. Leiter, E.H., Beamer, W.G., Coleman, D.L. and Longcope, C. (1987b) Androgenic and estrogenic metabolites in serum of mice fed dehydroepiandrosterone: relationship to antihyperglycemic effects. Metabolism, 36; 863–869. Leiter, E.H., Prochazka, M. and Shultz, L.D. (1987c) Effect of immunodeficiency on diabetogenesis in genetically diabetic (db/db) mice. J. Immunol., 138, 3224–3229. Leiter, E.H. and Chapman, H.D. (1994) Obesity-induced diabetes (Diabesity) in C57BL/KsJ mice produces aberrant trans-regulation of sex steroid sulfotransferase genes. J. Clin. Invest., 93, 2007–2013. Leiter, E.H. and Herberg, L. (1996) Aging, pancreatic islets, and glucose homeostasis in inbred mice. ILSI Press, Washington, D.C. I, 153–170. Leiter, E.H. and Herberg, L. (1997) The polygenetics of diabesity in mice. Diabetes Rev., 5, 131–148. Leiter, E.H., Kintner, J., Flurkey, K., Beamer, W.G. and Naggert, J.K. (1999) Physiologic and endocrinologic characterization of male sex-biased diabetes in C57BLKS/J mice congenic for the fat mutation at the carboxypeptidase E locus. Endocrine, 10, 57–66. Lemonnier, D., Winand, J., Furnelle, J. and Christophe, J. (1971) Effect of a high-fat diet on obese-hyperglycaemic and non-obese Bar Harbor mice. Diabetologia, 7, 328–333.

OB/DB IN MICE WITH MUTATIONS

91

Levine, A.S., Morley, J.E., Brown, D.M. and Handwerger, B.S. (1982) Extreme sensitivity of diabetic mice to naxoloneinduced suppression of food intake. Physiol. Behav., 28, 987– 989. Li, C., Ioffe, E., Fidahusein, N., Connolly, E. and Friedman, J.M. (1998) Absence of soluble leptin receptor in plasma from dbpas/dbpas and other db/db mice./ J. BioL Chem,, 273, 10078– 10082. Like, A.A. (1985) Spontaneous diabetes in animals. In The Diabetic Pancreas, edited by B.W. Volk and K.F.Wellmann, pp. 385–413. New York: Plenum Press. Lubec, B., Aufricht, C., Herkner, K., Hoeger, H., Adamiker, D., Gialamas, H., et al. (1994) Creatine reduces collagen accumulation in the kidneys of diabetic db/db mice. Nephron, 67, 214–217. Ludwig, S., Müller-Wieland, D., Goldstein, B.J. and Kahn, C.R. (1988) The insulin receptor gene and its expression in insulin-resistant mice. Endocrinology, 123, 594–600. McIntosh, C.H.S. and Pederson, R.A. (1999) Noninsulin-dependent animal models of diabetes mellitus. In Experimental Models of Diabetes, edited by J.H.McNeill, pp. 338–398. CRC Press LLC. MacDougald, O.A., Hwang, C.H., Fan, H. and Lane, M.D. (1995) Regulated expression of the obese gene product (leptin) in white adipose tissue and 3T3-L1 adipocytes. Proc. Natl. Acad.Sci., USA, 92, 9034–9037. Maffei, M., Fei, H., Lee, G.H., Dani, C., Leroy, P., Zhang, Y., et al. (1995a) Increased expression in adipocytes of ob RNA in mice with lesions of the hypothalamus and with mutations at the db locus. Proc. Natl. Acad. Sci. USA, 92, 6957–6960. Maffei, M., Halaas, J., Ravussin, E., Pratley, R.E., Lee, G.H., Zhang, Y., et al. (1995b) Leptin levels in human and rodent: measurement of plasma leptin and ob RNA in obese and weight-reduced subjects. Nat. Med., 1, 1155–1161. Marx, M., Trittenwein, G., Aufricht, C., Hoeger, H. and Lubec, B. (1995) Agmatine and spermidine reduce collegan accumulation in kidneys of diabetic db/db mice. Nephron, 69, 155–158. Mayer, J., Dickie, M.M., Bates, M.W. and Vitale, J.J. (1951) Free selection of nutrients by hereditarily obese mice. Science, 113, 745–747. Meade, C.J., Ashwell, M. and Sowters, C. (1979) Is genetically transmitted obesity due to an adipose tissue defect? Proc. R. Soc. Lond. B, 205, 3954410. Meade, C.J., Brandon, D.R., Smith, W., Simmonds, R.G., Harris, S. and Sowter, C. (1981) The relationship between hyperglycaemia and renal immune complex deposition in mice with inherited diabetes. Clin. Exp. Immunol., 43, 109–120. Memon, R.A., Grunfeld, C., Moser, A.H. and Feingold, K.R. (1994) Fatty acid synthesis in obese insulin resistant diabetic mice. Horm. Metab. Res., 26, 85–87. Memon, R.A., Fuller, J., Moser, A.H., Smith, P.J., Grunfeld, C. and Feingold, K.R. (1999) Regulation of putative fatty acid transporters and acyl-CoA synthetase in liver and adipose tissue in ob/ob mice. Diabetes, 48, 121–1227. Midena, E., Segato, T., Radin, S., di Giorgio, G., Meneghini, F., Piermarocchi, S., et al. (1989) Studies on the retina of the diabetic db/db mouse. I Endothelial cell-pericyte ratio. Ophthalmic Res., 21, 106–111. Mizuno, T.M., Bergen, H., Funabashi, T., Kleopoulus, S.P., Zhong, Y.G., Bauman, W.A., et al. (1996) Obese gene expression: Reduction by fasting and stimulation by insulin and glucose in lean mice, and persistent elevation in acquired (diet induced) and genetic (yellow agouti) obesity. Proc. Natl. Acad. Sci., USA, 93, 3434–3438. Moinat, M., Deng, C., Muzzin, P., Assimacopoulos-Jeannet, F., Seydoux, J., Dulloo, A.G., et al. (1995) Modulation of obese gene expression in rat brown and white adipose tiisues. FEBS Letters, 373, 131–134. Molina, J.M., Premdas, F.H., Klenck, R.E., Eddlestone, G., Oldham, S.B. and Lipson, L.G. (1984) The dynamic insulin secretory response of isolated pancreatic islets of the diabetic mouse. Evidence for a gene dosage effect on insulin secretion. Diabetes, 33, 1120–1123. Moon, B.C. and Friedman, J.M. (1997) The molecular basis of the obese mutation in ob2J mice. Genomics 42, 152–156. Moore, S.A., Peterson, R.G., Felten, D.L., Cartwright, T.R. and O’Connor, B.L. (1980) Reduced sensory and motor conduction velocity in 25-week-old diabetic [C57BL/Ks (db/ db)] mice. Exp. Neurol., 70, 548–555. Mu, J.L., Naggert, J.K., Svenson, K.L., Collin, G.B., Kim, J.H., McFarland, C., et al. (1999) Quantitative trait loci analysis for the differences in susceptibility to atherosclerosis and diabetes between inbred mouse strains C57BL/ 6J and C57BLKS/J. J. Lipid. Res., 40, 1328–1335.

92

L.HERBERG AND E.H.LEITER

Naggert, J.K., Mu, J.L., Frankel, W., Bailey, D.M. and Paigen, B. (1995) Genomic analysis of the C57BL/Ks mouse strain. Mamm Genome, 6, 131–133. Namae, M., Mori, Y., Yasuda, K., Kadowaki, T., Kanazawa, Y. and Komeda, K. (1998) New method for genotyping the mouse Lepob mutation, using a polymerase chain reaction assay. Lab. Anim. Sci., 48, 103–104. Neel, J.V. (1962) Diabetes mellitus a thrifty genotype rendered detrimental by “progress”. Am. J. Hum. Genet., 14, 353–362. Neel, J.V., Weder, A.B. and Julius, S. (1998) Type II diabetes, essential hypertension, and obesity as “syndromes of impaired genetic homeostasis”: the thrifty genotype hypothesis enters the 21st century. Perspect. Biol. Med., 42, 44–74. Nishina, P.M., Lowe, S., Wang, J. and Paigen, B. (1994a) Characterization of plasma lipids in genetically obese mice: the mutants obese, diabetes, fat, tubby, and lethal yellow. Metabolism, 43, 549–553. Nishina, P.M., Naggert, J.K., Verstuyft, J. and Paigen, B.(1994b) Atherosclerosis in genetically obese mice: the mutants obese, diabetes, fat, tubby, and lethal yellow. Metabolism, 43, 554–558. Norido, F., Canella, R., Zanoni, R. and Gorio, A. (1984) Development of diabetic neuropathy in the C56BL/Ks(db/db) mouse and ist treatment with gangliosides. Exp. Neurol., 83, 221–232. Parson, W., Camp, J.L. and Crispell, K.R. (1954) Dietary dilution studies in mice with gold thioglucose-induced obesity and in mice with the hereditary obesity-diabetes syndrome. Metabolism, 3, 351–356. Pelleymounter, M.A., Cullen, M.J., Baker, M.B., Hecht, R., Winters, D., Boone, T., et al. (1995) Effects of the obese gene product on body weight regulation in ob/ob mice. Science, 269, 540–543. Poitout, V., Rouault, C., Guerre-Millo, M., Briaud, I. and Reach, G. (1998) Inhibition of insulin secretion by leptin in normal rodent islets of Langerhans. Endocrinology, 139, 822– 826. Prochazka, M., Premdas, F.H., Leiter, E.H. and Lipson, L.G. (1986) Estrone treatment dissociates primary versus secondary consequences of “Diabetes” (db) gene expression in mice. Diabetes, 35, 725–728. Proietto, J. and Thorburn, A.W. (1994) Animal models of obesity theories of aetiology. Baillieres Clin. Endocrinol. Metab., 8, 509–525. Reichling, S., Patel, H.V., Freeman, K.B., Kates, A.L. and Himms-Hagen, J. (1988) Attenuated cold-induced increase in mRNA for uncoupling protein in brown adipose tissue of obese (ob/ob) mice. Biochem. Cell. Biol., 66, 193–196. Renold, A.E., Prote, D. and Shafrir, E. (1988) Definitions for diabetes types: use and abuse of the concept ‘animal models of diabetes mellitus’. In Frontiers in diabetes research Lessons from animal diabetes II, edited by E.Shafrir and A.E.Renold, pp. 3–4. London, Paris: John Libbey. Robertson, D.M. and Sima, A.A.F. (1980) Diabetic neuropathy in the mutant mouse[C57BL/ ks (db/db)]. A morphometric study. Diabetes, 29, 60–67. Roduit, R. and Thorens, B. (1997) Inhibition of glucose-induced insulin secretion by long-term preexposure of pancreatic islets to leptin. FEBS Letters, 415, 179–182. Runner, M.N. (1954) Inherited hypofunction of the female pituitary in the sterile-obese syndrome in the mouse. Rec. Genet. Soc. Am., 23, 63–64. Runner, M.N. and Gates, A. (1954) Sterile, obese mothers. J.Hered., 45, 51–55. Seufert, J., Kieffer, T.J., Leech, C.A., Holz, G.G., Moritz, W., Ricordo, C., et al. (1999). Leptin suppression of insulin secretion and gene expression in human pancreatic islets: implications for the development of adipogenic diabetes mellitus. J. Clin. Endocrinol. Metab., 84, 670–676. Shafrir, E. (1992) Animal models of non-insulin-dependent diabetes. Diabetes Metab. Rev., 8, 179–208. Sheena, J. and Meade, C.J. (1978) Mice bearing the ob/ob mutation have impaired immunity. Int. Arch. Allergy ImmunoL, 57, 263–268. Shimomura, Y., Oku, J., Glick, Z. and Bray, G.A. (1982) Opiate receptors, food intake and obesity. Physiol. Behav., 28, 4414445. Sima, A.A.F. and Robertson, D.M. (1979) Peripheral neuropathy in the diabetic mutant mouse. An ultrastructural study. Lab. Invest., 40, 627–632. Simpson, E.R., Merril, I.C., Hollub, A.J., Graham-Lorence, S. and Mendelson, C.R. (1989) Regulation of estrogen biosynthesis by human adipose cells. Endocr. Rev., 10, 136–148.

OB/DB IN MICE WITH MUTATIONS

93

Skoza, L., Giacomelli, F. and Wiener, J. (1980) Lysosomal enzymes in the heart of the genetically diabetic mouse. Lab. Invest. 43, 443–448. Smithberg, M. and Runner, M.N. (1957) Pregnancy induced in genetically sterile mice. J. Hered., 48, 97–100. Sprott, R.L. (1972) Long-term studies of feeding behavior of obese, diabetic, and viable yellow mutant mice under ad lib and operant conditions. Psychol. Rep., 30, 991–1003. Stephens, T.W., Basinski, M., Bristow, P.K., Bue-Valleskey, J.M., Burgett, S.G., Craft, L., et al (1995) The role of neuropeptide Y in the antiobesity action of the obese gene product. Nature, 377, 530–532. Stern, J.S., Hirsch, J., Drewnowski, A., Sullivan, A.C., Johnson, P.R. and Cohn, C.K. (1983) Glycerol kinase activity in adipose tissue of obese rats and mice: effects of diet composition. J. Nutr., 113, 714–720. Strohmayer, A.J. and Smith G.P. (1987) The meal pattern of genetically obese (ob/ob) mice. Appetite, 8, 111–123. Surwit, R., Seldin, M., Kuhn, C., Secor, C. and Feinglos, M. (1994) Diet-induced obesity and diabetes in C57BL/6J and C57BL/KsJ mice. Mouse Genome, 92, 523–525. Surwit, R.S. and Williams, P.G. (1996) Animal models provide insight into psychosomatic factors in diabetes. Psychosom Med, 58, 582–589. Swenne, I. and Andersson, A. (1984) Effect of genetic background on the capacity for islet replication in mice. Diabetologia, 27, 464–467. Swerdloff, R.S., Batt, R.A. and Bray, G.A. (1976) Reproductive hormonal function in the genetically obese (ob/ob) mouse. Endocrinology, 98, 1359–1364. Swerdloff, R.S., Peterson, M., Vera, A., Batt, R.A.L., Heber, D. and Bray, G. (1978) The hypothalamic-pituitary axis in genetically obese (ob/ob) mice: response to luteinizing hormone-releasing hormone. Endocrinology, 103, 542–547. Tartaglia, L.A., Dembski, M., Weng, X., Deng, N., Culpepper, J., Devos, R., et al. (1995) Identification and expression cloning of a leptin receptor, OB-R. Cell, 83, 1263–1271. Thurlby, P.L. and Trayhurn, P. (1978) The development of obesity in preweanling obob mice. Br. J. Nutr., 39, 3974402. Thurlby, P.L. and Trayhurn, P. (1979) The role of thermoregulatory thermogenesis in the development of obesity in genetically obese (ob/ob) mice pair-fed with lean siblings. Br. J. Nutr, 42, 377–385. Trayhurn, P. and James, W.P.T. (1978) Thermoregulation and non-shivering thermogenesis in the genetically obese (ob/ ob) mouse. Pflügers Arch, 373, 189–193. Trayhurn, P. (1979) Thermoregulation in the diabetic-obese (db/db) mouse. Pflügers Arch, 380, 227–232. Trayhurn, P., Jones, P.M., McGuckin, M.M. and Goodbody, A.E. (1982) Effects of overfeeding on energy balance and brown fat thermogenesis in obese (ob/ob) mice. Nature, 295, 323– 325. Trayhurn, P., Thomas, M.E.A., Duncan, J.S. and Rayner, D.V. (1995) Effects of fasting and refeeding on ob gene expression in white adipose tissue of lean and obese (ob/ob) mice. FEBS Letters, 368, 488–490. Trayhurn, P., Duncan, J.S., Hoggard, N. and Rayner D.V. (1998) Regulation of leptin production: a dominant role for the sympathetic nervous system? Proc. Nutr. Soc., 57, 413– 419. Truett, G.E., Tempelman, R.J., Walker, J.A. and Krause Wilson, J. (1998) Misty (m) affects growth traits . Am. J. Physiol., 275, R29-R32. Vicario, P.P., Slater. E.E. and Saperstein, R. (1989) The effect of ponalrestat on sorbitol levels in the lens of obese and diabetic mice. Biochem. Internat., 19, 553–561. Vitadello, M., Couraud, J.Y., Hässig, R., Gorio, A. and Di Gianberardino, L. (1983) Axonal transport of acetylcholinesterase in the diabetic mutant mouse. Exp. Neurol., 82, 143— 147. Vitadello, M., Filliatreau, G., Dupont, J.L., Hassig, R., Gorio, A. and Di Giamberardino, L. (1985) Altered axonal transport of cytoskeletal protein in the mutant diabetic mouse. J. Neurochem., 45, 860–868. Weide, L.G. and Lacy, RE. (1991) Hereditary hydronephrosis in C57BL/KsJ mice. Lab. Anim. Sci., 41, 415–418. Weigle, D.S., Bukowski, T.R., Foster, D.C., Holderman, S., Kramer, J.M., Lasser, G., et al. (1995) Recombinant ob protein reduced feeding and body weight in the ob/ob mouse. J. Clin. Invest., 96, 2065–2070. Weninger, M., Xi, Z., Lubec, B., Szalay, S., Höger, H. and Lubec, G. (1992) L-arginine reduces glomerular basement membrane collagen Nε-carboxymethyllysine in the diabetic db/db mouse. Nephron, 62, 80–83.

94

L.HERBERG AND E.H.LEITER

Wilkinson, M. and Moger, W.H. (1981) Transpubertal modulation of pituitary and testicular function in the ob/ob mouse. Hormone Res., 14, 95–103. Wilson, L.M. and Sinha, H.L. (1985) Thermal preference behavior of genetically obese (ob/ ob) and genetically lean (+/?) mice. Physiol. Behav., 35, 545–548. Wolf, G., Schroeder, R., Thaiss, F., Zyadeh, F.N., Helmchen, U. and Stahl, R.A.K. (1998) Glomerular expression of p27Kip1 in diabetic db/db mouse: role of hyperglycemia. Kidney Int., 53, 869–879. Yen, T.T.T., Lowry, L. and Steinmetz J. (1968) Obese locus in Mus musculus: a gene dosage effect. Biochem. Biophys. Res. Commun., 33, 883–887. Yen, T.T.T. and Acton, J.M. (1972) Locomotor activity of various types of genetically obese mice. Proc. Soc. Exp. Biol. Med., 140, 647–650. Yoon, J.W., Leiter, E.H., Coleman, D.L., Kim, M.K., Pak, C.Y., McArthur, R.G., et al. (1988) Genetic control of organreactive autoantibody production in mice by obesity (ob) diabetes (db) genes. Diabetes, 37, 1287–1293. Zhang, Y., Proenca, R., Maffei, M., Barone, M., Leopold, L. and Friedman, J.M. (1994) Positional cloning of the mouse obese gene and its human homologue. Nature, 372, 425– 432. Ziyadeh, F.N., Cohen, M.P., Guo, J. and Jin, J. (1997) RAGE mRNA expression in the diabetic mouse kidney. Mol. Cell Biochem., 170, 147–152.

4. THE ZUCKER DIABETIC FATTY (ZDF) RAT RICHARD G.PETERSON Indiana University School of Medicine; Anatomy, MS 5035; 635 N.Barnhill Dr.; Indianapolis, IN 46202–5120 and Genetic Models Inc.; P.O. Box 68737; Indianapolis, IN 46268–0737

ABSTRACT The data reviewed in this chapter support the obese male ZDF rat as a significant model for NIDDM and shows that it has many of the characteristics of human diabetes. Although diabetes has been extensively studied in the obese male ZDF rats, the obese females can be induced to develop diabetes with diets. The mechanistic changes, which accompany the development of diabetes in the male, have been described but there is still much characterization work to be done on the female, since it is likely that it will be different than the male in some respects. INTRODUCTION The obese male ZDF rat is considered to be a good model for NIDDM and has been used extensively since the late 1980s to investigate the mechanisms, which underlie diabetes. This model has been used for the testing many of the drugs which have been and are being developed and used for the treatment of NIDDM. Some of the long-term complications of diabetes have been studied in this model. The obese male ZDF rat develops diabetes rapidly between 6 and 10 weeks of age (Figure 1A). The obese female ZDF rat does not typically develop diabetes on commercially available rat chows (Figure 1A). However, when fed a diet high in animal fat it consistently develops diabetes within a matter of weeks (Figure 1A). The diabetic female promises to be an excellent NIDDM model; it appears to have several advantages over the diabetic male. This review lists and discusses many of the papers where obese ZDF males have been used. The diabetic female has not been available for long and thus there are not many publications available. Nevertheless the general characteristics of the diabetic ZDF female rat will be described.

96

R.G.PETERSON

Figure 1 This figure demonstrates some of the features of ZDF rats with different phenotypes. A demonstrates blood glucose levels that are seen in typical groups of diabetic and non-diabetic rats; B shows typical insulin levels found in various groups of ZDF rats.

ZDF rats are available for purchase from a single source by license agreement with Indiana University. For information on price and availability contact should be made with Genetic Models, Inc, P.O.Box 68737, Indianapolis, IN 46268–0737 USA. Phone: 317–824–7070, Fax: 317–824–7080, E-mail: [email protected], Web: http://www.geneticmodels.com/.

Address correspondence to: Richard G.Peterson, Anatomy, MS 5035; 635 N.Barnhill Dr.; Indianapolis, IN 46202–5120; Phone: 317–274–2436; Fax: 317–274–5756; E-Mail: [email protected] and Genetic Models Inc.; P.O. Box 68737; Indianapolis, IN 46268–0737; Phone: 317–824–7070; Fax: 317–824– 7080; E-Mail: [email protected]; Web: http://www.geneticmodels.com

THE ZUCKER DIABETIC FATTY (ZDF) RAT

97

BACKGROUND The fatty Zucker rat has been valued as a model of obesity since the genetic obesity trait was discovered by the Zuckers in a colony of rats they were maintaining (Zucker and Zucker, 1961). The Zuckers shared animals from their colony with a number of investigators across the world. Many of the individuals who received Zucker rats started their own colonies. Since that time the Zucker fatty rats have been used extensively for obesity studies and in pre-diabetic studies. Most of the work done with rats from Zucker colonies has demonstrated that they are insulin resistant and normoglycemic. One of the investigators who received rats from the Zuckers was Dr. Walter Shaw at Eli Lilly and Co. In 1977 several groups of Zucker rats were transferred from the Shaw colony to Indiana University School of Medicine and a colony was started under the supervision of Dr. Julia Clark. Shortly after transfer, obese male rats were recognized to be diabetic in both the Shaw and Clark colonies (Clark et al., 1983). After several years of trying to perpetuate the diabetic trait the project was abandoned and the colony was turned over to our laboratory at Indiana University School of Medicine. DEVELOPMENT OF THE ZDF MODEL The above Zucker colony was followed for several years. A number of diabetic obese males were seen and characterized, but the trait was not consistent in all obese males. Since the colony had not been Caesarian rederived, and it carried a number of potential pathogens, a decision was made to re-derive the colony. Several diabetic males and females with diabetic lineage were identified as breeders for re-derivation. Following selection, the rats were inbred for the diabetic condition that only the obese males expressed (Greenhouse et al., 1988; Peterson, 1994b; Peterson et al., 1990b). By 10 generations of inbreeding the diabetic trait expressed by the obese males was “fixed” in three sublines. One of the sublines was continued as the commercially available ZDF rat. A second subline appeared to be phenotypically identical to the above commercially available line; this line was dropped. A third less diabetic subline, the ZDFB, has been continued but has not been available commercially. Two colonies of “ZDF” rats exist. The first of the colonies was licensed by Indiana University to Genetic Models Inc. Indianapolis; the second was started from male rats that were purchased by Roger Unger (Dallas, TX). According to information we have, a cross of an obese diabetic male ZDF rat and a heterozygous (fa/+) Zucker female, from Charles River, started their colony. Further breeding of offspring with fatty males from the Indiana University colony resulted in the stock from which their current colony was perpetuated. Although there appear to be many similarities between the Genetic Models and Unger colonies, they are obviously not genetically and perhaps phenotypically identical. Nevertheless, we have not differentiated between the data from the two colonies since no direct comparisons have been made between rats. The easiest way to tell if animals are from the second colony is if the article has Roger Unger’s name on it or if the study is from University of Texas Southwestern Medical Center, Dallas, TX and is published after about 1991. The papers usually reference purchase of “breeders” from Peterson at Indiana University School of Medicine. PHENOTYPIC APPEARANCES The general characteristics of the ZDF male have been described in the literature (Peterson, 1994b; Peterson et al., 1990b). The characteristics of the model are: 1) hyperglycemia, that develops between 7 and 10 weeks of age (Figure 1A), 2) early hyperinsulinemia, that rapidly falls as the β-cells fail (Figure IB), β) fasting hyperglycemia, which first appears at 10 to 12 weeks of age and progresses with aging, 4) abnormal

98

R.G.PETERSON

glucose tolerance, which becomes progressively worse with age, 5) hyperlipidemia, 6) mild hypertension, 7) impaired wound healing and 8) kidney disease. Obese diabetic rats from the Genetic Models colonies also demonstrate less weight gain relative to non-diabetic obese and lean ZDF controls. For a number of years the obese female ZDF rat was not thought to be diabetic, as it did not become diabetic on rat chows. More recent studies have demonstrated that the female will become diabetic if it is fed “Western style” diets with high levels of fats from animal sources. The only published information on the female model is in abstract form, a brief reference in a review paper (Peterson, 1994b) and a recently published paper (Corsetti et al., 2000). Unpublished information from Genetic Models Inc. Laboratories, demonstrates several interesting characteristics of the development of diabetes in the female model: 1) when the obese females are fed a diabetogenic diet they become fully diabetic within a few weeks with a similar blood glucose curve to the obese male ZDF rat (Figure 1A), 2) diabetes can be initiated by diet anywhere from 6 weeks of age to about 25 weeks of age and, 3) diabetes is reversible if the rats are taken off the diet within the first few weeks of treatment. Unpublished data from Johnson et al., demonstrate several additional important characteristics of this model: 1) depletion of β-cells is slower than in the diabetic male (Figure 1B), which preserves blood insulin levels for a longer period of time, 2) it responds for a longer period of time to the antidiabetic drugs such as the thiazolidinediones, and 3) the diabetic ZDF female is hypertensive while the non-diabetic ZDF female is not. GENETICS The leptin receptor (Lepr or Ob-R) is the membrane receptor for leptin. Two mutations of the leptin receptor produce obesity in the rat. The mutations have been called by various names over the years. The first of the mutations was defined as fa as described by the Zuckers (1961). The fa mutation has subsequently been defined as Leprfa or Ob-Rfa; it has been and continues to be called just fa. The fa mutation results in shortened leptin receptor protein which does not effectively interact with leptin (Chua et al., 1996; White et al., 1997). The other mutation of the leptin receptor has been known cp, k, or the facp. In keeping with the above naming, it should now be referred to as the Leprk or Ob-RK mutation but the above terms (cp, k or facp) are often still used. It is a Tyr763Stop nonsense mutation in the gene, just before the transmembrane domain (Takaya et al., 1996; Wu-Peng et al., 1997); the result is an absence of the Lepr receptor. Both of the mutations in the leptin receptor are phenotypically expressed as obesity with high levels of normal leptin in the blood. The ZDF rat carries the Leprfa or Ob-Rfa mutation, which is typically referred to as just fa. PANCREATIC ABNORMALITIES Many characteristics of the type 2 diabetes found in the ZDF rat resemble those of the type 2 diabetes found in adult humans. With respect to some of the characteristics, significant abnormalities have been found in pancreatic structure and function. The majority of the studies utilizing the ZDF have concentrated on the pancreas and its functional responses, before and after diabetes was initiated. A number of changes in the islet cells have been observed which either proceed or accompany the diabetic state. In summary, changes were observed in glucose transporters (Unger, 1991a), in insulin release, fat metabolism in the islet (Unger, 1995), β-cell apoptosis, and other mechanisms which are related to the insulin release (see below).

THE ZUCKER DIABETIC FATTY (ZDF) RAT

99

Glucose Transport One of the early hypotheses regarding the development of diabetes in the ZDF model was related to the under-expression of the GLUT-2 transporter. The early demonstration that the diabetic ZDF male had little GLUT-2 transporter on pancreatic β-cells suggested that the high Km glucose transport in β-cells was impaired, and thus, glucose-stimulated insulin secretion was defective and it can not correct the hyperglycemia (Johnson et al., 1990; Orci et al., 1990). The initial studies were followed by several additional studies and reviews that extended and confirmed the early investigations (Milburn et al., 1993; Ohneda et al., 1993a; Ohneda et al., 1993b; Ohneda et al., 1994; Unger, 1991a; Unger, 1991b). In summary, GLUT-2 transporter depletion occurs, as the obese ZDF rats become diabetic. Ohneda et al. (1995) subsequently have shown that the above changes could be prevented by pair feeding ZDF obese rats with lean control littermates. Control of food intake prevented the loss of glucose-stimulated insulin secretion and the reduction of β-cell GLUT-2. They demonstrated that food restriction preserved βcell volume fraction. Another study demonstrated that diet, supplemented with magnesium, was able to reduce the GLUT-2 loss (Balon et al., 1995). Insulin Secretion All the above studies on the GLUT-2 transporter involved insulin release and related the level of GLUT-2 to the changes in insulin release. The studies demonstrated that there was a good correlation between the loss of transporter and the decrease in glucose stimulated insulin release. Zawalich et al. (1995) demonstrated in prediabetic islets that peak first and second phase insulin secretory responses to 20mM glucose were higher than in lean control islets. They showed that 3H-inositol efflux and inositol phosphate responses were greater from prediabetic rat islets than control islets. Zhou et al. (1998) demonstrated that prediabetic and diabetic ZDF islets had an increased capacity to esterify fatty acids. Hirose et al. (1996) found that islets from Wistar rats exposed to FFA increased their low Km glucose usage by 2.5X as opposed to no change in islets from 6 week-old lean and fatty diabetic ZDF rats. Insulin secretion increased significantly in Wistar islets cultured in high FFA and declined in islets from obese ZDF rats. They concluded that islets from young lean and obese ZDF rats lack the capacity for FFA to induce βcell function. Abnormalities in the release of insulin have been extensively studied in fully diabetic ZDF rats (Polonsky, 1995). Sturis et al. (1994) observed that insulin pulses can be entrained to the exogenous glucose oscillations in Wistar and fatty Zucker rats but not in diabetic ZDF rats. The results were seen in pancreas perfusions as well as in perifused islets. They observed that insulin secretory rates could not be stimulated in ZDF rat with high glucose in the perfusion medium. Thus they felt their data suggested that the defect, in the ZDF model, was at the cellular level. Some of the islet cell mechanisms, which have been suggested, are in the mRNA expression for proteins, which relate to the insulin release mechanisms such as those affecting Ca2+ and K+ channels (Roe et al., 1996; Tokuyama et al., 1995). This group (Stoffel et al., 1995; Tokuyama et al., 1996; Tokuyama et al., 1995) suggested that these and additional changes in the normal pattern of gene expression probably all contributed to the development of (β-cell dysfunction in the ZDF rat. The results showed that there were marked reductions in insulin release, when stimulation was done with glucose and KCl (Tang et al., 1996). With respect to the mechanisms, investigators (Cockburn et al., 1997; Tokuyama et al., 1995) have seen significant differences in glucokinase and hexokinase activity in -cells. Impairment of insulin secretion can be due to hyperglycemia (glucose toxicity) or elevated islet triglyceride content (lipotoxicity). One group (Seufert et al., 1998) has investigated how impairment of insulin secretion may be initiated by down-regulation of insulin gene transcription in the ZDF rat. They

100

R.G.PETERSON

found that both IDX-1 and C/EBPbeta are regulated in response to sustained hyperglycemia or hyperlipidemia and may be involved in the changes in insulin gene expression, which could in turn lead to defects in insulin release and diabetes mellitus. Leptin Receptor Effects Shimabuto et al. (1997a) reported that normal pancreatic islets are able to lower their triglyceride content in response to leptin. This is accomplished by preventing triglyceride formation and increasing fatty acid oxidation. Leptin did not have a lipid lowering effect on in vivo or in vitro islets from ZDF rats. As a result, ZDF islets have increased triglyceride content and esterification capacity (Unger et al., 1999). The same group demonstrated (Zhou et al., 1997) that when islets from lean rats were cultured with 2 mM free fatty acids (FFA) the proinsulin mRNA ratio rose while there was no change m fa/fa islets. Their findings showed a possible causal relationship between the loss of leptin action in islets and the response of -cells to FFA in ZDF rats. They (Zhou et al., 1998) demonstrated that ZDF islets have an increased capacity to esterify fatty acids. They found significant increases in acetyl Co A carboxylase and fatty acid synthetase in prediabetic and diabetic ZDF islets. Increased leptin levels reduced acetyl CoA carboxylase (ACC) and fatty acid synthetase (FAS) in control islets but did not decrease the high ACC and FAS expression in islets of fa/fa rats. Koyama et al. (1997) demonstrated that islets from ZDF rats were resistant to the lipopenic action of leptin. Nitric Oxide Effects Shimabukuro et al. (1997b) reported data that support the hypothesis that free fatty acid-induced suppression of insulin output in islets from prediabetic ZDF rats is mediated by a rise in nitric oxide (NO) levels in the islets. Exposing islets to agents, which lower NO by blocking nitric-oxide synthase expression, prevented fatty acid-induced suppression. In vivo treatment of prediabetic ZDF rats with the agents prevented the iNOS expression in islets and prevented β-cell loss and hyperglycemia. Wang et al. (1998b) overexpressed the normal leptin receptor (OB-Rb) in ZDF islets by perfusing ZDF pancreata with a vector containing the OB-Rb, cDNA. Islets isolated from the animals demonstrated normalized responses to leptin by: 1) lowering islet triglyceride, 2) blocking TG formation from free fatty acids, 3) suppression of the overproduction of NO, and 4) restoring the preproinsulin mRNA response to free fatty acids. The data support the hypothesis that defective leptin action is a contributing factor in the lipotoxic diabetes seen in ZDF rats. Leptin receptor dysfunction by itself can not be the only factor, since other rats with the defect, such as female fatty ZDF rats and normal fatty Zucker rats, do not become diabetic in the same way. In this review we do show that fatty female ZDF rats can become diabetic on the appropriate high-fat diets. Over-Expression of OB-R Several studies have been done after over-expressing the wild-type OB-R protein in islets from obese diabetic ZDF rats which do not have the normal receptor (Shimabukuro et al., 1998; Wang et al., 1998a; Wang et al., 1998b; Zhou et al., 1998). When the islets were studied, normal leptin receptor activity was restored in β-cells including: 1) raised phosphorylated STAT3 (Wang et al., 1998a), 2) prevention of FA induced Bcl-2 suppression (Shimabukuro et al., 1998), 3) lowered islet triglyceride by blocking its formation from free fatty acids (Wang et al., 1998b), 4) reducing the overproduction of NO (Wang et al., 1998b), 5) restored preproinsulin mRNA response to free fatty acids (Wang et al., 1998b), 6) raised GLUT-2 protein without leptin (Wang et al., 1998a), 7) raised GK protein without leptin and raised it

THE ZUCKER DIABETIC FATTY (ZDF) RAT

101

further with leptin (Wang et al., 1998a), 8) increased preproinsulin mRNA without leptin (Wang et al., 1998a), and 9) restored glucose stimulated insulin secretion (Wang et al., 1998a). Zhou et al. (1998) demonstrated the restoration of expression of peroxisome proliferator-activated receptor alpha and enzymes of fatty acid oxidation are corrected by over-expression of normal OB-R in islets of fa/fa rats. Another report (Thomas et al., 1992) provided evidence that relates to the results. They found that islets from pigs, transplanted under the kidney capsule of obese male ZDF rats, prevented diabetes. The rats, however, did not lose weight and they continued to have high insulin levels (both rat and pig insulin were detected). Effects of Free Fatty Acids Lee et al. (1994) demonstrated a rise in plasma free fatty acids (FFA) which correlated with the triacylglycerol content of prediabetic islets and plasma glucose concentrations. They showed that normal rat islets, which were cultured in medium containing FFA, had increased basal insulin secretion. Both normal islets and prediabetic islets exposed to FFA in the medium show reduced first-phase glucose-stimulated insulin secretion. Unger (1995) reviewed much of the literature which supports this idea. A number of additional ZDF studies have been published supporting and expanding the effects of free fatty acids on βcells (Hirose et al., 1996; Lee et al., 1997; Shimabukuro et al., 1997a; Zhou et al., 1997). Hirose et al. (1996) concluded that islets from ZDF rats lack the capacity for FFA to induce β-cell function by increasing sensitivity of the β-cell to glucose. Apoptosis of (β-cells Apoptosis may be due to a cascade of events which are related to free fatty acids levels and their metabolism in the β-cell (Lee et al., 1994; Unger, 1995). The whole process appears to hinge on accumulation of lipid in the β-cell (Lee et al., 1997). The initiating step in the β-cells may be due to the defective leptin receptor, which is expressed in the fa/fa genotype as discussed above. Leptin has a lipopenic action in normal islets; this effect is not present in the obese diabetic ZDF rat due to the missing actions of leptin. The missing leptin effects make them more vulnerable to FFA-induced functional impairment and triglyceride accumulation (Hirose et al., 1996; Unger, 1995; Wang et al., 1998a). Since obese females from the ZDF strain and other obese rat strains with the defective leptin receptor do not become diabetic on normal chows, there are probably other factors in the obese ZDF male which contribute to the development of the diabetic condition. Ceramide Metabolism Two reports (Shimabukuro et al., 1998; Shimabukuro et al., 1998b) have examined the relationship of ceramide metabolism as a mechanism behind β-cell loss. Ceramide is thought to be a messenger in cytokineinduced apoptosis (Shimabukuro et al., 1998b). Increases in NO are associated with β-cell loss (Shimabukuro et al., 1997b). Ceramide and NO accumulation were seen in islets when islets were cultured with increased levels of FFA. When Triacin C (a fatty acyl-CoA synthetase inhibitor) and troglitazone (an enhancer of FFA oxidation) are used, DNA laddering was blocked and NO production was reduced (Shimabukuro et al., 1998b). When the OB-R receptor was overexpressed in ZDF islets similar results were obtained (Shimabukuro et al., 1998).

102

R.G.PETERSON

Changes in β-Cell Mass Pick et al. (1998) demonstrated that the β-cell mass in prediabetic ZDF rats was similar to non-diabetic Zucker fatty rats and greater than that in lean control rats. However, after the onset of diabetes, the β-cell mass in the ZDF rats was significantly lower than that of the Zucker fatty rats. Although the β-cell proliferation rate remained the highest in the ZDF rats, they had a lower β-cell mass than the Zucker fatty controls. The authors suggested that lower cell mass was due to an increased rate of cell death by apoptosis in the ZDF rat. Shimabukuro et al. (1998b) demonstrated similar prediabetic proliferation of β-cells and loss of cells in the diabetic condition. They (Shimabukuro et al., 1997b) reported that free fatty acid-induced suppression of insulin output in islets from prediabetic ZDF rats was potentially mediated by a rise in nitric oxide (NO) levels. Agents that lower NO prevented the suppression of insulin output. ZDF rats exhibit early β-cell compensation (hyperplasia) to accommodate for their insulin resistance; the accommodation was followed by decompensation and β-cell loss (Shimabukuro et al., 1998b). The cell loss was similar to what was observed in the human condition. The mechanism by which this β-cell loss (apoptosis) occurred has been studied by two major laboratories (Pick et al., 1998; Shimabukuro et al., 1998; Shimabukuro et al., 1998b). Pick et al. (1998) observed that despite a higher proliferation rate, the diabetic ZDF rat had a lower β-cell mass than the fatty Zucker rat. Increased lipid levels in blood and tissues are a hallmark of the obesity trait expressed by adult Zucker fatty rats. One group (Berk et al., 1997) examined oleate uptake in hepatocytes, cardiac myocytes, and adipocytes from adult male Wistar (+/+), Zucker lean (fa/+) and fatty (fa/fa), and Zucker diabetic fatty (ZDF) rats. They found that there was a tissuespecific increase in oleate uptake in Zucker fatty (fa/fa) and ZDF (fa/fa) adipocytes, which was greater in the ZDF rats. Oleate uptake was studied in adipocytes from male +/+, fa/+, and fa/fa weanlings. Although there were no differences in obesity, blood glucose, or plasma fatty acids in the groups, there was already a difference in the oleate uptake in the adipocytes of the fa/fa group. COMPLICATIONS AND OTHER ABNORMALITIES Lipids and Lipoproteins Koyama et al. (1997) investigated how both insulin resistance and hyperinsulinemia might be secondary to increased tissue triglyceride. They studied correlations between triglyceride content of skeletal muscle, liver, and pancreas and plasma insulin, [insulin]×[glucose] product, and β-cell function. Their results are consistent with the hypothesis that tissue triglyceride content sets the level of both insulin resistance and insulin production. Vadlamudi et al. (1997) demonstrated that synthesis of total lipids by the liver slices, and glyceride synthesis in tissue homogenates, were significantly higher in the Zucker rats but unchanged in the ZDF animals when compared to lean controls. They found that FAS and LPL activities were elevated in the Zucker rats but unchanged in the ZDF rats and concluded that the measurements could not explain the hypertriglyceridemia observed in Zucker fatty and ZDF rats. Sparks et al. (1998) did a comprehensive study of the lipoprotein and apolipoprotein alterations that are seen in the ZDF rat during diabetic progression from hyperinsulinemia to insulinopenia. The results demonstrate significant changes in pre- triglyceride-rich lipoproteins at 10 weeks of age, while there were changes in all the lipoprotein fractions at 20 weeks. Between the times there was an increase in the amount of apo B which was accompanied by a shift in the amount (particularly the apo B 100) from the very-low-density fraction into the intermediate and LDL fractions. The above results and other data presented in the study demonstrate lipoprotein changes as type 2 diabetes progresses from the hyperinsulinemic to the insulinopenic stage.

THE ZUCKER DIABETIC FATTY (ZDF) RAT

103

Amylin Changes in amylin/insulin secretion have been suggested as a mechanism for the onset of type 2 diabetes. A couple of studies have supported the abnormal amylin secretion hypothesis using data gathered from ZDF rats (Gedulin et al., 1991; Pieber et al, 1993). Effects of Incretin Hormones Incretin hormones from the gut have an influence on the sensitivity that the β-cell has for glucose. Relative to this phenomenon, GLP-1 content of the intestine has been investigated (Berghofer et al., 1997) and it was found that incretin hormone expression (GIP and GLP-1) is unaltered by the diabetic state in the ZDF rat. Despite the normal state of hormone expression, additional GLP-1 is effective in sensitizing the pancreas of the ZDF rat. Hargrove et al. (1995) have shown that infusion of GLP-1(7–37) in hyperinsulinemic, hyperglycemie ZDF rats produced a transitory increase in plasma insulin concentration and normalized the plasma glucose concentration. In a study using pancreas perfusions, Shen et al. (1998) demonstrated a relative insensitivity of the ZDF pancreas from lean and obese males and females to 10mM glucose; GLP-1 treatment increased the sensitivity to glucose. In a more recent study (Young et al., 1999) a GLP-1 agonist (exendin-4) was able to both acutely and chronically lower blood glucose in ZDF obese male rats. Insulin Resistance One of the hallmarks of the prediabetic and type 2 diabetic condition is insulin resistance (Ishikawa et al., 1998). Decreased levels of GLUT4 transporter have been reported in some muscle groups in the diabetic ZDF rat (Danis and Yang, 1993; Dohm et al., 1993; Friedman et al., 1991; Handberg et al., 1993). Decreased transporter levels is consistent with the development of insulin resistance. Considine et al. (1995) reported that PKC isoforms are significantly increased in the ZDF and have concluded that it is possible that PKC-mediated phosphorylation contributes to the insulin resistance. Increased specific proteintyrosine-phosphatase activity has been associated with the insulin resistance in the ZDF rat (Ahmad and Goldstein, 1995). Vascular Changes Bohlen and Lash (1995) found that endothelial- and nonendothelial-mediated dilation of intestinal arterioles are not modified by acetylcholine, ADP, and nitroprusside in ZDF obese males. Treatments caused equivalent dilation in controls and diabetics for both large and intermediate diameter arterioles. Additional treatment of the diabetic with streptozotocin lowered their insulin concentration, but did not significantly effect the resting plasma glucose concentration. The additional level of diabetes caused a decrease in the endothelial-dependent vasodilation. In a later study this group (Jin and Bohlen, 1997) provided evidence for impaired flow-mediated dilation in ZDF rat. They demonstrated decreased resting blood flow after hyperglycemia in fatty Zucker (normoglycemic) rats. The results showed that both acute and chronic hyperglycemia can disturb endothelial regulation of blood flow in intestinal vasculature. Connors et al. (1997) examined the expression of the aldose reductase and IGF-I in ZDF rats. They were not able to determine any changes in aldose reductase and IGF-I expression between ZDF lean and fatty diabetic ZDF rats in various vascular tissues they looked at. Another study (Kawaguchi et al., 1999) has demonstrated that an inhibitor of nitric oxide synthase (NOS), 1-nitro-arginine methyl ester (L-NAME), prevented the decrease in blood pressure that is associated

104

R.G.PETERSON

with insulin treatment. These findings suggested that insulin reduced blood pressure in the ZDF rat by stimulating NOS activation and NO production. Retinal Changes A small number of studies have been done on retinas of diabetic ZDF rats. Quantitative assessment of capillary cell nuclear density showed that diabetic retinas were hypercellular and that the retinal capillary basement membranes were thicker when compared to lean rats (Danis and Yang, 1993). Ottlecz et al. (1993) demonstrated changes in the specific activities of retinal Na+, K+-ATPase (Na, K-ATPase) and Mg2+ATPase (Mg-ATPase) in the ZDF at different times after the onset of the disease. Na, K-ATPase specific activity and the Na, K-ATPase activity for the alpha 3 isozyme were significantly lower while the MgATPase activity was significantly increased. The same laboratory (Ottlecz et al., 1996) investigated the relationship of serum and/or retinal angiotensin-converting enzyme (ACE) activity with the decrease in Na, K-ATPase activity in the retina of experimentally diabetic ZDF rats. Although there was a decrease in both serum and retinal ACE, their interpretation of the data did not support the cause effect relationship between ACE and Na, K-ATPase activity. Glucose Transporters Changes in GLUT-2 levels in pancreatic β-cells were described above. Other tissues show changes in GLUT-2 and GLUT-4 levels in diabetic ZDF rats. One of the studies (Slieker et al., 1992) demonstrated that the diabetic ZDF rat showed multiple changes in glucose transporter levels in different tissues. Muscle, adipose tissue, and liver appeared to be regulated in a tissue-selective manner in response to changes in insulin and glucose. They noted that alterations in GLUT2 and/or GLUT4 protein levels appeared to be related to the severity of the diabetic state and not to obesity. Another laboratory (Dohm et al., 1993; Friedman et al., 1991) specifically looked at GLUT-4 in various muscle groups and found that some types of muscles changed and some did not. Mitochondrial Enzymes MacDonald et al. (1996) demonstrated that the enzyme activity of mitochondrial glycerol phosphate dehydrogenase (mGPD) and pyruvate carboxylase (PC) were both significantly reduced in the islets of 6and 12-week-old ZDF rats. Although they felt that the “modest” reductions in mGPD and PC might contribute to the diabetic condition, they did not feel that it was likely to represent the primary genetic defect in the ZDF rat. Gastric Mobility Green et al. (1997) demonstrated that ZDF rats emptied glucose from the stomach significantly faster than controls. They found that drugs, which stimulate cholecystokinin and secretin release, significantly slowed gastric emptying, and significantly reduced postprandial hyperglycemia in diabetic rats.

THE ZUCKER DIABETIC FATTY (ZDF) RAT

105

Wound Healing and Infections Wound healing appears to be significantly effected in the diabetic ZDF rat. Vrabec (1998) demonstrated that healing of acute tympanic membrane perforations in Zucker diabetic fatty rats demonstrated a significantly prolonged closure time when compared to streptozotocin-induced diabetes and normal control rats. Difficulties in healing have been observed in our laboratories in diabetic ZDF rats after surgically implanting blood pressure telemetry transmitters. The healing problems are not seen in control rats or if the surgery is performed before the development of diabetes in ZDF rat. One study has shown a difference in the effect of endotoxin on lipolysis and lipoprotein lipase activity (Lanza-Jacoby et al., 1995). This group demonstrated that exposure to endotoxin would cause an increase in the lipolytic rate in adipocytes from obese and an even greater increase in diabetic obese rats than in control lean rats. Nephropathy Although changes do occur in the kidney of the ZDF rat with the diabetes, the strain has a propensity to develop hydronephrosis and thus it is not a good model for the study of diabetic nephropathy (Vora et al., 1996). Neuropathy Two publications have reported peripheral nerve changes in obese ZDF males (Mathew et al., 1997; Peterson et al., 1990a). The first of these papers (Peterson et al., 1990a) demonstrated reduced conduction velocities, nerve edema, and morphological abnormalities in the nerve fibers. The second paper (Mathew et al., 1997) demonstrated metabolic abnormalities in peripheral nerve which are consistent with other diabetic animal models. The results showed changes in phospholipid metabolism, ATPase activity, protein kinase C expression and PKC activity. TREATMENT PPARγ Agonists A number of studies have been performed with the ZDF rat demonstrating the effectiveness of PPAR agonists. Drugs are most effective when given before diabetes is fully developed. The thiazolidinediones were the first of the agonists to be approved for use in human diabetes. The compounds have been effective in the ZDF rat; as a result they have been used in testing many new potential drugs (Brown et at., 1999; Pickavance et al., 1998; Shibata et al., 1999; Shimabukuro et al., 1998a; Sreenan et al., 1999; Sreenan et al., 1996; Upton et al., 1998; Zhang et al., 1996; Zierath et al., 1998). Effective compounds attenuate the development of overt diabetes through improved insulin sensitivity and maintenance of -cell function. The results lead to improvement in hyperglycemia and hyperinsulinemia. Treatment with some of TZD compounds lowers circulating and pancreatic lipids (Brown et al., 1999; Shimabukuro et al., 1998a). However, when these drugs are used in ZDF rats, there was a profound increase in body weight without an increase in hyperphagia (Zhang et al., 1996). Other classes of drugs, which are agonists for the PPAR receptor (Cobb et al., 1998), normalize glycemia in the ZDF rat. Sturis et al. (1995) observed that combined acarbose and pioglitazone treatment improved but did not normalize glucose levels, and it did not improve

106

R.G.PETERSON

entrainment of insulin release. Our laboratories at Genetic Models, Inc. have conducted experiments with troglitazone, metformin and combinations of both compounds in diabetic males and diabetic females (induced by diet). The studies demonstrated that metformin alone partially prevented hyperglycemia in both the males and females. Troglitazone and the combination of troglitazone and metformin were both effective. As in other experiments the troglitazone treated males put on more weight than their fully diabetic brothers did; this was true of both the troglitazone and combination treated rats. The females, however, were different; troglitazone treated fatty females put on weight at a comparable rate to treated males whereas females treated with the combination of troglitazone and metformin put on weight at a significantly lower rate. The experiment raises some interesting potential benefits of treatments with combinations of drugs to control weight gain. Troglitazone action in the ZDF rat islets has been reported in one study. One of the ways that leptin normally functions is by up-regulating the expression of uncoupling protein-2 (UCP-2) mRNA (Shimabukuro et al., 1997). Although this study has demonstrated up-regulation in islets of normal rats, they showed that it had no effect in islets of obese ZDF rats. The report demonstrated that troglitazone increased the UCP-2/beta-actin mRNA ratio in both lean and obese ZDF rats. Other Treatments Other treatments have been used in effectively treating the ZDF rats, including vanadium, magnesium, acarbose, conjugated linoleic acid, beta 3-adrenergic agonists, metformin, IGF-1 and monoxidine. Poucheret et al. (1998) reference an unpublished study, in a review on the effects of vanadium on diabetes, and comment on the results that demonstrate that vanadyl sulfate can lower elevated blood glucose, cholesterol and triglycerides in the Zucker diabetic fatty rat, as well as in many other models. Another vanadium compound, bis(maltolato)oxovanadium (Yuen et al., 1999) has been given acutely by gavage chronically in drinking water. Acute doses resulted in decreased blood glucose, insulin and plasma triglyceride. Chronic administration significantly reduced plasma glucose levels and preserved pancreatic βcell function. A magnesium-supplemented diet prevented deterioration of glucose tolerance and delayed the development of diabetes in the ZDF rat (Balon et al., 1995). The magnesium treatment preserved higher insulin and C-peptide concentrations and higher levels of expression of GLUT-2 and insulin mRNA. The disaccharidase inhibitor, acarbose was effective in glycemic control (Dohm et al., 1993; Peterson, 1994a) and in reducing some of the complications of diabetes (Peterson, 1994a; Peterson et al., 1993). One laboratory (Dohm et al., 1993; Friedman et al., 1991) showed significant increases in GLUT4 protein in gastrocnemius and red quadriceps muscles with acarbose treatment of diabetic rats. Houseknecht et al. (1998) have presented evidence that conjugated linoleic acid is able to normalize impaired glucose tolerance and improve hyperinsulinemia in the pre-diabetic ZDF rat. The study provided evidence that the effect is consistent with activation of PPAR gamma receptors and suggested that dietary CLA may prove to be an important therapy for the prevention and treatment of type 2 diabetes. Liu et al. (1998)used a selective beta 3-adrenergic agonist and found that treatment normalized hyperglycemia and reduced hyperinsulinemia and circulating free fatty acid (FFA) levels. It improved glucose tolerance and reduced insulin response during an intravenous glucose tolerance test by a mechanism similar to that induced by chronic cold exposure. Metformin has been used in studies (Handberg et al., 1993; Sreenan et al., 1996). The drug has significant effects on diabetes and some of the other conditions associated with diabetes (Handberg et al., 1993; Sreenan et al., 1996) but it did not improve the GLUT4 levels in skeletal muscles (Handberg et al., 1993). In another study Clark et al. (1997) demonstrated that treatment with IGF-1, alone or in combination with growth hormone releasing peptide or growth hormone, improved the diabetic state and stimulated growth.

THE ZUCKER DIABETIC FATTY (ZDF) RAT

107

Most recently monoxidine, an antihypertensive agent, has been shown to have a beneficial effect on abnormal glucose metabolism and renal protein excretion in ZDF rats (Yakubu-Madus et al., 1999). These data indicate that the obese male and female ZDF models have significant utility for testing new experimental drugs that are being developed for the treatment of diabetes. The models have also been very useful in defining the mechanisms by which diabetes develops and how the drugs modify the pathological metabolic condition. Based on our results and the results presented in this review we feel that the obese male and female ZDF rats both make distinct and important models for the study and treatment of diabetes. REFERENCE Ahmad, F. and Goldstein, B.J. (1995) Increased abundance of specific skeletal muscle proteintyrosine phosphatases in a genetic model of insulin-resistant obesity and diabetes mellitus. Metabolism, 44, 1175–1184. Balon, T.W., Gu, J.L., Tokuyama, Y., Jasman, A.P. and Nadler, J.L. (1995) Magnesium supplementation reduces development of diabetes in a rat model of spontaneous NIDDM. Am. J. Physiol., 269, E745–752. Berghofer, P., Peterson, R.G., Schneider, K., Fehmann, H.C. and Goke, B. (1997) Incretin hormone expression in the gut of diabetic mice and rats. Metabolism, 46, 261–267. Berk, P.D., Zhou, S.L., Kiang, C.L., Stump, D., Bradbury, M. and Isola, L.M. (1997) Uptake of long chain free fatty acids is selectively up-regulated in adipocytes of Zucker rats with genetic obesity and non-insulin-dependent diabetes mellitus. J. Biol. Chem., 272, 8830– 8835. Bohlen, H.G. and Lash, J.M. (1995) Endothelial-dependent vasodilation is preserved in non-insulin-dependent Zucker fatty diabetic rats. Am. J. Physiol., 268, H2366–2374. Brown, K.K., Henke, B.R., Blanchard, S.G., Cobb, J.E., Mook, R., Kaldor, I., et al. (1999) A novel N-aryl tyrosine activator of peroxisome proliferator-activated receptor- reverses the diabetic phenotype of the Zucker diabetic fatty rat. Diabetes, 48, 1415–1424. Chua, S.C.J., White, D.W., Wu-Peng, X.S., Liu, S.M., Okada, N., Kershaw, E.E., et al. (1996) Phenotype of fatty due to Gln269Pro mutation in the leptin receptor (Lepr). Diabetes, 45, 1141–1143. Clark, J., Palmer, C.J. and Shaw, W.N. (1983) The diabetic Zucker fatty rat. Proc. Soc. Exp. Biol. Med., 173, 68–75. Clark, R.G., Thomas, G.B., Mortensen, D.L., Won, W.B., Ma, Y.H., Tomlinson, E.E., et al. (1997) Growth hormone secretagogues stimulate the hypothalamic-pituitary-adrenal axis and are diabetogenic in the Zucker diabetic fatty rat. Endocrinology, 138, 4316–4323. Cobb, J.E., Blanchard, S.G., Boswell, E.G., Brown, K.K., Charifson, P.S., Cooper, J.P., et al. (1998) N-(2Benzoylphenyl)-L-tyrosine PPARgamma agonists. 3. Structure-activity relationship and optimization of the N-aryl substituent. J. Med. Chem., 41, 5055–5069. Cockburn, B.N., Ostrega, D.M., Sturis, J., Kubstrup, C., Polonsky, K.S. and Bell, G.I. (1997) Changes in pancreatic islet glucokinase and hexokinase activities with increasing age, obesity, and the onset of diabetes. Diabetes, 46, 1434–1439. Connors, B., Lee, W.H., Wang, G., Evan, A.P. and Bohlen, H.G. (1997) Aldose reductase and IGF-I gene expression in aortic and arteriolar smooth muscle during hypo- and hyperinsulinemic diabetes. Microvasc. Res., 53, 53–62. Considine, R.V., Nyce, M.R., Allen, L.E., Morales, L.M., Triester, S., Serrano, J., et al. (1995) Protein kinase C is increased in the liver of humans and rats with non-insulin-dependent diabetes mellitus: an alteration not due to hyperglycemia. J. Clin. Invest., 95, 2938–2944. Corsetti, J.P, Sparks, J.D., Peterson, R.G., Smith, R.L. and Sparks, C.E. (2000) Effects of dietary fat on the development of non-insulin dependent diabetes mellitus in obese Zucker diabetic fatty male and female rats. Atherosclerosis, 48, 231–241. Danis, R.P. and Yang, Y. (1993) Microvascular retinopathy in the Zucker diabetic fatty rat. Invest. Ophthalmol. Vis. Sci., 34, 2367–2371. Dohm, G.L., Friedman, J.E. and Peterson, R.G. (1993) Acarbose treatment of non-insulin-dependent diabetic fatty (ZDF/Drt-fa) rats restores expression of skeletal muscle glucose transporter GLUT4. In Drugs in Development,

108

R.G.PETERSON

Vol. 1. α-G/ucosidase Inhibition: Potential Use in Diabetes, edited by J.R.Vasselli, C.A.Maggio and A.Scriabine, pp. 173–180. Branford, Connecticut: Neva Press. Friedman, J.E., de Vente, J.E., Peterson, R.G. and Dohm, G.L. (1991) Altered expression of muscle glucose transporter GLUT-4 in diabetic fatty Zucker rats (ZDF/Drt-fa). Am. J. Physiol., 261, E782–788. Gedulin, B., Cooper, G.J. and Young, A.A. (1991) Amylin secretion from the perfused pancreas: dissociation from insulin and abnormal elevation in insulin-resistant diabetic rats. Biochem. Biophys. Res. Commun., 180, 782–789. Green, G.M., Guan, D., Schwartz, J.G. and Phillips, W.T. (1997) Accelerated gastric emptying of glucose in Zucker type 2 diabetic rats: role in postprandial hyperglycaemia. Diabetologia, 40, 136–142. Greenhouse, D.D., Michaelis, O.E.4. and Peterson, R.G. (1988) The development of fatty and corpulent rat strains. In New Models for Genetically Obese Rats for Studies in Diabetes, Heart Disease and Complications of Obesity, edited by C.T.Hansen and O.E.4. Michaelis, pp. 3– 6. Division of Research Services, Veterinary Resources Branch, Bethesda: NIH. Handberg, A., Kayser, L., Hoyer, P.E., Voldstedlund, M., Hansen, H.P. and Vinten, J. (1993) Metformin ameliorates diabetes but does not normalize the decreased GLUT 4 content in skeletal muscle of obese (fa/fa) Zucker rats. Diabetologia, 36, 481–486. Hargrove, D.M., Nardone, N.A., Persson, L.M., Parker, J.C. and Stevenson, R.W. (1995) Glucose-dependent action of glucagon-like peptide-1 (7–37) in vivo during short- or long-term administration. Metabolism, 44, 1231–1237. Hirose, H., Lee, Y.H., Inman, L.R., Nagasawa, Y., Johnson, J.H. and Unger, R.H. (1996) Defective fatty acid-mediated beta-cell compensation in Zucker diabetic fatty rats. Pathogenic implications for obesity-dependent diabetes. J. Biol. Chem., 271, 5633–5637. Houseknecht, K.L., Vanden Heuvel, J.P., Moya-Camarena, S.Y., Portocarrero, C.P., Peck, L.W., Nickel, K.P., et al. (1998) Dietary conjugated linoleic acid normalizes impaired glucose tolerance in the Zucker diabetic fatty fa/fa rat. Biochem. Biophys. Res. Commun., 24, 678– 682. Ishikawa, Y., Saito, M.N., Ikemoto, T., Takeno, H., Watanabe, K. and Tani, T. (1998) Actions of the novel oral antidiabetic agent HQL-975 in insulin-resistant non-insulin-dependent diabetes mellitus model animals. Diabetes Res. Clin. Pract., 41, 101–111. Jin, J.S. and Bohlen, H.G. (1997) Non-insulin-dependent diabetes and hyperglycemia impair rat intestinal flowmediated regulation. Am. J. Physiol., 272, H728–734. Johnson, J.H., Ogawa, A., Chen, L., Orci, L., Newgard, C.B., Alam, T., et al. (1990) Under-expression of β cell high Km glucose transporters in noninsulin-dependent diabetes. Science, 250, 546–549. Kawaguchi, M., Koshimura, K., Murakami, Y., Tsumori, M., Gonda, T. and Kato, Y. (1999) Antihypertensive effect of insulin via nitric oxide production in the Zucker diabetic fatty rat, an animal model for non-insulin-dependent diabetes mellitus. Eur. J. Endocrinol., 140, 341–349. Koyama, K., Chen, G., Lee, Y. and Unger, R.H. (1997) Tissue triglycerides, insulin resistance, and insulin production: implications for hyperinsulinemia of obesity. Am. J. Physiol., 273, E708–713. Lanza-Jacoby, S., Rose, G.L., Rosato, E.F., Sedkova, N. and Considine, R.V. (1995) In vitro effect of endotoxin on lipolysis and lipoprotein lipase in adipocytes from lean, obese and obese diabetic Zucker rats. J. Endotox. Res., 2, 449–454. Lee, Y., Hirose, H., Ohneda, M., Johnson, J.H., McGarry, J.D. and Unger, R.H. (1994) Beta-cell lipotoxicity in the pathogenesis of non-insulin-dependent diabetes mellitus of obese rats: impairment in adipocyte-beta-cell relationships. Proc. Natl. Acad. Sci. USA, 91, 10878–10882. Lee, Y., Hirose, H., Zhou, Y.T., Esser, V., McGarry, J.D. and Unger, R.H. (1997) Increased lipogenic capacity of the islets of obese rats: a role in the pathogenesis of NIDDM. Diabetes. 46, 408–413. Liu, X., Perusse, F. and Bukowiecki, L.J. (1998) Mechanisms of the antidiabetic effects of the beta 3-adrenergic agonist CL-316243 in obese Zucker-ZDF rats. Am. J. Physiol., 274, R1212–1219. MacDonald, M.J., Tang, J. and Polonsky, K.S. (1996) Low mitochondrial glycerol phosphate dehydrogenase and pyruvate carboxylase in pancreatic islets of Zucker diabetic fatty rats. Diabetes, 45 1626–1630.

THE ZUCKER DIABETIC FATTY (ZDF) RAT

109

Mathew, J., Bianchi, R., McLean, W.G., Peterson, R.G., Roberts, R.E., Savaresi, S., et al. (1997) Phosphoinositide metabolism, Na, K-ATPase and protein kinase C are altered in peripheral nerve from Zucker fatty diabetic rats (ZDF/Gmi-fa). Neurosci. Res. Comm., 20, 21– 30. Milburn, J.L.J., Ohneda, M., Johnson, J.H. and Unger, R.H. (1993) Beta-cell GLUT-2 loss and non-insulin-dependent diabetes mellitus: current status of the hypothesis. Diabetes Metab. Rev., 9, 231–236. Ohneda, M., Inman, L.R. and Unger, R.H. (1995) Caloric restriction in obese pre-diabetic rats prevents beta-cell depletion, loss of beta-cell GLUT 2 and glucose incompetence. Diabetologia, 38, 173–179. Ohneda, M.Johnson, J.H., Inman, L.R., Chen, L., Suzuki, K., Goto, Y, et al (1993a) GLUT2 expression and function in beta-cells of GK rats with NIDDM. Dissociation between reductions in glucose transport and glucose-stimulated insulin secretion. Diabetes, 42, 1065–1072. Ohneda, M., Johnson, J.H., Inman, L.R. and Unger, R.H. (1993b) GLUT-2 function in glucose-unresponsive beta cells of dexamethasone-induced diabetes in rats. J. Clin. Invest., 92, 1950–1956. Ohneda, M., Johnson, J.H., Lee, Y.H., Nagasawa, Y. and Unger, R.H. (1994) Post-GLUT-2 defects in beta-cells of noninsulin-dependent diabetic obese rats. Am. J. Physiol., 267, E968–974. Orci, L., Ravazzola, M., Baetens, D., Inman, L., Amherdt, M., Peterson, R.G., et al. (1990) Evidence that downregulation of β-cell glucose transporters in non-insulin-dependent diabetes may be the cause of diabetic hyperglycemia. Proc. Natl. Acad. Set. USA, 87, 9953– 9957. Ottlecz, A., Bensaoula, T., Eichberg, J. and Peterson, R.G. (1996) Angiotensin-converting enzyme activity in retinas of streptozotocin-induced and Zucker diabetic rats. The effect of angiotensin II on Na+,K( +)-ATPase activity. Invest. Ophthalmol. Vis. Sci., 37, 2157– 2164. Ottlecz, A., Garcia, C.A., Eichberg, J. and Fox, D.A. (1993) Alterations in retinal Na+, K(+)-ATPase in diabetes: streptozotocin-induced and Zucker diabetic fatty rats. Curr. Eye. Res., 12, 1111–1121. Peterson, R.G. (1994a) alpha-Glucosidase inhibitors in diabetes: lessons from animal studies. Eur.J.Clin. Invest., 24, Suppl. β, 11–18. Peterson, R.G. (1994b) The Zucker diabetic fatty (ZDF) rat. In Lessons From Animal Diabetes V, edited by E.Shafrir, pp. 225–230. London: Smith-Gordon. Peterson, R.G., Doss, D.I., Neel, M.-A., Little, L.A., Kincaid, J.C. and Eichberg, J. (1993) The effectiveness of acarbose in treating Zucker diabetic fatty rats (ZDF/Drt-fa). In Drugs in Development, Vol. I. -Glucosidase Inhibition: Potential Use in Diabetes, edited by J.R. Vasselli, C.A.Maggio and A.Scriabine, pp. 167–172. Branford, Connecticut: Neva Press. Peterson, R.G., Neel, M.-A., Little, L.A., Kincaid, J.C. and Eichberg, J. (1990a) Neuropathic complications in the Zucker diabetic fatty rat (ZDF/Drt-fa). In Frontiers in Diabetes Research. Lessons from Animal Diabetes III, edited by E.Shafrir, pp. 456–458. London: Smith-Gordon. Peterson, R.G., Shaw, W.N., Neel, M.-A., Little, L.A. and Eichberg, J. (1990b) Zucker diabetic fatty fat as a model fornon-insulin-dependent diabetes mellitus. ILAR News, 32, 16– 19. Pick, A., Clark, J., Kubstrup, C., Levisetti, M., Pugh, W., Bonner-Weir, S., et al. (1998) Role of apoptosis in failure of beta-cell mass compensation for insulin resistance and beta-cell defects in the male Zucker diabetic fatty rat. Diabetes, 47, 358–364. Pickavance, L., Widdowson, P.S., King, P., Ishii, S., Tanaka, H. and Williams, G. (1998) The development of overt diabetes in young Zucker Diabetic Fatty (ZDF) rats and the effects of chronic MCC-555 treatment. Br. J. Pharmacol., 125, 767–770. Pieber, T.R., Stein, D.T., Ogawa, A., Alam, T., Ohneda, M., McCorkle, K., et al. (1993) Amylin-insulin relationships in insulin resistance with and without diabetic hyperglycemia. Am. J. Physiol., 265, E446–453. Polonsky, K.S. (1995) Lilly Lecture 1994. The beta-cell in diabetes: from molecular genetics to clinical research. Diabetes, 44, 705–717. Poucheret, P., Verma, S., Grynpas, M.D. and McNeill, J.H. (1998) Vanadium and diabetes. Mol. Cell. Biochem., 188, 73–80. Roe, M.W., Worley, J.F.3., Tokuyama, Y., Philipson, L.H., Sturis, J., Tang, J., et al. (1996) NIDDM is associated with loss of pancreatic beta-cell L-type Ca2+channel activity. Am. J. Physiol., 270, E133–140.

110

R.G.PETERSON

Seufert, J., Weir, G.C. and Habener, J.F. (1998) Differential expression of the insulin gene transcriptional represser CCAAT/enhancer-binding protein beta and transactivator islet duodenum homeobox-1 in rat pancreatic beta cells during the development of diabetes mellitus. J. Clin. Invest., 101, 2528–2539. Shen, H.Q., Roth, M.D. and Peterson, R.G. (1998) The effect of glucose and glucagon-like peptide-1 stimulation on insulin release in the perfused pancreas in a non-insulin-dependent diabetes mellitus animal model. Metabolism, 47, 1042–1047. Shibata, T., Matsui, K., Nagao, K., Shinkai, H., Yonemori, F. and Wakitani, K. (1999) Pharmacological profiles of a novel oral antidiabetic agent, JTT-501, an isoxazolidinedione derivative. Eur. J. Pharmacol., 364, 211–219. Shimabukuro, M., Higa, M., Zhou, Y.T., Wang, M.Y, Newgard, C.B. and Unger, R.H. (1998) Lipoapoptosis in Betacells of Obese Prediabetic fa/fa Rats. Role of serine palmitoyltransferase overexpression. J. Biol. Chem., 273, 32487–32490. Shimabukuro, M., Koyama, K., Chen, G., Wang, M.Y, Trieu, F., Lee, Y., et al. (1997a) Direct antidiabetic effect of leptin through triglyceride depletion of tissues. Proc. Natl. Acad. Sci. USA, 94, 4637–4641. Shimabukuro, M., Ohneda, M., Lee, Y. and Unger, R.H. (1997b) Role of nitric oxide in obesity-induced beta cell disease. J. Clin. Invest., 100, 290–295. Shimabukuro, M., Wang, M.Y, Zhou, Y.T., Newgard, C.B. and Unger, R.H. (1998) Protection against lipoapoptosis of beta cells through leptin-dependent maintenance of Bcl-2 expression. Proc. Natl.Acad. Sci. USA., 95, 9558–9561. Shimabukuro, M., Zhou, Y.T., Lee, Y. and Unger, R.H. (1997) Induction of uncoupling protein-2 mRNA by troglitazone in the pancreatic islets of Zucker diabetic fatty rats. Biochem. Biophys. Res. Commun., 237, 359–361. Shimabukuro, M., Zhou, Y.T., Lee, Y. and Unger, R.H. (1998a) Troglitazone lowers islet fat and restores beta cell function of Zucker diabetic fatty rats. J. Biol. Chem., 273, 3547– 3550. Shimabukuro, M., Zhou, Y.T., Levi, M. and Unger, R.H. (1998b) Fatty acid-induced beta cell apoptosis: a link between obesity and diabetes. Proc. Natl. Acad. Sci. USA., 95, 2498– 2502. Slieker, L.J., Sundell, K.L., Heath, W.F., Osborne, H.E., Bue, J., Manetta, J., et al. (1992) Glucose transporter levels in tissues of spontaneously diabetic Zucker fa/fa rat (ZDF/drt) and viable yellow mouse (Avy/a). Diabetes, 41, 187–193. Sparks, J.D., Phung, T.L., Bolognino, M., Cianci, J., Khurana, R., Peterson, R.G., et al. (1998) Lipoprotein alterations in 10- and 20-week-old Zucker diabetic fatty rats: hyperinsulinemic versus insulinopenic hyperglycemia. Metabolism, 47, 1315–1324. Sreenan, S., Keck, S., Fuller, T., Cockburn, B. and Burant, C.F. (1999) Effects of troglitazone on substrate storage and utilization in insulin-resistant rats. Am. J. Physiol., 276, E1119– E1129 Sreenan, S., Sturis, J., Pugh, W., Burant, C.F. and Polonsky, K.S. (1996) Prevention of hyperglycemia in the Zucker diabetic fatty rat by treatment with metformin or troglitazone. Am. J. Physiol., 271, E742–747. Stoffel, M., Tokuyama, Y., Trabb, J.B., German, M.S., Tsaar, M.L., Jan, L.Y., et al. (1995) Cloning of rat KATP-2 channel and decreased expression in pancreatic islets of male Zucker diabetic fatty rats. Biochem. Biophys. Res. Commun., 212, 894–899. Sturis, J., Pugh, W.L., Tang, J., Ostrega, D.M., Polonsky, J.S. and Polonsky, K.S. (1994) Alterations in pulsatile insulin secretion in the Zucker diabetic fatty rat. Am. J. Physiol, 267, E250–259. Sturis, J., Pugh, W.L., Tang, J. and Polonsky, K.S. (1995) Prevention of diabetes does not completely prevent insulin secretory defects in the ZDF rat. Am. J. Physiol, 269, E786– 892 . Takaya, K., Ogawa, Y., Hiraoka, J., Hosoda, K., Yamori, Y., Nakao, K., et al (1996) Nonsense mutation of leptin receptor in the obese spontaneously hypertensive Koletsky rat. Nat. Genet., 14, 130–131. Tang, J., Pugh, W., Polonsky, K.S. and Zhang, H. (1996) Preservation of insulin secretory responses to P2 purinoceptor agonists in Zucker diabetic fatty rats. Am. J. Physiol, 270, E504–512. Thomas, F.T., Pittman, K. and Thomas, J.M. (1992) Reversal of type II diabetes in Zucker rats by xenogeneic pancreas islet transplant. Transplant. Proc., 24, 647–648. Tokuyama, Y., Fan, Z., Furuta, H., Makielski, J.C., Polonsky, K.S., Bell, G.I., et al. (1996) Rat inwardly rectifying potassium channel Kir6.2: cloning electrophysiological characterization, and decreased expression in pancreatic islets of male Zucker diabetic fatty rats. Biochem. Biophys. Res. Commun., 220, 532–538.

THE ZUCKER DIABETIC FATTY (ZDF) RAT

111

Tokuyama, Y., Sturis, J., DePaoli, A.M., Takeda, J., Stoffel, M., Tang, J., et al. (1995) Evolution of beta-cell dysfunction in the male Zucker diabetic fatty rat . Diabetes, 44, 1447–1457. Unger, R.H. (1991a) Diabetic hyperglycemia: link to impaired glucose transport in pancreatic beta cells. Science, 251, 1200–1205. Unger, R.H. (1991b) Role of impaired glucose transport by beta cells in the pathogenesis of diabetes. J. NIH Research., 3, 77–81. Unger, R.H. (1995) Lipotoxicity in the pathogenesis of obesity-dependent NIDDM. Genetic and clinical implications. Diabetes, 44, 863–870. Unger, R.H., Zhou, Y.T. and Orci, L. (1999) Regulation of fatty acid homeostasis in cells: novel role of leptin. Proc. Natl. Acad. Sci. USA., 96, 2327–2332. Upton, R., Widdowson, P.S., Ishii, S., Tanaka, H. and Williams, G. (1998) Improved metabolic status and insulin sensitivity in obese fatty (fa/fa) Zucker rats and Zucker Diabetic Fatty (ZDF) rats treated with the thiazolidinedione, MCC-555. Br. J. Pharmacol, 125, 1708– 1714. Vadlamudi, S., Dolan, P., Peterson, R.G., Dohm, G.L. and Barakat, H. (1997) The lipogenic potential of lean, obese and diabetic rats . Med. Sci. Res., 25, 543–544. Vora, J.P., Zimsen, S.M., Houghton, D.C. and Anderson, S. (1996) Evolution of metabolic and renal changes in the ZDF/Drt-fa rat model of type II diabetes. J. Am. Soc. Nephrol., 7, 113–117. Vrabec, J.T. (1998) Tympanic membrane perforations in the diabetic rat: a model of impaired wound healing. Otolaryngol. Head Neck Surg., 118, 304–308. Wang, M.Y., Koyama, K., Shimabukuro, M., Mangelsdorf, D., Newgard, C.B. and Unger, R.H. (1998a) Overexpression of leptin receptors in pancreatic islets of Zucker diabetic fatty rats restores GLUT-2, glucokinase, and glucosestimulated insulin secretion. Proc. Natl. Acad. Sci. USA, 95, 11921–11926. Wang, M.Y., Koyama, K., Shimabukuro, M., Newgard, C.B. and Unger, R.H. (1998b) OB-Rb gene transfer to leptinresistant islets reverses diabetogenic phenotype. Proc. Natl. Acad. Sci. USA, 95, 714–718. White, D.W., Wang, D.W., Chua, S.C.J., Morgenstern, J.P., Leibel, R.L., Baumann, H., et al. (1997) Constitutive and impaired signaling of leptin receptors containing the Gln → Pro extracellular domain fatty mutation. Proc. Natl. Acad. Sci. USA, 94, 10657–10662. Wu-Peng, X.S., Chua, S.C.J., Okada, N., Liu, S.M., Nicolson, M. and Leibel, R.L. (1997) Phenotype of the obese Koletsky (f) rat due to Tyr763Stop mutation in the extracellular domain of the leptin receptor (Lepr): evidence for deficient plasma-to-CSF transport of leptin in both the Zucker and Koletsky obese rat. Diabetes, 46, 513–518. Yakubu-Madus, F.E., Johnson, W.T., Zimmerman, K.M., Dananberg, J. and Steinberg, M.I. (1999) Metabolic and hemodynamic effects of moxonidine in the Zucker diabetic fatty rat model of type 2 diabetes. Diabetes, 48, 1093–1100. Young, A.A., Gedulin, B.R., Bhavsar, S., Bodkin, N., Jodka, C., Hansen, B., et al. (1999) Glucose-lowering and insulinsensitizing actions of exendin-4: studies in obese diabetic (ob/ob, db/db) mice, diabetic fatty Zucker rats, and diabetic rhesus monkeys (Macaca mulatta). Diabetes, 48, 1026–1034. Yuen, V.G., Vera, E., Battell, M.L., Li, W.M. and McNeill, J.H. (1999) Acute and chronic oral administration of bis (maltolato)oxovanadium(IV) in Zucker diabetic fatty (ZDF) rats. Diabetes Res. Clin. Pract., 43, 9–19. Zawalich, W.S., Zawalich, K.C., Kelley, G.G. and Shulman, G.I. (1995) Islet phosphoinositide hydrolysis and insulin secretory responses from prediabetic fa/fa ZDF rats. Biochem. Biophys. Res. Commun., 209, 974–980. Zhang, B., Graziano, M.P., Doebber, T.W., Leibowitz, M.D., White-Carrington, S., Szalkowski, D.M., et al. (1996) Down-regulation of the expression of the obese gene by an antidiabetic thiazolidinedione in Zucker diabetic fatty rats and db/db mice. J.Biol. Chem., 271, 9455– 9459. Zhou, Y.T., Shimabukuro, M., Lee, Y., Koyama, K., Higa, M., Ferguson, T., et al. (1998) Enhanced de novo lipogenesis in the leptin-unresponsive pancreatic islets of prediabetic Zucker diabetic fatty rats: role in the pathogenesis of lipotoxic diabetes. Diabetes, 47, 1904–1908. Zhou, Y.T., Shimabukuro, M., Lee, Y., Koyama, K., Trieu, F. and Unger, R.H. (1997) Leptin normalizes the impaired response of proinsulin mRNA to long chain fatty acids in heterozygous Zucker diabetic fatty rats. J. Biol. Chem., 272, 25648–25651.

112

R.G.PETERSON

Zhou, Y.T., Shimabukuro, M., Wang, M.Y., Lee, Y., Higa, M., Milburn, J.L., et al. (1998) Role of peroxisome proliferatoractivated receptor alpha in disease of pancreatic beta cells. Proc. Natl. Acad. Sci. USA, 95, 8898–8903. Zierath, J.R., Ryder, J.W., Doebber, T., Woods, J., Wu, M., Ventre, J., et al. (1998) Role of skeletal muscle in thiazolidinedione insulin sensitizer (PPARgamma agonist) action. Endocrinology, 139, 5034–5041. Zucker, L.M. and Zucker, T.F. (1961) Fatty, a new mutation in the rat. J.Heredity, 52, 275– 287.

5. KK AND KKAy MICE SHIGEHISA TAKETOMI1 and HITOSHI IKEDA2 1Discovery 2Pharmaceutical

Research Laboratories and

Research Laboratories, Takeda Chemical Ind., Ltd., Yodogawa-ku, Osaka 532–8686, Japan

ABSTRACT The KK strain of mice has two characteristics for the development of NIDDM. One is the fact that their diabetic traits are inherited by polygenes. This genetic feature is quite different from some other types of diabetic animals such as ob/ob or db/ db mice and Wistar fatty or fatty Zucker diabetic rats, whose diabetic states are induced by mutations of the gene. The other is that the KK strain has a latent diabetic state before the development of hyperglycemia and glucosuria. At this stage, glucose intolerance and insulin resistance are already observed. Therefore, KK mice are suitable for studying the causal genes of diabetes, effects of environmental factors and evaluation of antidiabetic and antiobesity drugs. KK and KKAy mice can be obtained from Clea Japan In. (Tokyo, Japan). INTRODUCTION Non-insulin-dependent diabetes mellitus (NIDDM) is a heterogenous disease with multiple etiologies. The development of diabetes characterized by hyperglycemia, insulin resistance in peripheral tissues and impaired insulin secretion to glucose is caused by genetic and environmental factors. A series of studies using KK mice has clearly pointed out that the close interaction of genetic and environmental factors such as diet and obesity plays an important role in the development of diabetes.

114

S.TAKETOMI AND H.IKEDA

HISTORY OF DERIVATION OF KK STRAIN Kondo et al. (1957) selected out and established many mouse strains from Japanese native mice. Among these inbred mouse strains, Nakamura (1962) found that the KK mouse strain, which was named “KK” for its habitat (Kasukabe in Saitama prefecture) (Kondo et al., 1957) is spontaneously diabetic. Therefore, several investigators (Nakamura, 1962; Nakamura et al., 1963, 1967; Iwatsuka et al., 1970a; Dulin et al., 1983) reported that the diabetic characteristics were associated with moderate obesity (30–35 g body weight at the age of 5 months), such as sluggishness, polyphagia, polyuria, persistent glucosuria, glucose intolerance, moderate hyperglycemia, hyperlipidemia, insulin resistance of peripheral tissues, hyperinsulinemia, histological changes in the pancreas and renal glomerular changes, suggested several similarities between the diabetic state in the KK mice and human NIDDM associated with obesity. Blood glucose levels were higher in males than in females, but there was no sex bias in glucose intolerance (Nakamura, 1962; Nakamura et al., 1963, 1967). A genetic study of KK mice indicated that diabetic traits were inherited by polygenes (Nakamura et al., 1963). Because diabetes and obesity in KK mice were relatively moderate, Nishimura (1969), one of Kondo’s coworkers transferred the yellow obese gene (Ay) into KK mice by the repeated crossing of yellow obese mice and KK mice. The Ay allele (dominant allele at the mouse agouti locus) is associated phenotypically with yellow fur, hyperphagia and obesity (Danforth, 1927; Bultman et al., 1992; Michaud et al., 1994). This congenic strain of KK mice has been named yellow KK or KKAy mice. KKAy mice (yellow) can be easily distinguished by the color of the fur from KK mice (black) at weaning. The diabetic characteristics of KKAy mice such as hyperglycemia, hyperinsulinemia and obesity were observed at young ages (6–8 weeks), but reverted apparently to normal after 40 weeks of age (Iwatsuka et al., 1970b, 1974a). This may be due to reduction in food intake, because disappearance of diabetic signs in the aged yellow KK mice is not associated with amelioration of insulin resistance or glucose dysmetabolism expected from reduction of adiposity. Therefore, the intrinsic abnormality such as insulin resistance and glucose dismetabolism is thought to be primarily responsible for the development of diabetes in the KK strain. STRAIN DIFFERENCE FOR DIABETOGENIC ACTION OF OBESITY Since the initial studies reported by Nakamura (1962), KK mice were used as the models for NIDDM. However, KK mice in our stock derived from the original KK colony at Nagoya University showed glucose intolerance and insulin resistance without hyperglycemia and glucosuria, as long as they kept on a laboratory chow (Iwatsuka et al., 1970a). Accordingly, KK mice could be defined as a strain with diabetic genes. When obesity was produced by transfer of the yellow obese gene (Ay), KK mice exhibited many diabetic symptoms, such as hyperglycemia, glucosuria, hypertrophy of pancreatic islets and degranulation of β cells (Iwatsuka et al., 1970b). Thus, the yellow KK (KKAy) mice developed more severe diabetes than the original KK mice. These phenomena were also observed in obese KK mice induced by feeding high calorie diets (Table 1) or hyperphagia due to hypothalamic lesion with goldthioglucose treatment (Matsuo et al., 1970, 1971a, b, 1972). When compared with C57B1 or ICR mice, dietary obesity induced hyperglycemia only in KK mice, while goldthioglucose-induced obesity induced hyperglycemia in KK and ICR mice, but not in C57BL mice. The degree of glucose intolerance was in the order: KK>ICR>C57BL. KK mice were also more susceptible to diabetogenic effects of obesity than hybrid mice of KK and C57BL Address correspondence to: Dr. Hitoshi Ikeda, Pharmaceutical Research Laboratories, Takeda Chemical Ind., Ltd., Yodogawa-ku, Osaka 532–8686, Japan.

KK AND KKAY MICE

115

which should carry half of the genes of the KK strain (Iwatsuka et al, 1970c, 1974b) (Table 2). These observations suggest that the difference observed among these mouse strains can be due to genetic factors and that diabetic characteristics in KK mice may be determined by concentration of polygenes, as reported by Nakamura (1962). Table 1 Effects of high calorie diets on body weight gain and blood glucose in KK and ICR mice. Strain Diet

KK

ICR

No. of mice Body weight gain (g/4 weeks)

Type

Total calorie (kcal./ 100g)

Fat content (%)

CE-2 B K L CE-2 B K L

342.2 386.1 386.1 476.1 342.2 386.1 386.1 476.1

3.5 10.5 2.5 20.5 3.5 10.5 2.5 20.5

3.7±0.6a 5.2±1.0 3.8±0.9 6.8±0.8 0.7±1.5 1.1±0.7 −1.8±0.9 0.8±1.5

6 6 6 6 6 6 6 6

Blood glucose (mg/ 100 ml)

211±26 (3/6)c 272±8b (5/6) 182±21 (3/6) 289±21b(5/6) 114±4 (0/6) 142±8 (0/6) 137±8 (0/6) 144±13 (0/6)

Animals used were 12 week-old. a Mean ± SE. b Significant compared with the corresponding CE-2 group (p < 0.05). c No. of animals with glucosuria/no. of animals tested. Table 2 Glucose tolerance in mice with varying concentration of KK genes. Genotype

Expected content of KK genes (%)

N

Glucose tolerance (mg/dl at 1h)

KK (F1×KK) (F1×KK)-Ay (F1×C57BL/6J) (F1×C57BL/6J)-Ay C57BL/6J

100 75 (50–100) 75 (50–100) 25 (0–50) 25 (0–50) 0

15 24 38 24 18 19

765±60 533±26 659±17 354±18 488±25 378±30

F1 hybrids with the Ay gene of KK and C57BL/6J mice were back-crossed to KK and C57BL/6J mice. At 4 weeks of age, all of six types of mice were not glucosuric, but glucose tolerance became lower in order: C57BL/6J= (F1×C57BL/6J)>(F1×C57BU6J)-Ay>(F1×KK)(F1×KK)-Ay> KK. Mean ± SE.

Furthermore, glucose intolerance and insulin resistance are thought to be good markers for predisposition to diabetes. INSULIN RESISTANCE NIDDM is the most common of metabolic disorders characterized by impaired response of peripheral tissues to insulin. The insulin resistance in liver, muscle and adipose tissue is thought to lead to perturbation of glucose homeostasis, as well as to impaired insulin secretion in response to glucose. Several investigators

116

S.TAKETOMI AND H.IKEDA

have reported insulin resistance and glucose intolerance of the KK strain. Such deterio ration was accompanied by hyperinsulinemia, degranulation of pancreatic β-cells and advanced insensitivity of peripheral tissues to insulin. Epididymal adipose tissues of KK mice were less sensitive to the stimulatory action of insulin on glucose oxidation (Iwatsuka et al., 1970a, 1974c). This phenomenon was confirmed by experiments using adipocytes prepared by Rodbell’s method (Taketomi et al., 1988; Rodbell, 1964). Glucose oxidation in KK adipose cells showed decreased insulin sensitivity and responsive ness, the dose-response curve being shifted to the right and the maximum activity relatively low when compared to adipose cells from C57BL/ 6J mice. Whereas, lipid synthesis from glucose or acetate was normal response to insulin in adipose tissues or adipocytes from KK mice. Such heterogeneity of hormone sensitivity may result in development of diabetes associated with obesity in this strain. Furthermore, in adipose tissue or adipocytes, lipolysis was less sensitive to epinephrine and insulin (Iwatsuka et al., 1970a, 1974c). Thus, the alterations in lipolysis may also be a risk factor leading to obesity. Generally, a negative correlation between the cell size and hormone sensitivity was reported (Herberg et al., 1970). However, this phenomenon was not valid in KK mice, because KK adipocytes were less sensitive to the hormone than C57BL/6J adipocytes of similar cell size (Iwatsuka et al., 1974c). Therefore, impaired hormonal sensitivity is more likely determined by genetic factors than simply by the changes in cellularity. The F1-hybrid mice of KK and C57BL mice showed intermediate responses of adipose tissue to insulin (Iwatsuka et al., 1970c). Introduction of the Ay gene into KK and C57BL mice decreased insulin sensitivity of adipose tissue due to increased cell size (Iwatsuka et al., 1974c). Therefore, this phenomenon may be caused by changes in cellularity. These findings clearly indicate that the insulin resistance of KK mice is inherent and susceptible to the diabetogenic influence of obesity. Taketomi et al. (1988) found that adipocytes of KK mice were less sensitive and less responsive to insulin with respect to glucose uptake and [1-l4C]-glucose oxidation, but normosensitive to insulin with [6l4C]-glucose oxidation or glyceride-glycerol and fatty acid synthesis from both [1–14C]- and [6-l4C]glucose. These findings indicate that, in the adipocytes from KK mice, hormone sensitivity is normal with regard to glycolysis and Krebs cycle, and the hormone resistance is restricted to the pentose shunt. The same type of hormone resistance has been reported in the large-size adipocytes from spontaneously obese and old rats (Olefsky, 1977; Richardson et al., 1978). However, the mechanism of the hormone resistance was quite different between KK mice and the old rats. In the former, [I-l4C]-glucose oxidation was insensitive to vitamin K5, which shows insulin-like actions without binding to insulin receptors, but fatty acid synthesis from glucose responded to the agent in a usual manner. In the latter, [1-l4C]-glucose oxidation was normally stimulated by vitamin K5, but fatty acid from glucose was decreased. In the large cells from the old rats, an accumulation of NADPH induced by decreased fatty acid synthesis seems to result in a feedback inhibition of the pentose shunt. Accordingly, cells from the old rats showed a normal response to vitamin K5, a powerful NADPH oxidant. On the other hand, the abnormalities in the adipocytes from KK mice were accounted for by the selective defect in the postinsulin binding system. [1–14Q-glucose oxidation in adipocytes of KK mice was also insensitive to insulin mimicking agents such as H2O2. The downregulation of the insulin receptor caused by insulin or insulin mimics in cultured adipose tissue of KK mice was also impaired. In conclusion, the adipocytes of KK mice are insulin-resistant with respect to glucose uptake and pentose pathway due to defects in the post-insulin binding signaling pathway. In KKAy mice with elevated plasma insulin levels, insulin resistance was associated with decreased expression of glucose transporter protein (GLUT4) in adipocytes and muscles (Hofmann et al.,1991). Interestingly, in mildly hyperglycemie KK mice, T3 (3,3’,5-triiodo-L-thyronine) treatment increased both GLUT4 and MyoD mRNA levels in skeletal muscle in vivo, indicating the possibility that stimulation of

KK AND KKAY MICE

117

muscle differentiation leads to amelioration of hyperglycemia through increased glucose metabolism due to activation of GLUT4 (Shimokawa et al., 1997). As for the regulation of the insulin receptor, adipocytes from C57BL/6J mice showed downregulation of insulin binding by insulin or vitamin K5, but KK cells did not (Taketomi et al., 1988). This finding indicates that the regulation of insulin binding seems to be closely related to the signaling pathway acting after insulin binding. In the skeletal muscles of KKAy mice, high glucose concentration in vitro reduced insulin binding, suggesting that high glucose levels in vivo may enhance insulin resistance through reduction of insulin binding (Kobayashi et al., 1992). Liver is also a target tissue of insulin resistance. However, hepatic glucose production based upon fasting blood glucose levels is likely to be normally regulated by insulin in prediabetes or mild diabetes, because fasting hyperglycemia is not observed (DeFronzo, 1997). In moderate/severe diabetes, fasting blood glucose levels are increased, indicating that insulin can not prevent excessive hepatic glucose production. The plasma glucose levels after 20 h fasting were normal in KK mice, but mildly increased in KKAy mice (Iwatsuka et al., 1970b). Wyse et al. (1974) reported that glucose production from pyruvate in the liver of obese and hyperinsulinemic KK mice was not suppressed during the fed state. KKAy mice with hyperinsulinemia showed higher activities of insulin-suppressive enzymes of gluconeogenesis (glucose-6-phosphatase and fructose-1, 6-diphosphatase), as compared with KK mice (Taketomi et al., 1973). From these findings, the response to insulin in the liver of KKAy mice is impaired. However, KKAy mice also showed high activities of insulin-inducible enzymes involving glycolysis (glucokinase and pyruvate kinase), pentose phosphate cycle (6-phosphogluconate dehydrogenase) and lipogenesis (ATP citrate lyase, malic enzyme and acetylCoA carboxylase), indicating that insulin normally regulates enzyme activities of glycolysis, pentose phosphate cycle and lipogenesis. Thus, there was a heterogeneity in the enzyme activity alterations in response to insulin. KKAy and KK mice showed a correlation between hepatic lipogenic activity and plasma insulin levels (Matsuo et al, 1971b; Taketomi et al., 1973). Predisposition to diabetes in KK strain may arise from a normal response to insulin in lipogenesis of liver and adipose tissue concomitant with insulin resistance in the glucose uptake and oxidation of adipose tissue (Table 3). Recently, many reports suggest that tumor necrosis factor (TNF-α) is involved in the development of insulin resistance in both animals and human NIDDM by inhibiting tyrosine kinase activity on the -subunit of the insulin receptor (Hotamisligil and Spiegelman, 1994; Hotamisligil et al, 1994). In muscle of KKAy mice, the RNA transcript encoding the p55 TNF-α receptor is elevated (Hofmann et al., 1994). Moreover, Lactobacillus casei (LC), which is known to inhibit the production of cytokines such as interferon- and interleukin 2, showed antidiabetic Table 3 The response to insulin in adipocytes and liver of KK mice. Adipocytes Glucose uptake Pentose Shunt Glycolysis Lipid synthesis Lipolysis

Impaired Impaired Normal Normal Impaired

Glycolysis Gluconeogenesis Lipid synthesis

Normal Impaired Normal

Liver

118

S.TAKETOMI AND H.IKEDA

Figure 1 Relation of SSBG or impedance-1 to SSPI in C57BL/6J and KK mice. C57BL/ 6J and KK mice after 24 hourfasting were injected s.c. with 0.25 ml saline containing epinephrine (2.5 mg), propranolol (125 mg), glucose (5 mg, [3H-3] glucose; 2 µCi) with or without different doses of insulin. Blood samples were taken 60 min after injection for determination of steady state of plasma glucose (SSBG) and insulin (SSPI).

effects in KKAy mice (Matsuzaki et al., 1997). These findings suggest that the elevated production of TNFor other cytokines is responsible for insulin resistance in KKAy mice. A convenient method for assessment of the overall insulin sensitivity in small animals (mice) was established using subcutaneous injection of epinephrine, propranolol and glucose with or without insulin (Taketomi et al., 1982). KK mice showed a reduction of steady state blood glucose levels in response to insulin, as compared with C57BL/6J mice (Figure 1). OTHER CHARACTERISTICS RESPONSIBLE FOR NIDDM The brown adipose tissue (BAT) is well-known to be involved in the regulation of energy metabolism. Thermogenesis in the interscapular BAT was lower in KKAy mice than in C57BL/6J mice at 7 weeks of age (Ohta et al., 1988). In addition, the responsiveness of BAT mitochondrial GDP binding to cold exposure, norepinephrine injection and sucrose intake was lower in KKAy mice than in C57BL/6J mice. These findings suggest that abnormalities in BAT may contribute to the development and maintenance of obesity. Recently, leptin mRNA levels in BAT as well as in peripheral white adipose tissues of KKAy mice were found to be higher than those of C57BL/6J mice, (Sakane et al., 1998). Furthermore, KKAy mice showed higher leptin mRNA levels in the white adipose tissue (WAT), as compared with KK mice (Hayase et al, 1996). In both mice, downregulation of leptin gene expression during fasting was recognized in mesenteric and subcutaneous WAT, but not in epididymal WAT, indicating the presence of regional differences in the regulation of leptin gene expression in adipose tissue. Leptin regulates appetite and body weight by reducing food intake and increasing energy expenditure. Most obese animals have been linked to defects in

KK AND KKAY MICE

119

the leptin system: deficiency of leptin in ob/ob mice which have a mutation in the leptin gene, and defects of leptin receptor in db/db mice and of all leptin receptor isoforms in Zucker fatty rats which show overexpression of leptin mRNA (Zhang et al., 1994; Lee et al., 1996; Chen et al., 1996; Phillips et al., 1996; Iida et al., 1996). These animals are also leptin resistant (Frederich et al., 1995). Although the response to leptin in KK or KKAy mice is unclear, these mice may be also leptin resistant. HISTOLOGICAL CHANGES AND COMPLICATIONS In KK and KKAy mice, plasma insulin levels were elevated after a meal, but not after glucose loading, indicating that β cells of the KK strain are selectively insensitive to glucose (Iwatsuka et al., 1970a,b). As for morphological changes, in KK mice, cell degranulation, glycogen deposition and hypertrophy were observed at the age of 16 weeks (Iwatsuka et al., 1970b; Shino et al., 1970). KKAy mice with severe hyperinsulinemia showed more prominent changes in pancreatic islets than KK mice (Shino et al., 1970). Degranulation, glycogen deposition and hypertrophy of β cells were observed at the age of 5–10 weeks. These changes became prominent at 16 weeks of age, and diminished with increasing age. By electron microscopy, degranulated β cells contained well-developed Golgi apparatus, abundant granular endoplasmic reticulum and fine granules of glycogen. cell granules were occasionally facing the cell membrane indicating emiocytosis. These changes suggest increased synthesis and release of insulin which correspond to high levels of plasma insulin. KK and KKAy mice showed renal lesions, similar to human diabetic nephropathy (Nakamura, 1962; Treser et al., 1968; Camerini-Davalos et al., 1970; Iwatsuka et al., 1970b; Wehner et al., 1972; Reddi et al., 1978; Emoto et al., 1982). Diffuse glomerulosclerosis, nodular changes and peripheral glomerular basement membrane Table 4 The motor nerve conduction velocity (MNCV) and content of nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), and neurotrophin-3 (NT-3) in the sciatic nerves of KKAy mice. Strain

Age (weeks) No. of mice MNCV (m/s) NGF (ng/g wet tissue)

BDNF (ng/g wet tissue)

NT-3 (ng/g wet tissue)

C57BL/6J KKAy

28 28

7.31±0.53 3.77 1 0.48a

7.88±0.56 3.52±0.79a

15 12

25.0±2.7 22.6±2.2a

1.49 ±0.41 0.97±0.29

Mean ± SE. a Significant compared with C57BL/6J mice (p < 0.001).

(GBM) thickening were recognized in KKAy mice. Diani et al. (1987) found that GBM thickening occured at the early age and developed rapidly in KKAy mice in comparison with other diabetic animals. Mesangial enlargement and hypercellularity were also observed in KK mice. Proteinuria and microalbuminuria were observed in KK and KKAy mice (Treser et al., 1968; Wehner et al., 1972; Reddi et al., 1978). However, it remains still unclear if the glomerulosclerosis is due to diabetes in the KK strain, because a high incidence of the same changes was observed in the prediabetic stage in KK mice. The additional factor(s) such as renal amyloidosis may be involved in the pathogenesis of glomerulosclerosis (Soret et al., 1977). Changes of nerve function were also investigated. Decreased motor nerve conduction velocity was observed in KKAy mice, but not in KK mice, as compared with C57BL/6J mice (Table 4). Furthermore, KKAy mice showed decreases in nerve growth factor, brain-derived neurotrophic factor and neurotrophin-3 contents of the sciatic nerve at 28 weeks of age. Similar findings have been observed in streptozotocininduced diabetic rats with severe hyperglycemia (Hellweg et al., 1990; Rodriguez-Pena et al., 1995; Hounsom et al., 1997). Therefore, these abnormalities in growth factors suggest that the development of

120

S.TAKETOMI AND H.IKEDA

Figure 2 Effect of pioglitazone on blood glucose in KKAy mice. Pioglitazone was given to male KKAy mice (9 weekold) as a dietary admixture for 4 days. Means ± SD (n=5). The different superscripts indicate significant difference (P < 0.05) in the interaction effect.

abnormal nerve functions in KKAy mice may be due to elevation of plasma glucose levels. Many studies on diabetic neuropathy were performed using streptozotocin-induced diabetic rats, a representative model of insulin-dependent diabetes mellitus (Hounsom et al., 1997). Further studies on KKAy mice may provide a useful model for the investigation of diabetic neuropathy. CHROMOSOME MAPPING To identify the genetic factors underlying NIDDM, the quantitative trait loci (QTL) analysis using 97 microsatellite markers was performed on KK and KKAy mice (Suto et al., 1998). Two kinds of F2 progeny comprising mice with a/a genotype at the agouti locus (chromosome 2) and mice with Ay/a genotype from mating between female C57BL/6J and male KKAy mice were used. In F2 a/a progenies, a significant locus on chromosome 6 responsible for control of fasting glucose and three loci on chromosomes 3, 5 and 14 were identified. In F2 Ay/a progenies, three suggestive loci were identified: a locus for fasting glucose on chromosome 9, and two loci for glucose intolerance on chromosome 1 and 8. Thus, only in F2 Ay/a progenies, were loci for glucose tolerance identified, indicating that these loci may interact with the Ay allele. This is consistent with the findings that glucose intolerance in KK mice is moderate and becomes overt in response to the Ay allele. Genetic dissection was also reported with two different polygenic rat model for NIDDM, GK and OLETF rats (Galli et al., 1996; Gauguier et al., 1996; Kanemoto et al., 1998). These combined results may suggest common genetic factors underlying NIDDM, a complex and multifactorial disease with multiple etiologies.

KK AND KKAY MICE

121

USEFUL FOR EVALUATION OF ANTIDIABETIC DRUGS Therapies for treatment of NIDDM can be grouped mainly into four categories: 1) insulin, 2) insulin secretagogues, 3) enhancers of insulin action (reduced insulin resistance) and 4) inhibitors of glucose absorption. In KK and KKAy mice, effects of antidiabetic compounds which reduce insulin resistance or increase insulin sensitivity have been evaluated. The first member of thiazolidinediones (TZDs), ciglitazone (ADD-3878) was discovered in the in vivo screening system using KKAy mice (Fujita et al., 1983). This compound decreased hyperglycemia, hyperlipidemia and hyperinsulinemia, accompanied by reduced insulin resistance in peripheral tissues. Subsequently, pioglitazone (AD-4833) (Figure 2) and troglitazone (CS-045) were selected using KKAy or KK mice, respectively (Ikeda et al., 1990; Fujiwara et al., 1988). TZDs were also reported to induce preadipocytes to differentiate into mature adipocytes, acting as a high affinity ligand for transcription factor PPAR (peroxisome-proliferation activated receptor γ) (Tontonoz et al., 1994). The affinity of many TZDs for PPAR has been known to correlate with antidiabetic effects (Berger et al, 1996; Willson et al, 1996). Although PPAR has been implicated as one of the critical factors responsible for the mechanisms of insulin sensitivity and obesity, changes in the expression of PPAR in adipose tissues were not reported in KK or KKAy mice. In two animal models of obesity (ob/ob mice and fatty Zucker rats), adipose tissue PPAR levels were not altered (Vidal-Puig et al, 1996; Shimoike et al., 1998). Recently, the mechanism(s) of pioglitazone underlying reduced insulin resistance are being investigated in liver and muscles of KKAy mice, to clarify the relationship between fatty acids and insulin resistance (Saha et al., 1994). Furthermore, KK or KKAy mice were used to study the effect of TZD on UCP mRNA levels in skeletal muscles and on GLUT4 expression in adipocytes and muscles, respectively (Shimokawa et al., 1998; Hofmann et al., 1991). Other insulinsensitizing compounds structurally unrelated to TZD have been tested using KKAy mice (Meglasson et al, 1998; Yoshida et al, 1996; Pill et al, 1999). As for development of insulin secretagogues, KK and KKAy mice have been used, because these mice showed impaired tolerance to oral glucose administration and postprandial hyperglycemia. Tolbutamide has been used for glycemic control after meal, but often induces severe hypoglycemia because of stimulation of insulin secretion independent of blood glucose levels. However, tolbutamide does not exert a hypoglycemic action in KKAy mice (Kameda et al, 1982). Non-sulfonylurea compounds, A-4166 or AD-4610 suppressed the rise in the postprandial glucose level through increased insulin secretion after glucose loading in KK mice (Sugiyama et al, 1988; Sato et al, 1991). Another mechanism for suppressing postprandial hyperglycemia is through inhibition of intestinal glucosidases. -Glucosidase inhibitors such as acarbose and voglibose were effective in KK and KKAy mice (Hamada et al., 1988; Matsuo et al., 1992; Odaka et al, 1992). In addition, voglibose prevented the development of diabetes and diabetic nephropathy in KKAy mice (Odaka et al, 1992). KKAy mice with impaired thermogenesis in response to sucrose intake were used to assess the effect of β3-adrenoceptor agonists, BRL 26830A and CL3l6, 243 on diabetes and obesity (Yoshida et al., 1991; Hioki et al., 1995; Sakane et al., 1998). BRL 26830A increased BAT thermogenesis and decreased hyperglycemia, and CL316,243 suppressed leptin mRNA expression in KKAy mice. Furthermore, the effects of apetite suppressor, mazindol, were evaluated in KKAy mice (Yoshida et al, 1996). The drug activated BAT thermogenesis via increased noradrenaline turnover.

122

S.TAKETOMI AND H.IKEDA

REFERENCES Berger, J., Bailey, P., Biswas, C., Cullinan, C.A., Doebber, T.W., Hayes et al. (1996) Thiazolidinediones produce a conformational change in peroxisomal proliferation-activated receptor-: Binding and activation correlate with antidiabetic actions in db/db mice. Endocrinology, 137, 4189–4195. Bultman, S., Michaud, E.J., and Woychik, R.P. (1992) Molecular characterization of the mouse agouti locus. Cell, 71, 1195–1204. Camerini-Davalos, R.A., Oppermann, W., Mittl, R. et al. (1970) Studies of vascular and other lesions in KK mice. Diabetologia, 6, 324–329. Chen, H.O., Charlat, L.A. Tartaglia, E.A. et al. (1996) Evidence that the diabetes gene encodes the leptin receptor: identification of a mutation in the leptin receptor gene in db/db mice. Cell, 84, 491–495. Danforth, C.H. (1927) Hereditary adiposity in mice. J. Hered., 18, 153–162. Defronzo, R.A. (1997) Pathogenesis of type II diabetes: metabolic and molecular implications for identifying diabetes genes. Diabet. Rev., 5, 177–269. Diani, A.R., Sawada, G.A., Zhang, N.Y. et al. (1987) The KKAy mouse: A model for the rapid development of glomerular capillary basement membrane thickening. Blood Vessels, 24, 297–303. Dulin, W.E., Gerritsen, G.C., and Chang, A.Y. (1983) Experimental and spontaneous diabetes in animals. In Diabetes mellitus: theory and practice, edited by M.Ellenberg and H.Rifkin, pp. 361–408. Medical Examination Publishing Co.. Emoto, M., Matsutani, H., Kimura, S. et al. (1982) Nephropathy in KK mice treated with a limited diet. J. Jap. Diab. Soc., 25, 1211–1219. Frederich, R., Hamann, A., Anderson, S. et al. (1995) Leptin levels reflect body lipid content in mice: evidence for dietinduced resistance to leptin action. Nature Med., 1, 1311–1314. Fujita, T., Sugiyama, Y., Taketomi, S. et al. (1983) Reduction of insulin resistance in obese and/or diabetic animals by 5-[4-(1-methylcyclohexylmethoxy)benzyl]thiazolidine-2, 4-dione (ADD-3878, U-63, 287, ciglitazone), a new antidiabetic agent. Diabetes, 32, 804– 810. Fujiwara, T., Yoshioka S., Yoshioka, T. et al. (1988) Characterization of new oral antidiabetic agent CS-045. Studies in KK and ob/ob mice and Zucker fatty rats. Diabetes, 37, 1549– 1558. Galli, J., Li, L.-S., Glaser, A. et al. (1996) Genetic analysis non-insulin-dependent diabetes mellitus in the GK rat. Nature, Genet., 12, 31–37. Gauguier, D., Froguel, P., Parent, V. et al. (1996) Chromosomal mapping of genetic loci associated with non-insulindependent diabetes mellitus in the GK rat. Nature Genet., 12, 38–43. Hamada, H., Iida, R., Ishimura, K. et al. (1988) Effect of acarbose on diabetes in KK mice. Jap. Pharmacol. Ther., 17, 17–28. Hayase, M., Ogawa, Y., Katsuura, G., Shintaku, H., Hosoda, K. and Nakao, K. (1996) Regulation of obese gene expression in KK mice and congenic lethal yellow obese KKAy mice. Am. J. Physiol., 271, E333—E339. Hellweg, R., and Hartung, H.-D. (1990) Endogenous levels of nerve growth factor (NGF) are altered in experimental diabetes mellitus: A possible role for NGF in the pathogenesis of diabetic neuropathy. J. Neurosci. Res., 26, 258–267. Herberg, L., Gries, F.A., and Hesse-Wortmann, Ch. (1970) Effect of weight and cell size on hormone-induced lipolysis in New Zealand obese mice and American obese hyperglycemic mice. Diabetologia, 6, 300–305. Hioki, Y., Itoh, Y., Nakajima, A., Fukuroda, T., Ohashi, H., Kamei, T., et al. (1995) A novel adrenaline derivative, AZ002, and its hypoglycemic action in yellow KK mice. Jpn. J. Pharmacol., 69, 251–258. Hofmann, C., Lorenz, K., and Colca, J.R. (1991) Glucose transport deficiency in diabetic animals is corrected by treatment with oral antihyperglycemic agent pioglitazone. Endocrinology, 129, 1915–1925. Hofmann, C., Lorenz, K., Braithwaite, S.S. et al. (1994) Altered gene expression for tumor necrosis factor-α and its receptors during drug and dietary modulation of insulin resistance. Endocrinology, 134, 264–270. Hotamisligil, G.S. and Spiegelman, B.M. (1994) Tumor necrosis factor: a key component of the obesity-diabetes link. Diabetes, 43, 1271–1278.

KK AND KKAY MICE

123

Hotamisligil, G.S., Budavari, A., Murray, D. and Spiegelman, B.M. (1994) Reduced tyrosine kinase activity of the insulin receptor in obesity-diabetes. Central role of tumor necrosis factor-α. J. Clin. Invest., 94, 1543–1549. Hounsom, L., and Tomlinson, D.R. (1997) Does neuropathy develop in animal models? Clinical Neurosci., 4, 380–389. Iida, M., Murakami, T., Ishida, K. et al. (1996) Substitution at codon 269 (glutamine → proline) of the leptin receptor (OB-R) cDNA is the only mutation found in the Zucker fatty (fa/ fa) rat. Biochem. Biophys. Res. Commun., 224, 597–604. Ikeda, H., Taketomi, S., Sugiyama, Y. et al. (1990) Effects of pioglitazone on glucose and lipid metabolism in normal and insulin resistant animals. Arzneim. Forsch./Drug Res., 40, 156– 162. Iwatsuka, H., Matsuo, T., Shino, A. et al. (1970a) Metabolic disturbance of KK mice in chemical diabetes. J. Takeda Res. Lab., 29, 685–692. Iwatsuka, H., Shino, A., and Suzuoki, Z. (1970b) General survey of diabetic features of yellow KK mice. Endocr. Jap., 17, 23–35. Iwatsuka, H., and Shino, A. (1970c) Studies on diabetogenic action of obesity in mice: congenital insulin resistance of KK mice. Endocr. Jap., 17, 535–540. Iwatsuka, H., Ishikawa, E., and Shino, A. (1974a) Remission of diabetic syndromes in advanced age of genetically obese and diabetic mice, yellow KK.J. Takeda Res. Lab., 33, 203–212. Iwatsuka, H., Ishikawa, E., Shimakawa, K. et al. (1974b) The role of genetic background of KK strain in pathogenesis of diabetes: Polygenic inheritance of a sensitivity to a diabetogenic action of obesity. J. Takeda Res. Lab., 33, 248–251. Iwatsuka, H., Taketomi, S., Matsuo, T. et al. (1974c) Congenitally impaired hormone sensitivity of the adipose tissue of spontaneously diabetic mice, KK.Validity of thrifty genotype in KK mice. Diabetologia, 10, 611–616. Kameda, K., Ono, S., and Abiko, Y. (1982) Hypoglycemic action of 2-[2-(4, 5-dihydro-IH-imidazo-2-yl)-1-phenylethyl] pyridine dihydrochloride sesquihydrate (DG-5128), a new hypoglycemic agent. Arzneim. Forsch./Drug Res., 32, 39–44. Kanemoto, N., Hishigaki, H., Miyakita, A. et al. (1998) Genetic dissection of OLETF, a rat model for non-insulindependent diabetes mellitus. Mammalian Genome, 9 419–425. Kobayashi, S., Naitoh, T., Nakadate, T., Kimura, I. and Kimura, M. (1992) Suppression by high glucose concentration of insulin receptor up-regulation in diaphragm and flexor digitorum brevis muscles from diabetic KK-CAy and streptozotocin-diabetic mice. J. Pharmacobio-Dyn., 15, 175–180. Kondo, K., Nozawa, K., Tomita, T. et al. (1957) Inbred strains resulting from Japanese mice. Bull. Exp. Animals, 6, 107–112 (In Japanese). Lee, G.H., Proenca, R., Montez, J.M. et al. (1996) Abnormal splicing of the leptin receptor in diabetic mice. Nature, 379, 632–635. Matsuo, T., Shino, A., Iwatsuka, H. et al. (1970) Induction of overt diabetes in KK mice by dietary means. Endocr. Jap., 17, 477–488. Matsuo, T., Furuno, K., and Shimakawa, K. (197la) Factors affecting development of hyperglycemia in KK mice. J. Takeda Res. Lab., 30, 307–313. Matsuo, T., Iwatsuka, H., and Suzuoki, Z. (197 1b) Metabolic disturbance of KK mice in overt diabetes. Endocr. Jap., 18, 501–506. Matsuo, T., and Shino, A. (1972) Induction of diabetic alterations by goldthioglucose-obesity in KK, ICR and C57BL mice. Diabetologia, 8, 391–397. Matsuo, T., Odaka, H., and Ikeda, H. (1992) Effect of an intestinal disaccharidase inhibitor (AO-128) on obesity and diabetes. Am. J. Clin. Nutr., 55, 314S-317S. Matsuzaki, T., Yamazaki, R., Hashimoto, S. and Yokokura, T. (1997) Antidiabetic effects of an oral administration of Lactobacillus casei in a non-insulin-dependent diabetes mellitus (NIDDM) model using KKAy mice. Endocrine J., 44, 357–365. Meglasson, M.D., Wilson, J.M., Yu, J.H., Robinson, D.D., Wyse, B.M. and de Souza, C.J. (1993) Antihyperglycemic action of guanidinoalkanoic acids: 3-Guanidinopropionic acid ameliorates hyperglycemia in diabetic KKAy and

124

S.TAKETOMI AND H.IKEDA

C57BL6job/ob mice and increases glucose disappearance in rhesus monkeys. J. Pharm. Exp. Therap., 266, 1454–1462. Michaud, E.J., Bultman, S.J., Klebig, M.L. et al. (1994) A molecular mode for the genetic and phenotypic characteristics of the mouse lethal yellow (A?) mutation. Proc. Natl. Acad. Set. USA, 91, 2562–2566. Nakamura, M. (1962) A diabetic strain of the mouse. Proc. Jap. Acad., 38, 348–352. Nakamura, M., and Yamada, K. (1963) A further study of the diabetic (KK) strain of the mouse: F1 and F2 offspring of the cross between KK and C57BL/6J mice. Proc. Jap. Acad., 38, 489–493. Nakamura, M., and Yamada, K. (1967) Studies on diabetic (KK) strain of the mouse. Diabetologia, 3, 212–221. Nishimura, M. (1969) Breeding of mice strains for diabetes mellitus. Exp. Animals (Tokyo), 18, 147–157. Odaka, H., Shino, A., Ikeda, H. et al. (1992) Antiobesity and antidiabetic actions of a new potent diasaccharidase inhibitor in genetically obese-diabetic mice, KKAy. J. Nutr. Sci. Vitaminol., 38, 27–37. Ohta, Y., Kitazaki, T., and Tsuda, M. (1988) Thermogenesis in brown adipose tissue of genetically obese, diabetic (KKAy) mice. Endocr. Jap., 35, 83–92. Olefsky, J.M. (1977) Mechanisms of decreased insulin responsiveness of large adipocytes. Endocrinology, 100, 1169–1177. Phillips, M.S., Liu, Q., Hammond, H.A. et al. (1996) Leptin receptor missense mutation in the fatty Zucker rat. Nature Genet., 13, 18–19. Pill, J., and Kühnle, H.-F. (1999) BM 17.0744: A structurally new antidiabetic compound with insulin-sensitizing and lipid-lowering activity. Metabolism, 48, 34–40. Reddi, A.S., Oppermann, W., Patel, D.G. et al. (1978) Diabetic microangiopathy in KK mice. III. Effect of prolonged glyburide treatment on glomerulosclerosis. Exp. Mol. Pathol., 29, 92–101. Richardson, D.K., and Czech, M.P. (1978) Primary role of decreased fatty acid synthesis in insulin resistance of large rat adipocytes. Am. J. Physiol., 234, E182-E189. Rodbell, M. (1964) Metabolism of isolated fat cells. I. Effects of hormones on glucose and lipolysis. J. Biol. Chem., 239, 375–380. Rodríguez-Peña, A., Botana, M., Gonzalez, M. et al. (1995) Expression of neurotrophins and their receptors in sciatic nerve of experimentally diabetic rats. Neurosci. Lett., 200, 37— 40. Saha, A.K., Kurowski, T.G., Colca, J.R. et al. (1994) Lipid abnormalities in tissues of the KKAy mouse: effects of pioglitazone on malonyl-CoA and diacylglycerol. Am. J. Physiol., 267, E95-E101. Sakane, N., Yoshida, T., Umekawa, T. et al. (1998) Acute and chronic regulation of ob mRNA levels by (βadrenoceptor agonists in obese yellow KK mice. Endocr. J., 45, 647–651. Sato, Y., Nishikawa, M., Shinkai, H. et al. (1991) Possibility of ideal blood glucose control by a new oral hypoglycemic agent, N(trans-4-isopropylcyclohexyl-carbonyl)-D-phenylalanine (A-4166), and its stimulatory effect on insulin secretion in animals. Diab. Res. Clin. Pract., 12, 53–60. Shimoike, T., Yanase, T., Umeda, F. et al. (1998) Subcutaneous or visceral adipose tissue expression of the PPAR gene is not altered in the Fatty (fa/fa) Zucker rat. Metabolism, 12, 1494–1498. Shimokawa, T., Kato, M., Shioduka, K., Irie, J. and Ezaki, O. (1997) Effect of triiodothyronine on muscle cell differentiation and blood glucose level in hyperglycemic KK mice. Biochem. Biophys. Res. Commun., 235, 790–793. Shimokawa, T., Kato, M., Watanabe, Y. et al. (1998) In vivo effects of pioglitazone on uncoupling protein-2 and -3 mRNA levels in skeletal muscle of hyperglycemic KK mice. Biochem. Biophys. Res. Commun., 251, 374–378. Shino, A., and Iwatsuka, H. (1970) Morphological observations on pancreatic islets of spontaneous diabetic mice, yellow KK. Endocr. Jap., 17, 459–476. Soret, M.G., Peterson, T., Wyse, B.M. et al. (1977) Renal amyloidosis in KK mice that may be misinterpreted as diabetic glomerulosclerosis. Arch. Pathol. Lab. Med., 101, 464–468. Sugiyama, Y., Taketomi, S., Tawada, H. et al. (1988) Selective potentiation of 5-methyl-2-(1-methylcyclohyxyl)-4oxazoleacetic acid (AD-4610) on glucose-induced insulin secretion. Horm. Metab. Res., 20, 145–149. Suto, J.S., Matsuura, S., Imamura, K. et al. (1998) Genetic analysis of non-insulin-dependent diabetes mellitus in KK and KK-Ay mice. Eur. J. Endocr., 139, 654–661.

KK AND KKAY MICE

125

Taketomi, S., Tsuda, M., Matsuo, T. et al. (1973) Alterations of hepatic enzyme activities in KK and yellow KK mice with various diabetic states. Horm. Metab. Res., 5, 333–339. Taketomi, S., Ikeda, H., Ishikawa, E. et al. (1982) Determination of overall insulin sensitivity in diabetic mice, KK.Horm. Metab. Res., 14, 14–18. Taketomi, S., Fujita, T., and Yokono, K. (1988) Insulin receptor and postbinding defects in KK mouse adipocytes and improvement by ciglitazone. Diab. Res. Clin. Pract., 5, 125– 134. Tontonoz, P., Hu, E., and Spiegelman, B.M. (1994) Stimulation of adipogenesis in fibroblasts by PPAR γ2, a lipid activated transcription factor. Cell, 79, 1147–1156. Treser, G., Oppermann, W., Ehrenreich, T. et al. (1968) Glomerular lesions in a strain of genetically diabetic mice. Pro. Soc. Exp. Biol. Med., 129, 820–823. Vidal-Puig, A., Jimenez-Linan, M., Lowell, B.B. et al. (1996) Regulation of PPAR gene expression by nutrition and obesity in rodents. J. Clin. Invest., 97, 2553–2561. Wehner, H., Hohn, O., Faix-Shade, U. et al. (1972) Glomerular changes in mice with spontaneous hereditary diabetes. Lab. Invest., 27, 331–340. Willson, T.M., Cobb, J.E., Cowan, D.J. et al. (1996) The structure-activity relationship between peroxisome proliferation-activated receptor agonism and the antihyperglycemic activity of thiazolidinediones. J. Med. Chem., 39, 665–668. Wyse, B.M., and Dulin, W.E. (1974) Further characterization of diabetes-like abnormalities in the T-KK mouse. Diabetologia, 10, 617–623. Yoshida, T., Hiraoka, N., Yoshioka, K. et al. (1991) Antiobesity and antidiabetic actions of a α3-adrenoceptor agonist, BRL 26830A, in yellow KK mice. Endocr. Jap., 38, 397–403. Yoshida, T., Umekawa, T., Wakabayashi, Y., Yoshimoto, K., Sakane, N. and Kondo, M. (1996) Anti-obesity and antidiabetic effects of mazindol in yellow KK mice: Its activating effect on brown adipose tissue thermogenesis. Clin. Exp. Pharm. Physiol., 23, 476–482. Zhang, Y.R., Proenca, R., Maffei, M. et al. (1994) Positional cloning of the mouse obese gene and its human homologue. Nature, 372, 425–432.

6. THE OBESE SPONTANEOUSLY HYPERTENSIVE RAT (SHROB, KOLETSKY RAT): A MODEL OF METABOLIC SYNDROME X RICHARD J.KOLETSKY, JACOB E.FRIEDMAN and PAUL ERNSBERGER Departments of Nutrition and Medicine, Case Western Reserve University School of Medicine, Cleveland, OH 44106–4935

SUMMARY The SHROB rat is a unique strain with genetic obesity, hypertriglyceridemia, hyperinsulinemia, renal disease with proteinuria, and genetically determined hypertension, characteristics paralleling human Syndrome X. The obese phenotype results from a single recessive trait, a mutation resulting in truncation of all forms of the leptin receptor, designated fak. The mutation is on the same allele as the Zucker fatty trait (fa), but the latter results only in an amino acid substitution. SHROB have exaggerated circadian rhythms and abnormalities in hypothalamic function leading to hypersecretion of corticosterone. SHROB have fasting insulin levels 20 fold greater than heterozygous or wild type SHR (lean) siblings. The SHROB are glucose intolerant compared to lean siblings, but retain fasting euglycemia even on a high sucrose diet. The absence of fasting hyperglycemia suggests that polygenic interactions with additional modifier genes are required for Type 2 diabetes in obese rat models. Insulin resistance in SHROB is due mainly to diminished expressions of insulin receptor and IRS-1 proteins with subsequent effects on downstream postreceptor insulin action. Despite multiple metabolic derangements and severe insulin resistance, hypertension is not exacerbated in SHROB compared to SHR. Thus, insulin resistance and hypertension are independent in this model. Dietary manipulations and pharmacological interventions, including treatments of hypertension, diabetes and hyperlipidemia, and modify the expressions of the various phenotypic components of metabolic syndrome X. The SHROB serves as a useful model to understand the interactions of the various metabolic abnormalities that make up Syndrome X including obesity, hypertenison and hyperinulinemia. Increased activity of the sympathetic nervous system may be a common factor leading by separate pathways to hypertension and to insulin resistance.

SHROB AND SYNDROME X

127

HISTORICAL BACKGROUND The obese spontaneously hypertensive rat or Koletsky rat strain, with the proposed designation of SHROB/ Kol, arose spontaneously in 1969 (Koletsky, 1972). A female spontaneously hypertensive rat (SHR), descended from early breeding stock obtained from the NHLBI colony in 1968, was mated with a male Sprague-Dawley rat. The resulting hybrid offspring that remained hypertensive (systolic blood pressure >150 mmHg) were inbred, and after several generations an abnormal phenotype was noted among some of the litters. These obese rats were not only hypertensive like their lean littermates but showed additional phenotypic characteristics, including hyperlipidemia, hyperinsulinemia, and proteinuria with kidney disease (Koletsky, 1975a; Koletsky & Ernsberger, 1996). The obese genotype involves both sexes and represents a homozygous recessive trait originally designated as f (Koletsky, 1972) but later reclassified as fak (Koletsky & Ernsberger, 1996; Takaya et al., 1996). The injurious gene responsible for SHROB appears to be recessive since its characteristic effect is not present when it is paired with its dominant (lean, wild type) allele (Fak). Male and female SHROB are infertile, and the recessive fak/fak genotype can only be inherited when both parents are heterozygous and each carries the same recessive allele (Fak/fak). Lean SHR heterozygous carriers (Fak/Fak) are indistinguishable from homozygous wild type littermates except through breeding of obese offspring. However, the consequences of heterozygosity for the fak trait have not been systematically investigated. The fak mutation in the SHROB is a nonsense mutation of the leptin receptor gene, resulting in a premature stop codon in the leptin receptor extracellular domain at position 763 (Takaya et al., 1996). As a result of this additional stop codon, none of the leptin receptor isoforms in SHROB should include a membrane-bound segment (Ishizuka et al., 1998). In contrast, the Zucker fatty (fa) rat has a missense mutation at position 269, which reduces its functionality (Yamashita et al., 1997). The mutation fak should truncate all forms of the leptin receptor and eliminate downstream events triggered by leptin. Lean rats from the original litter of SHR/Sprague Dawley hybrids that contained obese offspring were then bred continuously by brother-sister mating of lean heterozygotes in the same closed facility in the Animal Resource Center of Case Western Reserve University School of Medicine from 1971 to the present day. This period comprises at least 60 generations of continuous inbreeding. SHROB/Kol breeders were sent to the NIH in 1972, where they were backcrossed with other SHR strains and two normotensive strains (WKY/N and LA/N) in a complex scheme, resulting in at least five distinct strains (Greenhouse, Hansen et al., 1990). The recessive trait was for some time designated as cp (corpulent). The hypertensive phenotype was lost from the backcrossed SHR/N-cp, with obese animals showing systolic blood pressures <140. The relationship of the genotypes and phenotypes of various hybrid strains to the parent SHROB/Kol colony is unknown. Heterozygous Fak/fak breeders have been were crossed with heterozygous Fa/fa breeders from a Zucker Fatty Rat colony. The resulting obese offspring (fak/fa) were slightly (3%) lighter than an SHROB and slightly (14%) heavier than a Zucker fatty rat (Yen, Shaw et al., 1977). Blood pressure, lipid levels and other metabolic abnormalities were never reported. The success of this SHROB/Zucker cross led some authors to suggest that the recessive traits causing obesity in the Zucker fatty rat and the SHROB were due to mutations at the same allelic site. The trait in the Zucker rat is known as fa while the SHROB became known as fak (Yen et al., 1977). Address correspondence to: Paul Ernsberger, Ph.D., Department of Nutrition, Case Western Reserve University, School of Medicine, 10900 Euclid Avenue, Cleveland, OH 44106–4935 USA. Tel: (216) 368– 4748; Fax: (216) 368–4752; E-mail: [email protected]

128

R.J.KOLETSKY ET AL.

Differences have been reported between the Zucker fatty rat and the SHROB. The most notable is blood pressure. The SHROB is consistently hypertensive while the Zucker animal, according to most reports, is normotensive or at best has only slightly elevated blood pressure (Buñag & Barringer, 1988). Both SHROB (Koletsky, 1975b; Abramowsky et al., 1984; Koletsky et al., 1995) and Zucker fatty rats (Kasiske et al., 1985) develop renal disease. However, the renal disease is more rapid, severe and extensive in SHROB, leading to mortality in SHROB but not in Zucker rats. A further difference is the lack of circadian rhythms in food and water intake, and corticosterone excretion in Zucker rats (White et al., 1989) contrasted to the normal day-night variations in all these variables in SHROB rats (Koletsky & Ernsberger, 1992). Furthermore, SHROB are somewhat heavier than Zucker fatty rats, particularly for females (17%) (Yen et al., 1977). These differences between Zucker fatty rats and SHROB may reflect similar mutations on differing genetic backgrounds (SHR versus Sherman and Merck stock rats) or the effects of a nonsense mutation of the leptin receptor in SHROB versus a missense mutation in the Zucker. PHENOTYPIC FEATURES OF THE SHROB STRAIN Obesity SHROB weigh nearly the same as their lean littermates at 30d of age (Figure 1). At 4–6 weeks of age SHROB of both sexes are still not overweight relative to SHR littermates, but show a rounded contour of the lower trunk. This external trait first distinguishes SHROB from lean SHR. Soon afterwards, SHROB gain weight rapidly. Food intake in young adult SHROB is increased about 40% compared with lean SHR littermates (Ernsberger et al., 1994). Mature SHROB routinely reach peak weights between 750 and 1000g. Obese males are slightly but not significantly heavier than females at all ages. Lean SHR, in contrast, show a marked sex difference in body weight, as is typical for rodent species. By maturity SHROB surpass their lean littermates by a factor of about two for males and three for females. Enormous deposition of fat occurs throughout the body, most notably in the subscapular depot, which exceeds 30g in SHROB, but is nearly absent in SHR littermates (Table 1). Retroperitoneal deposits within the abdomen, the sex-specific epididymal and myometrial depots, as well as the mesenteric fat pads associated with the viscera are enlarged by up to 19-fold in female and 12-fold in male SHROB relative to SHR controls (Table 1). Only the head, face and distal portions of the fore and hind limbs lack massive fat deposits. The accumulated lipid must be synthesized endogenously, since the animals are fed standard chow containing 5% fat by weight. Consistent with this notion, incorporation of labeled glucose into lipids in the heart, diaphragm, skeletal muscle, and adipose tissues and liver in SHROB is three fold that of SHR (O’Dea & Koletsky, 1977). Similarly, glycogen synthesis is 2.8-fold greater in SHROB. Body length is increased 11% in females with no change in males, implying an acceleration of linear growth in females (Table 1). Females are shorter in length than males, for both SHROB and SHR. The Lee index, an established measure of rodent obesity, is greater in the obese genotype in both sexes, as is the body mass index (BMI, kg/m2 using length in place of height). Blood Pressure and Cardiovascular Characteristics SHROB and their SHR littermates develop high blood pressure spontaneously (Figure 2). In both SHROB and SHR, systolic blood pressure reaches hypertensive levels (>150mm Hg) beginning at about 3 months of age and then rises progressively between 6 and 9 months. Hypertension is maintained until shortly before death. Unlike SHR where males show higher blood pressure than females, SHROB show no sex difference

SHROB AND SYNDROME X

129

Figure 1 Evolution of body weight across the lifespan in SHROB/Kol and SHR/Kol rats. Rats were weighed weekly and body weight was sorted by phenotype, age and sex. Each point represents the mean ± standard error of 12 to 58 rats. Standard error bars are not visible on most points because they are smaller than the plot symbols.

in blood pressure. The standard errors for tail cuff blood pressure are small throughout adulthood (Figure 2), indicating that tail cuff measurement techniques are highly reproducible, and that the rats show a consistently hypertensive phenotype. Direct mean arterial pressure (MAP) under urethane anesthesia confirms the tail cuff data (Table 1). Left ventricular wall thickness is similar in SHROB male, SHROB female and SHR male, but is reduced in female SHR relative to the other three groups Table 1 Summary of phenotype data available on rats of the SHROB/Kol strain. Variable

Metabolic parameters Body weight (g) Nasoanal length (mm) Lee index (g1/3/mm×105) Retroperitoneal fat (g) Subscapular fat (g) Epididymal/myometrial fat pad (g) Mesenteric fat pad (g) Cardiovascular properties Mean arterial pressure (MAP) (mmHg)

SHROB Male (N=18)

SHROB Female (N=24)

SHR Male (N=12)

SHR Female (N=16)

608±29* 235±4 358±5* 36±2* 32+3* 15±1* 12±1*

576±15* 221±2*† 375±4*† 41±2* 46±3*† 19±1*† 15±1*

384±7 234±2 308±3 3.0±0.4 ND 5.3±1.3 1.3±0.2

237±5† 199±3† 308±3 2.2±0.4 ND 3.5±1.0 0.8±0.1

123±3*

125±4*

139±4

139±6

130

R.J.KOLETSKY ET AL.

Variable

SHROB Male (N=18)

SHROB Female (N=24)

SHR Male (N=12)

SHR Female (N=16)

MAP after ganglionic blockade (mmHg) Left ventricular wall thickness (mm) Heart weight (g) Organ weights Liver weight (g) Kidney weight (g) Adrenal weight (mg) Blood chemistry Glucose (mg%) Triglyceride (mg%) Cholesterol (mg%) Urea nitrogen (mg%) Creatinine (mg%) Uric acid (mg%) Phosphate (mg%) Albumin (g%)

60±4 4.1±0.09 1.5±0.06

71±3*† 4.0±0.09 1.5±0.03*

62±2 4.1±0.1 1.3±0.04

83±5* 3.5±0.1† 1.0±0.03*†

21±1.4* 3.4±0.16* 53+3*

17±1.4* 2.9±0.09*† 64±3*†

11±0.5 2.8±0.01 44±3

7±0.3† 1.7±0.03† 50±3

322±41* 553±85* 151±18* 41±4 0.49±0.01* 4.4±1.4* 10.5±1.3* 3.5±0.5*

337±29* 646±58* 121±13* 31±2† 0.43±0.04 2.4±0.4 4.3±0.9† 3.6±0.10*

198±23 67±1 59±7 32±3 0.31±0.03 1.7±0.1 5.5±0.3 2.1±0.1

238±18 60±4 59±6 25±2 0.42±0.05 2.5±0.3 4.7±0.7 2.4±0.1

* Significant difference between SHROB and SHR of the same sex, P<0.05 by analysis of variance. † Significant difference between males and females of the same phenotype, P<0.05 by analysis of variance. ND Not detectable.

(Table 1). Heart weight is not significantly affected by the obese genotype in male rats, while obese female rats had heavier hearts than their lean sisters. These changes may reflect either cardiac hypertrophy or a physiological adaptation to increased body size. Under basal conditions while maintained on regular chow ad libitum, the SHROB has slightly lower blood pressures than their lean littermates. This suggests that obesity does not contribute to hypertension in this model. Furthermore, caloric restriction and weight loss do not normalize blood pressure in SHROB and may even increase it (Koletsky & Puterman, 1976; Ernsberger et al., 1994). The mechanisms responsible for hypertension in SHROB and their lean littermates are not known, but presumably are polygenic in origin, as has been shown for other SHR substrains (Schork et al, 1995). Hyperlipidemia SHROB of either sex uniformly develop hyperlipidemia, characterized by markedly elevated plasma triglycerides and a moderate rise in plasma cholesterol relative to age-matched SHR (Table 1). Serum lipid levels in SHROB are elevated as early as 5 weeks of age, and continue to rise throughout life until the last few weeks of life when the animals are in a terminal decline (Koletsky, 1975b). Triglyceride and cholesterol values are not elevated in lean littermates of either sex. Lipid profiles are not different between sexes in either group. Possible mechanisms for elevated cholesterol in SHROB included increased hepatic synthesis and impaired catabolism of cholesterol (Tan et al, 1976).

SHROB AND SYNDROME X

131

Figure 2 Evolution of blood pressure across the lifespan in SHROB/Kol and SHR/Kol rats. Rats were place in restrainers and systolic blood pressure was determined by a tail cuff device as previously described (Ernsberger et al., 1994b). Each point represents the mean ± standard error of 6 to 36 rats.

Kidney Disease SHROB develop spontaneous glomerulopathy with proteinuria when maintained on a soy-based rat chow (Koletsky, 1975b; Abramowsky et al., 1984; Ernsberger 1993; Koletsky et al., 1995). Proteinuria is detected as early as 6 weeks of age and accelerates exponentially. By 6 months of age, SHROB exhibit severe protinuria (Figure 3). SHR do not exhibit significant proteinuria or kidney disease, despite having even more severe hypertension than SHROB. Raising blood pressure via weight cycling, or via a high salt diet (4% NaCl) exacerbates kidney disease in SHROB (Koletsky et al., 1995), as shown in Figure 3. Proteinuria increases in SHR fed a high salt diet, but to a far lesser extent. Morphologically the kidneys of SHROB show focal segmentai glomerulosclerosis and nephrosclerosis, resembling human diabetic and hypertensive glomerular and vascular damage. In spite of the proteinuria, the SHROB does not show increases in creatinine and blood urea nitrogen until terminally ill. A possible mediator of renal pathology in the SHROB is angiotensin II, because receptor sites for this hormone are down-regulated in SHROB with early renal pathology, consistent with overactivity of the renin-angiotensin system (Ernsberger et al., 1993). Angiotensin II may also contribute to kidney damage caused by diabetes in humans (Kasiske et al., 1993). Vascular Disease Originally, a proportion of obese animals developed vascular disease similar to atherosclerosis (Koletsky, 1975b). This pathologic finding ceased around 1980. The genetic defects leading to obesity and hypertension may be separate from the mutation that caused atherosclerosis on the background of these disorders. The vascular disease affected pancreatic, superior mesenteric and hepatic arteries along with their

132

R.J.KOLETSKY ET AL.

Figure 3 Urinary protein excretion in SHROB/Kol and SHR/Kol rats on control and high-salt diets and in weightcycled SHROB (VLCD). Rats aged 3–4 months were maintained for two months on control chow (about 1% NaCl) or identical chow containing 4% NaCl. Urine was collected in metabolic cages for 24h and assayed for protein as previously described (Koletsky et al., 1995).

branches. Occasional myocardial necrosis was observed. Vessels showed focal or diffuse nodular thickening, beading and tortuosity with aneurysms and thrombosis. Microscopically, the principal lesion consisted of smooth muscle cell hyperplasia involving medial and intimal layers, with abundant fat droplets and foam cell formation. Polyarteritis was often present in other arteries of SHROB with and without fatty fibrous plaques. Lean SHR developed vascular disease much later in life, similar in structure and location to SHROB but with smaller amounts of fat and foam cell formation. Prior to one year, vascular disease was present in 40% of SHROB but only 5% of SHR. Toward the end of life, 35% of SHR had some degree of vascular damage. The reason for the disappearance of the vascular disease is unknown. Interestingly, one of the many substrains derived from the SHROB by the NIH appears to develop vascular and myocardial lesions (Russel & Koeslag, 1990). SHROB show pathological changes in the blood vessels of the retina, which resemble those of human diabetic retinopathy (Huang et al., 1995). The SHROB retina shows signs of neovascularization and progressive capillary dropout. Retinopathy occurs in the absence of hyperglycemia, implicating other factors in its pathogenesis, such as hypertension, hyperlipidemia and hyperinsulinemia. Endocrine Function Leptin Circulating levels of the adipose tissue hormone leptin is elevated 170 fold in the plasma of the SHROB rat relative to SHR littermates (Friedman et al., 1997a). This presumably reflects the profound leptin resistance

SHROB AND SYNDROME X

133

resulting from the complete absence of functional leptin receptors. Remarkably, the expression of leptin mRNA in adipose tissue is only elevated by 3-fold. This implies that the clearance of leptin from the circulation may be impaired in SHROB rats. In addition to mediating the hormonal actions of leptin, the leptin receptor may mediate leptin transport from the circulation into the cerebrospinal fluid (Lynn et al., 1996). If so, then SHROB should have negligible levels of leptin in their cerebrospinal fluid. However, the level of leptin in cerebrospinal fluid samples from SHROB and SHR were similar (Ishizuka et al, 1998). Thus, despite a lack of leptin receptors, leptin was able to penetrate the blood brain barrier in SHROB. However, the ratio of plasma to cerebrospinal fluid leptin was far lower in SHROB, implying impaired leptin transport into the rbain. Hypothalamic-pituitary-adrenal axis SHROB produce twice as much corticosterone as lean littermates while maintaining diurnal rhythm (Koletsky & Ernsberger, 1992). Histologically, the adrenal zona fasciculata that produces corticosterone is enlarged in the SHROB (Koletsky, 1975b). A higher dose of the synthetic glucocorticoid dexamethasone is needed in SHROB than in SHR to suppress endogenous corticosterone production (Koletsky & Ernsberger, 1992). The SHROB has normal diurnal rhythms in urinary corticosterone, urine volume, food intake, and water consumption, but the amplitude of each rhythm is greater than for their SHR littermates. This partial profile of the hypothalamic-pituitary-adrenal axis suggests a setpoint abnormality at the level of the hypothalamus or pituitary gland. Removal of the pituitary gland from SHROB normalized food intake, induced weight loss, and lowered circulating triglyceride and cholesterol (Koletsky & Snajdar, 1979b). Additionally, proteinuria and renal disease were arrested. These effects that were greater than those seen with pair feeding of control animals, suggest an independent hormonal effect on lipid metabolism and renal disease. Endocrine pancreas Pancreatic islets are greatly enlarged in SHROB compared with SHR littermates (Koletsky, 1975b). Fasting insulin levels are 20-fold elevated in SHROB, but fasting hyperglycemia is absent. Oral glucose tolerance testing in SHROB demonstrated a sustained post-challenge elevation in plasma glucose at 180 and 240 min compared to lean SHR littermates despite a marked and prolonged insulin response, suggestive of glucose intolerance (Figure 4). The rate of insulin-stimulated 3-O-methylglucose transport was reduced 68% in isolated epitrochlearis muscles from the SHROB compared to SHR. Insulin-stimulated tyrosine phosphorylation of the insulin receptor β-subunit and IRS-1, in intact skeletal muscle of SHROB, were reduced by 36% and 23%, respectively, compared to SHR, due primarily to 32% and 60% decreases in insulin receptor and IRS-1 protein expression, respectively. The levels of p85 regulatory subunit of phosphatidylinositol-3-kinase and the glucose transporter GLUT4 were reduced by 28% and 25% in SHROB muscle compared to SHR. In the liver of SHROB, the ability of insulin to induce tyrosine phosphorylation of IRS-1 was not changed, but insulin receptor phosphorylation was decreased by 41% compared to SHR due to a 30% reduction in insulin receptor levels (Ernsberger et al., 1996b; Friedman et al., 1997a; Friedman et al., 1997b; Ishizuka et al., 1998; Friedman et al., 1998; Ernsberger et al., 1999a). Insulin-stimulated phosphorylation of tyrosine residues on the insulin receptor and on the associated docking protein IRS-1 are reduced in skeletal muscle and liver compared to SHR, due mainly to diminished expression of insulin receptor and IRS-1 proteins (Friedman et al., 1997). Reduced expression of IRS-1 may in particular represent the molecular basis for insulin resistance in this model.

134

R.J.KOLETSKY ET AL.

Figure 4 Plasma glucose and insulin during oral glucose tolerance testing in SHROB/Kol and SHR/Kol rats. Rats 4–6 months of age were fasted for 18h and given a 6g/kg glucose load by gavage with a feeding tube at time 0, then blood samples (0.2 ml) were taken at regular intervals. Glucose was assayed in whole blood using a colorometric glucose oxidase assay (One-Touch; Lifescan, Milpitas, CA). Insulin was measured by radioimmunoassay using rat insulin standards and an antibody directed against rat insulin (Linco, St. Charles, IL).

Reproductive function Both male and female SHROB have consistently failed to produce offspring when mated with each other or with their SHR siblings, even when male SHROB have been treated with daily injections of testosterone. Microscopic study of gonads demonstrates decreased levels of spermatogenesis or fewer mature ovarian follicles with ova (Koletsky, 1975b). We speculate that metabolic and neuroendocrine abnormalities may contribute to functional sterility.

SHROB AND SYNDROME X

135

Morphologic and Anatomic Data Autopsy studies of adult SHROB and their SHR littermates show that major organs are enlarged in association with the obese genotype (Table 1). The livers of male and female SHROB are more massive than SHR of either sex and contain abundant fat. The kidneys and adrenal glands are also enlarged in SHROB, roughly in proportion to body size. Lifespan and Cause of Death SHROB rats live about half as long as their SHR siblings (Figure 3). SHROB die between 225–375 days of age, while SHR siblings show <5% mortality at these ages (data not shown). Median life expectancy differs little by sex, with females living only about two weeks longer. The shortened lifespan of SHROB is primarily a result of kidney disease and protein wasting. The terminal course extends several weeks and is characterized by marked anorexia and weight loss. USE OF THE SHROB RAT AS AN EXPERIMENTAL MODEL Metabolic syndrome X consists of insulin resistance as a primary defect associated with compensatory hyperinsulinemia, impaired glucose tolerance, dyslipidemia, and hypertension (Zavaroni et al., 1994). Obesity is often present. A very similar constellation of abnormalities is found in SHROB. The relationships between the diverse phenotypic features of human Syndrome X are unknown. The use of SHROB as a model might provide information on the interactions between the various components of the syndrome. Dietary Manipulations Low calorie diet Weanling SHROB restricted to 1/3 of their usual consumption lost 30% of their body weight, but remained obese (Koletsky & Puterman, 1976; Koletsky & Puterman, 1977). Plasma triglyceride and cholesterol levels nearly normalized, but surprisingly blood pressure rose significantly during caloric restriction. Proteinuria and renal disease were significantly reduced. Fasting insulin fell in calorically restricted SHROB but still exceeded levels in SHR (O’Dea & Koletsky, 1977). Lipogenesis was reduced but remained elevated in calorically restricted SHROB, particularly in adipose tissue depots. Glycogen synthesis was actually increased in most tissues after caloric restriction. Thus, caloric restriction failed to correct all of the metabolic abnormalities in SHROB, suggesting intrinsic deficits. The life span of SHROB on low calorie intake was doubled. Weight cycling Obese humans commonly alternate between dietary restriction and bingeing and show repeated cycles of weight loss and regain (Folsom et al., 1996). The SHROB rat has been used to model this “yo-yo syndrome”, by alternating a very low calorie diet with ad libitum refeeding. The fluctuations of body weight have deleterious cardiovascular effects and increase mortality, by unknown mechanisms. SHROB were subjected to one or more cycles of a very low calorie diet (VLCD) followed by regain of lost weight after

136

R.J.KOLETSKY ET AL.

restoration of free access to food. SHROB lost weight on the restricted diet, but subsequently regained more than was lost despite similar total overall cumulative food intakes as ad lib fed SHROB. Excess fat accumulated selectively within abdominal depots. Cycled animals lost less weight after each succeeding cycle and metabolic efficiency was enhanced. Blood pressure in weight cycled obese animals were increased by 27 mmHg beyond baseline levels of ad lib fed SHROB (Ernsberger et al., 1994). In association with the rise in blood pressure, urinary protein excretion rose, and renal damage was enhanced (Koletsky & Ernsberger, 1996). The likely mechanism for the rise in blood pressure during weight regain is increased sympathetic activity. The evidence implicating the sympathetic nervous system in refeeding hypertension includes downregulation of cardiac -adrenergic receptors (Ernsberger et al., 1994b) and kidney -adrenergic receptors (Ernsberger et al., 1999b) in weight-cycled SHROB. Furthermore, ganglionic blockade selectively abolished the elevation in blood pressure (Ernsberger et al., 1994b; Ernsberger et al., 1996a). Moreover, urinary catecholamine levels were elevated during the weight regaining phase of weight cycling and declined during caloric restriction (Ernsberger et al., 1999b). High salt diet Elevated intake of salt can increase blood pressure in some SHR strains and some humans (Jin et al., 1989). After two weeks on 4% NaCl, SHROB showed a 47 mmHg increase in tail cuff blood pressure, while SHR showed a smaller rise (22 mmHg) (Ernsberger et al., 1994a). Blood pressure remained elevated in SHROB to the same level for at least 7 weeks on high salt. The hypertensive effect of dietary salt was confirmed by direct mean blood pressure under anesthesia. This difference was abolished by ganglionic blockade, suggesting a role for the sympathoadrenal system in promoting salt-sensitivity in SHROB. Other laboratories have implicated the sympathetic nervous system in the response to SHR to excess dietary salt (Wyss et al., 1992). SHROB fed the high salt diet showed nearly 3-fold higher excretion of protein in the urine and greater renal damage relative to SHROB fed regular chow (Figure 3). At autopsy, the size of the heart and the thickness of the left ventricular wall were increased by dietary salt loading in SHROB, but not in SHR. On the high salt diet SHROB suffered 57% mortality compared with 4% mortality of SHROB on regular chow. The cause of death was uncertain, but renal failure, stroke and cardiac death may all contribute. SHROB are markedly responsive to dietary NaCl and represent a novel model of salt-sensitive obese hypertension. High sucrose diet Excess dietary sucrose is thought to exacerbate the metabolic components of syndrome X in humans (Young & Landsberg, 1982). In some obese rat models, excess dietary sucrose can elicit or exacerbate diabetes (Velasquez et al., 1995). SHROB placed on a 60% sucrose diet sucrose diet did not gain additional weight or change their fat distribution. Despite fasting euglycemia, the high sucrose diet impaired glucose tolerance. Fasting insulin and the insulin response to glucose challenge were further increased from the elevated levels found in SHROB relative to SHR. The high sucrose diet did not effect blood pressure but did lower triglyceride levels. Proteinuria was reduced in SHROB on the high sucrose diet.

SHROB AND SYNDROME X

137

Response to Pharmacotherapy Antihypertensive drugs Treatment of SHROB with a combination of antihypertensive drugs consisting of hydralazine, hydrochlorothiazide and propranolol rendered the animals normotensive (Koletsky and Snajdar, 1979a). This treatment reduced incidence of vascular disease and prolonged lifespan despite a lack of effect on blood lipid levels. When the rats were treated with antihypertensive drugs and simultaneously maintained on a low calorie diet, which alleviated the hyperlipidemia, the result was complete elimination of vascular disease and increase in longevity to that of a normal rat. This supports the view that prophylactic management of high blood pressure and hyperlipidemia reduces the frequency of cardiovascular and renal damage. The ganglionic blocker chlorisondamine was administered to SHROB and SHR to eliminate the sympathoadrenal contribution to maintenance of blood pressure. The lower blood pressure associated with the obese genotype persisted following ganglionic blockade, suggesting that obesity per se and the sympathetic nervous system are not the causes of hypertension in this obesity model (Koletsky and Ernsberger, 1992; Koletsky and Ernsberger, 1996). Moxonidine treatment Overactivity of the SNS may be responsible for exacerbating obesity and other phenotypic components of Syndrome X such as hypertension and insulin resistance. Moxonidine, a centrally acting selective I1imida2oline receptor agonist that inhibits SNS activity, was administered to SHROB and SHR for 90 days in food at 8 mg/ kg/day. Moxonidine significantly reduced mean blood pressure in both groups (Friedman et al., 1998). Moxonidine treatment reduced fasting insulin levels by 71% in SHROB, and lowered plasma free fatty acids by 25%. In SHR, moxonidine treatment did not effect insulin levels but decreased free fatty acids by 17%. During an oral glucose tolerance test, blood glucose levels in moxonidine-treated SHROB were reduced relative to untreated SHROB from 60 minutes onwards. Insulin secretion was facilitated at 30 (83% greater) and 60 min (67% greater) post-challenge compared to control SHROB. In skeletal muscle, moxondine treatment increased the expression of the insulin receptor protein by 19% in SHROB, but was without effect in SHR. The level of IRS-1 protein was decreased by 60% in untreated SHROB compared with SHR. Moxonidine treatment enhanced the expression and insulin-stimulated phosphorylation of IRS-1 protein in skeletal muscle by 74% and 27%, respectively in SHROB, and by 40% and 56% in SHR. Moxonidine increased the levels of expression of IRS-1 protein in liver by 275% in SHR and 260% in SHROB, respectively. These findings indicate that chronic inhibition of sympathetic activity with moxonidine therapy can lower free fatty acids and significantly improve insulin secretion, glucose disposal, and expression of key insulin signaling intermediates in an animal model of obese hypertension. Conclusion The SHROB rat is a unique animal model expressing multiple abnormal phenotypes including genetic obesity, spontaneous hypertension, hyperinsulinemia and hyperlipidemia. These features closely resemble those found in the human “Syndrome X.” The SHROB also has a spontaneous and progressive nephrotic syndrome, that is a potential model for human diabetic and hypertensive nephropathies as well as a retinopathy that resembles diabetes. Overactivity of the SNS may be a common link between hypertension

138

R.J.KOLETSKY ET AL.

and insulin resistance. We have used dietary modifications and pharmacologic interventions to study the physiologic changes that alter the various phenotypic components of Syndrome X. Further studies directed at understanding the mechanisms and interactions between the multiple abnormal phenotypes and their underlying genotypes will lead to better understanding of these commonly intertwined clinical problems. Availability SHROB/Kol and SHR/Kol littermates are commercially available exclusively from Genetic Models, Inc. in Indianapolis Indiana (telephone: 1–317–824–7070, telefax: 1–317–824–7080, e-mail: [email protected]). Pricing has not been set, but costs are high owing to several factors. Most significantly, because the SHROB is completely infertile, all breeding must be carried out with heterozygotic genetic carriers. Only one in four offspring are SHROB with the obese phenotype, the remaining three being lean SHR. In addition, litter sizes with these hypertensive animals are characteristically small (4 to 8). Furthermore, the animals require additional veterinary support because of increased incidence of superficial skin infections, particularly if soft bedding is not provided and changed frequently. Surgical procedures require special care because of increased risk of infection, poor wound healing, and circulatory disturbances related to both hypertension and obesity. REFERENCES Abramowsky, C.R., Aikawa, M., Swinehart, G.L. and Snajdar, R.M. (1984) Spontaneous nephrotic syndrome in a genetic rat model. Am. J. Pathol., 117, 400–408. Buñag, R D. and Barringer, D.L. (1988) Obese Zucker rats, though still normotensive, already have impaired chronotropic baroreflexes. Clin. Exp. Hypertem. {A}, 10 Suppl 1, 257–262. Ernsberger, P., Ishizaka, T., Liu, S., Farrell, C.J., Bedol, D., Koletsky, R.J. and Friedman, J.E. (1999a) Mechanisms of antihyperglycemic effects of moxonidine in the obse spontaneously hypertensive Koletsky rat (SHROB). J. Pharmacol. Exp. Ther., 288, 139–147. Ernsberger, P., Koletsky, R.J., Baskin, J.S. and Foley, M. (1994B) Refeeding hypertension in obese spontaneously hypertensive rats. Hypertension, 24, 699–705. Ernsberger, P., Koletsky, R.J., Baskin, J.Z. and Collins, L. (1996a) Consequences of weight cycling in obese spontaneously hypertensive rats. Am. J. Physiol., 270, R864-R872. Ernsberger, P., Koletsky, R.J. and Collins, L.N. (1994a) Lethal consequences of a high salt diet in obese SHR. Hypertension, 24 376 (Abstract). Ernsberger, P., Koletsky, R.J., Collins, L.N. and Bedol, D. (1996b) Sympathetic nervous system in salt-sensitive and obese hypertension: Amelioration of multiple abnormalities by a central sympatholytic agent. Cardiovasc. Drug Ther., 10 Suppl. 1, 275–282. Ernsberger, P., Koletsky, R.J., Collins, L.A. and Douglas, J.G. (1993) Renal angiotensin receptor mapping in obese spontaneously hypertensive rats. Hypertension, 21, 1039–1045. Ernsberger, P., Koletsky, R.J., Kilani, A., Viswan, G. and Bedol, D. (1999b) Effects of weight cycling on urinary catecholamines: Sympathoadrenal role in refeeding hypertension. J. Hypertens., 16, 2001–2005. Folsom, A.R., French, S.N., Zheng, W., Baxter, J.E. and Jeffery, R.W. (1996) Weight variability and mortality: The Iowa Women’s Health Study. Int. J. Obes., 20, 707–709. Friedman, J.E., Ishizuka, T., Liu, S., Bedol, D., Koletsky, R.J. and Ernsberger, P. (1997a) Metabolic consequences of a nonsense mutation in the leptin receptor gene (fak) in the obese spontaneously hypertensive Koletsky rat (SHROB). Clin. Endocrinol, Diabetes, 106 (Suppl. 3), 82–84. Friedman, J.E., Ishizuka, T., Liu, S., Farrell, C.J., Bedol, D., Koletsky, R.J., et al. (1997b) Reduced insulin receptor signaling in the obese spontaneously hypertensive Koletsky rat. Am. J. Physiol., 273, E1014-E1023.

SHROB AND SYNDROME X

139

Friedman J.E., Ishizuka, T., Liu, S., Farrell, C.J., Koletsky, R.J., Bedol, D., et al (1998) Antihyperglycemic activity of moxonidine: Metabolic and molecular effects on obese spontaneously hypertensive rata. Blood Press., 1 (Suppl. 3), 32–39. Greenhouse, D.D., Hansen, C.T. and Michaelis, O.E. (1990) Development of fatty and corpulent rat strains. ILAR News, 32, 2–4. Huang, S.S., Khosrof, S.A., Koletsky, RJ., Benetz, B.A. and Ernsberger, P. (1995) Characterization of retinal vascular abnormalities in lean and obese spontaneously hypertensive rats. Clin. Exp. Pharmacol. Physiol., 22 Suppl. 1, S129–31. Ishizuka, T., Ernsberger, P., Liu, S., Bedol, D., Lehman, T.M., Koletsky, R.J., et al. (1998) Phenotypic consequences of a nonsense mutation in the leptin receptor gene (fak) inobese spontaneously hypertensive Koletsky rata (SHROB). J. Nutr., 128, 2299–2306. Jin, H., Chen, Y.-F., Yang, R.-H. and Oparil, S. (1989) Atrial natriuretic factor in NaCl-sensitive and NaCl-resiatant spontaneously hypertensive rats. Hypertension, 14, 404–412. Kasiske, B.L., Cleary, M.P., O’Donnell M.P. and Keane, W.F. (1985) Effects of genetic obesity on renal structure and function in the Zucker rat. J. Lab. Clin. Med., 106, 598–604. Kasiske, B.L., Kalil, R.S., Ma, J.Z., Liao, M. and Keane, W.F. (1993) Effect of antihypertensive therapy on the kidney in patients with diabetes: a meta-regression analysis. Ann. Intern. Med., 118, 129–138. Koletaky, RJ., Boccia, J. and Ernsberger, P. (1995) Acceleration of renal disease in obese SHR by exacerbation of hypertension. Clin. Exp. Pharmacol. Physiol., 22, S254-S256. Koletsky, R.J. and Ernsberger, P. (1992) Obese SHR (Koletsky Rat): A model for the interactions between obesity and hypertension. In Genetic Hypertension, edited by J.Sassard, pp. 373–375. London: John Libbey. Koletsky, R.J. and Ernsberger, P. (1996) Phenotypic characterization of a genetically obese and hypertensive rat strain: SHROB/Kol. Rat Genome, 2, 10–22. Koletsky, S. (1972) New type of spontaneously hypertensive rats with hyperlipemia and endocrine gland defects. In Spontaneous Hypertension: Its Pathogenesis and Complications, edited by K.Okamoto, pp. 194–197. Tokyo: Igaku Shoin Ltd. Koletsky, S. (1975a) Animal model obese hypertensive rat. Am. J. Pathol., 81, 463–466. Koletsky, S. (1975b) Pathologic findings and laboratory data in a new strain of obese hypertensive rats. Am. J. Pathol., 80, 129–142. Koletsky, S. and Puterman, D.I. (1976) Effect of low calorie diet on the hyperlipidemia, hypertension, and life span of genetically obese rats. Proc. Soc. Exp. Biol. Med., 161, 368– 371. Koletsky, S. and Puterman, D.I. (1977) Reduction of atherosclerotic disease in genetically obese rats by low calorie diet. Exp. Mol. Pathol., 26, 415424. Koletsky, S. and Snajdar, R. (1979b) Elimination of the hyperlipidemia and proteinuria of genetically obese rats by hypophysectomy. Lab. Invest., 41, 287–293. Koletsky, S. and Snajdar, R.M. (1979a) Reduction of vascular disease in genetically obese rats treated for hypertension and hyperlipidemia. Exp. Mol.Pathol., 30, 409–419. Lynn, R.B., Cao, G.Y., Considine, R.V., Hyde, T.M. and Caro, J.F. (1996) Autoradiographic localization of leptin binding in the choroid plexus of ob/ob and db/db mice. Biochem. Biophys. Res. Commun., 219, 884–889. O’Dea, K. and Koletsky, S. (1977) Effect of caloric restriction on basal insulin levels and the in vivo lipogenesis and glycogen synthesis from glucose in the Koletsky obese rat. Metabolism, 26, 763–772. Russel, J.C. and Koeslag, D.G. (1990) Jcr: LA-corpulent rat: A strain with spontaneous vascular and myocardial disease. ILAR News, 32, 27–32. Schork, N.J., Krieger, J.E., Trolliet, M.R., Franchini K.G., Koike, G, Krieger, E.M., et al. (1995) A biometrical genome search in rats reveals the multigenic basis of blood pressure variation. Genome Research, 5, 164–172. Takaya, K., Ogawa, Y., Hiraoka, J., Hosoda, K., Yamori, Y., Nakao, K., et al. (1996) Nonsense mutation of leptin receptor in the obese spontaneously hypertensive Koletsky rat. Nat. Genet., 14 130–131. Tan, E., Butkus, A. and Koletsky, S. (1976) Hepatic cholesterol metabolism in vitro in the obese spontaneously hypertensive, hyperlipemic and atherosclerotic rat. Exp. Mol. Pathol., 25 142–151.

140

R.J.KOLETSKY ET AL.

Velasquez, M.T., Abraham, A.A., Kimmel, P.L., Farkas-Szallasi, T. and Michaelis, O.E. (1995) Diabetic glomerulopathy in the SHR/N-corpulent rat: Role of dietary carbohydrate in a model of NIDDM. Diabetologia, 88, 31–38. White, B.D., Corll, C.B. and Porter, J.R. (1989) The metabolic clearance rate of corticosterone in lean and obese male Zucker rats. Metabolism, b 530–536. Wyss, J.M., Oparil, S. and Sripairojthikoon, W. (1992) Neuronal control of the kidney: Contribution to hypertension. Can. J. Physiol. Pharmacol., 70, 759–770. Yamashita, T., Murakami, T., Iida, M., Kuwajima, M. and Shima, K. (1997) Leptin receptor of Zucker fatty rat performs reduced signal transduction. Diabetes, 46, 1077–1080. Yen, T.T., Shaw, W.N. and Yu, P.L (1977) Genetics of obesity in Zucker rats and Koletsky rats. Heredity, 38 373–377. Young, J.B. and Landeberg, L. (1982) Diet-induced changes in synpathetic nervous system activity: possible implications for obesity and hypertension. J. Chronic. Dis., 36, 879–886. Zavaroni I., Bonini, L, Fantuzzi, M., Dall’aglio, E., Passeri, M. and Reaven, G.M. (1994) Hyperinsulinaemia, obesity, and syndrome X.J. Intern. Meet., 235, 51–56.

7. CHARACTERISTICS OF WISTAR FATTY RAT HIROYUKI ODAKA, YASUO SUGIYAMA AND HITOSHI IKEDA Pharmaceutical Research Division, Takeda Chemical Industries, Ltd, Yodogawa-ku, Osaka 532–8686, Japan

INTRODUCTION Many animal models of diabetes have been established and provided us with beneficial information on the causes and progress of diabetes in humans. Most of the experimentally diabetic models with insulin deficiency have been created by the chemical or surgical destruction of pancreatic B cells and have been applied for the studies on metabolic abnormalities and morphological changes in pancreas and other tissues related to diabetic complications. On the other hand, the spontaneously diabetic animals have been obtained due to mutation or selective breeding. Some spontaneously diabetic models like Wistar fatty rats were created by the crossbreeding of two strains with different genetic characteristics. These spontaneously diabetic models have been used for studies on the pathogenesis of diabetes and pharmacological evaluation of antidiabetic drugs. Needless to mention, we have to choose suitable models depending on the purpose of the study since neither of the models can exhibit all of the types and features of diabetes. ESTABLISHMENT AND UTILIZATION Wistar fatty rat was the first rat model of obese Type 2 diabetes (NIDDM) established on the basis of the hypothesis that both an environmental factor and genetic backgroud for diabetes were needed to develop diabetes (Ikeda et al., 1981). Among the environmental factors, obesity is assumed to be the most powerful risk factors related to metabolic and hormonal abnormalities. In respect of metabolism, the diabetogenic effect of obesity is of special interest, because obesity increases the insulin requirement and decreases the insulin sensitivity in the muscle, adipose tissue and liver. In fact, Type 2 diabetes is characterized by insulin resistance frequently associated with obesity in humans and animals. Zucker fatty rat develops obesity with

142

H.ODAKA ET AL.

hyperinsulinemia, hyperlipidemia, hyperphagia, and glucose intolerance (Zucker and Antoniades, 1972). However, their blood glucose level is close to normal throughout their life. This suggested that additional factors were required to provoke diabetes in obesity. Therefore, the crossbreeding of the fa-gene carrier of Zucker rat with the Wistar Kyoto rat was started because the Wistar Kyoto rat is less sensitive to insulin and less tolerant to glucose than the Zucker rat. At the 5th generation of backcrossing, male obese hybrids, Wistar fatty rats, were found to be hyperglycemic in addition to hyperlipidemic and hyperinsulinemic, in association with severe obesity. These characteristics have been maintained through the later generations of backcrossing. The characteristics of Wistar fatty rats as compared Table 1 Diabetes and obesity in male Zucker fatty (ZF) and male Wistar fatty (WF) rats at the age of 12–16 week. References Plasma glucose Plasma insulin Pancreatic insulin mRNA Plasma triglyceride Plasma cholesterol Glucose intolerance Insulin insensitivity Insulin response to glucose Body weight Food intake Mesenteric adipose tissue wt Subcutaneous adipose tissue wt Liver weight Hepatic enzyme activity Glucokinase G6Pase FDPase Malic enzyme G6PDH

ZF << WF ZF> WF ZF >> WF ZF > or=WF ZF< WF ZF< WF ZF < WF ZF< WF ZF< WF ZF < WF ZF < WF ZF< WF ZF << WF

a a b a, c c a a a a

c d

ZF >> WF ZF << WF ZF< WF ZF >> WF ZF >> WF

a: Ikeda et al. (1981), b: Seino et al. (1992), c: Suzuki et al. (1997), d: Ikeda et al. (1990)

with those of Zucker fatty rats are shown in Table 1. Wistar fatty rats have been contributed to the precise understanding of the nature of abnormal glucose and lipid metabolism in insulin target tissues in Type 2 diabetes, and are used for the evaluation of post-insulin-receptor signaling defects in insulin resistant tissues (Kobayashi et al., 1992; Hayakawa et al., 1996), and for the investigation of the role of the newly discovered bioactive substances related to obesity or diabetes such as leptin (Murase et al., 1998), uncoupling proteins (Matsuda et al., 1998). Diabetic complications such as nephropathy, neuropathy and microangiopathy deeply influence the quality of life in humans. Experimental animal models of diabetes such as streptozotocin-injected animals have been used to evaluate the morphological changes in these tissues or organs. However, the lesions of the tissues reflected by the insulin deficient diabetic condition were slight or moderate as compared with those of humans, and therefore more suitable animal models with obesity and diabetes were needed to evalute

CHARACTERISTICS OF WISTAR FATTY RAT

143

the development of complications in Type 2 diabetes. Wistar fatty rats developed nephropathy from their early life (Suzuki et al., 1997) and gave us useful information on the cause and progress of diabetic nephropathy. Because Wistar fatty (fa/fa) rats are infertile, they are obtained by breeding the fa-carrier of Wistar lean rats (+/fa). These breeders are not commercially available but are obtained from Takeda Chemical Industries. Investigators must be prepared to the expenses of large space and cost of breeding. INSULIN RESISTANCE Overall Insulin Resistance Wistar fatty rats have provided us a lot of information on abnormalities of glucose and lipid metabolism in the obese and insulin resistant condition. The rats showed glucose intolerance, hypersecretion of insulin in response to oral glucose load and impaired response to exogenous insulin (Ikeda et al., 1981; Sugiyama and Taketomi et al., 1990). Insulin-stimulated glycogen synthesis and glycolysis in isolated soleus muscles, and insulin-stimulated glucose oxidation and lipogenesis in adipocytes were impaired in Wistar fatty rats as compared with those in lean littermates (Sugiyama et al., 1990a). Hypertriglyceridemia was attributed to the overproduction and impaired catabolism of triglycerides (Kazumi et al., 1996) with increasing expression of lipogenesis promoting enzymes such as L-type pyruvate kinase (Noguchi et al., 1992). Plasma cholesterol level was also higher, and hepatic and intestinal 3-hydroxy-3-metyhylglutaryl (HMG)-CoA reductase activities were higher in Wistar fatty rats (Jiao et al., 1991). Sugiyama et al. (1990b) studied the peripheral and hepatic insulin resistance in Wistar fatty rats using an isotopic method combined with a euglycemic clamp technique (Figure 1). In Wistar lean rats, peripheral glucose utilization (PGU) was increased and hepatic glucose production (HGP) was decreased by elevating steady state plasma insulin (SSPI). In Wistar fatty rats, neither PGU was increased nor HGP suppressed even when 480 mU/h of insulin was infused and SSPI reached about 2000 µU/ml. These findings clearly indicated the presence of hepatic as well as peripheral insulin resistance. Abnormal glucose metabolism in liver was also identified by measuring hepatic enzyme activities related to glycolysis and gluconeogenesis: glucokinase: (GK), pyruvate kinase: (PK) and glucose-6-phosphatase: (G6Pase), fructose-1, 6diphosphatase: (FDPase) (Sugiyama et al., 1989). All these enzyme activities were higher in Wistar fatty rats than in lean rats, but G6Pase/ GK ratio became larger with advancing age in fatty rats suggesting the abnormal glucose handling in the liver of Wistar fatty rats. The impaired GK mRNA expression was also reported (Noguchi et al., 1993), but glucose transporter (GLUT) 2 gene expression was increased in Wistar fatty rats (Yamamoto et al., 1993). From these studies it was revealed that insulin resistance existed not only in peripheral tissues but also in the liver of Wistar fatty rats. Mechanism of Insulin Resistance Recent extensive investigations on the cellular insulin signal transduction have unveiled vital roles of signaling molecules, including insulin receptors (IR), insulin receptor substrate-1 (IRS-1) and phosphatidylinositol 3-kinase (PI3K), in the actions of insulin and defects in the signaling systems have been implicated as pivotal causes of insulin resistance in NIDDM. All the biological actions of insulin are elicited by insulin binding to IR, which is followed by insulin signal trasduction in the cell. When insulin bound to the IRα-subunit, the (β-subunit was immediately tyrosine phosphorylated and its kinase activity toward exogenous substances, such as IRS-1, was potentiated. Phosphorylated IRS-1 was found to bind to

144

H.ODAKA ET AL.

Figure 1 A comparison of the steady state plasma glucose (SSPG), peripheral glucose utilization and hepatic glucose production in male Wistar fatty rats (Sugiyama and Shimura et al., 1990). Mean ± SD (n=6). *: p < 0.01 vs the corresponding basal value.

intracellular signaling molecules, including PI3K. PI3K has been shown to be involved in glucose uptake followed by glucose transporter GLUT-4 translocation in response to insulin stimulation. In the muscle of Wistar fatty rats, total IR number and insulin binding to the receptor is decreased and insulin-stimulated tyrosine phosphorylation of IR and IRS-1, as well as PI3K activation are reduced or blunted (Table 2) (Kobayashi et al., 1992; Hayakawa et al., 1996). When insulin sensitizing agent, pioglitazone, was administered to Wistar fatty rats for the purpose of precise confirmation of the cause of insulin resistance in this rat model, insulin-stimulated tyrosine phosphorylation

CHARACTERISTICS OF WISTAR FATTY RAT

145

Table 2 Plasma components, Insulin receptor (IR), post-receptor signaling and TNF-α levels in Wistar lean (WL) and fatty (WF) rats with or without pioglitazone (pio) administration. References Plasma glucose Plasma triglyceride Plasma insulin Plasma leptin Total IR number Insulin binding to IR IR Tyr phosphorylation IRS-1 Tyr phosphorylation PI3K activity PI3K p85 content Plasma TNF-α Muscle TNF-α Muscle SMase

WL < WF-pio << WF WL=WF-pio << WF WL < WF-pio < WF WL << WF-pio=WF WL=WL-pio > WF—WF-pio WL=WL-pio > WF=WF-pio WL=WL-pio=WF-pio > WF WL=WL-pio=WF-pio > WF WL=WL-pio=WF-pio > WF WL=WL-pio=WF=WF-pio WL=WF-pio << WF WL=WF-pio << WF WL=WF-pio << WF

a, b, c a, b, c a, b, c c a b a, b a a a c c c

a: Hayakawa et al. (1996), b: Kobayashi et al. (1992), c: Murase et al. (1998)

of IR and IRS-1, and PI3K activity was restored to the level of normal Wistar lean rats affecting neither total IR number nor PI3K p85 content (Table 2). These results revealed not only the characteristics of the impaired post-insulin-receptor signaling but the usefulness for the evaluation of antidiabetic agents in Wistar fatty rats. Data from Wistar fatty rats indicate that pioglitazone acts at post-binding sites in insulin target tissues. Recent studies imply a correlation between tumor necrosis factor-α (TNF-α) and insulin resistance. In Wistar fatty rats, TNF-α levels in both plasma and skeletal muscle, the major target tissue of insulin, were significantly higher than those in lean rats (Table 2) (Murase et al., 1998). Intracellular mechanism for TNFα-induced insulin resistance has been partially somewhat clarified. TNF-α inhibits insulin signaling by binding to the p55 TNF-α receptor and by activating of neutral sphingomyelinase (SMase). TNF-α responsive neutral SMase hydrolyses membrane sphingomyelin to ceramide which activates membranebound ceramide-activated protein kinase (Kiu et al., 1994), and the kinase probably induces serine phosphorylation of IRS-1 and converts IRS-1 into an inhibitor of IR tyrosine kinase activity (Hotamisligil et al., 1996). The activity of SMase in Wistar fatty rats was much higher than that in lean rats (Table 2). The extent of this change was similar to that in the level of muscle TNFα- suggesting that elevation of neutral SMase is closely linked with insulin resistance. Moreover, administration of pioglitazone decreased plasma and muscle TNF-α levels followed by a decrease in plasma indicators of insulin resistance such as glucose, triglycerides and insulin (Table 2). There were significantly positive correlations between the muscle TNF-α level and the plasma glucose or triglyceride level among Wistar fatty rat groups (Murase et al., 1998). The change in muscle TNF-α production is one of the major causal factors for the alteration of insulin resistanceinduced metabolic abnormalities. These results indicate that the high TNF-α production in the muscle of fatty rats plays a principal role for the induction of insulin resistance, and abnormal glucose and lipid metabolism via interfering with the intracellular insulin signal transduction in Wistar fatty rats (Figure 2).

146

H.ODAKA ET AL.

Figure 2 Mechanism of insulin resistance in Wistar fatty rats.

NEPHROPATHY Several studies in humans with Type 1 (insulin-dependent diabetes mellitus, IDDM) and in experimentally insulin deficient animals have been performed to gain a better understanding of the pathogenesis of diabetic nephropathy. On the other hand, a few studies on nephropathy in type 2 (NIDDM) patients have shown that several factors such as hyperglycemia, hyperlipidemia and hypertension are associated with the development of renal lesions. In addition, obese and hypertensive SHR/N-cp rats develop glomerular lesions when they became hyperglycemic under a high sucrose feeding (Velasquez et al., 1989). However, there is no complete understanding which factor plays the principal role in the manifestation of nephropathy in NIDDM. To gain a better understanding of the contribution of hyperglycemia per se to the development of renal disease, indicators of diabetes and nephropathy of three types of fatty rats (male and female Wistar fatty rats and male Zucker fatty rats) were compared (Table 3). All fatty rats showed obesity with hyperphagia, Table 3 Comparison of nephropathy among fatty rats (Suzuki et al., 1 997). Plasma glucose Plasma triglyceride Plasma insulin Urinary protein Urinary albumin Urinary NAG

male ZF male WF female WF female WF female WF male ZF

< female WF < male ZF < male WF < male ZF < male ZF << female WF

<< male WF < female WF < male ZF < male WF < male WF < male WF

CHARACTERISTICS OF WISTAR FATTY RAT

147

hyperinsulinemia and hyperlipidemia. Among these fatty rats, male Wistar fatty rats developed hyperglycemia followed by a marked urinary excretion of protein and albumin, whereas female Wistar fatty rats and male Zucker fatty rats showed less proteinuria and albuminuria. These observations suggest that the change in mesangial cell function due to the direct action of hyperglycemia is related to the development of nephropathy in Wistar fatty rats. In fact, our unpublished data showed the glomerular filtration rate (GFR) to be markedly increased at the age of 7 weeks when hyperglycemia developed in Wistar fatty rats. GFR in Wistar fatty rats, however, decreased after the age of 22 weeks. A similar remission is observed during late stage of diabetic nephropathy in humans. In male Wistar fatty rats, the expanded glomerular mesangial area, the local formation of nodular-like and tubular lesions such as hyaline cast formation, elongation of the inside diameter and flattening of the epithelial cells were observed (Figure 3) (Suzuki et al., 1997). Therefore, the development of proteinuria and albuminuria in the male Wistar fatty rats is closely related to hyperglycemia. Another study showed that normoglycemic Zucker fatty rats were less albuminuric, than hyperglycemic Zucker fatty rats; it also indicated a close relationship between albuminuria and hyperglycemia (McCaleb et al., 1992). Urinary albumin excretion in mildly hyperglycemic female Wistar fatty rats was lower than that in normoglycemic Zucker fatty rats. The difference in urinary albumin excretion between these rats seems to be due to plasma insulin levels or other factors. N-acetyl-β-Dglucoanimidase (NAG) which was detected in the renal tubule at a high concentration was excreted into the urine of male and female Wistar fatty rats in large quantities. The amount of urinary NAG in male Wistar fatty rats tended to be larger than in female Wistar fatty rats. Therefore, hyperglycemia seems to induce marked tubular lesions in male Wistar fatty rats, and other factors such as hypertriglyceridemia may also be responsible for tubular lesions in female Wistar fatty rats. The strict control of blood glucose at an early stage of diabetes is known to be effective for the prevention of nephropathy (UK Prospective Diabetes Study Group, 1993). In male Wistar fatty rats, insulin sensitizing agent, pioglitazone (Suzuki et al., 1997), or α-glucosidase inhibitor, voglibose (Suzuki et al, 1994), suppressed the development of nephropathy (Table 4). Hyperglycemia enhances the non-enzymatic glycation of collagen and other extracellular matrix components in the glomerular basement membrane and the mesangial area and activates the polyol pathway in mesangial cells. Wistar fatty rats are slightly hypertensive (Shibouta et al, 1996 and Yoshimoto et al., 1997) and administration of antihypertensive drugs such as angiotensin II antagonist, candesartan cilexetil (TCV-116), suppressed the development of hypertension as well as GFR change in Wistar fatty rats (Table 4) (Shibouta et al., 1996). Some other factors such as impaired sodium balance (Suzuki et al., 1996) and high expression of ICAM-1 in glomeruli (Matsui et al., 1996) were reported to be related to the induction of nephropathy in Wistar fatty rats. Few animal models of spontaneous diabetes other than Wistar fatty rats showed such a marked proteinuria at a young age. Furthermore, nephropathy in Wistar fatty rats is similar to that in diabetic patients at an early and a middle stage (Diani et al., 1998). Therefore, we believe that Wistar fatty rats is a suitable model to investigate the pathogenesis of diabetic nephropathy. CONCLUSION At present, demand for diagnosis and treatment of diabetes particularly with insulin resistance and diabetic complications are increasing, and need for the animal models of spontaneous diabetes seem to become larger for the evaluation of molecular mechanisms of these disorders even though transgenic animals are now used and provide us with a lot of information concerning the importance of specific genes for the

148

H.ODAKA ET AL.

Figure 3 Renal glomeruli and tubuli from male Wistar fatty rats at 26 weeks of age (Suzuki et al., 1997). A: Renal tubuli with hyaline cast formation, elongation of the inside diameter and flattening of the epithelial cells in Wistar fatty rats.×l00. B: Renal glomeruli with an expanded mesangial area and local nodular lesions in Wistar fatty rats. ×300. C: Normal glomeruli in Wistar lean rats. ×300.

CHARACTERISTICS OF WISTAR FATTY RAT

149

Table 4 Effect of pioglitazone (pio), voglibose (vog) or candesartan cilexetil (can) on nephropathy in Wistar fatty rats (WF).

ND: not determined, a: Suzuki et al. (1997), b: Yoshimoto et al. (1997), c: Suzuki et al. (1994), d: Shibouta et al. (1996).

development of diabetes. Wistar fatty rat provides new insights on insulin resistance and diabetic nephropathy. Insulin resistance in this fatty rat seems to be closely related to the enhanced muscular TNF-α production following by the impairment of post-insulin-receptor signaling defects. This rat will provide us further information on the molecular mechanism of insulin resistance, and is also very useful for investigating the change in glucose and lipid metabolism for developing obesity. The pathogenesis of diabetic complications has been extensively studied in insulin deficient diabetes models such as streptozotocin-injected animals. A recent investigation shows, however, that the Wistar fatty rat is a very useful NIDDM model for investigating diabetic nephropathy. REFERENCES Diani, A.R., Ledbetter, S.R., Sawada, G.A., Copeland, E.J., Wyse, B.M., Hannah, B.A., et al. (1988) Structural and functional evidence of diabetic nephropathy in the Wistar fatty diabetic rat. In Frontiers in Diabetes Research. Lessons from Animal Diabetes II, edited by E.Shafrir and A.E.Renold, pp. 525–541. London, Paris: John Libbey. Hayakawa, T., Shiraki, T., Morimoto, T., Shii, K. and Ikeda, H. (1996) Pioglitazone improves insulin signaling defects in skeletal muscle from Wistar fatty (fa/fa) rats. Biochem. Biophys. Res. Commun., 223, 439–444. Hotamisligil, G.S., Peraldi, P., Budavari, A., Ellis, R., White, M.F., et al. (1996) IRS-1-mediated inhibition of insulin receptor tyrosine kinase activity in TNF-α and obesity-induced insulin resistance. Science, 271, 665–668. Ikeda, H., Shino, A., Matsuo, T., Iwatsuka, H. and Suzuoki, Z. (1981) A new genetically obese-hyperglycemic rat (Wistar fatty). Diabetes, 30, 1045–1050. Ikeda, H., Sugiyama, Y. and Matsuo T. (1990) Characterization of the Wistar fatty rat. In Progress in Obesity Research, edited by Y.Oomura et al., pp. 435–439- London, Paris: John Libbey. Jiao, S., Matsuzawa, Y., Matsubara, K., Kubo, M., Tokunaga, K., Odaka, H., et al. (1991) Abnormalities of plasma lipoproteins in a new genetically obese rat with non-insulin-dependent diabetes mellitus (Wistar fatty rat). Int. J. Obes., 15, 487–495.

150

H.ODAKA ET AL.

Kazumi, T., Hirano, T., Odaka, H., Ebara, T., Amano, N., Hozumi, T., et al. (1996) VLDL triglyceride kinetics in Wistar fatty rats, an animal model og NIDDM. Diabetes, 45, 806– 811. Kobayashi, M., Iwanishi, M., Egawa, K. and Shigeta, Y. (1992) Pioglitazone increases insulin sensitivity by activating insulin receptor kinase. Diabetes, 41, 476–483. Liu, J., Mathias, S., Yang, Z. and Kolesnick, R.N. (1994) Renaturation and tumor necrosis factor-α stimulation of a 97kDa ceramide-activated protein kinase. J. Biol. Chem., 269, 3047–3052. Matsuda, J., Hosoda, K., Itoh, H., Son, C., Doi, K., Hanaoka, I., et al. (1998) Increased adipose expression of the uncoupling protein-3 gene by thiazolidinediones in Wistar fatty rats and in cultured adipocytes. Diabetes, 47, 1809–1814. Matsui, H., Suzuki, M., Tsukuda, R., Iida, K., Miyasaka, M. and Ikeda, H. (1996) Expression of ICAM-1 on glomeruli is associated with progression of diabetic nephropathy in a genetically obese diabetic rat, Wistar fatty. Diabetes Res. Clin. Pract., 32, 1–9. McCaleb, M.L. and Sredy, J. (1992) Metabolic abnormalities of the hyperglycemic obese Zucker rat. Metabolism, 41, 522–525. Murase, K., Odaka, H., Suzuki, M., Tayuki, N., Ikeda, H. (1998) Pioglitazone time-dependently reduces tumor necrosis factor-α level in muscle and improves metabolic abnormalities in Wistar fatty rats. Diabetologia, 41, 257–264. Noguchi, T., Iritani, N. and Tanaka, T. (1992) Molecular mechanism of induction of key enzymes related to lipogenesis. Proc. Soc. Exp. Biol. Med., 200, 206–209. Noguchi, T., Matsuda, T., Tomari, Y., Yamada, K., Imai, E., Wang, Z., et al. (1993) The regulation of gene expression by insulin is differentially impaired in the liver of the genetically obese-hyperglycemic Wistar fatty rat . FEBS Lett., 328, 145–148. Seino, Y., Yamamoto, T. and Koh, G. (1992) Insulin and glucose transporter gene expression in obesity and diabetes. Proc. Soc. Exp. Biol. Med., 200, 210–213. Sibouta, Y., Chatani. F., Ishimura, Y., Sanada, T., Ohta, M., Inada, Y., et al. (1996) TCV-116 inhibits renal interstitial and glomerular injury in glomerulosclerotic rats. Kidney Int. Suppl., 55, S115–118. Sugiyama, Y., Shimura, Y. and Ikeda, H. (1989) Pathogenesis of hyperglycaemia in genetically obese-hyperglycemic rats, Wistar fatty: presence of hepatic insulin resistance. Endocrinol. Jap. 36, 245–251. Sugiyama, Y., Taketomi, S., Shimura, Y., Ikeda, H. and Fujita, T. (1990a) Effects of pioglitazone on glucose and lipid metabolism in Wistar fatty rats. Arzneim-Forsch/Drug Res., 40, 263– 267. Sugiyama, Y., Shimura, Y. and Ikeda, H. (1990b) Effects of pioglitazone on hepatic and peripheral insulin resistance in Wistar fatty rats. Arzneim-Forsch/Drug Res., 40, 436–440. Suzuki, H., Ikenaga, H., Hayashida, T., Otsuka, K., Kanno, Y., Ohno, Y., et al. (1996) Sodium balance and hypertension in obese and fatty rats. Kidney Int. Suppl., 55, 150–153. Suzuki, M., Kataoka, O., Odaka, H., Yamazaki, H., Hamajo, H., Tsukuda, R., et al. (1994) Inhibitory effects of voglibose (AO-128) on postprandial hyperglycemia and diabetic nephropathy in Wistar fatty rats. Jap. Pharmacol. Ther., 22, 3479–3489. Suzuki, M., Yamada, Y., Yamasaki, H., Anayama, H., Sasaki, S., Odaka, H., et al. (1997) Nephropathy in genetically obese-diabetic Wistar fatty rats—Characterization and prevention. Jpn. Pharmacol. Ther., 25, 363–371. UK Prospective Diabetes Study Group. (1993) X. Urinary albumin excretion over 3 years in diet-treated type 2, (noninsulin dependent) diabetic patients, and association with hypertension, hyperglycemia anf hypertriglyceridemia. Diabetologia, 36, 1021–1029. Velasquez, M.T., Kimmel, P.L., Michaelis, O.E. 4th., Carswell, N., Abraham, A. and Bosch, J.P. (1989) Effect of carbohydraye intake on kidney function and structure in SHR/N-cp rats; a new model of NIDDM. Diabetes, 38, 679–685. Yamamoto, T., Fukumoto, H., Koh, G., Yano, H., Yasuda, K., Masuda, K., et al. (1991) Liver and muscle-fat type glucose transporter gene expression in obese and diabetic rats. Biochem. Biophys. Res. Commun., 175, 995–1002. Yoshimoto, T., Naruse, M., Nishikawa, M., Naruse, K., Tanabe, A., Seki, T., et al. (1997) Antihypertensive and vasculo- and renoprotective effect of pioglitazone in genetically obese diabetic rats. Am.J.Physiol. 272, E989–996.

CHARACTERISTICS OF WISTAR FATTY RAT

151

Zucker, L.M. and Antoniades, H.N. (1972) Insulin and obesity in the Zucker genetically obese rat “fatty”. Endocrinology, 90, 1320–1330.

8. THE NEW ZEALAND OBESE MOUSE: A POLYGENIC MODEL OF TYPE 2 DIABETES SOFIANOS ANDRIKOPOULOS, ANNE W.THORBURN and JOSEPH PROIETTO Department of Medicine, University of Melbourne, Royal Melbourne Hospital, Parkville Victoria 3050 Australia

ABSTRACT The New Zealand Obese (NZO) mouse is an excellent model of Type 2 diabetes because it exhibits classical characteristics of the disease such as insulin resistance and impaired glucosemediated insulin secretion. The polygenic nature of the NZO mouse makes it an attractive model to unravel the mechanisms responsible for these defects since the human condition also appears to be polygenic. We believe that determining the cause of obesity will be the key to understanding the metabolic syndrome of the NZO mouse. KEY WORDS: obesity, insulin secretion, insulin resistance, glycogen synthase, gluconeogenesis, fructose-1, 6-bisphosphatase INTRODUCTION The New Zealand Obese (NZO) mouse is a model of Type 2 diabetes that displays the appropriate characteristics of obesity, impaired glucose-stimulated insulin secretion and insulin resistance in muscle, fat and liver (Proietto and Larkins, 1993). Unlike other animal models of Type 2 diabetes, such as the ob/ob, db/db, Ay mice and the falfa (Zucker) rat which have single-point genetic mutations, the syndrome in the NZO mouse is likely to be the result of defects in multiple genes. In this regard, the NZO mouse is a good model of human Type 2 diabetes.

NZO MOUSE

153

ORIGINS The NZO mouse was developed from a mixed mouse colony at the University of Otago Medical School in Dunedin New Zealand (Bielschowsky and Goodal, 1970). This mouse colony was originally obtained from the Imperial Cancer Research Fund Laboratory, London in 1930. Researchers at the University of Otago noticed three different coat colors in this mouse colony, namely agouti, tan and chocolate and in 1948 began inbreeding animals of the same color. It was noticed that increasing numbers of agouti inbred animals became obese such that between the 12th and 17th generations mice were selectively bred for this characteristic. Subsequently, this gave rise to a new strain of obese, agouti colored mice termed NZO/B1. The other line of mice developed from agouti inbreeding was the New Zealand Black (NZB/ B1). After the 3rd generation some mice had a distinct black color and a pair of these was used to give rise to the NZB/B1 line. Inbreeding of the tan colored mice gave rise to the NZY/B1 line and inbreeding of chocolate colored mice the NZC line. As the NZO mouse is an inbred model, the lack of a suitable control strain has been a problem. Researchers over the years have used a number of lean mice as controls including the C57BL/6, Balb c and the albino ICR mouse. In our laboratory we have used the lean NZC mouse as the control strain since it was derived from the same mixed population of mice as the NZO and has similar metabolic characteristics to other lean control strains. All animals in our laboratory are routinely maintained on regular laboratory chow containing 4.5% fat by weight to prevent effects of diet-induced obesity (Surwit et al., 1988). It was ascertained early on that NZO mice, like many other obese rodents, are poor breeders. So much so, that Bielschowsky and Bielschowsky feared of nearly losing the line (Bielschowsky and Bielschowsky, 1956). This is demonstrated by the fact that only 16 of 50 NZO females have their second litter 19–25 days after the first, compared with 45 of 50 NZC female mice (Bielschowsky and Bielschowsky, 1956). Furthermore, regular 4–5 day cycles are the exception rather than the rule in female mice. Nonetheless, the breeding performance of NZO mice is better than other lines of obese mice. The reproductive status of the female NZO mouse can be improved by treatment with a β3 adrenergic receptor agonist (CL316,243) at a concentration of 0.001% w/w in the diet for a month from weaning to retard weight gain (unpublished observations, Dr Edward Leiter). Diet restriction to reduce body weight may also be effective. METABOLIC CHARACTERISTICS OF NZO MICE NZO mice have birth weights similar to control mice (Crofford and Davis, 1965), but by the age of 4–6 weeks they are already heavier than lean NZC mice and remain heavier throughout their life-span (Table 1) (Andrikopoulos et al., 1996; Crofford and Davis, 1965). This is due in part to early hyperphagia in the NZO mouse (Larkins, 1972), and is probably also contributed to by inactivity (although this has not been formally characterized). In addition to being heavier at 4 weeks of age, NZO mice have increased body lipid composition, as well as elevated plasma free fatty acid concentrations and these differences are maintained into adulthood (Andrikopoulos et al., 1996; Crofford and Davis, 1965). Hyperglycemia and hyperinsulinemia also appear to be early defects in NZO mice being present at 4–6 weeks of age (Veroni et al., 1991). Fed hyperglycemia is more pronounced than in the overnight fasted state at 4 weeks of age, whereas in adulthood, elevated plasma glucose concentrations are found in both the fed Address reprint requests to: Sofianos Andrikopoulos, Ph.D., Department of Medicine, University of Melbourne, Royal Melbourne Hospital, Parkville Victoria 3050 Australia. Tel: +61 3 9344–5478; Fax: +61 3 9347–1863; e-mail: [email protected]

154

S.ANDRIKOPOULOS ET AL.

and fasted states (Table 1). However, wide variations in blood glucose have been reported and not all accounts in the literature show elevated plasma glucose levels for the Table 1 Body weight, plasma glucose, insulin and free fatty acid (ffa) concentrations in fed 4-week and 20 week old NZO and NZC control male mice. 4 week old mice Body Weight (g) Plasma Glucose (mmol/1) Plasma Insulin (mU/1) Plasma FFA (mmol/1)

20 week old mice

NZC

NZO

NZC

NZO

19.0±0.7 16.7±1.3 12.4±0.8 0.53±0.04

25.4±1.0* 22.1±1.0* 25.8±4.6* 0.82±0.06*

27.8±0.4 14.2±0.7 13.0±0.9 1.1±0.1

55.8±1.5† 20.7±1.5† 37.1±5.0† 1.8±0.3*

Results are presented as mean ± SE (n=8–10). *P < 0.05, †P < 0.005 compared with NZC

NZO mouse (Bielschowsky and Bielschowsky, 1956; Upton et al., 1980). This may be partly due to the method of blood sampling and glucose measurement as well as the choice of control mice. Moreover, glycemia in NZO mice has been shown to be weight dependent, with heavier animals displaying higher blood glucose concentrations than lighter mice (Herberg et al., 1970). Leiter and colleagues (1998) from Jackson Laboratories have shown that approximately 60% of male NZO animals developed overt hyperglycemia and a small percentage (approximately 5%) developed severe hyperglycemia in the presence of hypoinsulinemia suggesting that this obese rodent is a threshold model for diabetes. While we have not conducted such a study on the colony maintained at the Walter and Eliza Hall Institute (WEHI), average NZO males were found to be hyperglycemic compared with NZC control mice (Andrikopoulos and Proietto, 1995; Andrikopoulos et al., 1993; Andrikopoulos et al., 1996; Veroni et al., 1991). Hyperglycemia in female NZO mice has not been well characterized. Glucose tolerance has been assessed by i.p. injection of a glucose bolus (2 g/kg). NZO mice had increased basal blood glucose concentrations and their post bolus glucose excursion was also significantly higher compared with C57BL/6 control mice (Larkins, 1971). Basal plasma insulin concentrations were approximately 2-fold higher in NZO mice compared with control mice, whereas insulin secretion in response to glucose was significantly impaired. In contrast, an i.p. bolus of arginine induced an insulin secretory response that was several fold higher than that for control mice (Larkins and Martin, 1972) suggesting selective loss of the glucose recognition system for insulin secretion. Pancreatic islets from NZO mice have been shown to be large and composed mostly of β-cells occupying the whole islet—a peripheral zone formed by α and β cells is often not present (Bielschowsky and Bielschowsky, 1956). This is most likely due to compensation in response to the obese state and hyperglycemic demand. Restricting the diet of NZO mice such that their body weights are not different from control C57BL/6 mice resulted in an improvement in glucose tolerance with a stimulation of insulin secretion in response to a glucose bolus (Larkins, 1973). Thus, the impairment in glucose tolerance of the NZO mouse is at least partly due to its obese state. Recently, it was shown that NZO mice have elevated serum leptin concentrations, a hormone secreted by adipocytes which is capable of inducing satiety. However, NZO mice appear to be insensitive to their hyperleptinemia since peripheral administration of leptin does not cause reduced food intake (Igel et al., 1997). Neuropeptide Y, a hypothalamic signal that causes an increase in food intake was found to be overexpressed in NZO mice, probably as a result of leptin resistance (Rizk et al., 1998). In contrast, intracerebroventricular administration of leptin did result in reduced food intake in NZO mice (Halaas et

NZO MOUSE

155

al., 1997). Furthermore, the function of the NZO variant of the leptin receptor is apparently normal (Igel et al., 1997). These results suggest that the cause of hyperphagia in the NZO mouse could be in the transport of leptin across the blood-brain barrier. Hyperglucagonemia has also been reported in the NZO mouse, which may be an important contributor to their metabolic disturbances (Upton et al., 1980). In contrast, plasma growth hormone levels are not different in NZO compared with C57BL/6 control mice (Larkins, 1971). INSULIN SECRETION STUDIES Although the glucose tolerance tests were indicative of impaired insulin secretion, it was the use of more appropriate procedures such as the i.v. glucose tolerance test and in vitro pancreatic islet cultures that confirmed defective insulin secretion in NZO mice. The following sections summarize in vivo and in vitro studies characterizing impaired insulin release in the NZO mouse. In vivo The first comprehensive study on the rate of insulin secretion in response to various secretagogues administered i.v. was performed by Cameron and colleagues (1974). In response to a 1 g/kg glucose bolus, NZO mice displayed a markedly blunted early phase secretion of insulin, while the late phase of insulin release was depressed compared to randomly bred white control mice. This was later confirmed by Veroni and colleagues (1991) who also reported a defect in both early and late phase insulin secretion in response to an i.v. glucose bolus (0.6 g/kg). Furthermore, Veroni et al. showed that this defect exists early in the progression of the syndrome being present in young 4–5 week old NZO compared with NZC control mice (Figure 1). This suggests that the defect in insulin secretion may be primary, although at this age NZO mice already have increased plasma free fatty acid levels (Table 1). It is possible therefore that these defects in insulin secretion are secondary to perturbations in lipid metabolism. The rate of insulin secretion is also defective in response to other non-glucose secretagogues such as glucagon and aminophylline, whereas it is greatly exaggerated in response to arginine (Cameron et al., 1974). In vitro Basal insulin release from NZO islets cultured in vitro was 5 times higher than from islets of C57BL/6 control mice (Larkins, 1973) corroborating the basal hyperinsulinemia seen in vivo. In response to 8.4 mmol/1 glucose, islets from control mice showed a significant increase in insulin secretion whereas NZO mouse islets did not respond. Furthermore, although higher glucose concentrations (16.7 mmol/1) caused NZO mouse islets to secrete significant amounts of insulin, the fold stimulation was less compared to control islets (2 vs 5) (Larkins et al., 1980). The sulphonylurea tolbutamide elicited a significant response in control whereas it had no effect in NZO mouse islets. In contrast, NZO mouse islets were more sensitive to arginine as an insulin secretagogue than were islets from C57BL/6 mice, confirming the in vivo data (Cameron et al., 1974). Moreover, insulin secretion in response to the glycolytic intermediate glyceraldehyde elicited a large response from NZO mouse islets, at least equal to that of control mouse islets (Larkins et al., 1980). This result suggests that the defect in insulin secretion lies in glucose metabolism, somewhere between glucose transport into the β-cell and the triose-phosphate step, where glyceraldehyde enters the glycolytic pathway. It is therefore possible that defects in GLUT2 or the β-cell glucose sensor glucokinase are responsible for defective insulin secretion in the NZO mouse. Thus, as with

156

S.ANDRIKOPOULOS ET AL.

Figure 1 Intravenous glucose tolerance tests in 4 week old NZO (■, n=8) and NZC ( , n=6) control mice. Results are expressed as percent change above basal. * P<0.05 (reproduced with permission from (Veroni et al., 1991)).

human Type 2 diabetes, the NZO mouse displays impaired glucose-mediated insulin secretion and in this regard is a good animal model in determining the molecular mechanisms of this defect. INSULIN ACTION The first experiment to show insulin resistance in the NZO mouse was reported with the earliest characterization of the obesity syndrome in 1953 and performed with the relatively crude method of the insulin tolerance test (Bielschowsky and Bielschowsky, 1953). This test showed that NZO mice tolerated an i.p. injection of 2–8 units/kg of insulin whereas injections of 0.4 units/kg in NZC control mice invariably proved to be fatal (Bielschowsky and Bielschowsky, 1956). This was confirmed by Crawford and Davis Jr. (1965) who, using the insulin tolerance test, showed that NZO mice were 4–5 times more insulin resistant than randomly bred albino mice. Although this test showed that NZO mice were insulin resistant, it did not describe the site of resistance (e.g. hepatic or peripheral) nor did it give any information on the intracellular perturbations responsible for the insulin resistance. Muscle and Fat Insulin Resistance In 1991 we adapted the euglycemic/hyperinsulinemic clamp technique to study glucose appearance and glucose disappearance in NZO and NZC control mice (Veroni et al., 1991). Animals were anesthetized and a right jugular vein and left carotid artery catheter were inserted for infusion and blood sampling respectively. A tracheostomy was also performed to prevent upper airway obstruction. An infusion of 6-3Hglucose (0.14 µCi/min) with insulin (3 mU/kg.min) was administered through the jugular vein. Blood glucose concentrations were monitored every 10 min and maintained at basal levels by infusion of a 5% glucose solution through a Y-junction connected to the catheter in the jugular vein. When steady-state blood glucose concentrations were obtained (after 60–90 min) three blood samples 5 min apart were obtained from

NZO MOUSE

157

Figure 2 Active glycogen synthase activity in muscle of 1 day (n=21–25) and 20 week (n=9–10) NZO and NZC control mice. *P<0.02 (reproduced with permission from (Thorburn et al., 1995)).

the carotid artery for analysis. Red blood cells were resuspended in heparinized normal saline and reinfused into the mice to prevent anemic shock. To our knowledge this was the first report of an euglycemic/ hyperinsulinemic clamp in mice. Furthermore, we combined this with the infusion of 1-l4C-2-deoxyglucose that is transported into cells, phosphorylated and not metabolized further which provides a measure of the rate of glucose uptake in tissue. Thus, the euglycemic/hyperinsulinemic clamp provides information on rates of glucose disappearance, endogenous glucose production and, when combined with 2-deoxyglucose tracer, glucose uptake in individual tissues. It is a more sensitive technique than the insulin tolerance test, since it avoids hypoglycemia invoked counterregulation and the site of insulin resistance can be identified (liver or muscle/ fat) under conditions that control for plasma glucose and insulin concentrations. In the basal state, glucose uptake in white adipose tissue, heart, diaphragm white quadriceps and white gastrocnemius was increased in NZO mice whereas there was no difference in brown adipose tissue, soleus, red quadriceps and red gastrocnemius muscle in 20 week old mice. This is similar to what has been described in patients with Type 2 diabetes, i.e. increased post-absorptive glucose disposal probably as a result of the mass action effect of glucose (Yki-Jarvinen, 1990). When insulin was infused most tissues assayed from NZO mice displayed a defect in glucose uptake. There was no stimulation, over basal, in white adipose tissue whereas brown adipose tissue, soleus, red and white quadriceps and white gastrocnemius showed a smaller increment compared to NZC control mice. In contrast, at 4 weeks of age white adipose tissue and white quadriceps responded normally, while other tissues of NZO mice showed impaired sensitivity to insulin. This suggests progressive deterioration of peripheral insulin resistance in the NZO mouse. To determine whether the defect in glucose uptake in NZO mice was due to a decrease in glucose transport, we measured the level of the insulin-stimulatable glucose transporter GLUT4 by immunoblotting (Ferreras et al., 1994). At 20 weeks of age, all tissues that showed a decrease in 2-deoxyglucose uptake also had decreased levels of GLUT4, except for soleus muscle. At 4 weeks of age, there was a decrease in brown

158

S.ANDRIKOPOULOS ET AL.

adipose tissue GLUT4 but no difference in other tissues tested supporting the idea that impaired insulinstimulated glucose uptake in peripheral tissues is secondary to other metabolic perturbations in the NZO mouse. In in vitro experiments, 2-deoxyglucose uptake in isolated soleus muscle was defective at all insulin concentrations tested in NZO mice compared to lean Balb C mice, as was glucose utilization and glycogen synthesis (Veroni and Larkins, 1986). This suggests that although total GLUT4 levels were not decreased in NZO mouse soleus muscle, there may be a defect in transporter intrinsic activity or other distal step in glucose metabolism. We also determined the activity of glycogen synthase and phosphorylase in quadriceps muscle of 1-day and 20 week old NZO and lean NZC mice. Active glycogen synthase activity was decreased in NZO mice compared with NZC mice at both ages tested, whereas there was no difference in glycogen phosphorylase activity (Figure 2) (Thorburn et al., 1995). A defect in muscle glycogen synthesis has previously been reported by Stauffacher and Renold (1969) who showed that diaphragm and adipose tissue from NZO mice incorporated less glucose into glycogen, in response to the same insulin levels, than lean albino Swiss mice. Defects is muscle glycogen synthase have been described in patients with Type 2 diabetes and these have been proposed to be of primary importance to the syndrome (Groop et al., 1993). Adipose tissue insulin resistance was also characterized in a series of somewhat contradictory experiments in the 1970s. Adipocytes isolated from NZO mice were shown to have a markedly reduced rate of lipolysis (measured by glycerol release) in response to 0.1 µg/ml isoprenaline compared with adipocytes from NZY control mice (Lovell-Smith and Sneyd, 1973). In contrast, the opposite response was observed in response to 0.55 µmol/L epinephrine (Upton et al., 1979). This is most likely due to the fact that isoprenaline is a specific β adrenergic agonist, whereas epinephrine is an α as well as an β adrenergic agonist. Defective lipolysis was present only in young mice suggesting that this may be a primary cause of the development of obesity in NZO mice. This defect in lipolysis was attributed to diminished cellular cAMP concentrations, as a result of increased phosphodiesterase activity (Lovell-Smith and Sneyd, 1974). A more comprehensive study of adipocyte insulin resistance was conducted by Macaulay and Larkins (1988) showing decreased insulin-stimulated glucose transport and utilization in NZO compared to lean NZC mice. The key glucose oxidation enzyme, pyruvate dehydrogenase was also found to be unresponsive to the stimulatory effects of insulin in NZO mice. Importantly, this study determined that an intracellular mediator, identified as an inositol-containing membrane glycophospholipid, produced by insulin in control NZC mice, whereas insulin failed to stimulate production of this mediator in NZO adipocytes. Thus, muscle and fat insulin resistance in the NZO mouse seems to be due to defects in multiple sites, including a decrease in the expression of GLUT4, decreased activity of key glucose metabolism enzymes such as glycogen synthase and pyruvate dehydrogenase, and the production of insulin mediators. Hepatic Insulin Resistance The euglycemic/hyperinsulinemic clamp showed that NZO mice had increased endogenous glucose production compared to NZC mice in the basal state at both 4 and 20 weeks of age. Furthermore, insulin significantly inhibited endogenous glucose production in NZC control mice at both ages tested, whereas it did not have this effect in NZO mice (Veroni et al., 1991). To determine the substrates responsible for increased endogenous glucose production in the NZO mouse we examined the in vivo rate of alanine and glycerol gluconeogenesis. The rate of conversion of both these substrates to glucose was increased in NZO mice compared to lean NZC control mice (Andrikopoulos and Proietto, 1995). In the case of glycerol gluconeogenesis, this was due to both an increase in substrate availability as well as an intrahepatic mechanism (Figure 3), whereas with alanine gluconeogenesis this was largely due to an increased rate of an

NZO MOUSE

159

Figure 3 The rate of glycerol conversion to glucose in NZO (□) and NZC (■) control mice (n=7). *P<0.005 (reproduced with permission from (Andrikopoulos and Proietto, 1995)).

intrahepatic mechanism (Andrikopoulos and Proietto, 1995). Increased gluconeogenesis from alanine and glycerol has been described in patients with Type 2 diabetes (Nurjhan et al., 1992; Puhakainen et al., 1992). An earlier report (Rudorff et al., 1970) showed that basal alanine gluconeogenesis was similar between NZO and control white mice and whereas insulin inhibited gluconeogenesis in control mice it had no effect in NZO mice. The discrepancy in basal alanine gluconeogenesis with our results may be due to the fact that we performed the studies in vivo such that the hepatic glucose and insulin concentrations and neural input were not disturbed. Rudorff et al. (1970) used the liver perfusion technique in which the animals were sacrificed prior to the experiment and the concentrations of hormones and metabolites in the perfusion medium may not be physiologically relevant. To identify the intrahepatic mechanisms responsible for accelerated alanine and glycerol gluconeogenesis we measured the activity of key regulatory enzymes in the gluconeogenic and glycolytic pathways (Andrikopoulos et al., 1993). The in vitro measurement of enzyme activity may not a reliable indicator of cellular activity since it can be affected by the homogenization and centrifugation procedures. Furthermore the enzyme is diluted many fold by the time a reaction rate is determined and the concentrations of

160

S.ANDRIKOPOULOS ET AL.

substrates and cofactors in the test tube may not reflect those present within the cell. However, in combination with other in vivo measurements we believe that determination of enzyme activity is a powerful tool of identifying biochemical mechanisms. The activity of the key hepatic glycolytic enzymes glucokinase and pyruvate kinase was enhanced in NZO mice compared with NZC control mice. This is an appropriate response to the hyperinsulinemia. The activity of the hepatic gluconeogenic enzyme phosphoenolpyruvate carboxykinase (PEPCK) and glucose-6-phosphatase was depressed in NZO compared with NZC control mice. Moreover, since PEPCK is primarily regulated at the transcriptional level, we determined PEPCK mRNA levels, and found that these were also diminished in NZO mice. This decreased PEPCK activity is surprising since it has been shown that this enzyme is rate limiting (Rongstad, 1979) and transgenic animals overexpressing PEPCK display hepatic insulin resistance and elevated plasma glucose concentrations (Rosella et al, 1995; Valera et al, 1994). The reduced PEPCK activity and mRNA levels is the appropriate response to the hyperinsulinemic environment in the NZO mouse. In contrast, the activity of the gluconeogenic enzymes pyruvate carboxylase and fructose-1, 6-bisphosphatase was increased in NZO mice. The increase in pyruvate carboxylase activity is not surprising since this enzyme is allosterically activated by free fatty acids and we have shown that plasma free fatty acids are increased in NZO mice (Table 1) (Andrikopoulos et al., 1996). The increase in pyruvate carboxylase activity is probably responsible for the increased rate of alanine conversion to glucose in NZO mice. The increase in fructose-1, 6-bisphosphatase activity was surprising and was investigated further, since it explained the accelerated rate of glycerol conversion to glucose. Furthermore, elevated fructose-1, 6-bisphosphatase activity has been proposed to account for increased glycerol gluconeogenesis in patients with Type 2 diabetes (Nurjhan et al., 1992). This is supported by an early report showing increased fructose-1, 6-bisphosphatase activity in liver biopsied from patients with Type 2 diabetes (Willms et al., 1970). Western and Northern blot experiments revealed that while fructose-1, 6-bisphosphatase protein levels were increased in NZO mice, mRNA levels were similar to NZC control mice (Andrikopoulos et al., 1996). Furthermore, the activity and protein levels of this enzyme was decreased in 1-day old NZO compared to NZC mice suggesting that the defect is acquired in response to the disturbances in lipid metabolism. This is corroborated by experiments showing that feeding NZC control mice a high fat diet for 12 days resulted in increased activity and protein levels of hepatic fructose-1, 6-bisphosphatase (Andrikopoulos and Proietto, 1995). We have recently confirmed this finding in high-fat fed rats (SA, JP unpublished observations). Liver glycogen levels were found to be reduced in 1-day old, whereas they were significantly increased in adult 20 week old NZO mice compared to NZO control mice (Thorburn et al, 1995). This is probably due to increased gluconeogenesis and substrate flux through gluconeogenesis i.e. enhanced glycogen synthesis via the “indirect pathway”. Thus, we propose that hepatic glucose overproduction in the NZO mouse is secondary to defects in lipid metabolism. IMMUNE ABNORMALITIES Antibodies against the insulin receptor (Harrison and Itin, 1979), as well as against native DNA and denatured, single-stranded DNA (Melez et al., 1980) have been described in NZO mice. However, we do not believe that immune abnormalities are responsible for the obesity, insulin resistance and impaired insulin secretion in these mice for the following two reasons. Firstly, antibody levels are higher in female NZO mice and the levels decline after 6 months of age to normal levels (Melez et al., 1980). Secondly, while glucose tolerance is impaired in NZO mice, it is normal in the related NZB/W strain despite severe immunological abnormalities in these mice (Upton, 1984).

NZO MOUSE

161

CONCLUDING REMARKS The NZO mouse displays the classic features of Type 2 diabetes of obesity, impaired glucose-mediated insulin secretion and peripheral as well as hepatic insulin resistance. These abnormalities in combination with the polygenic nature of the NZO mouse make it a suitable model for the study of obesity-related diabetes. Despite years of study, the cause of the syndrome in this mouse model is not known. We believe that the abovementioned disturbances in glucose metabolism are secondary to obesity and defects in lipid metabolism. With advances in technology and knowledge of the cellular pathways that control feeding in mice, we foresee that discovering the cause of obesity in the NZO mouse is not far away. The clues that we gain from this obese model will likely bring us closer to determining the cause of human Type 2 diabetes. NZO mice are commercially available from the Jackson Laboratory, Bar Harbor, Maine 04609 USA, the Walter and Eliza Hall Institute Parkville Victoria 3050 Australia and Bomholtgard, Aarhus, Denmark. ACKNOWLEDGEMENTS This work has been supported by a program grant from the National Health and Medical Research Council of Australia and grants-in-aid from Diabetes Australia Research Trust Fund. We would like to thank Dr. Edward Leiter for critical review of this manuscript. REFERENCES Andrikopoulos, S. and Proietto, J. (1995). The biochemical basis of increased hepatic glucose production in a mouse model of type 2 (non-insulin-dependent) diabetes mellitus. Diabetologia, 38, 1389–1396. Andrikopoulos, S., Rosella, G., Gaskin, E., Thorburn, A., Kaczmarczyk, S., Zajac, J.D., et al. (1993). Impaired regulation of hepatic fructose-1, 6-bisphosphatase in the New Zealand Obese mouse model of NIDDM. Diabetes, 42, 1731–1736. Andrikopoulos, S., Rosella, G., Kaczmarczyk, S.J., Zajac, J.D. and Proietto, J. (1996). Impaired regulation of hepatic fructose-1, 6-bisphosphatase in the New Zealand Obese mouse: an acquired defect. Metabolism, 45, 622–626. Bielschowsky, M. and Bielschowsky, F. (1953). A new strain of mice with hereditary obesity. Proc. Univ. Otago Med. School, 31, 29–31. Bielschowsky, M. and Bielschowsky, F. (1956). The New Zealand strain of obese mice. Their response to stilboestrol and to insulin. Austral. J. Exp. Biol., 34, 181–1898. Bielschowsky, M. and Goodal, C.M. (1970). Origin of inbred NZ mouse strains. Cancer Res., 30, 834–836. Cameron, D.P., Opat, F. and Insch, S. (1974). Studies of immunoreactive insulin secretion in NZO mice in vivo. Diabetologia, 10, 649–654, Crofford, O.B. and Davis, C.K.J. (1965). Growth characteristics, glucose tolerance and insulin sensitivity of New Zealand Obese mice. Metabolism, 14, 271–280. Ferreras, L., Kelada, A.S.M.K., McCoy, M. and Proietto, J. (1994). Early decrease in GLUT4 protein levels in brown adipose tissue of New Zealand obese mice. Int. J. Obesity, 18, 760– 765. Groop, L.C., Kankuri, M., Schalin-Janti, C., Ekstrand, A., Nikula-Ijas, P., Widen, E., et al. (1993). Association between polymorphism of the glycogen synthase gene and non-insulin-dependent diabetes mellitus. N. Engl. J. Med., 328, 10–14. Halaas, J.L., Boozer, C., Blair-West, J., Fidahuser, N., Denton, D.A. and Friedman, J.M. (1997). Physiological response to long-term peripheral and central leptin infusion in lean and obese mice. Proc. Natl. Acad. Sci. USA, 94, 8878–8883. Harrison, L.C. and Itin, A. (1979). A possible mechanism for insulin resistance and hyperglycaemia in NZO mice. Nature, 279, 334–336.

162

S.ANDRIKOPOULOS ET AL.

Herberg, L., Major, E., Hennigs, U., Gruneklee, D., Freytag, G. and Gries, F.A. (1970). Differences in the development of the obese-hyperglycemic syndrome in obob and NZO mice. Diabetologia, 6 292–299. Igel, M., Becker, W., Herberg, L. and Joost, H.G. (1997). Hyperleptinemia, leptin resistance, and polymorphic leptin receptor in the New Zealand Obese mouse. Endocrinology, 138, 4234–4239. Larkins, R.G. (1973). Defective insulin secretion in the N.Z.O. mouse: in vitro studies. Endocrinology, 93, 1052–1056. Larkins, R.G. (1973). Defective insulin secretory response to glucose in the New Zealand Obese mouse. Improvement with restricted diet. Diabetes, 22 251–255. Larkins, R.G. (1972). Endocrine abnormalities in the NZO mouse. In Department of Medicine, Royal Melbourne Hospital, University of Melbourne. Larkins, R.G. (1971). Plasma growth hormone in the New Zealand obese mouse. Diabetologia, 7, 302–307. Larkins, R.G. and Martin, F.I.R. (1972). Selective defect in insulin release in one form of spontaneous laboratory diabetes. Nature, 235, 86–88. Larkins, R.G., Simeonova, L. and Veroni, M.C. (1980). Glucose utilization in relation to insulin secretion in NZO and C57B1 mouse islets. Endocrinology, 107, 1634–1638. Leiter, E.H., Reifsnuder, P.C., Flurkey, K., Partke, H.J., Junger, E. and Herberg, L. (1998). NIDDM genes in mice. Deleterious synergism by both parental genomes contributes to diabetogenic thresholds. Diabetes, 47, 1287–1295. Lovell-Smith, C.J. and Sneyd, J.G.T. (1973). Lipolysis and adenosine 39, 59-cyclic monophosphate in adipose tissue of the New Zealand Obese mouse. J Endocr, 56, 1–11. Lovell-Smith, C.J. and Sneyd, J.G.T. (1974). Lipolysis and adenosine 39, 59-cyclic monophosphate in adipose tissue of the New Zealand Obese mouse; the activities of adipose tissue adenyl cyclase and phosphodiesterase. Diabetologia, 10, 655–659. Macaulay, S.L. and Larkins, R.G. (1988). Impaired insulin action in adipocytes of New Zealand Obese mice: a role for postbinding defects in pyruvate dehydrogenase and insulin mediator activity. Metabolism, 37, 958–965. Melez, K.A., Harrison, L.C., Gilliam, J.N. and Steinberg, A.D. (1980). Diabetes is associated with autoimmunity in the New Zealand obese (NZO) mouse. Diabetes, 29, 835–840. Nurjhan, N., Consoli, A. and Gerich, J. (1992). Increased lipolysis and its consequence on gluconeogenesis in noninsulin-dependent diabetes mellitus. J. Clin. Invest., 89, 169–175. Proietto, J. and Larkins, R.G. (1993). A perspective on the New Zealand Obese mouse. In Lessons from Animal Diabetes, edited by E.Shafrir, 4, 65–74. London: Smith-Gordon. Puhakainen, I., Koivisto, V.A. and Uki-Jarvinen, H. (1992). Lipolysis and gluconeogenesis from glycerol are increased in patients with noninsulin-dependent diabetes mellitus. J. Clin. Endocrinol. Metab., 75, 789–794. Rizk, N.M., Liu, L.S. and Eckel, J. (1998). Hypothalamic expression of neuropeptide-Y in the New Zealand obese mouse. Int. J. Obes., 22, 1172–1177. Rongstad, R. (1979). Rate-limiting steps in metabolic pathways. J.Biol. Chem., 254, 1875– 1878. Rosella, G., Zajac, J.D., Baker, L., Kaczmarczyk, S.J., Andrikopoulos, S., Adams, T.E., et al. (1995). Impaired glucose tolerance and increased weight gain in transgenic rats overexpressing a non-insulin-responsive phosphoenolpyruvate carboxykinase gene. Mol. Endocrinol., 9, 1396–1404. Rudorff, K.H., Huchzermeyer, H., Windeck, R. and Staib, W. (1970). Uber den EinfluB von Insulin auf die Alaningluconeogenese in der isoliert perfundierten Leber von New Zealand Obese mice. Eur. J. Biochem., 16, 481–486. Stauffacher, W. and Renold, A.E. (1969). Effect of insulin in vivo on diaphragm and adipose tissue of obese mice. Am. J. Physiol., 216, 98–105. Surwit, R.S., Kuhn, C.M., Cochrane, C., McCubbin, J.A. and Feinglos, M.N. (1988). Diet-induced type II diabetes in C57BL/6J mice. Diabetes, 37, 1163–1167. Thorburn, A., Andrikopoulos, S. and Proietto, J. (1995). Defects in liver and muscle glycogen metabolism in neonatal and adult New Zealand Obese mice. Metabolism, 44, 1298–1302. Upton, J.D. (1984). Intravenous glucose tolerance tests in the New Zealand strains of mice. Horm. Metab. Res., 16, 290–292.

NZO MOUSE

163

Upton, J.D., Sneyd, J.G.T. and Livesey, J. (1980). Blood glucose, plasma insulin and plasma glucagon in NZO mice. Horm. Metab. Res., 12, 173–174. Upton, J.D., Sneyd, J.G.T. and Rennie, P.I.C. (1979). Insulin resistance in the New Zealand Obese mouse (NZO): lipolysis and lipogenesis in isolated adipocytes. Arch. Biochem. Biophys., 197, 139–148. Valera, A., Pujol, A., Pelegrin, M. and Bosch, F. (1994). Transgenic mice overexpressing phosphoenolpyruvate carboxykinase develop non-insulin-dependent diabetes mellitus. Proc. Natl. Acad. Sci. USA, 91, 9151–9154. Veroni, M.C. and Larkins, R.G. (1986). Evolution of insulin resistance in isolated soleus muscle of the NZO mouse. Horm. Metab. Res., 18, 299–302. Veroni, M.C., Proietto, J. and Larkins, R.G. (1991). Evolution of insulin resistance in New Zealand Obese mice. Diabetes, 40, 1480–1487. Willms, B., Ben-Ami, P. and Soling, H.D. (1970). Hepatic enzyme activities of glycolysis and gluconeogenesis in diabetes of man and laboratory animals. Horm. Metab. Res., 2, 135– 141. Yki-Jarvinen, H. (1990). Acute and chronic effects of hyperglycaemia on glucose metabolism. Diabetologia, 33, 579–585.

9. THE NSY MOUSE: AN ANIMAL MODEL OF HUMAN TYPE 2 DIABETES MELLITUS WITH POLYGENIC INHERITANCE HIRONORI UEDA, HIROSHI IKEGAMI, MASAO SHIBATA* and TOSHIO OGIHARA Department of Geriatric Medicine, Osaka University Medical School, 2–2 Yamadaoka, Suita, Osaka 565, Japan. Tel: +81–6–6879–3852, Fax: +81–6–6879–3859 and * Department of Health, Aichi-Gakuin University, College of General Education Iwasaki, Nishincho, Aichi-gun, Aichi 470–01, Japan

ABSTRACT The Nagoya-Shibata-Yasuda (NSY) mouse strain was established as an inbred animal model with spontaneous development of diabetes, by selective breeding for glucose intolerance from outbred Jcl: ICR mice. NSY mice closely mimic human Type 2 diabetes in that the onset is agedependent, the animals are not severely obese, and both insulin resistance and impaired insulin response to glucose contribute to disease development. The cumulative incidence of diabetes reached 98% in male NSY mice and 31% in female NSY mice at 48 weeks of age. Type 2 diabetes in NSY mice is transmitted to F1 hybrids in an autosomal dominant manner. Inheritance of diabetes in NSY mice is polygenic, and four major loci (Nidd1nsy, Nidd2my, Nidd3my and Nidd4my) are mapped on mouse chromosomes (Chr) 11, 14, 6 and 11, respectively. Nidd1nsy and Nidd4nsy appear to affect insulin secretion, whereas Nidd2my and Nidd3nsy appear to affect insulin sensitivity. A locus on Chr 6 was significantly linked to epididymal fat weight. The NSY mouse will serve as a useful model for studies on the etiology of late-onset type 2 diabetes with polygenic inheritance in humans. INTRODUCTION The Nagoya-Shibata-Yasuda (NSY) mouse strain was established in 1980 as an inbred polygenic animal model with spontaneous development of Type 2 (non-insulin-dependent) diabetes mellitus, by selective breeding for glucose intolerance from outbred Jcl: ICR (Shibata et al., 1980). The NSY mouse closely mimics human Type 2 diabetes in that the onset is age-dependent, the animals are not severely obese, and

THE NSY MOUSE

165

Figure 1 Genealogy of NSY (Nagoya-Shibata-Yasuda), CTS (cataracta Shionogi), NOD (nonobese diabetic), and NON (nonobese nondiabetic) mice.

both insulin resistance and impaired insulin response to glucose contribute to disease development (Ueda et al., 1995). We summarize the pathogenesis and genetic features of Type 2 diabetes in this model. ORIGIN The NSY (Nagoya-Shibata-Yasuda) mouse was established as the descendants of streptozotocin-induced diabetic mice from outbred Jcl: ICR from which NOD and NON mice were also derived (Figure 1). The NOD mouse is a well-known animal model of autoimmune Type 1 diabetes mellitus (Makino et al., 1980). NON mice are reported to show impaired glucose tolerance without inflammatory changes in pancreatic islets (Makino et al., 1992). Diabetes was induced by administration of streptozotocin (STZ) in 5 (1 male, 4 female) out bred Jcl: ICR mice in 1971. A STZ-induced diabetic Jcl: ICR male mouse were mated with 4 STZ-induced diabetic Jcl: ICR female mice to obtain F1 mice. The Fl mice with impaired glucose tolerance were mated with siblings. In each generation, i.p. glucose tolerance test (i.p.GTT) was performed and the same process was repeated over several generations. Area under the glucose curve (AUC) was calculated according to the trapezoid rule from the glucose measurements at baseline (0 min), 60 and 120 min after glucose load in each generation (Figure 2) (Shibata et al., 1980). Glucose tolerance of these mice was almost normal in F2-F5 generations. From F6 generation, glucose tolerance of these mice was markedly impaired. From F7 generation, NSY mice were consistently hyperglycemic. Administration of STZ in F0 mice is unlikely to have affected the glucose tolerance in the next generations; rather, mice with a predisposition to glucose intolerance were selected by administration of STZ, and subsequent selection of mice with glucose intolerance resulted in the establishment of NSY mice. Three pairs of NSY mice (F36) were originally obtained from the Branch Hospital of Nagoya University School of Medicine in 1992, and the colony of NSY mice was maintained in the animal facilities of Osaka University Medical School by brother-sister mating with selective breeding for glucose intolerance. The colony in Osaka University Medical School (NSY/osk) is at F53 generation as of 1999.

Address correspondence to: Hironori Ueda, Department of Medical Genetics, Wellcome Trust Centre for the Study of Molecular Mechanisms in Disease, Cambridge Institute for Medical Research, Addenbrooke’s Hospital, Cambridge, CB2 2XY. Tel: 44 (0) 1223 762105; Fax: 44 (0) 1223 762102; E-mail: [email protected]

166

H.UEDA ET AL.

Figure 2 Blood glucose AUC during i.p. GTT in each generation of NSY mice. Values are mean ± SEM. Values in parentheses indicate numbers of mice.

PHENOTYPIC CHARACTERISTICS Incidence of Type 2 Diabetes At 4 weeks of age, the incidence of diabetes as assessed by 2-hour plasma glucose on i.p. GTT in NSY mice was 0% both in males (0/14) and females (0/17). NSY mice spontaneously developed Type 2 diabetes in an age-dependent manner. The cumulative incidence of diabetes reached 98% (49/50) in male NSY mice and 31% (9/29) in female NSY mice at 48 weeks of age (Figure 3). A gender difference in the incidence of diabetes was observed, with a higher incidence in males than in females, as in the case of other models of Type 2 diabetes. Pathogenesis of Type 2 Diabetes Growth curve of NSY mice Body weight of male NSY mice was moderately greater than that of female NSY mice (Figure 4). Severe obesity was not observed at any age in this strain. Body weight (g) of male NSY mice at 48 weeks of age was 48.9±0.7 (n=40), which is similar to that of male Jcl: ICR mice at the same age, 48.3±0.9 (n=50) (data from Clea Japan Inc., Tokyo, Japan). Therefore, unlike oblob, db/db and KKAy mice, which are characterized by severe obesity and glucose intolerance, NSY strain is an animal model of Type 2 diabetes without severe obesity.

THE NSY MOUSE

167

Figure 3 Cumulative incidence of diabetes mellitus among male (closed circles) and female (open circles) NSY mice at 4 to 48 weeks of age. Fourteen to 50 mice were examined at various ages.

Insulin secretion Figure 5 shows the blood glucose and insulin concentrations during i.p. GTT in male NSY and control C3H/ He mice at 8, 24 and 36 weeks of age. Fasting blood glucose concentration and blood glucose concentrations after injection of glucose were comparable in the two strains at 8 weeks of age. Blood glucose concentrations after injection of glucose were significantly higher in NSY mice than those in C3H/ He mice at 24 and 36 weeks of age. Insulin response to glucose was comparable in the two strains at 8 weeks of age. At 24 and 36 weeks of age, however, fasting insulin level was increased and glucosestimulated insulin secretion was markedly impaired in NSY mice. Glucose-stimulated insulin secretion was also impaired in NSY mice in in vitro studies using isolated islets in NSY mice (Ueda et al., 1995). Furthermore, insulin secretion was impaired in pancreatic islets in response not only to glucose but also L-arginine, glibenclamide and BayK8644 (voltage-dependent Ca channel opener) (Hamada et al., unpublished). In contrast, the pancreatic insulin content of NSY mice was maintained up to 48 weeks of age. These observations suggest that the islets of NSY mice retain the ability to synthesize insulin, but insulin response to glucose is impaired. In addition, fasting and non-fasting hyperinsulinemia was observed in NSY mice. It is possible that the regulatory pathway rather than the constitutive pathway in the pancreatic β-cells of NSY mice is impaired. Morphologically, neither islet hypertrophy nor lymphocytic infiltration into or around islets was observed in NSY mice (Ueda et al., 1995). These findings indicate that diabetes in NSY mice is not caused by autoimmunity, but that functional changes in insulin secretion from pancreatic β-cells in response to glucose may contribute to the development of Type 2 diabetes in the NSY mouse.

168

H.UEDA ET AL.

Figure 4 Body weight changes in male NSY (closed circles) and female NSY (open circles) mice from 4 to 48 weeks of age. n=4–5 in each group. Values are mean±SEM. *p<0.05 **p<0.01 ***p<0.01 vs. female NSY mice.

Insulin action Body weight of male NSY mice was moderately larger than that of male C3H/He mice after 6 weeks of age. Body mass index (BMI) and weight of epididymal fat pads were significantly higher in NSY mice than those in C3H/He mice (Ueda et al., 1995). Fasting plasma insulin of NSY mice was significantly higher than that of C3H/He mice after 24 weeks of age. These data suggest the existence of insulin resistance in NSY mice. Indeed, insulin tolerance test (ITT) revealed that insulin action was significantly impaired in NSY mice (Ueda et al., 2000). These findings implicate the existence of insulin resistance in the NSY mouse, and that insulin resistance in this strain may contribute to the development of Type 2 diabetes in NSY mice. Fatty liver was also observed in male NSY mice after 36 weeks of age. Fatty liver was not observed in female NSY mice at any age (H.Ueda et al., 2000). INHERITANCE OF DIABETES To determine the mode of inheritance of diabetes in NSY mice, female NSY (F41) mice were crossed with male C3H/He mice, and male NSY mice were crossed with female C3H/He mice to obtain NSY x C3H/He F1 (NC3F1) and C3H/He x NSY F1 (C3NF1) mice, respectively. The cumulative incidence of diabetes was 100% (25/25) in male C3NF1 mice and 97% (29/30) in male NC3F1 mice at 48 weeks of age, indicating that diabetes in NSY mice was transmitted to F1 hybrids in an autosomal dominant manner. These data also

THE NSY MOUSE

169

Figure 5 Longitudinal analysis of glucose tolerance and insulin response to glucose during intraperitoneal glucose tolerance test (2 g/kg body weight) in male, overnight-fasted C3H/ He (open circles) and NSY (closed circles) mice from 8 to 36 weeks of age. n=5–6 in each group. Values are mean ± SEM. *p<0.05 **p<0.01 ***p<0.001 vs. male C3H/He mice. + p<0.05 vs. basal.

suggest that X-linked genes, Y-linked genes, mitochondrial genes, or imprinted genes seemed unlikely to be a major genetic component of diabetes in NSY mice. Fatty liver also showed an autosomal dominant mode of inheritance. In contrast, epididymal fat accumulation and impaired insulin secretion showed an autosomal recessive mode of inheritance. BMI showed a co-dominant mode of inheritance. Paternalmaternal effects associated with the severity of diabetes were observed. Insulin resistance in male F1 mice was much Table 1 Four QTLs for glucose intolerance and its related phenotypes. Locus Phenotype Glucose Basal insulin

Nidd1 nsy (Chr11)

Nidd2nsy (Chr14)

Nidd3nsy (Chr6)

Nidd4nsy (Chr11)

9.50 (3.2×10–10) –

4.88 (1.3×10−5) 4.52 (3.0×10–5)

3.43 (3.7×10−4) 4.69 (2.0×10−5)

5.79 (1.6×10–6) –

170

H.UEDA ET AL.

Locus Nidd1 nsy (Chr11)

Nidd2nsy (Chr14)

Nidd3nsy (Chr6)

Nidd4nsy (Chr11)

Weight of epididymal fat pads Insulin/glucose ratio

–

6.75 (1.8×10−7) –

–

–

3.32 (4.7×10–4) –

Body mass index

–

–

–

5.35 (4.5×10–6) –

Highest Lod scores (>2.0) and p values in parentheses calculated by MAPMAKER/QTL are shown. The genome-wide threshold for significance in an F2 intercross is an lod score of 4.3, and for suggestive linkage is an lod score of 2.8.

more severe than that in the parental NSY strain. Severe hypertrophy of pancreatic islets was observed in male F1 mice at 48 weeks of age (Ueda H et al., 2000 and unpublished). SUSCEPTIBILITY LOCI FOR DIABETES MELLITUS By multipoint linkage analysis of glucose tolerance in F2 intercrosses (n=307) with microsatellite markers throughout the genome, three major loci (Nidd1 nsy, Nidd2nsy and Nidd3nsy) were mapped on mouse chromosomes (Chr) 11, 14 and 6, respectively (Table 1). The existence of a fourth locus (Nidd4nsy) with an age-dependent effect was suggested by longitudinal, but not cross sectional, analysis of linkage data (Ueda et at., 1999). Nidd1nsy Strong evidence of linkage was found around the central region of chromosome 11, with MLS (maximum lod score) of 9.50 (nominal P value of 3.2×10−10) for glucose AUC and 8.49 for blood glucose level at 120 min (glucose: 120 min), and the locus was designated Nidd1nsy (for NIDDM locus 1 in NSY mice). Linkage data were analyzed using the MAPMAKER/QTL computer package (Lander et al., 1989 and Paterson et al., 1987). Nidd1nsy showed no significant effect on basal insulin level (Table 1), suggesting that Nidd1nsy affected glucose tolerance probably through impaired insulin secretion rather than insulin resistance. Significant evidence of linkage of Nidd1nsy with glucose tolerance was observed at 12, 24, 36 and 48 weeks of age. The inheritance pattern at Nidd1nsy was consistent with the NSY alleles acting in a dominant or additive, but not recessive, manner to increase blood glucose at an older age. The inheritance pattern at Nidd1nsy acted at a younger age in a recessive or additive, but not dominant, manner, which may reflect the effect of Nidd4nsy on the same chromosome. Nidd2nsy Significant evidence of linkage was also found around a 20cM region of chromosome 14 (Nidd2nsy) with MLS of 4.88 for glucose: 120 min. Nidd2nsy was significantly linked to basal insulin level (MLS: 4.52), which is indicative of insulin resistance. Suggestive evidence of linkage with epididymal fat weight was found at this locus (MLS: 3.32). NSY allele at Nidd2nsy affected glucose tolerance in a recessive or additive, but not dominant manner.

THE NSY MOUSE

171

Nidd3nsy There is suggestive evidence of linkage around a broad central region of chromosome 6 (MLS: 3.43) for glucose: 120 min. Nidd3nsy was significantly linked to basal insulin level (MLS: 4.69), which is also indicative of insulin resistance. Significant evidence of linkage with epididymal fat weight was found in this locus (MLS: 6.75). NSY allele at Nidd3nsy affected glucose tolerance in a dominant manner. Significant evidence of linkage with fatty liver was also found in this locus (Ueda et al., unpublished). These data suggest that this locus may contribute to insulin resistance as well as fat metabolism in this strain. Nidd4nsy Significant evidence of linkage with glucose tolerance was observed at a young age (12 weeks) with MLS of 5.79 for glucose: 120 min in the proximal region of chromosome 11, but no significant linkage was observed at an older age. This additional locus was provisionally designated Nidd4nsy. Nidd4nsy affected glucose tolerance in a recessive or additive manner. Nidd4nsy showed no effect on basal insulin level, as in the case of Nidd1nsy, but unlike Nidd1nsy, significant evidence of linkage with insulin/glucose ratio was observed at this locus with MLS of 5.35. These data suggest that Nidd4nsy contributed to the sensitivity of β-cells to glucose in insulin secretion in the basal state. Summary of Mapped Loci None of the four loci showed significant linkage with body weight or BMI (Table 1). Nidd1nsy, Nidd2nsy and Nidd3nsy explained 29%, 16% and 10%, respectively, of the genetic variance and 14%, 8% and 5% of phenotypic variance. The three loci explained 54% of the genetic variance under a three-locus model constructed by MAPMAKER/QTL program. Furthermore, F2 mice homozygous for NSY alleles at both Nidd1nsy and Nidd2nsy marker loci showed markedly impaired glucose tolerance to a degree similar to that in NSY mice, whereas F2 mice homozygous for C3H alleles at both loci showed similar glucose tolerance to control C3H mice (Ueda et al., 1999). These two loci showed strong effects on glucose intolerance in this strain. A Candidate Gene for Nidd1nsy A potential candidate gene for Nidd1nsy is Tcf2 (Karolyi et al., 1992), encoding hepatic nuclear factor 1β (HNF1β), whose mutation has recently been reported in a family with a rare, early-onset, monogenic subtype of diabetes, MODY, in humans (Horikawa Y et al., 1997 and Nishigori H et al., 1998). Tcf2 gene was expressed in the pancreas and a pancreatic β-cell line, MIN6 (Ueda et al., 1999). Both of the two alternatively spliced forms of Tcf2 transcripts were expressed in the pancreas and a pancreatic β-cell line as well as liver and kidney. The nucleotide sequence of Tcf2 in NSY mouse was found to be allelically variant from that in the C3H mouse, including a nucleotide substitution at codon 222 which led to an amino acid change from hydrophilic threonine to hydrophobic alanine (T222A) in the DNA binding domain (Ueda et al., 1999). Due to the broad Nidd1nsy interval, however, it is too premature to conclude that this sequence variant is Nidd1 nsy itself. Further genetic and functional studies by making congenic strains are necessary to determine whether or not HNF1β is responsible for the effect of Niddlnsy.

172

H.UEDA ET AL.

DIABETIC NEPHROPATHY NSY mice develop glomerular abnormality consisting of thickening of the capillary basement membrane and increase of the mesangial matrix. These changes are basically the same as those observed in human diabetic glomerulosclerosis (Shibata et al., 1983), suggesting that NSY mice could be a model of human diabetic nephropathy. Administration of elastase or lysozome to NSY mice was reported to induce an inhibitiory effect on glomerular capillary basement membrane thickening (Shibata et al., 1981 and Shibata et al., 1986). Possibly, elastase prevented thickening by acting on collagen in the basement membrane, and lysozyme prevented thickening by accelerating the degradation of glycoproteins of the thickened glomerular basement membrane (Shibata et al., 1981 and Shibata et al., 1986). PRACTICAL INFORMATION NSY mice are suitable to study the genetics, pathogenesis and molecular mechanism of Type 2 diabetes mellitus. NSY mice are also useful for the development of new drugs for diabetes treatment, and understanding the mechanism of insulin resistance and insulin secretion. In particular, NSY mice are useful for the genetic dissection of polygenic traits and studies on the influence of environmental factors. The development of diabetes mellitus in NSY mice is late, and fasting plasma glucose and non-fasting plasma glucose concentrations are slightly elevated after 36 weeks of age. Development of diabetes in NSY mice can be accelerated by giving water supplemented with 30% sucrose from 4 weeks of age. Almost all NSY mice given 30% sucrose develop diabetes mellitus by 12 weeks of age (Nojima et al., unpublished data). This method would be useful to test the effect of drugs or other therapies on diabetes in NSY mice. The NSY mouse is an animal model of Type 2 diabetes of polygenic, but not monogenic, inheritance. In particular, two major loci (Nidd1nsy and Nidd2nsy) strongly affect the disease (Ueda et al., 1999). Although these two loci are located on autosomes, sexual dimorphism in the expression of diabetic phenotypes is also observed in this strain, with a higher cumulative incidence of diabetes in males than females. Therefore, male mice should be used for studies on diabetic phenotypes. We used C3H/He strain as a non-diabetic control strain. As a preliminary study, we studied glucose tolerance in several inbred strains of mice, such as BALB/c, DBA/2, C3H/He, and C57BL/6 mice at 24 weeks of age. All these strains showed similar glucose tolerance, except for C57BL/6 mice which were very glucose intolerant as previously reported by others (Kaku K et al., 1988 and Surwit RS et al., 1988). Since C3H/He strain is the second most common strain used as experimental mice after C57BL/6 (Festing MFW et al., 1979), we chose C3H/He mice as the control strain. Jcl: ICR mice, from which NSY mice were derived, may also be used as a non-diabetic control strain. Jcl: lCR mice, however, are outbred and the background of this strain is heterogeneous. Thus, Jcl: ICR mice are not recommended as a non-diabetic control strain, in paticular for genetic studies. Several inbred strains derived from Jcl: ICR mice could be used as a control strain. The average life span of NSY mice is one and a half years. A colony of NSY mice can be maintained normally as in the case of other laboratory strains of mice, and the cost of maintenance is the same as that for other laboratory strains. It is recommended that selective breeding for glucose intolerance is continued to maintain the high incidence of diabetes in each NSY colony. NSY mice are not available from commercial breeders, but they are maintained at Osaka University Medical School, Nagoya University Daiko Medical Center, Clea Japan Inc. and the Institute for Laboratory Animal Research, Nagoya University School of Medicine in Japan. Currently, we are in the process of making NSY mice available from commercial breeders.

THE NSY MOUSE

173

REFERENCES Festing, M.F.W. (1979) Inbreeding and its consequences, and the history of the inbred strains. In Inbred strains in biomedical research, edited by M.F.W.Festing, p. 10. London: The Macmillan Press Ltd. Horikawa, Y., Iwasaki, N., Hara, M., Furuta, H., Hinokio, Y., Cockburn, B.N., et al. (1997) Mutation in hepatocyte nuclear factor-1β gene (TCF2) associated with MODY. Nat. Genet., 17, 384–385. Kaku, K., Fiedorek, F.T. Jr., Province, M. and Permutt, M.A. (1988) Genetic analysis of glucose tolerance in inbred mouse strains . Diabetes, 37 707–713. Karolyi, I.J., Guenet, J.L., Rev-Campos, J., Camper, S.A. (1992) The gene coding for variant hepatic nuclear factor 1 (Tcf-2), maps between the Edp-1 and Erba genes on mouse chromosome 11. Mamm. Genome., 3, 184–185. Lander, E.S., Green, P., Abrahamson, J., Barlow, A., Daly, M.J., Lincoln, S.E., et al. (1987) MAPMAKER: an interactive computer package for constructing primary genetic linkage maps of experimental and natural populations. Genomics, 1, 174–181. Makino, S., Kunimoto, K., Moraoka, Y., Mizushima, Y., Katagiri, K., Tochino, Y. (1980) Breeding of a non-obese, diabetic strain of mice. Exp. Anim., 29, 1–13. Makino, S., Yamashita, H., Kunimoto, K., Tsukahara, K., and Uchida, K. (1992) Breeding of the NON mouse and its genetic characteristics. In Current Concepts of a New Animal Model: The NON mouse, edited by N.Sakamoto, N.Hotta and K.Uchida, pp. 51–59. Tokyo: Elsevier. Nishigori, H., Yamada, S., Kohama, T., Tomura, H., Sho, K., Horikawa, Y., et al. (1998) Frameshift mutation, A263fsinsGG, in the hepatocyte nuclear factor-1β gene associated with diabetes and renal dysfunction. Diabetes, 47, 1354–1355. Paterson, A.H., Lander, E.S., Hewitt, J.D., Peterson, S., Lincoln, S.E., Tanksley, S.D. (1988) Resolution of quantitative traits into Mendelian factors by using a complete linkage map of restriction fragment length polymorphisms. Nature, 335, 721–726. Shibata, M., and Yasuda, B. (1980) New experimental congenital diabetic mice (N.S.Y. mice). Tohoku J. Exp. Med., 130, 139–142. Shibata, M., Kawanishi, A., Kishi, T., Kobayashi, K., Kuno, T., Sasaki, M., et al. (1981) Inhibitory effect of elastase on the glomerular capillary basement membrane thickening of the experimental congenital diabetic mice (N.S.Y. mice). Nagoya J. Med. Sci., 43 113– 115. Shibata, M. (1983) Microangiopathy in diabetic N.S.Y. mice. In Diabetic microangiopathy, edited by H.Abe and M.Hoshi, pp. 457–466. Tokyo: University of Tokyo Press. Shibata, M., Kishi, T., Yasuda, B., and Kuno, T. (1986) The inhibitory effect of lysozyme on the glomerular basement membrane thickening in spontaneous diabetic mice (N.S.Y. mice). Tohoku J. Exp. Med., 149, 39–46. Surwit, R.S., Kuhn, C.M., Cochrane, C, McCubbin, J.A. and Feinglos, M.N. (1988) Diet-induced type II diabetes in C57BL/6J mice. Diabetes, 37, 1163–1167. Ueda, H., Ikegami, H., Yamato, E., Fu, J., Fukuda, M., Shen, G-Q., et al. (1995) The NSY mouse: a new animal model of spontaneous NIDDM with moderate obesity. Diabetologia, 38, 503–508. Ueda, H., Ikegami, H., Kawaguchi, Y., Fujisawa, T., Yamato, E., Shibata, M., et al. (1999) Genetic analysis of lateonset Type 2 diabetes in a mouse model of human complex trait. Diabetes, 48 1168–1174. Ueda, H., Ikegami, H., Kawaguchi, Y., Fujisawa, T., Nojima, K., Babaya, N., et al. (2000) Age-dependent changes in phenotypes and candidate gene analysis in a polygenic animal model of Type II diabetes; NSY mouse. Diabetologia, in press. Ueda, H., Ikegami, H., Kawaguchi, Y., Fujisawa, T., Nojima, K., Babaya, N., et al. (2000) Paternal-maternal effects on phenotypic characteristics in spontaneously diabetic NSY mice. Metabolism, in press.

10. THE GOTO-KAKIZAKI RAT CLAES-GÖRAN ÖSTENSON Department of Molecular Medicine, The Endocrine and Diabetes Unit, Karolinska Institute and Hospital, SE-171 76 STOCKHOLM, Sweden.

KEY WORDS: Type 2 diabetes mellitus; Inbreeding; Glucose intolerance; Pancreatic islet morphology; Insulin secretion; Insulin resistance; Diabetic complications INTRODUCTION The Goto-Kakizaki (GK) rat is a nonobese substrain of Wistar rat origin, developing non-insulin dependent diabetes mellitus early in life. Glucose intolerance is most likely primarily due to impaired B-cell function on the background of a polygenic inheritance. The GK rat may be regarded as one of the best available rodent strains for the study of inherited type 2 diabetes. ORIGIN AND BREEDING OF THE GK RAT In the 1970s, Goto and co-workers in Sendai, Japan, initiated the GK sub-strain by selecting among 211 healthy Wistar rats nine animals of each gender with the highest (normal) blood glucose levels during an oral glucose tolerance test (OGTT) (Goto et al, 1975; Goto et al, 1988; Suzuki et al, 1992). Repeated selection of rats with tendency to enhanced OGTT response for breeding, resulted in clear-cut glucose intolerance after five generations (F6). Since the ninth generation (F8) sister-brother mating was consequently performed (Suzuki et al., 1992), implying that the GK rat today—after at least 70 generations —is an highly inbred substrain. This has been verified by the non-complicated result of inter-animal skin transplantation (unpublished data), as well as microsatellite mapping of the GK rat genome (Galli et al., 1996). In the F8 generation of GK rat, impaired insulin response to glucose was demonstrated in the perfused isolated pancreas (Kimura et al, 1982). Thereafter, glucose intolerance and impairment of glucose-induced

GOTO-KAKIZAKI RAT

175

insulin secretion have been rather constant features of the GK rat, also when bred in colonies outside Japan (Suzuki et al., 1992; Östenson et al., 1996). However, as pointed out below, despite these similarities, other characteristics such as islet cell morphology and islet metabolism seem to differ between some of the GK rat colonies. GK Rat Colonies Until the end of the 1980s, GK rats were bred only in Sendai, Japan. The colonies then started in Paris (Portha et al., 1991) and Stockholm (Östenson et al., 1993a) with breeding pairs from Japan; these colonies are still existing. Other colonies existed only shorter periods during the 1990s in London (Hughes et al., 1994) and Aarhus, Denmark. At present (1999) there are also GK rat colonies in e.g. Tampa FL. (derived from Paris; Villar-Palasi & Farese, 1994), Seattle WA (derived from Tampa; Metz et al., 1999), Cardiff UK (Lewis et al., 1996), Brussels, Belgium (derived from Paris; Sener et al., 1996), and, in addition to Sendai, in several Japanese universities such as Kyoto (Tsuura et al., 1993), Osaka (Yoshida et al., 1996), and Matsumoto (Suzuki et al., 1997). Also, GK rats have been made available commercially by a Danish producer of laboratory animals, M&B A/S (www.m-b.dk). Maintenance and Breeding of the GK Rat; Some Practical Considerations Although originally suggested by the Japanese investigators that GK rats ideally should be fed oriental laboratory chow, it appears that the rat feeds and breeds perfectly well also on conventional Western lab rodent diet and tap water. Experience from the Stockholm GK rat colony, initiated in 1989 with F40 breeding pairs from Sendai, has indicated the importance of regulated environmental properties for optimal breeding. Thus, a humidity of 40–60% and temperature of 25–26°C in the breeding room seems to be advantageous for pregnancy and minimizes the number of stillbirths, which is reported to be greatly increased in GK rats (Malaisse-Lagae et al., 1997). It is likely that a too dry environment may reduce fertility by impairing the tubar ciliar activity. Attempts with insulin treatment of pregnant GK rats as well as foster mothers of Wistar rat strain have not improved the litter size considerably (Östenson, unpublished data). In the Stockholm GK rat colony, the litter size presently ranges between 2 and 12 pups, with a mean of about 8–9. It should be emphasised that all GK rats display diabetic features which furthermore are present at an early age and even in the fetuses (Abdel-Halim et al., 1994; Movassat et al., 1997; Serradas et al., 1998). Consequently, Wistar rats, preferably of out-bred origin, have been used as control animals by most investigators. In most studies, the body weights of GK rats (in the age range of 1 week to 4 months) have been 10–30% lower than those of age-matched control animals (cf. Sener et al., 1993; Hughes et al., 1994; Movassat et al., 1997; Giroix et al., 1999).

Address correspondence to: Claes-Göran Östenson, Department of Molecular Medicine, The Endocrine and Diabetes Unit, Karolinska Institute and Hospital, SE-171 76 STOCKHOLM, Sweden. E-mail: [email protected]; Fax: +46–8-51776280.

176

C.-G.ÖSTENSON

Glucose Tolerance and Plasma Insulin Levels In the original GK rat colony, oral glucose tolerance tests were performed by administering glucose (2g/kg) per os in overnight fasted animals. It has been reported from the Sendai GK rat colony that glucose tolerance continued to decrease until the 35th generation, whereafter it has been maintained at a rather stable level (Suzuki et al., 1992). In other laboratories, glucose tolerance has been assessed after i.p. injection of glucose in solution (2 g/kg) (cf. Östenson et al., 1993a; Abdel-Halim et al., 1994; Hughes et al., 1994) or i.v. (0.5–1.0 g/kg) (Portha et al., 1991; Gauguier et al., 1994). In male GK rats, irrespective of Stockholm or Paris origin, fasting blood glucose levels have been typically 7–9 mM as compared to 3–5 mM in agematched Wistar controls (Abdel-Halim et al., 1994) and in the fed state 10–18 mM and 6–8 mM, respectively (Sener et al., 1993; Ling et al., 1998; Giroix et al., 1999). In female GK rats, somewhat lower blood glucose concentrations have been noted (unpublished). Non-fasting plasma insulin levels in GK rats from all colonies have been similar or somewhat increased as compared to age-matched control Wistar rats (Portha et al., 1991; Suzuki et al., 1992; Östenson et al., 1993b; Sener et al., 1993; Tsuura et al., 1993; Okamoto et al., 1995 ; Movassat et al., 1997; Salehi et al., 1999; Metz et al., 1999), while fasting plasma insulin levels have been lower relative to control animals (Galli et al., 1996). Fetal plasma insulin concentrations have been reported significantly lower in GK rats from the Paris colony compared to control fetuses (Serradas et al., 1995). GENETIC CONSIDERATIONS The hereditary nature of diabetes in the GK rat is obvious from the way the strain was produced (Goto et al., 1975; Goto et al., 1988; Suzuki et al., 1992). Studies on the offspring in crosses between GK and Wistar rats demonstrated that these F1 hybrid rats, regardless whether the mother was a GK or a Wistar rat, exhibit glucose intolerance and glucose-induced insulin secretion intermediate to their parents (Abdel-Halim et al., 1994; Gauguier et al., 1994; Serradas et al., 1998). These findings indicate that conjunction of GK genes from both parents is required for defective insulin response and glucose tolerance to be fully expressed. They furthermore lead to the conclusion that maternal hyperglycemia in utero plays no or minimal role in the development of defects in the offspring. In this context it is noteworthy that no mitochondrial DNA deletion or polymorphism, which may be reponsible for maternal inheritance, was detected in GK rats (Gauguier et al., 1994; Serradas et al., 1995). However, mitochondrial DNA content was decreased in pancreatic islets of adult, but not fetal, GK rats, probably reflecting the effect of metabolic dysfunction in the rat. Repeated backcrossing of F1 hybrids of GK and Wistar rats with Wistar rats successively improved glucose tolerance, suggesting that several additive genes are responsible for glucose intolerance in the GK rat (Abdel-Halim et al., 1994). This observation was strongly supported by two genetic studies employing linkage analysis (Galli et al., 1996; Gauguier et al., 1996). Thus with a combination of phenotypic and genotypic studies on F2 offspring following initial crosses between diabetic GK and healthy Fischer or Brown Norway rats, respectively, at least six different independently segregating loci predisposing to glucose intolerance and impaired insulin secretion were found. In both studies a major locus, Niddm1 or Nidd/gk1, was found on chromosome 1 and estimated to contribute mainly to postprandial and less to fasting hyperglycemia. The somewhat less important Niddm2 or Nidd/gk2, influencing both postprandial and fasting glycemia, was localized to chromosome 2 and another locus linked to body weight, Weight 1 or bw/gk1, was shown on chromosome 7 of the rat. So far, no obvious candidate genes of diabetes have been assigned to the described chromosomal loci. Interestingly, however, the genes coding for two uncoupling proteins, Ucp2 and Ucp3, were positioned to a region of chromosome 1 linked to glucose intolerance

GOTO-KAKIZAKI RAT

177

Figure 1 An islet from a 3 month-old GK rat. The islet is round in shape with clearly defined borders (magnification ×180; Dr. A.Höög, Stockholm).

(Kaisaki et al., 1998). In addition, the mitochondrial glycerol-3-phosphate dehydrogenase gene was mapped to a part of chromosome 3 that contains a region linked to glucose intolerance in the GK rat (Koike et al., 1996). ISLET MORPHOLOGY AND HORMONE CONTENT Islet Structure and Composition One striking morphologic feature of GK rat pancreatic islets is the occurrence of so called star-fish shaped islets (Goto et al., 1988; Suzuki et al., 1992; Guenifi et al., 1995). These islets are characterised by disrupted structure with cords of pronounced fibrosis, i.e. connective tissue separating strands of endocrine cells, thereby resembling the appearance of a star-fish (Figures 1, 2). Accordingly, the mantle of glucagon and somatostatin cells is disrupted and these cells are found intermingled between B-cells. These changes are not present, or rare, in the pancreas of young GK rats, but increase in prevalence with aging (Suzuki et al., 1992). The parallel increase in IGF-2 expression with age in GK rat islets, but not in control rat islets (Höög et al., 1996), may at least partly account for these changes in islet morphology. Early observations from the Japanese colony (Suzuki et al., 1992) suggested degeneration and final paucity of B-cells in the ageing GK rat. B-cell replicatory rates, studied as labeling index after administration of 3H-thymidine, were however similar in GK and Wistar rats at 1, 3 and 6 months of age, indicating that the described changes with aging in GK rat islets cannot be accounted for by altered B-cell

178

C.-G.ÖSTENSON

Figure 2 A star-fish shaped islet from the same GK rat as in Fig. 1. Note the irregular shape with ill-defined borders and fibrous strands traversing the islet, as well as the typically enlarged size (magnification ×180; Dr. A.Höög, Stockholm).

replication (Östenson et al., 1996). This conclusion was, at least partly, supported by a recent study from France (Movassat & Portha, 1999). Immunohistochemical methods have been used to assess the distribution of various islet endocrine cell types in GK rats. From such studies, it appears that this distribution differs between some of the GK rat colonies. Thus, in the Stockholm colony B-cell density and relative volume of islet endocrine cells were alike in 2– 3-month-old GK rat and control Wistar rats (Östenson et al., 1993b; Guenifi et al., 1995). Similar results were reported in a study in Dallas with GK rats obtained directly from Sendai (Ohneda et al., 1993). In contrast, GK rats from the Paris colony reportedly display already at fetal stage a more than 50% reduction of the B-cell mass, which is maintained throughout the adult life and apparently precedes onset of hyperglycemia at about 3rd week after birth (Movassat et al., 1997). Incidentally, in the Stockholm colony, GK pups were hyperglycemic already at first week (Abdel-Halim et al., 1994). Concerning the glucagon-producing A-cell, in the Paris GK rat colony their relative volume was reported to be normal during fetal and early neonatal period, while reduced by approximately 1/3 at older ages (Movassat et al., 1997). No corresponding study has been performed regarding the somatostatin-producing D-cells, or other islet cell types. Pancreatic and Islet Hormone Content As for islet cellular composition, there is also some controversy regarding content of pancreatic hormones in GK rats. In animals from the Paris colony, markedly reduced pancreatic and islet insulin content, expressed per islet DNA, (less than 40% of control) has been consistently reported, also in the fetal GK rat (Sener et

GOTO-KAKIZAKI RAT

179

al., 1996; Movassat et al, 1997; Giroix et al, 1999). In other GK rat colonies, corresponding insulin levels have been similar, or more moderately decreased, as compared with control rats (Östenson et al., 1993b; Abdel-Halim et al., 1993; Suzuki et al., 1997; Salehi et al., 1999). Incidentally, islet insulin content was also normal in GK rats of the Seattle colony, which originally was derived from Paris via Tampa in Florida (Metz et al., 1999). Apparently, male GK rats are more prone than females to develop reduced islet insulin content, which may be accounted for by the more marked hyperglycemia in the males (unpublished observations). F1 hybrids of GK and Wistar rats did not show reduced contents of pancreatic insulin (AbdelHalim et al., 1994). The biosynthesis of insulin in isolated islets of both Paris and Stockholm colonies has been generally normal (Giroix et al., 1993a; Hutton et al, 1994). No major alteration in pancreatic glucagon content, expressed per pancreatic weight, has been demonstrated in GK rats (Abdel-Halim et al., 1993), although the A-cell mass was decreased, by about 35%, in adult GK rats of the Paris colony (Movassat et al., 1997). Pancreatic somatostatin content was slightly but significantly increased in Stockholm GK rats (Abdel-Halim et al., 1993). No studies on glucagon and somatostatin biosynthesis have been published. Isolation of Pancreatic Islets Pancreatic islets can be rather easily isolated by collagenase digestion from the pancreas of young GK rats; however, from animals older than 4 months the increasing prevalence of star-fish shaped islets (Figure 2) may reduce the yield of islets per pancreas. In fact, GK rat islets seem to resist the action of collagenase well, and may need slightly higher enzyme concentration, or longer exposure to the enzyme, than Wistar rat islets (Östenson, unpublished observation). ISLET AND BETA CELL FUNCTION Glucose-induced Insulin Secretion In GK rats, impaired glucose-stimulated insulin secretion has been demonstrated in vivo (Figure 3) (Portha et al., 1991; Gauguier et al., 1994; Galli et al., 1996; Gauguier et al., 1996; Salehi et al., 1999), in the perfused isolated pancreas (Portha et al., 1991; Östenson et al., 1993a; Abdel-Halim et al., 1993; AbdelHalim et al., 1994; Abdel-Halim et al., 1996) and in isolated pancreatic islets (e.g. Östenson et al, 1993a; Giroix et al, 1993a; Giroix et al, 1993b; Hughes et al, 1994). It is generally accepted that glucose stimulates insulin secretion by virtue of its metabolism. After rapid transport through the B-cell membrane, aided by the glucose transporter GLUT-2, the hexose is phosphorylated by glucokinase/hexokinase to glucose-6-phosphate, and then further metabolized through glycolysis and Krebs cycle to be oxidized with a yield of ATP. An increased cytosolic ATP/ADP ratio closes the ATP-regulated K+-channels and, when most of these channels (>99%) are closed, the cell membrane depolarises and voltage-dependent L-type Ca2+-channels open. The resulting increase in cytoplasmic calcium ions stimulates exocytosis of insulin granules (Efendic et al, 1991). A number of alterations or defects have been shown in the stimulus-secretion coupling for glucose in GK rat islets. GLUT-2 was found to be underexpressed, however, not likely to the extent that it could explain the impairment of insulin release (Ohneda et al, 1993). This assumption is supported by the fact that glucokinase/hexokinase activities were normal in GK rat islets (Östenson et al., 1993b; Tsuura et al., 1993). In addition, glycolysis rates in GK rat islets have been unchanged or even increased compared with control islets (Östenson et al., 1993a; Giroix et al., 1993a; Giroix et al., 1993b; Giroix et al., 1993c; Hughes et al.,

180

C.-G.ÖSTENSON

Figure 3 Glucose- and insulin-responses in vivo to an i.v. glucose challenge in GK rats (filled circles) and control Wistar rats (unfilled circles). ***) p < 0.001, and **) p < 0.01 between GK and control. (Modified after Salehi et al., 1999).

1994; Hughes et al., 1998; Ling et al., 1998). Furthermore, oxidation of glucose has been reported to be decreased (Giroix et al., 1993a), or unchanged (Östenson et al., 1993a; Hughes et al, 1994; Hughes et al, 1998; Giroix et al, 1993b) or even enhanced (Ling et al, 1998). Also lactate production has been shown to be increased in GK rat islets (Ling et al, 1998). These contradictory results may at least partly be due to differences between the different GK rat colonies. Although it may be hypothesised that the defective insulin response to glucose in GK rat B-cells is accounted for by an impaired ATP production and closure of the ATP-regulated K+-channels (Tsuura et al., 1993), the rate of ATP production in islet mitochondria was similar in Stockholm GK and control rats (Ling et al, 1998). Also in GK rats of Paris and Seattle colonies, overall ATP/ADP levels were found to be normal (Giroix et al, 1993b; Metz et al, 1999). Other defects in GK islet glucose metabolism include increased glucose cycling, due to increased glucose6-phosphatase activity in B-cells (Östenson et al., 1993a; Ling et al., 1998), impaired glycerol phosphate shuttle due to markedly reduced activity of the FAD-linked glycerol

GOTO-KAKIZAKI RAT

181

phosphate dehydrogenase (Östenson et al., 1993b; MacDonald et al., 1996), and decreased pyruvate carboxylase activity (MacDonald et al, 1996). It is possible that these alterations may affect ATP concentrations locally of importance for B-cell secretion. However, the enzyme dysfunctions were restored by normalization of glycemia in GK rats (MacDonald et al, 1996; Ling et al unpublished observations) with simultaneous but only partial improvement of glucose-induced insulin release. Hence, it is likely that these altered enzyme activities mediate a glucotoxic effect rather than being primary causes behind the impaired secretion. Abnormalities in the function of the ATP-regulated K+-channels and L-type Ca2+ channels (Kato et al., 1996) do not seem to account for the major impairment in insulin release in the GK rat (Hughes et al, 1998). Indeed, glucose-stimulated insulin secretion was markedly impaired in GK rat islets also when the islets were depolarised by a high concentration of potassium chloride and the ATP-regulated K+-channels kept open by diazoxide (Abdel-Halim et al., 1996). Similar results were obtained when insulin release was induced by exogenous calcium in electrically permeabilized GK rat islets (Okamoto et al, 1995). These findings, together with a recent study on animals from the Seattle colony (Metz et al, 1999), indicate that important defect(s) resides late in signal transduction, i.e. in the exocytotic machinery. Another intriguing aspect on possible mechanisms behind defective glucose-induced insulin release is the finding of dysfunction of lysosomal, glycogenolytic enzymes in GK rat islets (Salehi et al, 1999). In the light of a normal ATP generation by glucose, it could be proposed that a primary defect in glucose metabolism, with a key role in impaired insulin response to the hexose, resides in glycolysis. It still remains, however, to find that link. An attractive possibility to be explored is that increased activity of adenylate cyclase III (AC-III), due to functional mutations in the promoter region of the AC-III gene (Abdel-Halim et al., 1998), results in increased cAMP production and thereby enhanced activity of lactate dehydrogenase (Derda et al., 1980). This would in turn lead to increased glucose utilization and reduced cytoplasmic pool of NADH. It is well documented that altered NADH/NAD ratio is associated with impaired glucose-induced insulin release (Eto et al., 1999). The increased cAMP production has also offered the possibility to fully restore insulin response to glucose in the presence of an AC-III stimulator, such as forskolin (Abdel-Halim et al., 1996; Abdel-Halim et al., 1998) or a phosphodiesterase inhibitor (Guenifi et al., 1998). Insulin Response to Nonglucose Secretagogues It is of interest that the insulin response to sulfonylurea was decreased in GK rats compared with control rat islets (Giroix et al, 1993a). Among other nonglucose insulin stimulators, arginine has been shown to induce a normal, or even augmented insulin response from the GK rat pancreas (Kimura et al., 1982; Portha et al., 1991; Abdel-Halim et al, 1994; Hughes et al, 1994). Since preperfusion for 50–90 min in the absence of glucose reduced the insulin response to arginine in the GK, but not in the control pancreas (Portha et al, 1991), it is likely that previous exposure to glucose (in vivo or during the perfusion experiment) potentiates arginine-induced insulin secretion. Therefore, arginine-induced insulin release may also be impaired in GK rats like in type 2 diabetic patients, as was shown after normalization of glycemia (Porte Jr, 1991). Interestingly, using islets from F1 hybrids of GK and Wistar rats, impaired potentiation by glucose of arginine-stimulated insulin release was shown (Guenifi et al., 1998). Insulin responses to another amino acid, leucine and its metabolite, α-ketoisocaproate (KIC) have been shown to be diminished in GK rats from the Paris and Stockholm colonies (Giroix et al., 1993a; Guenifi et al., 1998; Giroix et al., 1999). This was attributed to defective catabolism of the amino acids, i.e. reduced generation of acetyl-CoA (Giroix et al., 1999). However, in GK rat islets from London and Kyoto colonies KIC induced normal insulin responses

182

C.-G.ÖSTENSON

(Hughes et al., 1994; Tsuura et al., 1993). Also in islets from mildly diabetic F1 hybrids of GK and Wistar rats, where glucotoxicity was shown to play a minor role, if any, in defective insulin release (Abdel-Halim et al., 1995), KIC stimulated insulin release normally in the absence, but not in the presence of glucose (Guenifi et al., 1998). Furthermore, KIC but not glucose exerted similar stimulation in GK and control islets, that were depolarized by potassium chloride. The latter findings further support the view that an abnormal B-cell metabolism of glucose proximal to the Krebs cycle is likely to account for the impairment of insulin release. ISLET MICRO-CIRCULATION Both pancreatic and islet blood flow, estimated with microsphere technique, were shown to be increased in GK rats (Svensson et al., 1994a; Atef et al., 1994) and in glucose intolerant F1 hybrids of GK and Wistar rats (Svensson et al., 1994b). Similar enhancement of islet microcirculation has been noted in other conditions with increased functional demand on the B-cells, e.g. pregnancy and hyperglycemia induced by glucose infusion (Svensson, 1994c). The increased islet blood flow in GK rats may be accounted for by an altered vagal nerve regulation, mediated by nitric oxide (NO), since vagotomy as well as inhibition of NO synthase normalized GK islet flow (Svensson et al., 1994a, 1994b). In addition, islet capillary pressure was increased in GK rats (Carlsson et al., 1996); however, this defect was restored after two weeks of normalisation of glycemia by phlorizin treatment. The possible relationship between islet microcirculation and B-cell secretory function remains to be established. INSULIN SENSITIVITY Studies of insulin sensitivity in the GK rat have been performed both in vivo, using clamp techniques, and in vitro in muscle, liver and adipose tissue. Thus, in hyperinsulinemic, euglycemic clamp studies combined with tracer determination of hepatic glucose production (HGP) in Sendai rats, the main defect responsible for a rather mild insulin resistance in GK rats was shown to be increased HGP in connection with dysregulation of hepatic fructose-2, 6-bisphosphate (Suzuki et al., 1992). These findings were later confirmed in clamps studies in animals of the Paris colony; however in addition to hepatic insulin resistance both at basal and hyperinsulinemic state, decreased insulin sensitivity in extrahepatic tissues was demonstrated (Bisbis et al., 1993). The defect in liver was characterized by decreased insulin receptor number but normal tyrosine kinase activity. In extrahepatic tissues, defective activation of glucose incorporation into glycogen by skeletal muscle, perhaps due to chronic activation of protein kinase C, has been suggested to contribute to insulin resistance and hyperglycemia in GK rats (Villar-Palasi & Farese, 1994; Avignon et al., 1996). Also in insulin receptor signaling, both in adipocytes and skeletal muscle of GK rats, specific defects have been demonstrated. Thus, impaired insulin-stimulated tyrosine phosphorylation of insulin-receptor substrate-1 (IRS-1) and inhibited effect of insulin on MAP kinase activation, probably due to altered serine/threonine phosphatase regulation, could account partly for the impaired insulin effect on glucose uptake and glycogen synthesis as seen in GK rat adipocytes (Begum & Ragolia, 1998). Furthermore, in skeletal muscle of GK rats defective post-receptor signaling was characterized by impaired insulin-stimulated glucose transport as well as PI-3kinase activated Akt kinase (protein kinase B) (Krook et al., 1997). However, the latter changes were fully restored, when glycemia was nearly normalized in GK rats by phlorizin treatment (reducing hyperglycemia by inhibiting renal glucose reabsorption). These observations strongly suggest that impaired insulin action in the GK rat, at least in extrahepatic tissues, is secondary to hyperglycemia. Such a glucotoxic effect is also

GOTO-KAKIZAKI RAT

183

supported by studies in F1 hybrids of GK and Wistar rats, in which insulin-mediated muscle glucose transport at 1 month of age was normal, while that of 2-month-old hybrid rats was moderately decreased compared with age-matched Wistar rats (Nolte et al., 1995). Altogether, these observations indicate the primacy of a B-cell secretory defect in the pathogenesis of diabetes in the GK rat. DIABETIC COMPLICATIONS At an early stage, GK rats were used in Sendai for studies of late diabetic complications. Signs of neuropathy were noted as reduced motor nerve conduction velocity (MNCV) in the tail nerve as early as in 2-month-old GK relative to Wistar rats (Goto et al., 1982; Goto et al., 1988; Suzuki et al., 1992). In addition, morphological alterations such as axonal degeneration and segmental demyelination as well as increased sorbitol and decreased myoinositol levels were observed in sciatic nerves in 6-month-old GK rats. Some of these defects were prevented by treatment with aldose reductase inhibitor (Goto et al., 1982). Reduced MNCV in the femoral nerve was also reported in 8-month-old GK rats of the Stockholm colony (Östenson et al., 1997) in which also signs of diabetic osteopathy was noted. In this context it is of interest that vitamin D metabolism has been reported impaired in the GK rat (Ishimura et al., 1995). Morphological changes indicating retinopathy seem to develop rather late in GK rats. Thus, an altered retinal endothelial cell/pericyte ratio was demonstrated in animals more than one year old, but not in 8month-old GK rats (Agardh et al., 1997; Agardh et al., 1998). Biochemical and functional alterations in the GK rat retina have been described as earlier phenomena, such as reduced glutathione levels (Agardh et al., 1998), increased tissue levels of vascular endothelial growth factor (VEGF) (Sone et al., 1997), and impaired retinal blood flow (Miyamoto et al., 1996). Regarding studies of experimental nephropathy, a gradual increase in glomerular basal membrane with aging has been shown in GK rats from 3 months of age (Suzuki et al., 1992). GK rats have also been found to develop impaired renal function (by 50–75%), reflected as increased serum creatinine and urea nitrogen levels, as compared to Wistar rats (Vesely et al., 1999). Ventromedial hypothalamic (VMH) lesions in GK rats accentuated the hyperglycemia and hypertriglyceridemia with visceral fat accumulation and reduced pancreatic insulin content (Yoshida et al., 1996). In addition, the VMH lesion enhanced proteinuria and glomerular basal membrane thickening as well as induced morphological changes in the aortal intima characteristic of an early stage of atherosclerosis. Thus, the VMH-lesioned GK rat has been suggested to be a model of both microangiopathy and macroangiopathy. CONCLUDING REMARKS In conclusion, an increasing amount of studies has demonstrated that the GK rat is a useful animal model of non-obese type 2 diabetes, in which the primary defect most likely resides in the B-cell. In addition, the nature of diabetes heredity in the GK rat, i.e. the additive action of several genes, may well resemble the polygenic basis for disease in the majority of type 2 diabetic patients. During the long-term inbreeding of GK rats—more than 20 years—the animals appear to have maintained rather stable levels of glucose intolerance and impairment of glucose-induced insulin response, also when studied in substrains of GK rats in various laboratories. However, other characteristics such as islet morphology and insulin content, as well as islet metabolism, have been shown to differ considerably between different substrain colonies, suggesting that newly introduced genetic mutations account for contrasting phenotypic properties.

184

C.-G.ÖSTENSON

REFERENCES Abdel-Halim, S.M., Guenifi, A., Efendic, S. and Östenson, C.-G. (1993) Both somatostatin and insulin responses to glucose are impaired in the perfused pancreas of the spontaneously non-insulin dependent diabetic GK (GotoKakizaki) rat. Acta Physiol, Scand., 148, 219– 226. Abdel-Halim, S.M., Guenifi, A., Grill, V., Luthman, H., Efendic, S. and Östenson, C.-G. (1994) Impact of diabetic inheritance on glucose tolerance and insulin secretion in spontaneously diabetic GK-Wistar rats. Diabetes, 43, 281–288. Abdel-Halim, S.M., Guenifi, A., Jansson, L., Andersson, A., Östenson, C.-G. and Efendic, S. (1995) A defective stimulus-secretion coupling rather than glucotoxicity mediates the impaired insulin secretion in the mildly diabetic F1 hybrids of GK-Wistar rats. Diabetes, 44, 1280–1284. Abdel-Halim, S.M., Guenifi, A., Khan, A., Larsson, O., Berggren, P.-O., Östenson, C.-G., et al. (1996) Impaired coupling of glucose signal to the exocytotic machinery in diabetic GK rats; a defect ameliorated by cAMP. Diabetes, 45, 934–940. Abdel-Halim, S.M., Guenifi, A., He, B., Yang, B., Mustafa, M., Höjeberg, B., et al. (1998) Mutations in the promoter of adenylyl cyclase (AC)-III gene, overexpression of AC-III mRNA, and enhanced cAMP generation in islets from the spontaneously diabetic GK rat model of type-2 diabetes. Diabetes, 47, 498–504. Agardh, C.-D., Agardh, E., Zhang, H. and Östenson, C.-G. (1997) Altered endothelial/ pericyte ratio in Goto-Kakizaki rat retina. J. Diab. Compile., 11, 158–162. Agardh, C.-D., Agardh, E., Hultberg, B., Qian, Y. and Östenson, C.-G. (1998) Glutathione levels are reduced in GotoKakizaki rat retina, but are not influenced by aminoguanidine treatment. Curr. Eye Res., 17, 251–256. Atef, N., Portha, B. and Pénicaud, L. (1994) Changes in islet blood flow in rats with NIDDM. Diabetologia, 37, 677–680. Avignon, A., Yamada, K., Zhou, X., Spencer, B., Cardona, O., Saba-Siddique, S., et al. (1996) Chronic activation of protein kinase C in soleus muscles and other tissues of insulin-resistant type II diabetic Goto-Kakizaki (GK), obese/ aged, and obese/Zucker rats. A mechanism for inhibiting glycogen synthesis. Diabetes, 45, 1396–1404. Begum, N. and Ragolia, L. (1998) Altered regulation of insulin signaling components in adipocytes of insulin-resistant type II diabetic Goto-Kakizaki rats. Metabolism, 47, 54– 62. Bisbis, S., Bailbe, D., Tormo, M.-A., Picarel-Blanchot, F., Derouet, M., Simon, J., et al (1993) Insulin resistance in the GK rat: decreased receptor number but normal kinase activity in liver. Am. J. Physiol., 265, E807-—E813. Carlsson, P.O., Jansson, L., Östenson, C.-G., Källskog, Ö. (1997) Islet capillary blood pressure increase mediated by hyperglycemia in NIDDM GK rats. Diabetes, 46, 947–952. Derda, D.F, Miles, M.F., Schweppe, J.S. and Jungmann, R.A. (1980) Cyclic AMP regulation of lactate dehydrogenase. J. Biol. Chem., 225, 11112–11121. Efendic, S., Kindmark, H. and Berggren, P.-O. (1991) Mechanisms involved in the regulation of the insulin secretory process. J. Intern. Med., 229 (suppl 2), 9–22. Eto, K., Tsubamoto, Y., Terauchi, Y., Sugiyama, T., Kishimoto, T., Takahashi, N., et al. (1999) Role of NADH shuttle system in glucose-induced activation of mitochondrial metabolism and insulin secretion. Science, 283, 981–985. Galli J., Li, L.-S., Glaser, A., Östenson, C.-G., Jiao, H., Fakhrai-Rad, H., et al. (1996) Genetic analysis of non-insulin dependent diabetes mellitus in the GK rat. Nature Genet., 12, 31– 37. Gauguier, D., Nelson, I., Bernard, C., Parent, V., Marsac, C., Cohen, D., et al. (1994) Higher maternal than paternal inheritance of diabetes in GK rats. Diabetes, 43, 220–224. Gauguier, D., Froguel, P., Parent, V., Bernard, C, Bihoreau, M.-T., Portha, B., et al. (1996) Chromosomal mapping of genetic loci associated with non-insulin dependent diabetes in the GK rat. Nature Genet., 12, 38–43. Giroix, M.-H., Vesco, L. and Portha, B. (1993a) Functional and metabolic perturbations in isolated pancreatic islets from the GK rat, a genetic model of noninsulin-dependent diabetes. Endocrinology, 132, 815–822. Giroix, M.-H., Sener, A., Bailbe, D., Leclercq-Meyer, V., Portha, B. and Malaisse, W.J. (1993b) Metabolic, ionic, and secretory response to D-glucose in islets from rats with acquired or inherited non-insulin-dependent diabetes. Biochem. Med. Metab. Biol., 50, 301–321.

GOTO-KAKIZAKI RAT

185

Giroix, M.-H., Sener, A., Portha, B. and Malaisse, W.J. (1993c) Preferential alteration of oxidative relative to total glycolysis in pancreatic islets of two rat models of inherited or acquired Type 2 diabetes mellitus . Diabetologia, 36, 305–309. Giroix, M.-H., Saulnier, C. and Portha, B. (1999) Decreased pancreatic islet response to L-leucine in the spontaneously diabetic GK rat: enzymatic, metabolic and secretory data. Diabetologia, 42, 965–977. Goto, Y., Kakizaki, M. and Masaki, N. (1975) Spontaneous diabetes produced by selective breeding of normal Wistar rats. Proc. Jpn. Acad., 51, 80–85. Goto, Y., Kakizaki, M. and Yagihashi, S. (1982) Neurological findings in spontaneously diabetic rats. Excerpta Medica ICS, No. 581, pp. 26–38. Goto, Y, Suzuki, K.-I., Sasaki, M., Ono, T. and Abe, S. (1988) GK rat as a model of nonobese, noninsulin-dependent diabetes. Selective breeding over 35 generations. In: Lessons from Animal Diabetes II, edited by E.Shafrir, and A.E.Renold, 2, 301–303. London: Libbey. Guenifi, A., Abdel-Halim, S.M., Höög, A., Falkmer, S. and Östenson, C.-G. (1995) Preserved β-cell density in the endocrine pancreas of young, spontaneously diabetic Goto-Kakizaki (GK) rats . Pancreas, 10, 148–153. Guenifi, A., Abdel-Halim, S.M., Efendic, S. and Östenson, C.-G. (1998) Preserved initiatory and potentiatory effect of α-ketoisocaproate on insulin release in islets of glucose intolerant rats. Diabetologia, 41, 1368–1373. Höög, A., Sandberg-Nordqvist, A.-C., Abdel-Halim, S.M., Carlsson-Skwirut, C., Guenifi, A., Tally, M., Östenson, C.-G., Falkmer, S., Sara, V., Efendic, S., Schalling, M. and Grimelius, L. (1996) Increased amounts of a high-molecularweight insulin-like growth fcator-II (IGF-II) peptide and IGF-II messenger ribonucleic acid in pancreatic islets of diabetic Goto-Kakizaki rats. Endocrinology, 137, 2415–2423. Hughes, S.J., Suzuki, K. and Goto, Y. (1994) The role of islet secretory function in the development of diabetes in the GK Wistar rat. Diabetologia, 37, 863–870. Hughes, S.J., Faehling, M., Thorneley, C.W, Proks, P., Ashcroft, F.M. and Smith, P.A. (1998) Electrophysiological and metabolic characterization of single β-cells and islets from diabetic GK rats. Diabetes, 47, 73–81. Hutton, J.C., Guest, P.C., Clark, A. and Östenson, C.-G. (1994) Proinsulin biosynthesis and processing in the GK rat model of type 2 diabetes. Diabetologia, 37 (Suppl. 1), A42. Ishimura, E., Nishizawa, Y., Koyama, H., Shoji, S., Inaba, M. and Morii, H. (1995) Impaired vitamin D metabolism and response in spontaneously diabetic GK rat. Miner. Electrolyte Metab., 21, 205–210. Kato, S., Ishida, H., Tsuura, Y., Tsuji, K., Nishimura, M., Horie, M., et al. (1996) Alterations in basal and glucosestimulated voltage-dependent Ca2+ channel activities in pancreatic β cells of non-insulin-dependent diabetes mellitus GK rat./. Clin. Invest., 97, 2417–2425. Kaisaki, P.J., Woon, P.Y, Wallis, R.H., Monaco, A.P., Lathrop, M. and Gauguier, D. (1998) Localization of tub and uncoupling proteins (Ucp) 2 and 3 to a region of rat chromosome 1 linked to glucose intolerance and adiposity in the Goto-Kakizaki (GK) type 2 diabetic rat. Mamm. Genome, 9, 910–912. Kimura, K., Toyota, T., Kakizaki, M., Takebe, K. and Goto, Y. (1982) Impaired insulin secretion in the spontaneous diabetes rats. Tohoku J. Exp. Med., 137, 453–459. Koike, G., Van Vooren, P., Shiozawa, M., Galli, J., Li, L.S., Glaser, A., et al. (1996) Genetic mapping and chromosome localization of the rat mitochondrial glycerol-3-phosphate dehydrogenase gene, a candidate for non-insulindependent diabetes mellitus. Genomics, 38 96–99. Krook, A., Kawano, Y., Song, X.M., Efendic, S., Roth, R.A., Wallberg-Henriksson, H., et al. (1997) Improved glucose tolerance restores insulin-stimulated Akt kinase activity and glucose transport in skeletal muscle from diabetic Goto-Kakizaki rats. Diabetes, 46, 2110– 2114. Lewis, B.M., Ismail, I.S., Issa, B., Peters, J.R. and Scanlon, M.F. (1996) Desensitisation of somatostatin, TRH and GHRH responses to glucose in the diabetic (Goto-Kakizaki) rat hypothalamus. J. Endocrinol., 151, 13–17. Ling, Z.-C., Efendic, S., Wibom, R., Abdel-Halim, S.M., Östenson, C.-G., Landau, B.R., et al. (1998) Glucose metabolism in Goto-Kakizaki rat islets. Endocrinology, 139, 2670–2675. MacDonald, M.J., Efendic, S. and Östenson, C.-G. (1996) Normalization by insulin treatment of low mitochondrial glycerol phosphate dehydrogenase and pyruvate carboxylase in pancreatic islets of the GK rat. Diabetes, 45, 886–890.

186

C.-G.ÖSTENSON

Malaisse-Lagae, E., Vanhoutte, C., Rypens, F., Louryan, S. and Malaisse, W.J. (1997) Anomalies of fetal development in GK rats. Acta Diabetol., 34, 55–60. Metz, S.A., Meredith, M., Vadakekalam, J., Rabaglia, M.E. and Kowluru, A. (1999) A defect late in stimulus-secretion coupling impairs insulin secretion in Goto-Kakizaki diabetic rats. Diabetes, 48, 1754–1762. Miyamoto, K., Ogura, Y., Nishiwaki, H., Matsuda, N., Honda, Y., Kato, S., et al. (1996) Evaluation of retinal microcirculatory alterations in the Goto-Kakizaki rat. Invest. Ophtalmol. Vis. Sci., 37, 898–905. Movassat, J. and Portha, B. (1999) Beta-cell growth in the neonatal Goto-Kakizaki rat and regeneration after treatment with streptozotocin at birth. Diabetologia, 42, 1098–1106. Movassat, J., Saulnier, C., Serradas, P. and Portha, B. (1997) Impaired development of pancreatic beta-cell mass is a primary event during the progression to diabetes in the GK rat. Diabetologia, 40, 916–925. Nolte, L.A., Abdel-Halim, S.M., Martin, I.K., Guenifi, A., Zierath, J.R., Östenson, C.-G., et al. (1995) Development of decreased insulin-induced glucose transport in skeletal muscle of glucose intolerant hybrids of diabetic GK rats. Clin. Sci., 88, 301–306. Ohneda, M.Johnson,J.H., Inman, L.R., Chen, L., Suzuki, K.-L, Goto, Y., et al. (1993) GLUT2 expression and function in β-cells of GK rats with NIDDM. Diabetes, 42, 1065–1072. Okamoto, Y., Ishida, H., Tsuura, Y., Yasuda, K., Kato, S., Matsubara, H., et al. (1995) Hyperresponse in calciuminduced insulin release from electrically permeabilized pancreatic islets of diabetic GK rats and its defective augmentation by glucose. Diabetologia, 38, 772–778. Östenson, C.-G., Khan, A., Abdel-Halim, S.M., Guenifi, A., Suzuki, K., Goto, Y., et al. (1993a) Abnormal insulin secretion and glucose metabolism in pancreatic islets from the spontaneously diabetic GK rat. Diabetologia, 36, 3–8. Östenson, C.-G., Abdel-Halim, S.M., Rasschaert, J., Malaisse-Lagae, F., Meuris, S., Sener, A., et al. (1993b) Deficient activity of FAD-linked glycerophosphate dehydrogenase in islets of GK rats. Diabetologia, 36, 722–726. Östenson, C.-G., Abdel-Halim, S.M., Andersson, A. and Efendic, S. (1996) Studies on the pathogenesis of NIDDM in the GK (Goto-Kakizaki) rat. In: Lessons from Animal Diabetes VI, edited by E.Shafrir, pp. 299–315. Boston: Birkhäuser. Östenson, C.-G., Fière, V., Ahmed, M., Lindström, P., Brismar, K., Brismar, T., et al. (1997) Decreased cortical bone thickness in spontaneously non-insulin-dependent diabetic GK rats. J. Diab. Compile., 11, 319–322. Porte Jr., D. (1991) β cells in type II diabetes mellitus. Diabetes, 40, 166–180. Portha, B., Serradas, P., Bailbé, D., Suzuki, K.-L, Goto, Y. and Giroix, M.-H. (1991) β-Cell insensitivity to glucose in the GK rat, a spontaneous nonobese model for type II diabetes. Diabetes, 40, 486–491. Salehi, A., Henningsson, R., Mosen, H., Östenson, C.-G., Efendic, S. and Lundquist, I. (1999) Dysfunction of the islet lysosomal system conveys impairment of glucose-induced insulin release in the diabetic GK rat. Endocrinology, 140, 3045–3053. Sener, A., Malaisse-Lagae, F., Östenson, C.-G. and Malaisse, W.J. (1993) Metabolism of endogenous nutrients in islets of Goto-Kakizaki (GK) rats. Biochem. J., 296, 329–334. Sener, A., Malaisse-Lagae, F., Ulusoy, S., Leclercq-Meyer, V. and Malaisse, W.J. (1996) Contrasting secretory behaviour of pancreatic islets from old rat in two models of non-insulin dependent diabetes. Diabetes Res., 31, 67–76. Serradas, P., Giroix, M.-H., Saulnier, C., Gangnerau, M.-N., Borg, L.A.H., Welsh, M., et al. (1995) Mitochondrial deoxyribonucleic acid content is specifically decreased in adult, but not fetal, pancreatic islets of the Goto-Kakizaki rat, a genetic model of noninsulin-dependent diabetes. Endocrinology, 136, 5623–5631. Serradas, P., Gangnerau, M.N., Giroix, M.H., Saulnier, C. and Portha, B. (1998) Impaired pancreatic beta cell function in the fetal GK rat. Impact of diabetic inheritance. J. Clin. Invest., 101, 899–904. Sone, H., Kawakami, Y., Okuda, Y., Sekine, Y., Honmura, S., Matsuo, K., et al. (1997) Ocular vascular endothelial growth factor levels in diabetic rats are elevated before observable retinal proliferative changes. Diabetologia, 40, 726–730. Suzuki, K.-L, Goto, Y. and Toyota, T. (1992) Spontaneously diabetic GK (Goto-Kakizaki) rats. In: Lessons from Animal Diabetes, edited by E.Shafrir, 4, 107–116. London: Smith-Gordon.

GOTO-KAKIZAKI RAT

187

Suzuki, N., Aizawa, T., Asanuma, N., Sato, Y., Komatsu, M., Hidaka, H., et al. (1997) An early insulin intervention accelerates pancreatic β-cell dysfunction in young Goto-Kakizaki rats, a model of naturally occurring noninsulindependent diabetes. Endocrinology, 138, 1106–1110. Svensson, A.M., Östenson, C.-G., Sandier, S., Efendic, S. and Jansson, L. (1994a) Inhibition of nitric oxide synthase by NG-nitro-L-arginine causes a preferential decrease in pancreatic islet blood flow in normal rats and spontaneously diabetic GK rats. Endocrinology, 135, 849–853. Svensson, A.M., Abdel-Halim, S.M., Efendic, S., Jansson, L. and Östenson, C.-G. (1994b) Pancreatic and islet blood flow in F1-hybrids of the non-insulin-dependent diabetic GK-Wistar rat. Eur. J. Endocrinol., 130, 612–616. Svensson, A.M. (1994c) Pancreatic islet blood flow in the rat. Thesis, Acta Universitatis Upsaliensis, Vol. 483, Uppsala. Tsuura, Y., Ishida, H., Okamoto, Y., Kato, S., Sakamoto, K., Horie, M., et al. (1993) Glucose sensitivity of ATPsensitive K+ channels is impaired in β-cells of the GK rat. Diabetes, 42, 1446–1453. Vesely, D.L., Gower, W.R. Jr., Dietz, J.R., Overton, R.M., Clark, L.C., Antwi, E.K., et al. (1999) Elevated atrial natriuretic peptides and early renal failure in type 2 diabetic Goto-Kakizaki rats. Metabolism, 48 771–778. Villar-Palasi, C. and Farese, R.V. (1994) Impaired skeletal muscle glycogen synthase activation by insulin in the GotoKakizaki (GK) rat. Diabetologia, 37, 885–888. Yoshida, S., Yamashita, S., Tokunaga, K., Yamane, M., Shinohara, E., Keno, Y., et al. (1996) Visceral fat accumulation and vascular complications associated with VMH lesioning of spontaneously non-insulin-dependent diabetic GK rat. Int. J. Obes. Relat. Metab. Disord., 20, 909–916. Zhou, Y.-R, Östenson, C.-G., Ling, Z.-C. and Grill, V. (1995) Deficiency of pyruvate dehydrogenase activity in pancreatic islets of diabetic GK rats. Endocrinology, 136, 3546–3551.

11. THE OLETF RAT KAZUYA KAWANO, TSUKASA HIRASHIMA, SHIGEHITO MORI, ZHI-WEI MAN AND TAKASHI NATORI Tokushima Research Institute, Otsuka Pharmaceutical Co., Ltd.

A spontaneously diabetic rat with polyuria, polydipsia and mild obesity was discovered in 1984 in an outbred colony of Long-Evans rats, which had been purchased from Charles River Canada (St. Constant, Quebec, Canada) in 1982. A strain of rats developed from this colony by selective breeding has since been maintained at the Tokushima Research Institute Otsuka Pharmaceutical (Tokushima, Japan), and is now referred to as the Otsuka Long-Evans Tokushima Fatty (OLETF) rat. The characteristic features of OLETF rats include: 1) late onset hyperglycemia (after 18 weeks of age), 2) a chronic disease state, 3) Increased urinary protein excretion at about 30 weeks of age. 4) higher glomerular filtration rate (GFR) compared with that of LETO rat, a strain derived from the same colony of Long-Evans rats that does not exhibit the diabetic syndrome; 4) Increased kidney weight and glomerular hypertrophy; 5) Decreased number of polyethyleneimine (PEI) in the glomerular basement membrane (GBM) in OLETF rats compared with LETO rats. Foot processes are irregularly arranged and retraction. Furthermore, the GBM showed marked thickening and rupture. 6) histological change of the kidney was focal mesangial lesion with proliferation of mesangial cells at 23 weeks of age. Glomerular damage progresses chronically and exudative lesions appear at 29 weeks of age. Most of glomeruli become obsolescent with severe atrophy of tubules after 90 weeks of age, resembling the advanced stage of human diabetic nephropathy. The glomerular lesions found in OLETF rats are similar to human diabetic nephropathy, and are characterized by fibrin cap, capsular drop and aneurysmal dilatations of intraglomerular vessels. Thus, the clinical and pathological features of the disease state in OLETF rats resemble those of human renal complications in type 2 diabetes.

THE OLETF RAT

189

INTRODUCTION Diabetes mellitus [DM] is a disease that is associated with several different pathogenetic origins. DM shows heterogeneous origin in spontaneous diabetic animal models in which the diabetic syndrome is linked to sex in some and obesity in others. The various animal models that express typical characteristics are useful for analyzing the complex forms of human diabetes. We previously reported a new inbred strain of Long-Evans Tokushima Lean [LETL] rats with IDDM without lymphopenia (Kawano et al., 1989 and Kawano et al., 1991). This LETL strain was established in 1989 from an outbred colony of Long-Evans rats that had been purchased from Charles River Canada Inc. From the same colony of rats, we established another inbred strain, the Otsuka Long-Evans Tokushima Fatty [OLETF] rat, which develops spontaneous persistent hyperglycemia, but can survive without insulin therapy for some period of time (Kawano et al., 1992). Male OLETF rats were diagnosed with diabetes in almost all cases at 24 weeks of age by oral glucose tolerance test [OGTT] with hyperglycemia and elevated levels of plasma cholesterol and triglyceride. As the stage of diabetes mellitus progressed, renal complication was inevitable for all animals (Kawano et al., 1992). In advanced countries, diabetic nephropathy mainly causes terminal renal failure. In Japan, patients with diabetic nephropathy comprise about 30% of patients receiving dialysis. The ratio has been increasing year by year. In 20 to 40% of patients with insulin-dependent diabetes mellitus [IDDM], nephropathy causes terminal renal failure. Even among patients with non-insulin-dependent diabetes mellitus [NIDDM], who comprise approximately 99% of diabetics in Japan, 10 to 20% of patients develop terminal renal failure. However, the etiology of these complications remains to be clarified in many aspects. It has been reported that deterioration of diabetic nephropathy depends on the duration of diabetes and success or failure in controlling blood sugar. Based on this concept, we performed sucrose loading in OLETF rats demonstrating exudative lesions with fibrin-cap in the kidney glomerulus, which resembled that in humans, to exacerbate diabetic conditions and cause deterioration of nephropathy. As a result, exacerbation of diabetes was observed, while nephropathy did not differ from that in untreated OLETF rats (Mori et al., 1996 and Mori et al., 1996). In sucrose-loaded animals, both food intake and protein intake were decreased by the rate of increase in sucrose intake. The decrease in protein intake may have influenced deterioration of nephropathy. Since protein intake may influence deterioration of nephropathy in OLETF rats, we conducted protein loading in OLETF rats to examine deterioration of nephropathy and clarify the pathogenesis. Furthermore, our recent data revealed that restriction of feeding for this strain suppressed the development and progression of both diabetes mellitus and renal complications (Mori et al., 1996). PROCESS OF BREEDING OLETF RATS The Long-Evans rats purchased from Charles River Canada Inc. in December 1982 were selectively bred (based on the hypothesis proposed by Goto et al.) by performing an OGTT on the male and female animals through the third generation, and then selectively mating the rats with high total plasma glucose. The same procedure was repeated in the offspring, but their plasma glucose levels were not elevated. However, a rat that differed from a spontaneous LETL rat was discovered. This rat was obese and demonstrated glucosuria, yet developed only mild subsequent weight loss. An OGTT performed on all the sibling rats revealed that all the males and one of the females were diabetic. Offspring of these brother—sister rats had already been born. We attempted to develop an NIDDM model from these siblings. Selective brother—sister mating was then initiated using males of more than 400g and apparently normal females at 9–10 weeks old. At least 20 pairs were mated in each generation. The males used for mating were examined by the OGTT when 25 weeks old. Only rats and their offspring with diabetes were bred,

190

K.KAWANO ET AL.

while rats of the normal type and their offspring were sacrificed. By June 1989, this process had been repeated through 20 generations. This established strain of NIDDM model animals has since been named the OLETF strain. The LETO strain was obtained by different original mating from that for OLETF rats, but both strains originated from the same colony of Long-Evans rats. The LETO line has not shown the diabetic syndrome from the F1 to the F20 generation. GENETIC PROFILE The genetic profiles determined by 22 biochemical markers and the RT1 class I and II phenotypes of OLETF, LETO and LETL rats. Although there were no differences within each strain, some interstrainstrain differences between OLETF and LETO were found in the esterases (Es-2, 3, 8, 9, 10). The RT1 haplotype of the three strains is used for both RT1 A and RT1 H, RT1 B, RT1 D loci. Skin graft experiments to determine homogeneity and histocompatibility differences showed delayed type rejection between OLETF and LETO, and LETO and LETL combinations. Permanent acceptance of skin grafts was observed in intra-strain combinations. GENERAL PATHOPHYSIOLOGY Serial data of the quantity of food intake, body weight, nonfasting plasma glucose, triglyceride, and cholesterol levels determined in the OLETF and LETO rats. The food intake of the male OLETF rats increased to approximately 30 g/day from 10 to 70 weeks of age. However, because food intake was greater in animals demonstrating glucosuria after 30 weeks of age, the standard deviation for food intake was greater after 50 weeks of age. Food intake in the male LETO rats was roughly 22 g/day by age 70 weeks. From 10 to 70 weeks of age, food intake increased to approximately 20 g/day in female OLETF rats, and approximately 16 g/day in female LETO rats. Average body weight of male OLETF rats at age 10 weeks was 381 g, approximately 80 g more than that of LETO rats at the same age. Thereafter, both strains continued to gain weight. However, because OLETF rats manifesting glucosuria after 30 weeks of age began to lose weight, the standard deviation for body weight was greater after 50 weeks of age. Average body weight of female OLETF rats at 10 weeks of age was 244 g, approximately 40 g more than LETO rats at the same age. Subsequently, both strains of rats continued to gain weight. By 70 weeks of age, average body weight of female OLETF and LETO rats was 532 g and 333 g, respectively, a difference of almost 200 g. The fed plasma glucose level in male OLETF rats began to rise after 30 weeks of age, with values at 50 and 70 weeks of 262 mg/dl and 326 mg/dl, respectively. On the other hand, plasma glucose in the male LETO rats at 10 weeks of age was approximately 120 mg/dl. Plasma glucose in these rats did not increase with age. Plasma glucose in female OLETF rats at 50 weeks of age was 140 mg/dl, and at 70 weeks, 126 mg/dl. Unlike the males, plasma glucose did not increase with age in females. Plasma glucose in female LETO rats at 10 weeks of age was 142 mg/dl. By 70 weeks of age, this value was roughly 110 mg/dl. At 10 weeks of age, the plasma triglyceride level in male OLETF rats was almost twice that of LETO rats. By 70 weeks of age, the triglyceride level was 426 mg/dl, approximately 4.3 times that of LETO rats at the same age. The triglyceride level in male LETO rats at 10 weeks and 70 weeks of age was 68 mg/ dl and 90 mg/dl, respectively. Triglycerides increased with age in both strains of female rats. However, the degree of increase was greater in the OLETF rats, with an increase from 155 mg/dl at age 10 weeks to 520 mg/dl

THE OLETF RAT

191

by 50 weeks of age. The triglyceride level in female LETO rats at 30 weeks, 50 weeks, and 70 weeks of age was 85 mg/dl, 113 mg/dl, and 168 mg/dl, respectively. Plasma cholesterol level in male OLETF rats increased with age. At 30 weeks, 50 weeks, and 70 weeks of age, these values were 122 mg/dl, 174 mg/dl, and 263 mg/dl, respectively. Plasma cholesterol in the LETO rats at 50 weeks and 70 weeks of age was 95 mg/dl and 141 mg/dl, respectively. On the other hand plasma cholesterol did not increase with age in the female OLETF rats. The cholesterol level was 100 mg/dl at 70 weeks of age. Plasma cholesterol in the female LETO rats followed the same pattern. ORAL GLUCOSE TOLERANCE TEST (OGTT) Changes in plasma glucose and insulin concentration during the OGTT determined in male and female OLETF and LETO rats at 10, 30, 50, and 70 weeks of age. Peak and 120-minute plasma glucose levels increased with age in the male OLETF rats during the OGTT. At 30 weeks of age, peak plasma glucose was greater than 300 mg/dl, with a value still exceeding 200 mg/dl at 120 minutes. However, plasma glucose did not increase with age in male LETO rats. Furthermore, these values did not exceed 300 mg/dl or 200 mg/dl, respectively. Plasma insulin level during the OGTT did not differ between the strains at 10 weeks of age. However, plasma insulin was markedly higher in the OLETF rat at 30 and 50 weeks of age. In contrast, plasma insulin in 70-week-old OLETF rats with severe diabetes was even lower than the LETO rats. Peak plasma glucose during the OGTT in female OLETF rats did not change significantly from 10 to 70 weeks of age. Nevertheless, the 120-minute values did increase with age, from 128 mg/dl at 10 weeks of age to 171 mg/ dl at 70 weeks of age. Plasma insulin showed a marked increase in the 70-week-old female OLETF rats compared to the corresponding level in LETO rats. Insulin levels in the LETO rat did not rise with age. RENAL COMPLICATION Weight of OLETF rat kidneys was greater than that of LETO rats at all ages. Water intake and urine volume gradually increased after 40 weeks of age and marked polydipsia and polyuria appeared after 60 weeks of age. Urinary protein excretion began to increase more than in LETO rats from approximately 30 weeks and drastically increased thereafter, ultimately reached to more than 1.0 g/dl. Histopathologically, early changes consisted of focal and segmentai lesions of the glomeruli. The slight widening of mesangial matrix with mesangial cell proliferation was observed at 23 weeks of age before the elevation of urinary protein level. At 29 weeks of age, besides mesangial lesions, a few glomeruli showed segmentai lesions with PAS-positive deposits in the mesangium or capillaries that resembled to the fibrin (hyalin) cap of the exudative lesion commonly observed in human diabetic glomerulopathy. Besides, at an early stage before 30 weeks of age, a number of glomeruli without mesangial and exudative lesions often showed Table 1 Glomerular lesions of male OLETF rat kidney. Age in weeks

23

30

55

80

100

Mesangial expansion Exudative lesion Obsolescence

16.9%

26.9%

33%

0%

3.4%

1.5% 0%

3% 0%

16.9% 2.3%

55.3% 19.2%

44.6% 31.5%

The proportion of each type of glomerular injuries is depicted. The mean value among five cases, is shown. The mesangial lesion includes glomeruli showing widened mesangial matrix with proliferated mesangial cells

192

K.KAWANO ET AL.

Age in weeks 23 30 55 80 100 without coexistent exudative lesions with and without coexistent mesangial lesions. The obsolescence includes glomeruli showing both segmentai and global obsolescence.

numerous eosinophilic granular deposits entrapped in epithelial cells (podocytes) which were darkly stained by PAM stain. After 50 weeks of age, glomeruli showing the exudative lesion with prominent fibrin caps became numerous and a more dominant finding than the mesangial lesions. Most of these lesions were characteristic in that they arose as focal and segmentai lesions in the early stage and showed segmentai obsolescence in the later stage. However, a few lesions of them advanced to global exudative lesions. Finestructurally, the fibrin caps in the exudative lesions were observed as accumulation of amorphous electron dense deposits located at the luminal side of slightly thickened basement membrane, some of which occluded the capillary lumen. Similar deposits were often observed inside the Bowmann’s capsule which showed the same finestructure with fibrin caps which were located between the basement membrane and parietal epithelial cells of the capsule. On the other hand, the aneurysmal dilatation of intraglomerular vessels, another characteristic finding in glomeruli, was frequently observed. These dilated vessels were contoured with thickened wall of mesangial tissue. However, vascular changes of the interstitium or vascular pole of the glomerulus such as arteriolo-hyalinosis and -sclerosis were not seen. Through all stages, the severity of mesangial lesions increased with age, some of which showed nodular expansion of the mesangial matrix. After 55 weeks of age, the segmentai or global obsolescence (sclerosis) of glomeruli involved numerous glomeruli which advanced to ultimately constitute the end stage kidney at 96 weeks of age. The glomerular changes of LETO rats were limited to a few mesangial lesion of minimal severity at an early stage and progressed slightly at a later stage to show mild focal mesangial lesions and occasional segmentai exudative lesions. The degree and time course of glomerular injury in OLETF rat kidneys were quantitatively analysed [Table 1]. The 3 types of glomerular injury were optically discerned; the glomerulus showing only mesangial lesion with widened matrix and proliferated cells, the glomerulus showing exudative lesion and the glomerulus showing obsolescence. The number of each type of glomerular injuries within a cross sectioned kidney was counted in each of 5 cases at 23, 30, 55, 80 and 100 weeks of age and the proportion of each type of injuries to all glomeruli was described. Before 30 weeks, the mesangial lesion was the predominant type of injury and showed 16.5% at 23 weeks of age and 26.2% at 30 weeks of age. The proportion of the exudative lesion increased to 16.7% at 55 weeks of age and reached maximal value of 52. 7% at 80 weeks. The obsolescent glomeruli appeared at 55 weeks of age and increased to 18.4% at 80 weeks of age. At 100 weeks of age, damaged glomeruli including both exudative lesions and obsolescence exceeded 70%. Since protein intake may influence deterioration of nephropathy in OLETF rats, we conducted protein loading in OLETF rats to examine deterioration of nephropathy and clarify the pathogenesis. We used male OLETF and LETO rats bred at the Tokushima Institute of Otsuka Pharmaceutical Co., Ltd. as well as F344 and BN rats purchased from Japan Charles River Inc. [Yokohama] [total: 4 strains]. In all strains, we established control groups in which standard pelleted food [CRF-1; Oriental Yeast Co., Ltd., containing 23% crude protein] was given and a 40% protein-loaded group [40% crude protein was contained in food]. Glomerular filtration rate [GFR] in OLETF rats was higher than those in other strains. However, there was no significant difference between the protein-loaded group and the control group. Neither LETO nor BN rats showed any significant differences between 40% protein-loaded groups and control groups. However, in F344 rats given the proteinloaded diet, GFR was significantly increased compared to that in the control

THE OLETF RAT

193

group. Among all strains, renal plasma flow [RPF] in the protein-loaded groups was slightly higher than that in control groups, although there were no significant differences. In the protein-loaded group consisting of OLETF rats, the number of polyethyleneimine [PEI] was significantly decreased compared to that in the control group. However, among LETO rats, there was no significant difference between the protein-loaded group and the control group. The number of PEI in the control group of OLETF rats was significantly decreased compared to that in LETO rats. Basement membrane thickness in the OLETF rat control group was significantly larger than that in LETO rats. However, there was no significant difference between the protein-loaded group and the control group. On electron microscopy, foot processes in LETO rats showed regular heights and intervals. However, in OLETF rats, both the height and interval of foot processes were irregular and retracted. In the proteinloaded group consisting of OLETF rats, ruptures of the basement membrane were observed, although the incidence was low. DISCUSSION A hyperglycemic, mildly obese, glycosuric rat model was developed by chance in an outbred colony of Long-Evans rats originally obtained from Charles River Canada. The diabetic phenotype was inherited. By successive mating of the male rats, we established the OLETF rat that demonstrates the NIDDM syndrome. The characteristic features of OLETF rats are as follows: 1) late onset of hyperglycemia (after 18 weeks of age), 2) a chronic disease course, 3) mild obesity, 4) inheritance, and 5) islet hyperplasia by newly formed cells. These characteristic pathological features closely resemble those of human NIDDM (Kawano et al., 1992). Human diabetes mellitus is classified into two types, IDDM and NIDDM, [WHO, 1985]. However, each type can be further classified clinically into several subtypes, including infectious or autoimmune type IDDM, as well as NIDDM type with or without obesity. The clinical course of DM in OLETF rats resembles that of db/db mice in which NIDDM changes to the IDDM type (Coleman et al, 1974). This conversion is the most distinctive characteristic of OLETF rats. In OLETF rats, we observed an increase in body weight following weaning accompanied by high plasma insulin and hyperplasia of B-cells in the islets. After 40 weeks of age, pancreatic islets began to deteriorate. After this stage, OLETF rat’s gradually developed IDDM-type diabetic syndrome with accompanying polyphagia, polyuria, decreased body weight, hyperglycemia and low plasma insulin. Plasma glucose returned to the normal range in these rats following insulin therapy, and their body weight increased gradually accompanied by a general improvement of symptoms. However, in db/ db mice, marked hyperglycemia {22. 4–33.6mM} develops at an earlier stage than in OLETF rats, and the clinical course progresses more rapidly with degeneration and atrophy of the islets. Other diabetic animals, such as ob/ob mice (Coleman et al., 1973), KK mice (Nakamura et al., 1962), Zucker fatty rats (Zucker et al., 1972) and GK rats (Goto et al, 1975) show NIDDM with accompanying polyphagia, hyperinsulinemia, and obesity, and no transformation into the IDDM type. The male-linked development of DM is another characteristic clinical feature of OLETF rats such as that observed in Wistar fatty rats and ZDF fatty rats (Ikeda et al., 1981 and Peterson et al., 1990). However, our previous results suggested that diabetes might be induced in females, as demonstrated by a diabetic change in these rats following early ovariectomy (Hirashima et al., 1996). Moreover, females might carry one of the genes responsible for inducing DM, because some (OLETF rats×LETO rats)F1 hybrids developed DM (Hirashima et al., 1995). We previously demonstrated that multiple recessive genes are involved in the induction of NIDDM-type diabetic syndrome in OLETF rats. These diabetogenic genes have been assigned

194

K.KAWANO ET AL.

to chromosome X [odb1] and 14 [Qdb2] (Hirashima et al, 1995 and 1996). Both genes [Odbl, Odb2] are essential for induction of the diabetic syndrome in OLETF rats (Hirashima et al, 1995 and 1996). The present results suggest that OLETF rats provide a useful animal model for analyzing the pathogenesis and complications of NIDDM as well as to study the development of drugs to regulate high plasma glucose associated with this disease. RENAL COMPLICATION We found that renal dysfunction and glomerular damage of OLETF rats progressed to an advanced stage, simulating the end stage kidney of advanced human diabetic nephropathy. The renal dysfunction of OLETF rats began to appear around 30 weeks of age, and continued to progress. The early light microscopic change of the kidney consisted of focal mesangial matrix widening with slight proliferation of mesangial cells at 23 week of age before the stage of clinical diabetic nephropathy. The glomerular lesions with fibrin caps, observed in a few glomeruli at 29 weeks of age, were thought to be identical to the exudative lesion described in human diabetic glomerulopathy (Olson et al, 1992), although they differed from the exudative lesion in humans in their focal and segmentai pattern of involvement. Although this lesion in human has been referred to as exudative in nature (Horsfield et al, 1965), most studies have emphasized an insudative process in their pathogenesis in which there is a permeation of glomerular capillary walls by plasma constituents (Anderson et al., 1989, Chung et al., 1966, Kawano et al., 1991, Salinas et al., 1970). The earliest glomerular change of human diabetic glomerulopathy was described as a generalized thickening of the glomerular basement membrane (Bergstrand et al., 1959). Thickening of glomerular basement membrane in OLETF rats was not prominent at any stage and was restricted to exudative lesions. Although this finding of limited glomerular basement membrane thickening seems to be in contrast with high levels of urinary protein, the excretion of urinary protein has failed to show a clear relationship with structural changes of the glomerular basement membrane (Mauer et al., 1984 and Osterby et al., 1986). After 50 weeks of age, the glomeruli showing the exudative lesions with prominent fibrin caps were numerous and became a more predominant finding than the mesangial lesion, and subsequently collapsed to become obsolescent at the late stage. The exudative lesion is often present in other human nephropathies such as focal glomerular sclerosis and lupus nephritis and does not have major diagnostic value for diabetic glomerulopathy (Olson et al., 1992). The exudative lesions in OLETF rats seemed to be significant in that they were found without exception in all animals with diabetes mellitus and frequently accompanied the capsular lesion that is unusual in other diseases (Olson et al., 1992, Brrie et al., 1952 and Zollinger et al., 1987). These lesions are believed to be derived exclusively from the prolonged diabetic condition. In the late stage, mesangium of OLETF rats showed mild to moderate matrix widening in a non-specific pattern. Mesangial lesions showing nodular accent were rare and, if present, seemed not to be identical with the typical nodular lesion of Kimmelstiel-Wilson (Saito et al., 1988). The pathogenesis of structural nodular lesions remains incompletely understood. Previous studies have endowed a suggestion that mesangiolysis and subsequent capillary microaneurysm may produce Kimmelstiel-Wilson nodules (Saito et al., 1988, Stout et al., 1993 and Yajima et al., 1976). Mesangiolytic change was often found in the mesangium of OLETF rats but was obscured by the superimposed exudative lesion. Typical mesangiolytic change was rarely seen. It is evident from our observations that exudative lesions bearing mesangiolytic change of OLETF rats fails to evolve to typical nodular lesion. Based on suggestions of previous authors mentioned above, the reason why the typical nodular lesion did not develop in OLETF rats is probably the absence of mesangiolysis preceding mesangial expansion. The reason for the lack of mesangiolysis in OLETF rats remains to be clarified.

THE OLETF RAT

195

The glomerular lesions found in OLETF rats, similar to those in human diabetic nephropathy, are exudative lesions characterized by fibrin caps, capsular drop and aneurysmal dilatation of intraglomerular vessels. However, the pathogenesis of diabetic nephropathy in OLETF rats demonstrating these findings remains to be clarified. In a previous study, we performed sucrose loading in OLETF rats based on the hypothesis that deterioration of diabetic nephropathy depends on the duration of diabetes and success or failure in blood sugar control. As a result, deterioration of diabetes was noted, but nephropathy did not deteriorate (Mori et al., 1996 and Mori et al., 1996). A corresponding factor was that, food and protein intakes were decreased because sucrose-loaded rats preferred to ingest sucrose. The decrease in protein intake may have influenced the progression of nephropathy. Furthermore, in humans, it has been reported that a protein-limited diet is effective in treating diabetic nephropathy. Therefore, we speculated that there might be an association between deterioration of diabetic nephropathy and protein intake. We therefore performed protein loading in OLETF rats, and examined the involvement of protein in deterioration of diabetic nephropathy. After protein loading in OLETF rats, urinary protein and albumin levels, which are indices of diabetic nephropathy, were increased. In LETO, F344 and BN rats without diabetes, there were no marked changes in urinary protein or albumin levels. The internal and external hyaline layers of the glomerular basement membrane function as a charge barrier via heparan sulfate, and inhibit the passage of negatively charged substances such as albumin (Tuttle et al., 1991). However, it has been reported that heparan sulfate levels are decreased under diabetic conditions, destroying the charge barrier and enhancing the permeability of the basement membrane (Zats et al., 1985). Protein loading in OLETF rats significantly decreased the number of PEI on the basement membrane, suggesting destruction of the charge barrier. On electron microscopy, foot processes in LETO rats showed regular heights and intervals on the basement membrane. However, foot processes in OLETF rats were irregularly arranged and fused. In addition, thickening of the basement membrane was observed in the protein-loaded OLETF rats, while rupture of the basement membrane was observed in a small number of rats. Therefore, it is suggested that the pathogenesis of diabetic nephropathy in OLETF rats involve enhanced permeability related to destruction of the charge and size barriers. In humans, it has been reported that GFR and RPF are increased after intravenous administration of amino acids (Zats et al., 1985). Concerning the mechanism involved in this finding, it is speculated that protein loading increases blood amino acid levels, increasing vasodilatory prostaglandin (PG) levels and elevating GFR and RPF. Pretreatment with a PG production-inhibiting agent, indomethacin, inhibits increases in GFR and RPF even when amino acid is administered (Anderson et al., 1995). In rats, it has also been reported that protein loading increases GFR (Fukuzawa et al., 1996). These increases in GFR and intraglomerular pressure are considered to promote formation of the mesangial matrix (Mogensen et al., 1982). Based on these studies, we speculated that protein loading might cause increases in GFR and intraglomerular pressure, exacerbating diabetic nephropathy in OLETF rats. However, protein loading in OLETF rats did not cause any changes in GFR or RPF. This may be because GFR was measured at 30 weeks of age in this study; the urinary protein level in the control group consisting of OLETF rats was also increased at this age, suggesting that GFR had already increased. Therefore, there may not have been any differences between the protein-loaded group and the control group. However, in OLETF rats, glomerular hypertrophy, expansion of the mesangial area related to increased mesangial matrix, and renal hypertrophy was noted together with albuminuria during the early stage (Fukuzawa et al., 1996). In this experiment, kidney weights were 2.2 g (0.47% of body weight) in the control group consisting of LETO rats and 3.6 g (0.52%) in the control group consisting of OLETF rats. In OLETF rats, kidney weight was markedly increased compared to that in LETO rats. In addition, in the proteinloaded group consisting of OLETF rats, kidney weight (3.9 g, 0.59%) was markedly increased compared to

196

K.KAWANO ET AL.

that in the control group (3.6 g, 0.52%) (P<0.01), suggesting that renal function was enhanced during the early stage in the protein-loaded group consisting of OLETF rats. In patients with diabetic nephropathy, the initial physiological change is glomerular hyperfiltration, while the initial morphological change is glomerular hypertrophy. Mogensen et al. (1982) initially examined glomerular hyperfiltration and renal hypertrophy in patients with early-stage diabetic nephropathy, and found increases in kidney weight in patients with IDDM. They indicated that increases in the glomerular filtration rate during the initial phase correlated with glomerular volume and glomerular loop area in patients with IDDM and that mesangial volume gradually increased with GFR. In patients with NIDDM, Yoshinouchi et al., reported that renal hypertrophy was detected when tests were negative for urinary microalbumin. Many studies have indicated that the state of blood sugar control is closely involved in the onset of diabetic nephropathy. In this experiment, blood sugar and insulin levels in the protein-loaded group consisting of OLETF rats on OGTT at 25 weeks of age were decreased compared to those in the control group consisting of OLETF rats, although there were no significant differences. Despite this finding, urinary protein levels and glomerular morphological changes exacerbated. Therefore, protein intake and genetic factors other than blood sugar control may be involved in deterioration of diabetic nephropathy, since protein loading did not induce diabetic nephropathy in LETO, BN or F344 rats without diabetes. Protein loading in OLETF rats caused deterioration of diabetic nephropathy at 30 weeks of age. Therefore, it was shown that protein intake and genetic factors other than blood sugar control were important factors for deterioration of diabetic nephropathy in OLETF rats. Furthermore, the mechanism by which nephropathy deteriorates may involve enhancement of renal function and increased permeability related to destruction of the charge barrier and the size barrier of the glomerular basement membrane. The preventive effect of food restriction on diabetes mellitus and its complications has not been sufficiently studied in animal models. It has been demonstrated in OLETF rats that the restriction of food reduces the levels of plasma glucose and eventually delay the onset of diabetes (Mori et al., 1996). In this study, OLETF rats fed with 30% restricted feeding during 6–80 weeks of age showed reduced levels of urinary protein and much less injury to glomeruli than those without food restriction. Therefore, the glomerular injury of OLETF rats is postulated to be derived exclusively from the prolonged diabetic condition, and the onset of diabetes mellitus and renal complications of OLETF rats correlate closely with the food restriction. In conclusion, the present report clearly demonstrates that OLETF rats provide a useful animal model for analyzing the pathogenesis of NIDDM and diabetic nephropathy. Supply system of OLETF rats A diabetic model rat, “OLETF”, has been established from outbred Long-Evans rats obtained from Charles River Canada, 1982, at Tokushima Research Institute, Otsuka Pharmaceutical Co. Ltd. The rat has obtained a good reputation as a model of human NIDDM from a variety of scientists, though we could not meet the demand of OLETF rats due to a relatively limited production. This time we are pleased to inform you that we are ready to distribute the OLETF rats, because we have developed an additional production unit for the rats in Ichiba Breeding Center which started April 1, 1992. Your request of rats will be considered if your experimental project is appropriate and valuable, although the number of rats will be limited for each laboratory. We hope that the OLETF rats may contribute to the basic study of NIDDM of humans. Basic principles

THE OLETF RAT

197

1. The OLETF rats will be supplied from the Ichiba branch of Tokushima Research Institute. (Ichiba-cho, Awa-gun, Tokushima, Japan) 2. Four to six weeks old rats will be supplied. 3. No females will be supplied. 4. Each laboratory, which needs older rats, is recommended to maintain those until use. Procedures to apply for rats 1. Submit the application form, memorandum, experimental design or protocol to the Laboratory Animal resources, Tokushima Research Institute. 2. We will inform acceptance of the application with a shipment order. 3. Animal costs will not be billed but users should pay transportation fees. Requests to be sent to Dr. K.Kawano, Section of Laboratory Animal Resources, Tokushima Research Institute, Otsuka Pharmaceutical Co., Ltd. 463–10 kagasuno Kawauchi-cho Tokushima 771–0192, Japan REFERENCES Anderson, S. and Vora, J.P. (1995) Current concept of renal hemodynamics in diabetes. J. Diabetes Complications, 9, 304–307. Anderson, W.R. and Jahnke, W.R. (1989) Insudative hyaline cap lesion of diabetic glomerulosclerosis. Human Pathol., 20, 388–390. Bergstrand, A. and Bucht, H. (1959) The glomerular lesions of diabetes mellitus and their electron-microscopic appearances. J. Pathol Bact., 77, 231–242. Brrie, H.J., Askanazy, C.L. and Smith, G.W. (1952) More glomerular changes in diabetes. Can. Meet. Assoc. J., 66, 428–435. Churg, J. and Dachs, S. (1966) Diabetic renal disease: Arteriosclerosis and glomerulosclerosis. Pathol. Annual, 1, 148–171. Coleman, D.L. and Hummel, K.P. (1974) Hyperinsulinemia in pre-weaning diabetes (db) mice. Diabetologia, 10, 607–610. Coleman, D.L. and Hummel, K.P. (1973) The influence of genetic background on the expression of the obese (ob) gene in the mouse. Diabetologia, 9, 287–293. Fukuzawa, Y., Watanabe, Y., Inaguma, D. and Hotta, N. (1996) Evaluation of glomerular lesion and abnormal urinary findings in OLETF rats resulting from a long-term diabetic state. J. Lab. Clin. Med., 128, 568–578. Goto, Y., Kakizaki, M. and Masaki, N. (1975) Spontaneous diabetes produced by selective breeding of normal wistar rats. Proc. Japan Acad., 51, 80–85. Hirashima, T., Kawano, K., Mori, S., et al. (1996) The influence of testosterone on the spontaneous hyperglycemic Otsuka Long-Evans Tokushima fatty (OLETF) rat. Int. J. Diabetes, 4, 85–92. Hirashima, T., Kawano, K., Mori, S., et al. (1995) A diabetogenic gene (ODB-1) assigned to the X-chromosome in OLETF rats. Diabetes Research and Clinical Practice, 27, 91–95. Hirashima, T., Kawano, K., Mori, S., et al. (1996) A diabetogenic gene, ODB2, identified on chromosome 14 of the OLETF rat and its synergistic action with ODB1. BBRC, 224, 420–425. Hirschberg, R.R., Zipser, R.D., Slomowitz, L.A., et al. (1988) Glucagon and prostaglandins are mediators of amino acid-induced rise in renal hemodynamics. Kidney International, 33, 1147–1155. Horsfield, G.I. and Lannigan, R. (1965) Exudative lesion in diabetes mellitus. J Clin. Pathol., 18, 47–53.

198

K.KAWANO ET AL.

Ikeda, H., Shino, A., Matsuo, T., Iwatsuka, H. and Suzuoki, Z. (1981) A new genetically obese-hyperglycemic rat (Wistar Fatty). Diabetes, 30, 1045–1050. Kawano, K., Hirashima, T., Mori, S., Abe, F., Kurosumi, M. and Saitoh, Y. (1989) A new rat strain with insulin-dependent diabetes mellitus, “LETL”. Rat News Lett., 22, 14–15. Kawano, K., Hirashima, T., Mori, S., Saitoh, Y., Kurosumi, M. and Natori, T. (1991) New inbred strain of Long-Evans Tokushima Lean rats with IDDM without lymphopenia. Diabetes, 40 (11):1375–1381. Kawano, K., Hirashima, T., Mori, S., et al. (1992) Spontaneous long-term hyperglycemic rat with diabetic complications; Otsuka Long-Evans Tokushima Fatty (OLETF) strain. Diabetes 41, 1422–1428. Kawano, K., Hirashima, T. and Mori, S. (1991) A new rat strain with non-insulin dependent diabetes mellitus, “OLETF” rat. Rat News Lett., 25, 24–26. Mauer, S.M., Steffes, M.W., Ellis, E.N., et al. (1984) Structural-functional relationship in diabetic nephropathy. J. Clin. Invest, 74, 1143–1155. Mogenson, C.E. (1982) Diabetic mellitus and the kidney. Kidney Int, 21, 673–675. Mori, S., Kawano, K. and Hirashima, T. (1996) Study on progress of diabetes induced sucrose-load in OLETF rats. 73th Japan association of animal diabetes research. Wakayama, Tokyo, February, 1996. Mori, S., Kawano, K. and Hirashima, T. (1996) Study on progress of diabetes and diabetic nephropathy induced sucrose-load in OLETF rats. The Journal of the Japan Diabetes Society. Omiya, May, 1996. Mori, S., Kawano, K., Hirashima, T., et al. (1996) Relationship between diet control and the development of spontaneous type II diabetes and diabetic nephropathy in OLETF rats. Diabetes Research and Clinical Practice, 33, 145–152. Nakamura, M. (1962) A diabetic strain of the mouse. Proc. Japan Acad. 38, 348–352. Olson, J.L. (1992) Diabetes mellitus. In Pathology of the kidney, edited by R.H.Heptinstall, pp. 1715–1763. Boston: Little, Brown and Company. Osterby, R., Anderson, A.R., Gregersen, H.J., et al. (1986) Quantitative structural data on the development of diabetic nephropathy. Diab. Nephropath., 5, 10–11. Peterson, R.G., Shaw, W.N., Neel, M.-A. and Eichberg, J. (1990) Zucker diabetic fatty rat as model for non-insulindependent diabetes mellitus. ILAR News (Institute of Laboratory Animal Resources) 32, 16–19. Saito, Y., Kida, H., Takeda, S., et al. (1983) Mesangiolysis in diabetic glomeruli: Its role in the formation of nodular lesions. Kidney Int., 34, 389–396. Salinas-Madrigal, L., Pirani, C.L. and Pollak, V.E. (1970) Glomerular and vascular “insudative” lesions of diabetic nephropathy: electron microscopic observations. Am. J. Pathol., 59, 369–397. Stout, L.C., Kumar, S. and Whorton, E.B. (1993) Focal mesangiolysis of the Kimmelstiel-Wilson nodule. Human Pathol., 24, 77–89. Tuttle, K.R., Bruton, J.L., Perusek, M.C., et al. (1991) Effect of strict glycemic control on renal hemodynamic response to amino acids and renal enlargement in insulin-dependent diabetes mellitus. The New England Journal of Medicine, 324, 1626–1632. Yajima, G. A histopathological study of diabetic nephropathy-light and electron microscopic observations. Acta Pathol. Jpn., 26, 47–62. Zatz, R., Meyer, T.W., Rennke, H.G., et al. (1985) Predominance of hemodynamic rather than metabolic factors in the pathogenesis of diabetic glomerulopthy. Proc. Natl. Acad. Sci., 82, 5963–5967,. Zollinger, H.U., Mihatsch, M.J., Gudat, F., et al. (1987) In: Renal pathology in biopsy. Diabetic glomerulosclerosis, pp. 391–399. Berlin: Springer-Verlag. Zucker, L. and Antoniades, H. (1972) Insulin and obesity in Zucker genetically obese rat “Fatty”. Endocrinology, 90, 1320–1330.

12 THE JCR: LA-cp RAT: AN ANIMAL MODEL OF OBESITY AND INSULIN RESISTANCE WITH SPONTANEOUS CARDIOVASCULAR DISEASE J.C.RUSSELL and S.E.GRAHAM Department of Surgery, University of Alberta, Edmonton, Alberta, Canada

ABSTRACT The JCR: LA-cp strain is a unique rodent model possessing all of the important elements of the insulin resistance/obesity/hypertriglyceridemia syndrome seen in humans, including endstage cardiovascular disease. The autosomal recessive cp gene has been shown to result in an absence of membrane-bound leptin receptors. This results in an early hyperphagia and hypersecretion of very-low-density lipoproteins (VLDL) and hypertriglyceridemia. A profound peripheral insulin resistance then develops, and vasculopathy ensues as the rats mature. The strain is stable genetically and phenotypically and is now well characterized. As such, it offers the possibility of relatively inexpensive experimental study of the pathophysiology of insulin resistance at all levels. The spontaneous development of atherosclerosis and the rats’ sensitivity to dietary cholesterol allow for experimental study of atherogenesis and putative antiatherosclerotic treatments. The rats also show spontaneous ischemic damage to the heart that can be experimentally exacerbated. This may allow for research into protective or modulatory measures for myocardial infarct, for which no satisfactory small-animal model has yet been established. In all these areas, the strain can provide a practical way to screen significant numbers of new candidate drugs still at the drug discovery stage. The development of effective new approaches to type 2 diabetes and its severe cardiovascular sequelae will require the use of this or similar animal models.

200

J.C.RUSSELL AND S.E.GRAHAM

HISTORICAL BACKGROUND A growing awareness of the importance of the obesity/insulin resistance syndrome as a primary cause of vascular disease has led to a heightened interest in animal models that mimic the human syndrome. In particular, there is a need for a smallanimal model for both research into the underlying pathophysiological mechanisms and the assessment of pharmaceutical interventions. Such a model should not only be metabolically similar to humans with either insulin resistance or overt type 2 diabetes, but should also exhibit the associated pathological sequelae, atherosclerosis, and ischemic damage. The JCR: LA-cp is a unique strain that meets these criteria and provides a means to both address basic mechanistic questions and screen putative new treatment approaches. The autosomal recessive gene, now designated as cp (corpulent), was first isolated by Simon Koletsky (1973) in a cross between rats of the Sprague Dawley strain (a very common albino rat) and the SHR (spontaneously hypertensive rat) strain, another mutation that arose in the Wistar colony at Kyoto University in Japan. Rats that were homozygous for the cp gene (cp/cp) were not only obese, but were hypertensive and exhibited a malignant atherosclerosis that was accompanied by the development of aortic aneurysms (Koletsky, 1975). In addition, male cp/cp rats had a lifespan of only some 10 months. Rats heterozygous (cp/+) or homozygous normal (+/+) were lean and normal. Koletsky designated the strain of rats as Obese SHR and sent some animals to Hansen at the National Institutes of Health in Bethesda (NIH), who created a defined and stabilized strain. This proved difficult to do due to the extensive load of deleterious genes in the complex background genome. Hansen crossed the Obese SHR with two in-house inbred strains at NIH, the SHR/N and the LA/N, and then backcrossed repeatedly to the parent strains to create (after 12 backcrossings) two congenic strains, the SHR/N-cp and the LA/ N-cp (Hansen, 1983; Greenhouse et al., 1988). These two strains were inbred and retained only the cp gene from the original Obese SHR. At the fifth backcross to the LA/N strain, initial breeding stock was sent from NIH to the University of Alberta, where they were the founding animals of a new closed outbred colony retaining approximately 3% of the genome derived from the Obese SHR. At the seventh backcross to the SHR/N strain, breeding stock was sent to the G.D.Searle Company in Indianapolis, where it became the origin of a colony now designated as SHHF/Mcc. These two non-congenic strains, however, both exhibited pathophysiological characteristics not evident in the two related congenic strains. Most importantly, whereas the JCR: LA-cp rats developed atherosclerosis and ischemic myocardial lesions, the SHHF/Mcc rats developed a fatal cardiomyopathy, apparently due to a single gene. The SHHF/Mcc strain has been inbred by McCune for the myocardiopathic trait and now exhibits a very high frequency of cardiomyopathy with ultimate congestive heart failure (Haas et al., 1995). The symbolic designation of the genetic mutations leading obesity in rats has not been consistent and some investigators have adopted their own symbols. The first mutation found was named fatty (fa) by Zucker and Zucker (1961). The mutatation later identified by Koletsky (1973) was given the name corpulent (cp) by Hansen (Greenhouse et al., 1988). This was later claimed to be allelic, in the sense that animals heterozygous for both the cp and fa mutations were obese, while those heterozygous for only one mutation were lean (Yen et al., 1977). There was an opinion that the mutations were therefore the same, although the independent origins and metabolic differences between the cp/cp and fa/fa rats suggested that was not the case. In fact, cp and fa are different mutations on the same gene and thus not allelic in the classical sense. McCune (1994) adopted the symbolism facp before the molecular basis of the two mutations became known. More recently Leibel’s group have adopted another symbol (f) for the cp mutation (WuPeng et al., 1997), perhaps after the suggestion of Yen et al. (1977). All of these refer to the same mutation, and the symbol cp has been the most widely and consistently used by those working with rats with the mutation.

THE JCR: LA-CP RAT

201

The susceptibility to vascular disease in the JCR: LA-cp rats is multifactorial and polygenetic in origin. Other strains, such as the LA/N-cp, do not show this trait and have lost some of the severity of the metabolic dysfunction in the course of inbreeding to create the congenic status. The JCR: LA-cp strain is therefore maintained as a closed outbred colony in order to retain the unknown genetic elements that have led to the cardiovascular disease in this animal model. PATHOPHYSIOLOGY OF THE JCR: LA-cp RAT Leptin Receptors The mutated gene first found by Koletsky and designated as cp by Hansen (1983) has recently been shown to consist of a nucleotide substitution leading to a stop codon in the extracellular domain of the leptin receptor (ObR) (Wu-Peng et al., 1997). As a consequence, cp/cp rats have no leptin receptor in the plasma membrane. Although the functions of leptin are not entirely clear at the time of this writing, a primary element is the inhibition of the hypothalamic neuropeptide Y (NPY) mechanism that is the single most powerful mediator of eating. The cp/cp rat has elevated levels of NPY in the arcuate nucleus and median eminence, the hypothalamic areas involved in the control of eating (Williams et al., 1990, 1992), and the rats are highly hyperphagic (Russell et al., 1990). Leptin also inhibits the release of insulin from pancreatic β-cells (Emilsson et al., 1997). The cp/cp rat has a massive hypersecretion of insulin that greatly exceeds that of the other obese rat, the fa/ fa Zucker (Pederson et al., 1991), which exhibits a reduced binding affinity of leptin to the receptor and is thus merely leptin resistant. Hyperphagia and Body Weight On weaning at 3 weeks of age, the cp/cp rats are no heavier than their lean +/? and +/+ littermates. Nonetheless, they have a detectably more round phenotype and can be unequivocally identified. Their food intake is greater at 3 weeks and increases rapidly as the animals mature (Figure 1). Their body weight also increases more rapidly and by 8 weeks of age they are already markedly obese. By 9 months of age, the cp/ cp male rats normally weigh 800–900 grams compared to approximately 400–450 grams for the lean male rats. Beyond 12 months of age, these rats have reduced their food intake and are starting to show deterioration in their condition and a decline in weight. Although a reduction of food intake to that of the cp/ + or +/+animals (~20 g/day at 12 weeks) results in a lower body weight, it does not normalize their weight to quite that of the lean rats. If cp/cp rats are restricted to 12 g/day (60% of the intake of +/+ animals), their weight is approximately normalized, but the phenotype retains its rounded and obese shape and the animal is simply a smaller rat. Insulin Resistance and Hyperlipidemia The primary metabolic abnormalities of the cp/cp rat are a profound insulin resistance and hypertriglyceridemia. A major question now under study is, which comes first, the insulin resistance and hyperinsulinemia or the hyperlipidemia? Plasma triglyceride concentrations are elevated in male cp/cp rats at 3 weeks of age and continue to rise rapidly until 12 weeks of age. This is accompanied by an elevated intracellular triglyceride content and precedes the rapid, substantive rise in plasma insulin levels that occurs at about 51/2 weeks of age (Russell et al., 1998a). The hyperlipidemia is due to a marked hypersecretion of VLDL, resulting in high VLDL levels and elevated triglyceride concentrations (Dolphin et al., 1987;

202

J.C.RUSSELL AND S.E.GRAHAM

Figure 1 Food intake and body weight of male +/? (○) and cp/cp (●) rats as a function of age.

Russell et al., 1989b). In 12-week-old cp/cp rats, these are typically 280 mg/100 ml in males and much higher in females, at 500 mg/100 ml. More modest increases occur in cholesterol and phospholipid concentrations, probably due to VLDL requirements for particle stability. There is a prominent flowthrough to the highdensity lipoprotein (HDL) fraction in these rats; however, lipoprotein lipase activity and VLDL clearance are not reduced. The hyperinsulinemia that is fully developed at 8 weeks of age in the cp/cp rat is a reflection of a profound insulin resistance that is not present at 4 weeks of age. The insulin resistance results in a complete absence of insulin-mediated glucose uptake by peripheral tissues or clearance from the blood (Russell et al., 1994, 1998a). The cp/cp rats are essentially normoglycemic (fasting plasma glucose levels are somewhat higher than in other strains of rat in all genotypes), but do have an impaired glucose tolerance under a high glucose load (Russell et al., 1987). The commonly used indices of abnormal glucose and insulin metabolism are the intravenous glucose tolerance and oral glucose tolerance tests, which are not sensitive to changes in the sensitivity to insulin in these animals (Russell et al., 1999). We have found that a standardized meal tolerance test, consisting of a test meal of a 5-gram pellet of rat chow given after an overnight fast and subsequent measurement of both plasma insulin and glucose, is the most sensitive and efficient method of assessing insulin sensitivity in these animals. As seen in Figure 2, the insulin response at 30 minutes postprandially provides all the information needed for an index of insulin sensitivity. As also shown in Figure 2, treatment with an insulin-sensitizing agent, in this case fenofibrate, results in an unchanged plasma glucose time course without hyperglycemia, but without the marked hyperinsulinemic reponse seen in the control animals.

THE JCR: LA-CP RAT

203

Figure 2 Plasma glucose and insulin response of male rats at 12 weeks of age to a meal tolerance test consisting of 5 g of rat show. , +/? control rats; ●, cp/cp control; ■, cp/cp treated for 4 weeks with fenofibrate at 100 mg/kg.

Cardiovascular Disease In metabolic terms, the cp/cp rats strongly mimic the human syndrome of abdominal obesity, with a concomitant insulin resistance, hyperinsulinemia, and hypertriglyceridemia. The cp/cp rat does not progress to a state where it can no longer maintain the excessive rate of insulin release from the pancreatic β-cell and resultant development of insulin-dependent diabetes. However, the pre-diabetic, insulin-resistant state is highly associated with cardiovascular disease in humans, and the rats similarly spontaneously develop marked vasculopathy, atherosclerosis, and ischemic lesions. Early atherosclerotic changes are evident in very young adult rats (by 12 weeks of age), and these increase in severity with age such that by 9 months essentially all cp/cp males have advanced intimal lesions of the aortic arch. A typical example of such a lesion visualized by scanning electron microscopy is shown in Figure 3. Thrombi of various sizes and ages are frequently found on the arterial surface, examples of which are shown in Figure 4. The presence of thrombi is probably related to the increased levels of plasminogen activator inhibitor-1 (PAI-1) in the cp/cp rat (Schneider et al., 1998). As shown in Figure 5, adherent macrophages are often present, particularly over areas of abnormal endothelium. Transmission electron microscopy shows the intimal lesions as being very similar to atherosclerotic lesions in humans, encompassing lipids, proteoglycan, collagen, macrophages or smooth muscle cells (SMC), and cellular debris in the intimal space (Figure 6). The endothelium overlying the lesions is usually intact and often not highly abnormal. The SMC of the media appear to be activated and migratory, moving through breaks in the internal elastic lamina into the intimal space. This is consistent with the observation that the aortic SMC of the male cp/cp rat are hyperproliferative and hyper-responsive to various cytokines, particularly insulin (Absher et al., 1997). The hyperactivity of the SMC is largely prevented by experimental manipulations that greatly reduce the hyperinsulinemia (Absher et al., 1999).

204

J.C.RUSSELL AND S.E.GRAHAM

Figure 3 Scanning electron micrograph of the aortic arch of a 9-month-old male cp/cp rat. There is a large raised intimal lesion over the lesser curve of the arch, with intact overlying endothelium.

The vascular lesions and medial SMC abnormalities are accompanied by a vascular dysfunction in the form of an enhanced contractile response to noradrenaline or phenylephrine and impaired nitric oxidemediated relaxation (O’Brien et al., 1997; O’Brien and Russell, 1998). In addition to the resultant vasospastic tendency, there is an increased activity of PAI-1 in the vessels of the male cp/cp rat, which leads to less lysis of intravascular thrombi. In confirmation of this tendency, we have frequently seen both large and small thrombi in the arteries by scanning electron microscope, as illustrated in Figure 4, and in coronary arteries by histology. Both the vascular lesions and the vascular dysfunction are much less severe in the cp/cp females than in the males, despite VLDL levels approximately twice those of the cp/cp males. Atherosclerotic lesions, intravascular thrombi, and other manifestations of cardiovascular disease do develop in cp/cp females, but only at advanced ages. The differences in disease severity between the sexes are not affected by early castration in either sex, although plasma lipid levels do respond to castration (Russell et al., 1993). The mechanism underlying the sexual dimorphism in the cardiovascular disease remains obscure, but may be related to lower insulin levels in the female cp/cp rat. We have recently found that the JCR:LA-cp rat is very sensitive to dietary cholesterol. Normal rat chow is essentially cholesterol free. The addition of 0.25% cholesterol leads to markedly exacerbated intimal lesions in male cp/cp, but not in +/?, rats at 6 months of age. A diet containing 1% cholesterol leads to advanced atherosclerotic lesions at 12 weeks of age. The dietary cholesterol results in increased VLDL cholesterol concentrations that appear to be pathogenic.

THE JCR: LA-CP RAT

205

Figure 4 Scanning electron micrograph of one of the branches of the arch of the aorta of a 9-month-old male cp/cp rat showing a large organized thrombus in the lumen of the branch. There is also another old thrombus on the flow divider at the orifice of the branch. The endothelium is abnormal, with adherent macrophages and underlying intimal lesions.

COMPARISON WITH OTHER OBESE RAT STRAINS Comparisons between the two genetically determined rat models of obesity are inevitable and valuable. The oldest and best known of the rat obesity mutations is that of the fa or fatty Zucker rat. Because the fa/fa rat is leptin resistant, only, the fatty rats are less hyperinsulinemic and, consistent with our findings of the importance of high insulin levels in vascular disease, do not develop intimal lesions or myocardial ischemic damage (Amy et al., 1988). The Zucker rat colonies show frequent infection with Mycoplasma species, most commonly Mycoplasma pulmonis, and this appears to be associated with abnormalities of the endothelium. This makes the use of Zucker rats in cardiovascular studies problematic unless the animals can be shown to be mycoplasma free. The rat strains with the cp gene provide an interesting demonstration of the multifactorial and polygenetic basis of the development of both type 2 diabetes and the related cardiovascular disease. The fully congenic LA/N-cp and SHR/N-cp strains have been backcrossed many times (n>16) and have lost all genetic contribution from the original Koletsky strain. As a consequence, neither strain exhibits the end-stage cardiovascular disease that is prominent in the JCR:LA-cp rat (Russell et al., 1991). The strains have been valuable for studies of the pathophysiology of insulin resistance and type 2 diabetes, especially in the hands

206

J.C.RUSSELL AND S.E.GRAHAM

Figure 5 Scanning electron micrograph of an area of the aorta with extensive adherent and activated macrophages.

of O.E.Michaelis and his coworkers (Recant et al., 1989; Triana et al., 1991; Yamini et al., 1992; Velasquez et al., 1995). The SHHF/Mcc-fa(cp) strain, developed by McCune from an SHR/N-cp colony originally held by the G.D.Searle Company, is unique in spontaneously developing a cardiomyopathy that progresses to fatal congestive heart failure. The colony has been inbred for the cardiomyopathic trait and has proven to be an excellent model for the study of a complex disease process (see, for instance, Ondera et al., 1998). MAINTENANCE OF THE STRAIN The maintenance of a closed outbred colony such as that of the JCR:LA-cp rat requires a clear plan and allowance for the inevitable tendency for genetic drift. Such drift can and will result in the loss of some of the unique genetic characteristics of a colony if allowed to occur. In the case of other strains with the cp/cp gene, this has led to the reduced severity of a metabolic dysfunction. Most importantly, the inbreeding needed to create the congenic strains also bred out the propensity for the development of the severe cardiovascular disease seen in the original Koletsky Obese SHR rat (Koletsky, 1975). In the case of the JCR:LA-cp rat, we ensure genetic stability of the colony through a formal assortive breeding program with 10 separate lines and using the technique described by Poiley (1960). Every effort is made to ensure that all lines are preserved and that inadvertent selection is not practised in the breeding protocol.

THE JCR: LA-CP RAT

207

Figure 6 Transmission electron micrograph of the aorta of a 9-month-old male cp/cp rat showing an advanced atherosclerotic lesion with the lumen on the upper left. The internal elastic lumina is intact, visible as a light band across the lower portion, with underlying smooth muscle cells and collagen deposits. The intimal space contains collagen, proteoglycan, migratory smooth muscle cells with lipid inclusions, and a macrophage/foam cell in the centre.

The breeding colony is maintained in a modified barrier facility to prevent the transmission of infection from the flowthrough of rodents coming from large commercial breeders. In our experience, animal centres can expect the occasional shipment of rats or mice harbouring one of the rodent viral pathogens or pinworms. In many cases, the only cure is to kill all of the affected rats and repopulate with clean stock, perhaps from cesarean-derived colonies. In the case of a unique colony such as that of the JCR:LA-cp rat, this would not be possible, or would be possible only at great expense. Contamination, then, would result in loss of the strain and all future research based on it. Our colony is maintained solely by members of our research group, who of necessity avoid all contact with other rodents. All our staff wear only facility clothing and footwear in the unit as well as hats, masks, and gloves. Food and other items are autoclaved before being brought into the unit. These precautions have been largely successful in preventing infection. In the early stages of development of the strain, when we were in an insecure facility, rat pinworm (Syphacia obvelata) was introduced from a commercial supplier. This parasite is extremely difficult to eradicate, especially in a large breeding colony. Three separate attempts have now been made to do so using Ivermectin, without success. We now treat the colony intermittently to keep the infestation down. Also at an early stage, the colony was contaminated with

208

J.C.RUSSELL AND S.E.GRAHAM

pneumonia virus of mice (PVM). This virus is virulent and is a serious problem in mice. In rats, it causes a transient bronchitis in suckling animals, who then go on to develop a strong immunity, thus spreading the virus in the breeding colony alone. We were only successful in eradicating PVM by moving adult rats, only, when we relocated to our current barrier facility, and have been virus antibody free since. The rats are bred by placing two females with one male and separating them 3 weeks later. Although they do produce ova and normal sperm, the cp/cp rats of either sex are not fertile: they are sufficiently obese so as to pose mechanical problems in mating and, in any event, seem quite uninterested. Repeated attempts to breed cp/cp animals of either sex have been unproductive. As a consequence, we must breed heterozygotes (cp/+). This results in progeny that are 25% cp/cp and obese, 25% +/+, and 50% cp/+, the two latter being lean. The lean animals of unknown genotype are therefore 2:1 cp/+ and +/+ and are characterized as +/?. New (+/?) breeders are subjected to a test breeding with a known cp/+ animal and are identified as having the cp/+ genotype by the presence of one or more cp/cp offspring in the resulting litter. Normally, male rats are bred at 10 weeks of age and can continue to function until 15–20 months of age. Females are bred at 10– 12 weeks of age and will produce 4, and possibly 5, litters of 8 pups each before being retired. In order to minimize stress in these animals, no breeding is done during the postpartum estrous period. All rats are maintained on aspen wood chip bedding in polycarbonate cages. Mothers with pups are housed in lower hamster-type cages so that the pups can reach the feed and water bottle. Wood shavings and 11×13-cm black polyvinylchloride tubes are added to enhance the environment and increase the sense of security of the dam and pups. All pregnant and suckling females are given a calcium supplement in the drinking water. It is important not to overcrowd the holding rooms of pregnant and suckling rats. Crowding leads to increased odour levels and a sense of competition or threat among the rats, with resultant eating of the young by their dams. Detailed records are kept of all breeding activity, and each weaned and retained rat is given an identification number. Although lean and obese rats do not differ in body weight at weaning (3 weeks), animals that are cp/ cp can already be recognized by phenotypic differences: the cp/cp rats have a more rounded abdomen and slightly enlarged jowls. Upon separation from the dam, weanling littermates are normally housed in a group for a week to reduce stress. Adult rats are normally housed in pairs, as they are social animals and this can reduce stress levels. However, it is not good practice to house a breeding-age lean male with a cp/cp male as the lean rat will show aggression toward the obese animal. Careful humidity control is essential in the breeding rooms and is a significant problem in western Canada, where humidity levels are low. Under low relative humidity (50% or less) suckling rats develop a condition known as ring tail, which consists of the development of annular constrictions on the tail, with or without sloughing of the distal portion of the tail. Rats with ring tail are stressed and unsuitable for experimental use. Although the mechanism of this disorder is unknown, prophylaxis can be achieved through keeping relative humidity in the breeding colony at 60%. EXPERIMENTAL TECHNIQUES Animal Handling and Drug Administration The cp/cp rats are more stress sensitive than normal rats and are metabolically compromised. Even simple relocation from one room to another can have significant metabolic effects, and transportation on a noisy metal cart can induce ischemic damage to the heart in these stress-sensitive animals. We have thus developed specific experimental approaches to minimize stress and to improve the quality of our data. Drugs and other agents are best administered to the rats in their feed or drinking water, rather than by gavage. Agents that are soluble and stable can be placed in the water bottle, which is weighed daily or twice

THE JCR: LA-CP RAT

209

weekly, and their consumption calculated. Because the rats, also, are weighed, the concentration of the agent in the water can easily be adjusted to achieve a desired dosage. In our laboratory, drugs are incorporated into powdered (unpelleted) rat chow, available from the manufacturer, or into standard chow. The calculated amount (based on body weight and expected or actual food intake) is mixed into the feed using a small commercial pasta maker, often with premixing in a mortar and pestle. The feed and drug mixture is then moistened with water and extruded through a custommade die. The resulting pellets are air dried without heat in a convection oven or on an open rack. Anesthesia Conventional barbiturate and other injection anesthesia do not work well on the cp/cp rats: the large fat stores result in rapid clearance of the drugs so that the animals tend to be too lightly anesthetized for surgery. Higher doses of anesthetic, however, typically result in anesthesia that is too deep and subsequent respiratory arrest. Inhalation anesthesia with halothane or isoflurane largely avoids these problems. With young rats, anesthetic induction can be accomplished by holding the rat on its back in the hand, with one’s fingers under the animal’s jaw, and placing a cone over the muzzle. Alternatively, the rat can be placed in a large jar or even left in a cage and halothane in oxygen blown in. Induction is performed using 3.5% halothane in oxygen, reduced to 2.5% once the animal is unconscious, with approximately 2% required to maintain a surgical plane. This concentration is adjusted from time to time to give the minimum level required during long procedures such as euglycemic insulin clamps or VLDL secretion rate studies. For any procedure lasting more than a short period, these animals need to be placed on a warmed table to prevent hypothermia. When animals are euthanized at the end of a protocol, blood and tissues are frequently collected for assay. It is important to ensure that the anesthesia and accompanying procedures are performed so as to minimize any stress on the rats and negative effects on metabolic parameters. We perform all anesthetic administration and procedures in a room that is located well away from the other rats in order to prevent distress in any animals that will subsequently undergo surgical manipulation. It needs to be remembered that rats communicate widely in the ultrasonic sound range above human hearing, and any upset over frightening or even simply new occurrences spreads rapidly throughout the group. Our research team makes ongoing efforts to ensure that the rats are never handled roughly or given adverse experiences. The animals are familiar with the technical staff from the breeding and maintenance unit, and this results in extremely sociable and unstressed behaviour. The experimenters are thus able to pick up a naive 6-month-old rat and take a blood sample from its tail without struggle by the animal or the need for restraint. Diurnal Cycle Rats are nocturnal, sleeping during the light phase of their diurnal cycle. This activity pattern is reflected in their metabolism, and it is thus preferable to conduct metabolic studies during the dark period of the rats’ cycle. Animals which are to be subjected to such studies are normally maintained on a reversed light cycle, with lights off from 0600 to 1800 hours. The timing of lights is often adjusted around the experimenter’s day for the convenience of performing procedures. All manipulations of the rats are performed under a dim light in a dedicated procedure room.

210

J.C.RUSSELL AND S.E.GRAHAM

Assessment of Insulin and Glucose Metabolism The conventional method of assessing insulin and glucose metabolism has been the intravenous glucose tolerance test (IVGTT), with the euglycemic insulin clamp being the ultimate approach. Both methods present serious problems in the cp/cp rat. Because the IVGTT shows relatively small differences in the rate of clearance of an injected glucose load (typically 0.5 g/kg) (Russell et al., 1999), it is difficult to detect experimental changes in the severity of insulin resistance using this test. A euglycemic clamp potentially gives detailed information on the glucose clearance and hepatic output and is practical in the cp/cp rat; however, the procedure is complex and lengthy and requires either that the rat be anesthetized throughout or that indwelling cannulae be surgically implanted in advance. Both the IVGTT and the euglycemic insulin clamp lead to very significant stress in these rats, which are more stress sensitive than normal strains (Leza et al., 1998; Mc Arthur et al., 1998). The resulting variable responses of the rats to stress can lead to significant variations in insulin and glucose metabolism, again tending to obscure experimental changes. These same considerations apply to the simpler IVGTT. In contrast to the IVGTT, which bypasses the complex gut hormone responses to food intake, a meal tolerance test (MTT) is much more sensitive to changes in insulin and glucose metabolism. We have developed a standardized MTT which minimizes stress on the rats and can be administered repeatedly to the same animal (Russell et al., 1999). The procedure uses a tail bleeding method to obtain blood samples without restraint and with minimal disturbance. A rat is accustomed to the procedure through an initial dummy procedure, usually done one week in advance. The animal is deprived of food over the light period (normally overnight if animals are kept on a reversed light/dark cycle). Two hours into the dark period, the rat is placed on a plate warmed to 37°C for 10 minutes to increase circulation through the tail. The animal is then held lightly by the base of the tail with one hand (the rat often trying to crawl under the upper arm of the experimenter). The tip of the tail is snipped off with a pair of fine scissors (the tail is pinched in the dummy procedure) and blood is milked from the tail into micro blood-collecting tubes (Microtainer, Becton Dickinson, Franklin Lakes, New Jersey, USA). An 0.8 ml blood sample is taken during each repeated sampling, and up to 1.5 ml can be collected at one time without distress. The rat is then returned to its cage and given a 5-gram pellet of its regular diet. The pellet is usually eaten within 15 minutes, and the clock is started when it is half consumed. Further tail bleed samples are taken at intervals, up to 120 minutes, and all are assayed for insulin and glucose. As shown in Figure 2, the rats maintain euglycemia under this protocol, but at the expense of a very large postprandial insulin response that can reach 1000 mU/L. The insulin response is short-lived and the plasma concentrations at 30 minutes provide an excellent index of insulin sensitivity, which is essentially the plasma insulin level required to control the glucose level. Treatments that lower insulin resistance result in a reduced 30-minute insulin concentration and may essentially prevent the postprandial response (Russell et al., 1998b, 1999). There is a well-established practice in metabolic studies to measure plasma lipid and insulin/glucose concentrations in the fasted state (in the rat, after a 16-hour starvation period). This results in relatively consistent values, but does not reflect the metabolic status of an animal in its normal day-to-day situation. In contrast, measurements on blood samples taken 2–3 hours into the light period, when the rat has eaten its first meal of the day, are much more representative of the animal’s ongoing metabolism. Plasma triglyceride and insulin levels will be significantly higher in the cp/cp rat, but glucose will not show any increase. The development of the hyperinsulinemia and hypertriglyceridemia as the juvenile cp/cp rat matures can be followed accurately in the fed state (Russell et al., 1998b), but is obscured in the fasted state.

THE JCR: LA-CP RAT

211

METABOLIC AND PATHOPHYSIOLOGICAL RESPONSES TO DRUGS An animal model of type 2 diabetes, such as the JCR:LA-cp rat, would have limited usefulness if its metabolism and pathological sequelae were invariant. One of the values of animal models is to enable the exploration of underlying mechanisms of disease and using the insights gained to develop new preventative or therapeutic treatments. To this end, we have conducted numerous studies of the effects of both pharmaceutical and other treatments on the metabolism and cardiovascular disease of the cp/cp rat. Fortunately, the JCR:LA-cp rat strongly mimics the obese insulin-resistant human in these studies, strengthening the justification for the use of the strain in screening or experimental verification of new antiobesity and insulin-sensitizing agents. The effectiveness of some agents in preventing end-stage cardiovascular disease in this animal model is highly encouraging in terms of our ultimate ability to discover effective treatments for the human population. The simplest treatments for the obesity/insulin resistance syndrome are food restriction and substantive levels of physical activity. Severely restricting the food of the cp/cp male rat (to 60% of the intake of a +/? control, or 12 g/day) is effective in reducing plasma insulin and triglyceride levels and is cardioprotective (Russell et al., 1990). However, this is a quite unrealistic and severe approach in humans, who, in any event, have greater choice in their behaviour. It is possible to induce rats to run voluntarily in running wheels through mild restriction of their food supply (Morse et al. 1995), and we have been able to achieve running of 6000– 8000 m/day for prolonged periods under this condition. Rats that run such distances, or are weight paired to runners, do show prevention of myocardial lesions (Russell et al., 1989a) and normalization of the vascular SMC (Absher et al., 1999). These results parallel reports of cardioprotection through exercise in humans, and our data suggest that the effects are related to a reduction in the hyperinsulinemia. The rats provide a model to further test these concepts experimentally. Alteration of the diet with the aim of reducing cardiovascular disease remains an attractive notion. Such strategies can be readily assessed in the cp/cp rat. For example, supplementation of the low-fat standard rat chow (less than 4% total fat and negligible cholesterol) with either olive oil or Ω-3 fatty acid-containing fish oil, has led to marked reductions in triglycerides, but no change in insulin levels or cardiovascular outcomes (Dolphin et al., 1988). In contrast, rats that drank 4% ethanol in the drinking water exhibited a 50% reduction in plasma insulin concentrations, greatly reduced hyperplasia of the islets of Langerhans, and a dramatic reduction in myocardial lesions (Russell et al., 1989c). These results provide confirmation of the well-established inverse correlation between moderate consumption of ethyl alcohol and cardiovascular disease in the human population. A number of pharmacological agents have been found to have either beneficial metabolic effects or antiatherosclerotic and cardioprotective effects in the cp/cp rat. The anorectic drugs benfluorex and Dfenfluramine reduce food intake and body weight, increase insulin sensitivity, reduce the seventy of vascular lesions and dysfunction, and are cardioprotective (Russell et al., 1997, 1998b). The highly effective inhibitor of triglyceride synthesis, MEDICA 16, strongly inhibits the development of insulin resistance and is antiatherogenic and cardioprotective (Russell et al., 1995, 1998a). Other compounds, such as captopril and probucol, are cardioprotective in the absence of any improvement in metabolism or reduction in vascular lesions (Russell et al., 1998c, d). These effects confirm that this animal model responds to a variety of interventions in ways that strongly parallel effects seen in the human disease state. POTENTIAL USES OF THE JCR:LA-cp RAT The underlying mechanisms leading to the insulin-resistant state remain elusive, although it is clear that, in the cp/cp rat, the insulin resistance is acquired after weaning and is not directly genetically controlled. The

212

J.C.RUSSELL AND S.E.GRAHAM

processes leading to the insulin resistance in the cp/cp rat are amenable to study and offer the prospect of understanding the slower sequences leading to insulin resistance in humans. Our group and others are pursuing this approach. The vasculopathy and atherosclerosis in the cp/cp rat of the strain are unique in animal models and provide a complementary model to the Watanabe Heritable Hyperlipidemic rabbit with its LDL receptor defect and hypercholesterolemic cardiovascular disease. The origins of the lesions in the JCR:LA-cp rat are clearly polygenetic and related to some components of the genome derived from the Koletsky Obese SHR. The identification of the genes responsible would provide an important indication of the currently unknown factors that lead to cardiovascular disease in humans. Regardless of the mechanisms responsible for the insulin resistance, there is a real interest in developing new pharmaceutical agents with insulin-sensitizing and/ or antiatherosclerotic or cardioprotective properties. The assessment of new candidate drugs requires an animal model that mimics the human disease state. The JCR:LA-cp rat is a unique model of the obesity/insulin resistance syndrome, possessing all of the disease characteristics seen in humans, including atherosclerosis and ischemic lesions of the heart. There is a growing interest in using this economical small-animal model in new drug screening and development programs, both for insulin-sensitizing and antiatherosclerotic compounds. AVAILABILITY AND SHIPPING At the time of writing, the rats are available in limited quantities from the original breeding colony at the University of Alberta. The address is: Department of Surgery 275 Heritage Medical Research Centre University of Alberta Edmonton, Alberta T6G 2S2 Canada (email: [email protected]) At the time of writing, commercial sales of rats are made at a cost of $125/ rat of any genotype. Shipping charges are on the order of $100 per shipment by guaranteed airfreight. In our laboratory, we have estimated the cost of maintenance of a cp/cp rat at $0.50/day. This is higher than normal as the cp/cp animals require more frequent cage changes and eat more food than do lean rats. The demand for these animals is growing as the strain is becoming increasingly accepted as a tool and model of the obesity/insulin resistance syndrome. Should demand become sufficiently great, planning for a commercial breeding operation would become feasible. Our laboratory has engaged in many collaborative experiments with other groups over the years. It has proven practical to both send animals to distant laboratories and to conduct the experimental work in our own laboratory and ship tissues or blood samples elsewhere. These collaborations have been highly productive and have permitted studies that extend beyond the abilities of any single research group. Long distance shipping of the rats requires much attention to detail. Airfreight operations vary widely between airlines. Some are unwilling to carry live animals or take them in transit from a connecting carrier. Some aircraft, including some of very recent design from Airbus Industrie, do not have sufficient installed heater capacity to allow for the winter shipment of live animals. Reliable shipment of these delicate and stresssensitive cp/cp rats involves prebooking the space and communication of the preassigned air waybill number to the recipient. It is important to avoid interlining (i.e., routing involving more than one carrier), if

THE JCR: LA-CP RAT

213

possible, as this can be a potent source of trouble, different carriers often having different procedures and policies. Shipments across international borders are best handled through specialized brokers, who can expedite Customs and veterinary clearance. All well-run laboratory animal centres have their own rules regarding the importation of animals. Generally, prior permission from the veterinary staff is required and is subject to certification of the health of the animals, especially as to their microbiological status. The presence of rat pinworm is common in rodent colonies. Its eradication is very difficult and animal units are wary of contaminated stock. The most effective procedure is to maintain pinworm-positive rats in isolation from others, which any modern facility should be able to achieve. REFERENCES Absher, P.M., Schneider, D.J., Baldor, L.C., Russell, J.C. and Sobel, B.E. (1999) The retardation of vasculopathy induced by attenuation of insulin resistance in the corpulent JCR:LA-cp rat is reflected by decreased vascular smooth muscle cell proliferation in vivo. Atherosclerosis, 143, 245–251. Absher, P.M., Schneider, D.J., Russell, J.C. and Sobel, B.E. (1997) Increased proliferation of explanted vascular smooth muscle cells: A marker presaging atherogenesis. Atherosclerosis, 131, 187–194. Amy, R.M., Dolphin, P.J., Pederson, R.A. and Russell, J.C. (1988) Atherogenesis in two strains of obese rats: The fatty Zucker and LA/N-corpulent. Atherosclerosis 69, 199–209. Dolphin, P.J., Amy, R.M., Koeslag, D.G., Limoges, B.E and Russell, J.C. (1988) Reduction of hyperlipidemia in the LA/ N-corpulent rat by dietary fish oil containing omega-3 fatty acids. Eiochim. Biophys. Acta, 962, 317–329. Dolphin, P.J., Stewart, B., Amy, R.M. and Russell, J.C. (1987) Serum lipids and lipoproteins in the atherosclerosis prone LA/N-corpulent rat. Biochim. Biophys. Acta, 919, 140–148. Emilsson, V., Lim, Y.L., Cawthorne, M.A., Morton, N.M. and Davenport, M. (1997) Expression of the functional leptin receptor in RNA in pancreatic islets and directly inhibitory action of leptin on insulin secretion. Diabetes, 46, 313–316. Gao, J., Sherman, W.N., McCune, S.M. and Osei, K. (1994) Effects of acute running exercise on whole body insulin action in obese male SHHF/Ncc-facp rats. J. Appl. Physiol., 77, 534–541. Greenhouse, D.D., Michaelis, O.E. IV, and Peterson, R.G. (1988) The development of fatty and corpulent rat strains. In New Models of Genetically Obese Rats for Studies in Diabetes, Heart Disease and Complications of Obesity, edited by C.T.Hansen, and O.E.Michaelis IV, pp. 3—6. National Institutes of Health, Bethesda. Haas, G.J., McCune, S.A., Brown, D.M. and Cody, R.J. (1995) Echocardiographic characterization of left ventricular adaptation in a genetically determined heart failure rat model. Am. Heart J., 130, 806–811. Hansen, C.T. (1983) Two new congenic rat strains for nutrition and obesity research. Fed. Proc., 42, 537. Koletsky, S. (1973) Obese spontaneously hypertensive rats: A model for the study of atherosclerosis. Exp. Mol. Pathol, 19, 52–60. Koletsky, S. (1975) Pathological findings and laboratory data in a new strain of obese hypertensive rats. Am. J. Pathol., 80, 129–142. Leza, J.C., Salas, E., Sawicki, G., Russell, J.C. and Radomski, M.W. (1998) The effect of stress on homeostasis in JCR:LA-cp rats. Role of nitric oxide. J. Pharmacol. Exp. Therap., 28, 1397–1403. McArthur, M.D., Graham, S.E., Russell, J.C. and Brindley, D.N. (1998) Exaggerated stress-induced release of nonesterified fatty acids in JCR:LA-corpulent rats. Metabolism, 47, 1383–1390. Morse, A.D., Hunt, T.W.M., Wood, G.O. and Russell, J.C. (1995) Diurnal variation of intensive running in fooddeprived rats. Can. J. Physiol. Pharmacol., 73, 1519–1523. O’Brien, S.F. and Russell, J.C. (1997) Insulin resistance and vascular wall function: Lessons from animal models (Review). Endocrin. Metab., 4, 155–162.

214

J.C.RUSSELL AND S.E.GRAHAM

O’Brien, S.F., McKendrick, J.D., Radomski, M.W., Davidge, S.T. and Russell, J.C. (1998) Vascular wall reactivity in conductance and resistance arteries: Differential effects of insulin resistance. Can. J. Physiol. Pharmacol., 76, 72–76. Ondera, T., Tamura, T., Said, S., McCune, S.A. and Gerdes, A.M. (1998) Maladaptive remodeling of cardiac myocyte shape begins long before failure in hypertension. Hypertension, 32, 753– 757. Pederson, R.A., Campos, R.V., Buchan, A.M.J., Chisholm, C.B., Russell, J.C. and Brown, J.C. (1991) Comparison of the enteroinsular axis in two strains of obese rats: The fatty Zucker and the JCR:LA-corpulent. Int. J. Obes., 15, 461–470. Poiley, S.M. (1960) A systematic method of breeder rotation for non-inbred laboratory animal colonies. Animal Care Panel, 10, 159–161. Recant, L., Voyles, N.R., Timmers, K.I., Bathena, S.J., Solomon, D., Wilkins, S., et al. (1989) Comparison of insulin secretory patterns in obese nondiabetic LA/N-cp and obese diabetic SHR/N-cp rats. Role of hyperglycemia. Diabetes, 38, 691–697. Russell, J.C., Amy, R.M., Graham, S. and Dolphin, P.J. (1993) Effect of castration on hyperlipidemic, insulin resistant JCR:LA-corpulent rats. Atherosclerosis, 100, 113–122. Russell, J.C., Amy, R.M., Graham, S.E., Dolphin, P.J., Wood, G.O. and Bar-Tana, J. (1995) Inhibition of atherosclerosis and myocardial lesions in the JCR:LA-cp rat by ββ9 tetramethylhexadecanedioic acid (MEDICA 16). Arterioscler. Thromb. Vase. Biol., 15, 918– 923. Russell, J.C., Amy, R.M., Manickavel, V., Ahuja, S.K. and Rajotte, R.V. (1987) Insulin resistance and impaired glucose tolerance in the atherosclerosis prone LA/N-corpulent rat. Arteriosclerosis, 7, 620–626. Russell, J.C., Amy, R.M., Manickavel, V. and Dolphin, P.J. (1989c Effects of chronic ethanol consumption in atherosclerosis-prone JCR:LA-corpulent rat. Arteriosclerosis, 9, 122–128. Russell, J.C., Amy, R.M., Manickavel, V., Dolphin, P.J., Epling, W.F., Pierce D., et al. (1989a) Prevention of myocardial disease in the JCR:LA-corpulent rat by running. J. Appl. Physiol., 66, 1649–1655. Russell, J.C., Bar-Tana, J., Shillabeer, G., Lau, D.C.W., Richardson, M., Wenzel, L.M., et al (1998a) Development of insulin resistance in the JCR:LA-cp rat: Role of triacylglycerols and effects of MEDICA 16. Diabetes, 47, 770–778. Russell, J.C, Dolphin, P.J., Graham, S.E., Amy, R.M. and Brindley, D.N. (1998b) Improvement of insulin sensitivity and cardiovascular outcomes in the JCR:LA-cp rat by D-fenfluramine. Diabetologia, 41, 380–389. Russell, J.C., Graham, S.E., Amy, R.M. and Dolphin, P.J. (1998c) Cardioprotective effect of probucol in the atherosclerosis-prone JCR:LA-cp rat. Eur. J. PharmacoL, 350, 203–210. Russell, J.C., Graham, S.E., Amy, R.M. and Dolphin, P.J. (1998d) Inhibition of myocardial lesions in the JCR:LAcorpulent rat by captopril. J. Cardiovasc. PharmacoL, 31, 971–977. Russell, J.C., Graham, S.E, and Dolphin, P.J. (1999) Glucose tolerance and insulin resistance in the JCR:LA-cp rat: Effect of miglitol (Bay m1099). Metabolism, 48, 701–706. Russell, J.C., Graham, S.E., Dolphin, P.J., Amy, R.M., Wood, G.O. and Brindley, D.N. (1997) Antiatherogenic effects of long-term benfluorex treatment in male insulin resistant JCR:LA-cp rats. Atherosclerosis, 132, 187–197. Russell, J.C., Graham, S. and Hameed, M. (1994) Abnormal insulin and glucose metabolism in the JCR:LA-corpulent rat. Metabolism, 43, 538–543. Russell, J.C., Koeslag, D.G., Amy, R.M. and Dolphin, P.J. (1989b) Plasma lipid secretion and clearance in the hyperlipidemic JCR:LA-corpulent rat. Arteriosclerosis, 9, 869–876. Russell, J.C., Amy, R.M., Michaelis, O.E., Mcune, S.M. and Abraham, A.A. (1991) Myocardial disease in the corpulent strains of rats. In Frontiers in diabetes research. Lessons from animal diabetes III, edited by E.Shafrir, pp. 402–407. London: Smith-Gordon. Russell, J.C., Manickavel, V., Koeslag, D.G. and Amy, R.M. (1990) Effects of advancing age and severe food restriction on pathological processes in the insulin resistant JCR:LA-corpulent rat. Diabetes Res., 53, 53–62. Schneider, D.J., Absher, P.M., Neimane, D., Russell, J.C. and Sobel, B.E. (1998) Fibrinolysis and atherogenesis in the JCR:LA-cp rat in relation to insulin and triglyceride concentrations in blood. Diabetologia, 41, 141–147.

THE JCR: LA-CP RAT

215

Triana, R.J., Suits, G.W., Garrison, S., Prazma, J., Brechtelsbauer, P.B., Michaelis, O.E., et al (1991) Inner ear damage secondary to diabetes mellitus. I. Changes in adolescent SHR/ N-cp rats. Arch. Otolaryn., 117, 635–640. Velasquez, M.T., Abraham, A.A., Kimmel, P.L., Farkas-Szallasi, T. and Michaelis, O.E. IV (1995) Diabetic glomerulopathy in the SHR/N-corpulent rat: role of dietary carbohydrate in a model of NIDDM. Diabetologia, 38, 31–38. Williams, G., Cardoso, H., Domin, J., Ghatei, M.A., Russell, J.C. and Bloom, S.R. (1990) Disturbances of regulatory peptides in the hypothalamus of the JCR:LA-corpulent rat. Diabetes Res., 15, 1–7. Williams, G., Shellard, L., Lewis, D.A., McKibbin, RE., McCarthy, H.D., Koeslag, D.G., et al. (1992) Hypothalamic neuropeptide Y disturbances in the obese cp/cp JCR:LA-corpulent rat. Peptides, 13, 537–540. Wu-Peng, X.S., Chua, S.C. Jr., Okada, N., Liu, S.-M., Nicolson, M. and Leibel, R.L. (1997) Phenotype of the obese Koletsky (f) rat due to Tyr763Stop mutation in the extracellular domain of the leptin receptor (Lepr). Diabetes, 46, 513–518. Yamini, S., Carswell, N., Michaelis, O.E. IV, and Szepesi, B. (1992) Adaptation in enzyme (metabolic) pathways to obesity, carbohydrate diet and to the occurrence of NIDDM in male and female SHR/N-cp rats. Int. J. Obes., 16, 765–774. Yen, T.T., Shaw, W.N. and Yu, P.C. (1977) Genetics of obesity in Zucker rats and Koletsky rats. J. Heredity, 38, 373–376. Zucker, L.M. and Zucker, T.F. (1961) Fatty: A new mutation in the rat. J. Heredity, 52, 275– 278.

13. THE NEONATALLY STREPTOZOTOCIN-INDUCED (nSTZ) DIABETIC RATS, A FAMILY OF NIDDM MODELS 1BERNARD

PORTHA, 1M.H.GIROIX, 1P.SERRADAS, 1J.MOVASSAT, 1D.BAILBE and 2M.KERGOAT

1Lab.

Physiopathologie Nutrition, CNRS ESA 7059, Universite Paris 71 D.Diderot, 2 place Jussieu, 75251 Paris Cedex 05, France. 2MERCK-LIPHA,

Centre de Recherche, 91380 Chilly-Mazarin, France

Syndromes resembling human diabetes occur spontaneously in some animal species. Alternatively they can also be induced by treating animals with drugs or viruses, excising their pancreases or manipulating their diet. Of course, none of the known animals models can be taken to reproduce human diabetes, but they are believed to illustrate various types of aetiological and pathogenic mechanisms that most probably also operate in humans. Among these models, diabetes induced in rats by neonatal streptozotocin administration (the so called n-STZ models) has been recognized during the last two decades, as adequate tools to study the long-term consequences of a gradually reduced B-cell mass [Weir et al., 1986; Grill and Östenson, 1988; Portha et al., 1990]: we and others have found that defects in insulin secretion and action which in many ways resemble those described in human NIDDM, develop in these n-STZ models. This present review aims to decribe the methods necessary to generate the n-STZ rodents, to sum-up the informations so far collected in this family of diabetic models, and to highlight their potential as well as their limitations for future research. NEONATAL STREPTOZOTOCIN MODELS (n-STZ) The diabetic syndrome in this model was described for the first time after injecting rats (Sherman or Wistar strains) on the day of their birth (n0=birth) i.v. (saphenous vein) or i.p. with 100 mg/kg streptozotocin (STZ) [Portha et al., 1974]. STZ is a 2-deoxymethyl-nitrosoureaglycopyranose molecule that produces a selective toxic effet in βcells and induces diabetes mellitus in most adult laboratory animals [Rerup, 1970]. Although the exact mechanism of its toxicity is still a matter of debate, one proposed site for the action of STZ is the nuclear DNA. During decomposition of STZ, highly reactive carbonium ions are formed, which cause alkylation of DNA bases {LeDoux et al., 1986}. In the following phase of excision DNA repair, the nuclear enzyme poly

N-STZ DIABETIC RATS

217

Figure 1 Evolution of STZ effect in the Wistar rat neonate after one injection (100 mg/ kg) on day of birth. The values for the STZ-injected animals are expressed as a percentage of those observed in control animals which were killed on the same day. See text for comments on pathogenic progression toward NIDDM in the adults.

(ADP-ribose) synthetase becomes activated to such an extent that cellular levels of its substrates NAD become critically depleted, leading to cell death [Okamoto, 1981], The neonatal rats treated with STZ at birth exhibit an insulin-deficient acute diabetes 3–5 days after birth (Figure 1). It was confirmed that they are diabetic during this period by checking the following criteria: plasma glucose is high (345+37 mg/dl), pancreatic insulin stores show a 93% decrease, plasma insulin is low considering the high glucose level, plasma glucagon is high despite unchanged pancreatic glucagon content [Portha et al., 1974; Portha et al, 1979]. All the surviving pups can be easily kept (mortality was <30%) to adulthood. It is remarkable that the marked hyperglycemia observed in the neonates following STZ is only transient (Figure 1) [Portha et al., 1974; Portha et al., 1979]. This may explain why some authors [Junod et al., 1969] unduly reported that neonatal rats were resistant to STZ. By the end of the first postnatal week, plasma glucose and insulin values no longer differ significantly from those of controls. Such a transitory nature of overt diabetes is a unique characteristic of STZ diabetes in the newborn rat (n0STZ) as compared with STZ diabetes induced in adult rat. At 3–4 weeks of age, i.e. at weaning, body weight and basal plasma glucose values in the n-STZ rats cannot be distinguished from control values. However by Address correspondance and reprint requests to: Prof. B.Portha, Lab. Physiopathology of Nutrition, CNRS ESA 7059, Universite Paris 7/D. Diderot, 2 place Jussieu, Tour 33 75251 Paris Cedex 05, France. Tel: +33 1 44 27 50 11; Fax: +33 1 44 27 78 91; email: [email protected]

218

B.PORTHA ET AL.

8 weeks of age and thereafter, n0-STZ rats showed mild basal hyperglycemia (150–180 mg/dl), abnormal responses to i.p. or i.v. glucose tolerance tests and a 50% decrease in pancreatic insulin stores without change in pancreatic glucagon stores (Figure 1) [Portha et al., 1979]. An interesting but more severe variant of this model has been reported by Bonner-Weir and Weir [Bonner-Weir et al., 1981; Weir et al., 1981]. Sprague Dawley pups are injected i.p. on day 2 after birth with 90 mg/kg STZ (n2-STZ). Only n2-STZ treated pups with plasma glucose values of 275 mg/dl or higher are selected and allowed to be raised. By 6 weeks of age these animals show a marked basal hyperglycemia (>200 mg/dl) and abnormal glucose tolerance. Mild hypoinsulinemia also becomes established by this time [Bonner-Weir et al., 1981]. Since the studies of Bonner-Weir et al. differed from ours not only with respect to different times of STZ administration but also with regard to the strain of rat employed, one cannot discern whether the difference in subsequent severity of diabetes is due to a difference in the timing of exposure to the toxin or to a difference in strain-related capacity for spontaneous remission. In view of this heterogeneity, we compared Wistar rats made diabetic with a STZ injection on day 2 (n2STZ version) and on day 5 after birth (n5-STZ version). As a consequence of such an approach, three models exhibiting non-insulin-dependent diabetes with graded severity, were obtained in the adult rat. The version n2-STZ exhibited characteristics (growth, basal plasma glucose and insulin levels, lack of insulin release in response to glucose in vivo, glucose intolerance, depletion of the pancreatic insulin stores) which are very similar to those obtained in the version n0-STZ [Portha et al., 1974; Portha et al., 1979]. By contrast the version n5-STZ was shown to exhibit a frank basal hyperglycemia with glucose intolerance, a raised glycated haemoglobin, a strong reduction of the pancreatic insulin stores, a decreased (50%) basal plasma insulin level and a lack of plasma insulin response to glucose in vivo. The development and progression of the hyperglycemia in the Wistar n5-STZ version demonstrated many similarities to those described by others [Levy et al., 1984;Schaflfer et al., 1985; Grill and Rundfeldt, 1986; Grill et al., 1986; Fantus et al., 1987} using the procedure described by Bonner-Weir et al. [1981] i.e. the i.p. administration of 90 mg/kg STZ on day 2 of age in female Sprague-Dawley rats: the onset of hyperglycemia occurs after 4 weeks of age; a consistent decrease in the non-fasting plasma insulin levels is found from this time; the pancreatic insulin stores are markedly decreased to 11% of control values (10 weeks of age); the perigonadal and retroperitoneal adipose tissue masses and adipocyte volumes are significantly smaller at 10 weeks of age [Blondel et al., 1989]. At adult age male n2-STZ rats are more severely affected than females [BonnerWeir et al., 1981]. Such difference is not related to sex-related dissimilarity in B-cell susceptibility to STZ during the neonatal period. It is rather the rise in androgens starting with puberty which is, at least partly, responsible for the more severe diabetic state in the males [Ostenson et al, 1989; Iwase et al, 1996]. Note that in mice also (NMRI strain), neonatal (day 2 and day 3 after birth) STZ (150 mg/kg, I.P.) has been reported to result in a permanent diabetes with impaired glucose-induced insulin secretion in vivo [Ahren and Skoglund, 1989]. The advantage of these n-STZ rodent models are several fold. First, they provide interesting models for the study of β-cell growth since β-cells partially regenerate during the spontaneous remission occuring after over neonatal diabetes. Second, they allow to study the long-term consequences of a gradually reduced Bcell mass. We and others [Weir et al., 1981; Levy et al., 1984; Schaffer et al, 1985; Grill and Rundfeldt, 1986; Blondel and Portha, 1989] have found that defects in insulin secretion and action develop in these nSTZ models, which in many ways resemble those described in human diabetes, and the informations from these models have provided new insights into the pathogenesis of NIDDM. These informations will be briefly summarized thereafter.

N-STZ DIABETIC RATS

219

β-CELL FUNCTION IN n-STZ MODELS Insulin Secretion in Response to Glucose and Non-glucose Secretagogues We first showed that the n0-STZ rats when adult are chaacterized by a low insulin release in vivo in response to glucose or aminoacids [Portha et al., 1979]. Because the B-cell number and the insulin stores in the pancreas of these diabetic rats were low, the defective insulin response observed in vivo could be attributed to these quantitative abnormalities of the islets. Additionally, alterations of the β-cell responsiveness to the stimuli might be present and be variable according to the nature of the stimulus. Insulin secretion studies were carried out primarily in 10–16 week-old n0-STZ rats using the isolated perfused pancreas technique [Giroix et aL, 1983]. Insulin response to glucose stimulation over the range 5. 5–22 mM was lacking, thus indicating loss of β-cell sensitivity to glucose. In contrast, glyceraldehyde elicited an insulin release as high as that obtained in the control pancreases. Mannose stimulated insulin secretion less in diabetics than in controls. The insulin secretion obtained in response to isoproterenol indicated that the ability of adenylate cyclase to generate cAMP in the β-cells of the diabetic was not decreased. In the absence of glucose, β-cells of diabetics were unexpectedly hypersensitive to arginine and leucine, and also the insulinotropic action of acetylcholine was increased as compared with controls [Giroix et al., 1983; Kergoat et al., 1987]. While some disagreement exists about quantitative results obtained in the different diabetic versions (Wistar n5-STZ, Sprague-Dawley n2-STZ) [Giroix et al., 1983; Levy et al., 1984; Fantus et al., 1987; Okabayashi et al., 1989], there is universal agreement that the β-cells in the n-STZ models are essentially insensitive to glucose but retain their responsiveness to non glucose secretagogues. Not only the response of β-cells to glucose alone was abnormal, but also the glucose potentiating effect on non-glucose secretagogues was lost: in control pancreas the response to arginine 19 mM in the presence of low glucose concentration was much smaller than that found at higher levels; in marked contrast, the insulin responses in n0-STZ rats were similar at both low and high glucose concentrations [Kergoat et al., 1987]. An unexpected result first reported in the Sprague Dawley n2-STZ model [Leahy and Weir, 1985] has been the discovery that insulin release also fails to fall appropriately when the glucose concentration is reduced: a paradoxical stimulation of insulin release (“off response”) is in fact elicited under these conditions. Similar observations have been obtained in the Wistar n5-STZ model [Serradas et al., 1991], but not in the Wistar n0-STZ model. In islets isolated from n0-STZ rats, the β-cells have been found to exhibit poor sensitivity to glucose [Halban et al, 1983; Portha, 1985]. However, it is clear that a high glucose concentration was able to trigger a significant release from the diabetic rat islets, in contrast to findings with perfused pancreas [Giroix et al., 1983]. Such a discrepancy between the perfused pancreas and isolated islets was also reported in the Sprague-Dawley n2-STZ model [Halban et al., 1983]. One possible explanation for this intriguing observation could be that, when islets are studied during static incubations, there is a higher intra-islet concentration of glucagon than in islets which have an intact vascularisation and are perfused in vivo or in vitro [Bonner-Weir and Orci, 1982]; a raised intracellular cyclic AMP concentration induced by glucagon would then be responsible for the stimulatory effect of glucose on insulin release. Nevertheless, one cannot presently eliminate the occurrence of an abnormal selection of islets after collagenase digestion of the pancreas from diabetic rats, as possible explanation for the difference between islet and perfused pancreas data. In islets of n0-STZ rats, the insulin biosynthesis (measured by the incorporation of 3H-phenylalanine into immunoprecipitable proinsulin material) also appeared to be less stimulated by glucose than in control islets [Portha, 1985; Portha et al, 1988].

220

B.PORTHA ET AL.

In the n2-STZ and n5-STZ diabetic rats, it has also been reported that the ratio of secreted amylin to insulin, is dramatically increased compared to non-diabetics [Inoue et al., 1992]. To what extent the higher proportion of amylin in secretory granules might participate in the mechanism of β-cell dysfunction is not known. The abnormal secretion found in the n-STZ models bears a resemblance to the insulin secretory characteristics found in human diabetes. It is noteworthy however that in the perfused rat pancreas both first and second-phase insulin responses are severely impaired, whereas in human NIDDM there is usually significant preservation of second-phase release. The reason for this difference is unknown. Incompetence of β-cells to Glucose in n-STZ Models: Intracellular Mechanisms The reasons for the glucose incompetence referred to above in β-cells of n-STZ rats are presently only partially understood. In the normal β-cell, most evidences suggest that an increased rate of glucose metabolism is necessary for initiation of glucose-stimulated insulin release [Malaisse, 1991]. Based on the observation that the n-STZ models have a reduction in the amount of GLUT-2 glucose transporter protein in their β-cells [Thorens et al., 1990], it has been proposed that impaired glucose entry into pancreatic β-cells causes the deterioration of glucose-induced insulin secretion. Our measurements of 3–0-methyl-glucose uptake by intact islets cells of n0-STZ diabetic rats, support the view that there is indeed a defect of the β-cell hexose transport system [Giroix et al., 1992; Giroix et al., 1992]. However, such a conclusion needs to be considered with caution. First, our results could reflect a change in the relative abundance of β and non-β islet cells since the fraction of the islets mass occupied by insulin-producing cells is indeed decreased by 50% [Giroix et al., 1992]. Second, it is not obvious that the anomaly of hexose transport plays any major role in the impaired secretory response of the β-cell to glucose in the n0-STZ model, since it has been repeatedly emphasized that the efficiency of glucose transport should be decreased by at least one or two orders of magnitude to become a rate-limiting step in glucose catabolism [Malaisse, 1991]. The finding that during 90–120 min incubation the overall rates of glucose utilization are little or not affected in the islets of diabetic rats [Portha et al., 1988], further suggests that the anomaly of hexose transport does not result, under steady-state conditions, in any marked change in the intracellular glucose concentration. The preferential alteration of the β-cell response to glucose in the n0-STZ model is not attributable to either a decrease in hexokinase and glucokinase activities (islet homogenates) or an altered binding of these isoenzymes to mitochondria [Giroix et al., 1990]. In the case of the high Km glucokinase, the sole perturbation in the diabetic animals resided in an apparently decreased affinity for glucose [Giroix et al., 1990]. These findings emphasize the view that the transport of glucose across the plasma membrane and its subsequent phosphorylation by glucokinase should not be considered as significantly responsible for the impairment of the functional response to glucose of the diseased β-cells in the n0-STZ model. Our results indicate that the rate of total glycolysis, as judged from the production of 3H2O from [5-3H] glucose, is not significantly different in islets from n0-STZ and control rats [Portha et al., 1988; Giroix et al., 1990], at least when related to the protein or DNA content of the islets. By contrast, the oxidative response to a high concentration of glucose was severely affected, especially in terms of [6-14C]glucose oxidation, an estimation of the oxidation of the glucose-derived acetyl residues in the Krebs cycle. Thus, the ratio [6–14C]glucose oxidation/ [5–3H]glucose utilization was much less markedly increased in response to a rise in hexose concentration [Giroix et al., 1990; Giroix et al., 1993]. This metabolic defect contrasts with a close to normal or even increased capacity of n0-STZ islets to oxidize [6–14C]glucose, [2–l4C]pyruvate, [U– 4C]glutamine and [U–l4C]leucine at low, non-insulinotropic concentrations of these substrates and a lesser impairment of the oxidation of [U-14C]leucine tested at a high concentration [Giroix et al., 1990].

N-STZ DIABETIC RATS

221

We have shown that in mitochondria of n0-STZ islets there was a severe decrease in both the basal (no Ca2+) generation of 3H2O from [2–3H]glycerol-3-phosphate and in the Ca2+ induced increment in [3H] glycerophosphate detritiation [Giroix et al., 1991; Giroix et al., 1992]. This coincided with the fact that a high glucose concentration failed to increase significantly [2–3H]glycerol conversion to 3H2O in intact islets from n0-STZ rats, in contrast to the situation found in nondiabetic rats [Giroix et al., 1992]. The n0-STZ mitochondria were also less efficient than those of control animals in generating 14CO2 from [1–14C]-2ketoglutarate. However, the alteration of the 2-ketoglutarate dehydrogenase was less marked than that of the FAD-linked glycerophosphate dehydrogenase and a normal to slightly elevated glutamate dehydrogenase activity was found in n0-STZ islet mitochondria [Giroix et al., 1991 ; Giroix et al., 1992]. Therefore it is suggested that in the n0-STZ model, the impairment of glucose-induced insulin secretion is, at least in part, due to a deficiency in the activity of mitochondrial FAD-linked glycerophosphate dehydrogenase, leading to an altered transfer of reducing equivalents into the mitochondria by the glycerol phosphate shuttle. We have also investigated the possibility that a defect on the islet cAMP production and/or in the islet phosphoinositide metabolism could be involved in the failure of the glucose-induced insulin secretion in the n0-STZ model [Dachicourt et al., 1996; Morin et al., 1996; Morin et al., 1997]. Concerning the first possibility we have shown that while there is no major alteration of the functionality of the adenylate cyclase/phosphodiesterase/cAMP system in the n0-STZ β-cells (in term of its modulation by forskolin, glucagon, pertussis toxin, glucagon, GIP, GLP-1 or IBMX), there is a defective glucose-induced cAMP generation [Dachicourt et al., 1996] that could be explained by a block in the step(s) linking glucose metabolism and activation of adenylate cyclase and/or the lack of expression of the Golf protein isotype despite increased expression of the ACII and ACIII adenylate-cyclase isoforms [Frayon et al., 1999]. Concerning the second possibility, we have reported that the glucose-induced polyphosphoinositide (PPI) hydrolysis was severely diminished in the n0-STZ β-cells. We have proposed that a reduced phosphatidylinositol kinase activity, concomitantly with a decreased Ca2+-stimulated phospholipase C activity, may participate to the alteration of the phosphoinositide pathway, the limitation of the inositol phosphate production and finally the impairment of the glucose-induced insulin release [Morin et al., 1996; Morin et al., 1997]. Is the function of ionic channels defective in the n-STZ β-cell? The closure of the K+–ATP channels is now widely recognized to be a key step in the mechanism of insulin secretion in response to glucose. The resultant depolarization allows Ca2+ entry through voltage-dependent Ca2+ channels, thereby triggering insulin secretion. Recently Tsuura et al. [1992] have studied the properties of the K+-ATP channels in single β-cells of n-STZ diabetic rats using the patch-clamp technique. The unitary conductance of the channel in the diabetic rat β-cells was virtually identical to that in control β-cells and there was no difference in the sensitivity of K+-ATP channels to ATP and glibenclamide between the diabetic and control groups. In response to glucose, the activity of the K+-ATP channels was diminished in a concentration dependent manner in both control and diabetic intact cells. However, the inhibition of the K+-ATP channels in intact βcells of n-STZ rats was significantly less than that in control cells. Even in the presence of high glucose, the openings of a few single K+-ATP channels were consistently observed in cell-attached patch membranes of diabetic, but not control β-cells. Finally, it appears that the impaired insulinotropic action of glucose in βcells of n-STZ rats is associated with a reduced sensitivity of the K+-ATP channel to extracellular glucose, but not to intracellular ATP. These conclusions fit very well with our observation of an impaired reduction of the 86Rb efflux when perifused n0-STZ islets are exposed to glucose [Giroix et al., 1993]. Therefore, the glucose insensitivity of the K+-ATP channels in n-STZ β-cells may be the result of an insufficient amount of ATP production caused by impaired glucose metabolism in diabetic β-cells, rather than by a defect in the

222

B.PORTHA ET AL.

K+-ATP channel itself. Within the frame of this hypothesis we have found in n0-STZ islets that the ATP/ ADP ration at high glucose concentration was lower than in non-diabetic islets [Giroix et al., 1993]. Using a complementary approach aimed to understand the diminished oxidative capacity of the diabetic βcell, new information related to the expression of the mitochondial genome in the n0-STZ β-cells has recently been gained. The contents of mitochondrial cytochrome b mRNA and mitochondrial 12S rRNA were lower (50–30%) in n0-STZ islets as compared with non-diabetic Wistar rats [Welsh et al., 1991]. The content of mitochondrial DNA (Southern blotting) was markedly decreased (70%) in islets from n0-STZ rats, but no deletion was detected [Welsh et al, 1991]. Therefore, since a lower mitochondrial RNA content may result in a diminished oxidative capacity, it is conceivable that this deficiency may contribute to insulin deficiency in the n0-STZ models. Is there a glucose/glucose-6-phosphate futile cycling in the β-cell? An additional explanation for the preferential defect of the β-cell secretory response to glucose in the n0-STZ rats has been proposed by Kahn et al. [1990]. It is related to the induction of glucose-6-phosphatase activity in n0-STZ islets resulting in an ATP-wasting futile cycling between glucose and glucose-6-phosphate. In islets from normal rats, such futile cycling is negligible. We were unable to confirm Khan’s observations since we found that the activity of islet glucose-6-phosphatase is equally low in diabetic and control rats [Giroix et al., 1992]. Are sympathetic and parasympathetic influences on the β-cell impaired? The parasympathetic nervous system exerts an important influence on insulin secretion. The receptors mediating the cholinergic effect are muscarinic in nature since acetylcholine-induced insulin secretion is readily blocked by atropine. We [Giroix et al., 1983; Kergoat et al, 1987] and others [Bonner-Weir et al, 1988; Grill and Östenson, 1988] have found a significantly enhanced response to acetylcholine in n-STZ rats when insulin release is tested using islets or perfused pancreas. This was associated with an increase of muscarinic receptors in the pancreatic islets as proposed by Östenson and Grill [1987]. Moreover insulin treatment of the n-STZ rats for three days lowered blood glucose, diminished binding of [3H]methylscopolamine and abolished the insulin secretory hyperresponse to carbamylcholine. Since, when normal islets had been kept in tissue culture for 3 days, binding was higher when in the culture medium a high rather than a low concentration of glucose was included [Östenson and Grill, 1985], these authors have suggested that in the n-STZ model, the hyperglycemia in vivo is a determinant of the number of muscarinic receptors in β-cells and that such regulation is associated with the increase in the cholinergic induced insulin secretion. We have also tested the notion that increased α-adrenergic receptor activity could contribute to suppress the glucose-induced insulin secretion in the perfused pancreas of n0-STZ rats by the addition of the selective α-adrenergic blocking agent UK-14304 to the perfusion medium. In the diabetic pancreas, the inhibitory effect of UK-14304 upon maximally stimulated-insulin release (by a glucose/arginine mixture) was significantly higher than that exerted on the non-diabetic pancreas, thus suggesting that the reactivity of the B-cell to sympathetic neural activation is enhanced in the n0-STZ model. Such an assumption is also supported by the data from Östenson et al [1989] and by the demonstration by Kurose et al [1992] that electrical splanchnic nerve stimulation decreases insulin secretion (perfused pancreas) more efficiently in nSTZ rats than in normal rats. Finally, it is interesting to note that α-cells and δ-cells in the n-STZ models have been found insensitive to some modalities of glucose regulation (despite the lack of any significant changes in glucagon and somatostatin stores), suggesting that incompetence to glucose is qualitatively similar for glucagon, somatostatin and insulin secretion [Giroix et al, 1984; Weir et al, 1986; Östenson et al, 1990; Giroix et al, 1992].

N-STZ DIABETIC RATS

223

Loss of Glucose-induced Insulin Release in n-STZ Models: Is it Reversible? The available evidences related to the etiology of the inability of β-cells to respond to glucose (the so-called glucose incompetence) in the n-STZ model, are contradictory. It has been first proposed that it is irreversibly damaged in the n-STZ rats: this was based on experiments in which islets of adult Wistar-Furth n0-STZ rats were transplanted under the kidney capsule of syngeneic normal or diabetic recipients and subsequently tested (perfusion of the graft-bearing kidneys) for glucose-induced insulin release [Inoue et al., 1994]. Against such a view, we have brought several convergent arguments which allow to exclude the possibility that the pathogenesis of loss of response of β-cells to glucose in n-STZ models can be entirely explained by a permanent toxic action of STZ. We have directly evaluated the possibility that STZ treatment per se causes an irreversible impairment of the insulin response to glucose by testing the insulin response at intervals after STZ treatment, from day 1 to day 21 using perifusion of pancreatic fragments [Portha and Kergoat, 1987]. While the glucose-induced insulin release was completely abolished on day 1 after STZ, it could be demonstrated after day 3. Moreover, it increased as a function of age. This restoration of the insulin response to glucose closely paralleled the recovery of pancreatic insulin stores. In sharp contrast with the lack of glucose response obseved in vitro in the adult, glucose-induced insulin release was still detected on day 21 [Portha and Kergoat, 1987]. Accordingly, these data support the notion that the pathogenesis of the β-cell lesion in n0-STZ rats is caused mainly by factors other than a primary permanent toxic action of STZ. Then, the sustained glucose load acting on a reduced β-cell mass (the so called glucotoxicity) could represent one among these factors leading to the functional abnormalities of the β-cells in this diabetic model. To test this hypothesis, n0-STZ islets were cultured five days at 5.5 or 11 mM glucose to determine whether their β-cell derangements could be modified by changing the environmental conditions. The insulin release and the (pro)insulin biosynthesis, measured either in basal or stimulated states, were then found to be similar in the islets of diabetic rats and controls after the 5.5 mM glucose culture period. By contrast, after the 11 mM glucose culture period the insulin release and the (pro)insulin biosynthesis in islets of diabetic rats were found significantly less stimulated by 16.5 mM glucose than in control islets [Portha, 1985]. According to these data and those of Welsh and Hellerström using a very similar protocol [1990], it seems likely that chronic hyperglycemia is responsible for these defects. However, reservation must be kept in mind, that isolated islets may not behave like islets in vivo, particularly after a period of tissue culture. A different experimental approach to determine whether the glucose defect could be counteracted was designed by normalizing the diabetic state in n0-STZ rats by insulin therapy [Kergoat et al., 1987]. Mixte lente insulin (5 U.kg−1, day−l) was given daily over one or five consecutive days. Insulin secretion was studied the morning after the last insulin injection with the isolated perfused pancreas preparation. Basal plasma glucose levels decreased in diabetic rats from 183 ± 8 to 136 ± 10 mg/dl after the 5-day insulin treatment (vs. 116 ± 3 mg/dl in control rats). Although the 1-day insulin treatment did not modify the lack of glucose response in diabetic rats, the 5-day insulin treatment improved their glucose-induced insulin secretion. Moreover, insulin therapy improved the priming effect of glucose on a second stimulation with glucose. The return of this glucose effect was hardly detectable after the 1-day insulin therapy but was clearly present after the 5-day treatment. The hyperresponse to arginine characteristic of the untreated diabetic rats, returned similar to that in controls after a 1-day insulin therapy, and it was again amplified at high glucose levels, although amplification remained lower than that of control rats. This indicates that the potentiating effect of glucose on the response to arginine was regained more precociously than the acute insulin response to glucose after insulin therapy. Nevertheless, the improvement of the β-cell secretory function in the n0STZ rats remained strictly dependent upon correction of the hyperglycemic-hypoinsulinemic pattern by

224

B.PORTHA ET AL.

insulin treatment, since 10 days after exogenous insulin impaired glucose-stimulated insulin secretion and hyperresponse to arginine-stimulated insulin secretion were back. However, these results do not allow the distinction whether the abnormality of the insulin secretion is due to high levels of glucose per se or to other factor(s) associated with the diabetic state. It was therefore investigated whether a long-lasting in vitro glucopenia during perfusion of pancreas could restore β-cell sensitivity to glucose in n0-STZ and n5-STZ rats [Portha et al., 1990]. In non-diabetic rats, after a 50 min perfusion with a medium containing no glucose, the integrated insulin response to a subsequent stimulation with 16 mmol/1 glucose was not significantly different from that obtained after a 20 min glucose omission period, this last condition representing our current procedure in the pancreas perfusion experiments. Conversely, in the n0-STZ rats, the incremental insulin response to 16 mmol/1 glucose after 50 min glucose omission, was enhanced 5-fold relative to the response obtained 20 min after isolation. Moreover, the normal biphasic pattern of response to glucose was restored [Portha et al., 1990]. The effect of glucose omission seems specific for the diabetic pancreas since glucose omission failed to exert a significant effect on the subsequent response to glucose in pancreas from non-diabetic rats. Finally, we have recently shown that the glucose-incompetent n0-STZ β-cells can be indeed rendered glucose competent by in vitro GLP-1 exposure [Dachicourt et al., 1997] and more generally by artificially raising their intracellular cAMP [Dachicourt et al., 1996]. Accordingly, in the adult n0-STZ rats, we are inclined to exclude the possibility that the pathogenesis of the β-cell incompetence to glucose can be entirely explained by a permanent toxic action of STZ since the βcells in adults are mostly regenerated β-cells resulting from a neogenesis process taking place at a time when STZ is no longer present in the body fluids. We have directly evaluated the possibility that the STZ treatment per se causes a chronic impairment of the insulin response to glucose by testing the insulin response at intervals after STZ treatment from day 1 to day 21 using perifusion of pancreatic fragments. While the glucose-induced insulin release was completely abolished (2% of the normal response) on day 1 after STZ, it could be demonstrated after day 3. Moreover, it increased as a function of age (6% and 36% of the normal response, on day 5 and day 14, respectively). This restoration of the insulin response to glucose closely paralleled the recovery of pancreatic insulin stores (6% and 51% of normal values, on day 5 and day 14, respectively). In sharp contrast with the lack of glucose response observed in vitro in the adult, glucoseinduced insulin release was still detected on day 21. Accord ingly, these data support the notion that the pathogenesis of the β-cell lesion in n0-STZ rats is caused mainly by factors other than a primary toxic effect of STZ. Then, the sustained glucose load acting on a reduced β-cell mass (glucotoxicity) could represent one among these factors leading to the functional abnormalities of the β-cells in this diabetic model. Nevertheless, as far as the reversibility of the β-cell secretory lesion is concerned, different conclusions have been reported in the n2-STZ and n5-STZ models: the defective glucose-induced insulin secretion in the Sprague-Dawley n2-STZ rat was not restored by insulin treatment [Leahy et al., 1985; Grill and Rundfeldt, 1986] (only the ability of glucose to potentiate arginine-induced insulin release was normalized). We also failed to improve significantly the glucose-induced insulin secretion in the Wistar n5-STZ model by chronic insulin treatment [Serradas and Portha, unpublished results]. Moreover, in n5-STZ rats, the incremental insulin release in response to 16 mM glucose after the 50 min glucose omission period remained unchanged as compared with the 20 min period of glucose omission [Portha et al., 1990; Serradas et al., 1991]. We are left without an explanation as to why the defective glucose-induced insulin secretion in the n5-STZ model cannot be reversed. It is, however, pertinent to notice that the basal plasma glucose level was consistently more elevated in the n5-STZ rat (17 mM) than in the n0-STZ rat (7 mM). In the case of the n5-STZ Wistar and the n2-STZ Sprague-Dawley models, a circumstantial explanation cannot be eliminated as it is not possible to exclude a chronic STZ effect on the remaining islets, besides the acute destruction of

N-STZ DIABETIC RATS

225

β-cells. In vitro findings in the adult mouse islets suggest that after a cytotoxic STZ injury there remains a population of partially damaged β-cells with a severely impaired ability to recognize glucose as a stimulus [Eizirik et al., 1988]. Such a possibility in the n5-STZ rats is in fact supported by the lack of significant βcell regeneration in their pancreases during the 2 weeks following the β-cell insult. In contrast, in the n0STZ Wistar rats, spontaneous β-cell regeneration after the STZ exposure enables a partial replenishment of the β-cell mass by newly formed β-cells [Cantenys et al., 1981; Portha et al., 1990; Wang et al., 1994; Movassat et al., 1997]. INSULIN ACTION IN n-STZ MODELS It is recognized that a severe reduction in β-cells, as obtained in subjects with Type I diabetes or animals after alloxan or STZ injection, is associated with resistance of target tissues to the action of insulin [Reaven et al., 1977; De Fronzo et al., 1982; Bevilacqua et al., 1985]. However at the time the n-STZ models were introduced, there was no clear answer to the question whether or not a more modest reduction in β-cell mass was associated with insulin resistance and on a more general background, whether or not a primary reduction of the β-cell mass necessarily leads to the development of insulin resistance. In adult n0-STZ females, we have shown that the hepatic glucose production measured in the basal state was higher in the diabetics than in controls despite similar peripheral insulin levels in both groups. The factors responsible for maintaining this elevated rate of basal hepatic glucose production in the n0-STZ rats remain to be identified. That this could be a reflection of an hepatic insulin resistance was eliminated since the liver of the n0-STZ diabetic rats was in fact hyperresponsive to submaximal insulin levels [Melin et al., 1991]. During the clamp studies, the sensitivity of the liver to insulin’s suppressive effect on glucose production was enhanced [Kergoat and Portha, 1985]. An abnormality in the mechanisms of suppression of hepatic glucose production in diabetic rat by hyperglycemia can be implied and it is also possible that other factors are operative that stimulate hepatic glucose production in diabetics by increasing glycogenolysis and/ or gluconeogenesis. The properties of the liver insulin receptor in the n0-STZ model have been studied [Portha et al., 1983; Kergoat et al., 1988; Melin et al., 1991]: liver insulin receptors are not upregulated and the tyrosine kinase activity remains unaffected, suggesting that the increased insulin effect in the liver of n0STZ rats is probably distal to the insulin receptor kinase. The insulin-mediated glucose uptake by the whole body as estimated during the clamp studies was found to be normal in n0-STZ rats [Kergoat and Portha, 1985]. In fact data related to glucose utilization in vivo by individual peripheral tissues show that even if the insulin action is not increased in the whole-body, this effect could indeed be greater in some tissues [Kergoat et al., 1991]: this is the case at the level of the white adipose tissue (paraovarian and inguinal) and at the level of the brown adipose tissue. In these tissues, an increased sensitivity and an increased responsiveness to insulin action are detected when comparing diabetic females to control ones. By contrast, insulin action is found normal in skeletal muscles and diaphragm of the same adult diabetic females. This observation of a normal insulin action in muscles together with an enhanced insulin action in the liver and white and brown adipose tissues, indicates that glucose is preferentially channeled towards the liver and adipose tissues in the n0-STZ females. Finally in the n0-STZ model, we were unable to demonstrate any insulin resistance. In the n2-STZ rats, glucose utilization rate induced by hyperinsulinemia was found normal, whereas the hepatic glucose production rate was significantly higher in the basal state, but it was normally suppressed by hyperinsulinemia [Blondel et al., 1989]. The data obtained in the basal state under postabsportive conditions could be interpreted as suggesting a discrete alteration of hepatic insulin action. However, one cannot

226

B.PORTHA ET AL.

presently eliminate the possibility that an increased glucagon secretion or some indirect mechanism elicited by the β-cell deficiency could be responsible for such an alteration. In contrast, in n5-STZ rats, clamp studies indicate that the glucose utilization rate induced by hyperinsulinemia was significantly reduced and the hepatic glucose production rate was less efficiently suppressed by hyperinsulinemia. Thus, insulin resistance was present in vivo at the level of both peripheral tissues and the liver [Blondel et al., 1989]. Our data related to impairment of insulin action in vivo in the Wistar n5-STZ model are therefore consistent with the experience with the Sprague-Dawley n2-STZ model having a poor response of the heart or adipocytes to insulin in vitro [Trent et al., 1984; Fantus et al., 1987], or a reduced rate of glucose disposal in vivo during an insulin suppression test [Levy et al., 1984]. Through the comparison of these models, it is therefore possible to put into light some correlations which delineate the conditions necessary for the emergence of insulin resistance (Figure 2). First, peripheral insulin action was kept normal, despite reduction of the β-cell mass by 50%, loss of the insulin secretion in response to glucose, very mild hypoinsulinemia and mild hyperglycemia. Second, insulin resistance developed only under the conditions expressed in n5-STZ rats reflecting a dramatic reduction of the β-cell mass, an overt basal hypoinsulinemia and a frank basal hyperglycemia. Taken together these data lend support to the suggestion that a certain degree of insulin deficiency is necessary to induce a clear-cut decrease of insulin action [Portha et al, 1995]. Finally, the n-STZ models have contributed to evaluate whether a metabolic derangement (hyperglycemia?) that occurs secondarily to the insulin deficiency or insulin deficiency per se (hypoinsulinism) is responsible for the development of insulin resistance. The impact of chronic hyperglycemia per se was evaluated in n5-STZ rats receiving a continuous infusion of phlorizin for 4 weeks [Blondel et al., 1990]. Phlorizin treatment of diabetic rats efficiently decreased their basal plasma glucose levels from 16 to 7 mM. Their glycohemoglobin percentage returned to normal level. In contrast, the basal plasma insulin levels and the glucose-stimulated insulin secretion remained as low as in the untreated diabetic rats. The basal glucose utilization and the glucose production rates were normalized by the phlorizin treatment and, when measured following submaximal hyperinsulinemia, both returned to normal values. Since phlorizin treatment in control rats did not affect any of the above parameters, our data demonstrate that the sole restoration of normoglycemia in the n5-STZ rats (in the absence of any improvement of the endogenous insulin release), can completely correct the impairments of the insulin action upon the glucose uptake by the peripheral tissues and the glucose production by the liver. Therefore in the n5-STZ insulin resistant diabetic model, hyperglycemia can be viewed as a pathogenic factor of its own (glucotoxicity) in the development of insulin resistance. To evaluate the consequences of chronic hypoinsulinism per se (in the absence of hyperglycemia), clamp experiments were carried out in young n0-STZ rats at 4 wk of age, at a time when basal plasma glucose is still normal. In these rats, the basal plasma insulin was significantly reduced and the glucose-induced insulin secretion in vivo was markedly decreased [Kergoat et al., 1991]. The clamp studies revealed that the overall glucose utilization in the n0-STZ rats, was significantly higher in the basal state and after submaximal hyperinsulinemia [Kergoat et al., 1991]. We also verified that this was correlated with an increased stimulation of glucose utilization in soleus muscle, diaphragm and white and brown adipose tissues. The in vivo observations related to white adipose tissue agree well with our in vitro studies or inguinal adipocytes [Kergoat et al., 1991]. Specifically, the subcutaneous inguinal adipocytes were more sensitive to insulin in 4-wk-old n0-STZ females than in control females, with respect to glucose conversion to total lipids only. As the insulin-induced increase in glucose oxidation was the same in adipocytes from n0-STZ females and controls, this suggests that the enhancement of the rate of glucose metabolism in the n0-STZ rat adipocytes reflects a change beyond glucose transport and glucose metabolism along the glycolytic pathway. From

N-STZ DIABETIC RATS

227

Figure 2 Scheme of how progressive decline in B-cell mass, as induced by injecting neonatal rats with streptozotocin (n-STZ), may lead to sequential appearance of abnormalities found in NIDDM. Events (pancreatic insulin stores, basal plasma glucose and insulin levels, glucose disappearance rates [K], and mean incremental insulin areas [∆I}) are based on comparison of data obtained from three models of neonatally induced diabetes in Wistar rat: n0-STZ, 100 mg/kg STZ at birth; n2-STZ, 80 mg/kg STZ two days after birth; n5-STZ, 80 mg/kg STZ five days after birth. See text for comments on pathogenic progression toward impairment of insulin action in peripheral tissues.

these data, it was concluded that, in young normoglycemic n0-STZ rats, the insulin-dependent glucose

228

B.PORTHA ET AL.

utilization by white and brown adipose tissues is clearly increased as a consequence of the mild chronic (4 wk duration) hypoinsulinism. This suggests that in young n0-STZ rats glucose is preferentially channeled toward adipose tissues, resulting in hypertrophied adipocytes and raises the possibility that the emergence of an increased body fat mass may result from a primary mild insulin deficiency. This is not the generally accepted sequence of events, since it is rather recognized that obesity predisposes to the development of diabetes. However, the possibility that a glucose-intolerant state in rats may enhance the glucose tolerance during diet-induced hyperphagia demonstrated an increased efficiency in weigh gain, whereas weight loss occurred in rats with normal glucose tolerance [Cunningham et al., 1983]. INTERACTIONS BETWEEN DIET, OBESITY OR GESTATION IN n-STZ MODELS n-STZ Diabetic Models and Diet The appropriate amount of carbohydrate and fat to be included in the diet of diabetic patients still remains highly controversial [Mann, 1980; Reaven, 1980]. Experimental data obtained from the diabetic animals are conflicting or inappropriate for several reasons: investigations are always carried out in severely diabetic animals with no insulin response and are based on short periods of observation because of the high death rate of acutely diabetic rats without insulin treatment. Data were also obtained in spontaneously diabetic rodents [Gutzeit et al., 1979; Leiter et al., 1981], but in these models, obesity is also present and as a consequence it is difficult to dissociate a specific metabolic consequence due to dietary factors from those linked to obesity per se. The n-STZ models are appropriate to circumvent these limitations. The effects of chronic high sucrose feeding for one month on in vivo and in vitro insulin secretion and on in vivo insulin action have been studied in Wistar n0-STZ rats [Kergoat et al., 1987]. Data obtained clearly showed a significant deterioration of glucose tolerance in these rats. This impairment results from two additive changes in insulin’s effect upon the target tissues: the insulin-mediated glucose uptake by peripheral tissues is decreased and the liver becomes resistant to insulin action due to diminished ability of insulin to suppress hepatic glucose output. These results suggest that high sucrose (instead of complex carbohydrate) feeding in rats with mild diabetes to metabolic events likely to develop insulin resistance in target tissues. Such a pattern is directly related to the insulin-deficient state in these rats, since the same sucrose diet induces an enhanced insulin-mediated uptake in non-insulin-deficient animals [Kergoat et al., 1987]. High lipid diet may also be regarded as an aggravating factor of glucose handling in Wistar n0-STZ rats: we have shown that in n0-STZ rats receiving a high lipid feeding for one month, the glucose-induced insulin secretion was not significantly enhanced (at variance with the effect observed in control rats) and the glucose tolerance was deteriorated (while it remained normal in control rats) [Portha et al, 1982]. n-STZ Diabetic Models and Obesity In most of the spontaneous animal models of NIDDM, it is not easy to assess the effect of obesity on diabetes since these two pathological states are always associated. In experimental diabetes induced in adult rats by pancreatectomy, alloxan or STZ, the modulatory effect of obesity upon the evolution of diabetes could not hitherto be studied since obesity was never obtained [Young et al., 1965; York and Bray, 1972; Friedman, 1972]. We have used n0-STZ rats to answer two questions: 1) can obesity be obtained in these animals and 2) if so, does obesity modify the course of diabetes? Two experimental designs were used to associate obesity with experimental diabetes in the adult rat. One used an electrolytic lesion of the ventromedial hypothalamus

N-STZ DIABETIC RATS

229

in adult diabetic Wistar n0-STZ rats. The other used genetically obese rats from the Zucker falfa strain made diabetic by a neonatal injection of STZ [Portha et al., 1983]. In the diabetic rats, weight gain was similar to that in the non-diabetic rats, whether hyperphagia was due to a ventromedial hypothalamic lesion or to a genetic factor. Glucose-induced insulin release in vivo was increased in obese diabetic rats as compared with non obese-diabetic rats. Although glucose-induced insulin secretion was increased in diabetic rats becoming obese (in both n0-STZ fa/fa Zucker rats and n0-STZ rats with hypothalamic obesity), these animals clearly did not ameliorate their glucose tolerance, suggesting the development of a state of insulin resistance related to obesity. Indeed, in some of these animals overt diabetes with permanent or transient glycosuria and basal hyperglycemia developed. This was noted in almost one-third of the diabetic obese rats. Thus, obesity may be looked upon as a contributing factor in the development of diabetes in insulin-deficient rats. The obese rat with mild nSTZ diabetes appears to be a suitable tool for the study of the relationship between obesity and diabetes. Note that alternative procedures to obtain an obese diabetic rat model has been proposed by Kawai et al. [1991] using administration of both STZ (90 mg/kg on day 2 after birth) and monosodium glutamate (2g/ kg, from day 1 to day 5). Also the combination of neonatal administration (60 mg/kg on day 2) and postnatal overnutrition produced by breeding rats in small litters (4 pups/litter) [Mende et al, 1996] or ICR mice on cafeteria-diet [Hasegawa et al., 1989] has been shown to promote higher body weight as well as increased blood glucose levels. n-STZ Diabetic Models and Pregnancy The effect of pregnancy on the course of diabetes has been studied in Wistar n0-STZ females [Triadou et al., 1982]. In the pregnant n0-STZ rats (late pregnancy) the glucose-induced insulin secretion was found increased compared with the virgin state. Basal plasma glucose was decreased, but plasma glucose levels after glucose load were similar to values found in the virgin state, thus suggesting decreased glucose tolerance. Glucose tolerance remained impaired one and two months post-partum while insulin secretion returned to the range found in the virgin state. These findings indicate that in n0-STZ females glucose tolerance is and remains impaired by pregnancy, while in normal female rats it is and remains unchanged. Thus, despite increased insulin response to glucose during late gestation in the n0-STZ rats, the diabetogenicity of pregnancy is confirmed with this experimental model. NEONATAL STREPTOZOTOCIN DIABETES: SPONTANEOUS REMISSION AND βCELL REGENERATION The unique characteristic of STZ diabetes in rat neonates is the transitory nature of overt β-cell deficiency as compared with STZ diabetes in the adult rat. In the n0-STZ model, we and others have shown that the recovery of plasma glucose to normal value as soon as one week after birth, was related to recovery of the pancreatic insulin content and the β-cell mass [Portha et al., 1974; Portha et al., 1979; Cantenys et al., 1981; Wang et al, 1994; Ferrand et al, 1995; Movassat et al, 1997]. From postnatal day 4 onward, signs of regeneration became apparent, in that numerous insulin positive cells were found throughout the acinar parenchyma and within the duct epithelium [Cantenys et al, 1981]. The apparent budding of islets from the ducts was a common feature [Dutrillaux et al, 1982]. A study of the mitotic rate in colchicine-treated animals suggested that many β-cells were formed by mitosis of undifferentiated cells [Dutrillaux et al., 1982]. However, the long term impairment of the plasma glucose homeostasis [Portha et al, 1979] and the persistence of a 50% reduced β-cell mass in adult animals [Portha et al, 1989] are proof that the

230

B.PORTHA ET AL.

regeneration process was incomplete. Notably we have shown that insulin therapy in the n0-STZ model during the neonatal hyperglycemic phase markedly improves the recovery of the insulin stores in the pancreas [Portha and Picon, 1982] and the total β-cell mass [Movassat et al, 1997], and leads to a significantly improved glucose tolerance in adult life [Portha and Picon, 1982]. Another factor which seems to be important to the efficiency of the regeneration process, is the timing of the STZ injection since it coincides with the normal development of the islet mass in the rat [McEvoy and Madson, 1980]. Indeed we found that Wistar rats injected with STZ at 7 days of age did not recover from the insult [Dutrillaux et al, 1982], and in the n5-STZ rats, the lack of significant reaccumulation of insulin of the pancreas during the 2 weeks following the β-cell insult contrasts with the recovery of the insulin stores as found in the pancreas of the n2STZ and n0-STZ rats. This confirms that there is some capacity of β-cell regeneration in the neonatal rat pancreas and that the capacity for β-cell regeneration in the Wistar strain decreases quickly during the first postnatal week and thereafter is not longer significant. It is noticeable that the Wistar n2-STZ model exhibits less severe diabetes than the Sprague-Dawley n2-STZ model. This difference is probably not related to difference in gender selection as female Wistar rats were used in our studies whereas the other authors have used male Sprague Dawley rats: we have checked that the hyperglycemia in the adult male Wistar n2-STZ rats was similar to that in the corresponding females (unpublished data). More realistic is the possibility that the difference in diabetes severity should be linked to the difference in the selection of the pups after STZ injection: in our study, all injected pups were investigated at adult age whereas in other studies only those with blood glucose over 12 mM two days after injection were accepted [Bonner-Weir et al, 1981]. An alternative but not exclusive explanation is that regeneration of β-cells in the Sprague-Dawley neonates is less efficient than in the Wistar ones. This assumption that the capacity for β-cell regeneration is dependent on the strain of rat used is supported by the observation that severity of diabetes after neonatal STZ is increased in the spontaneously hypertensive rats as compared to the Wistar rats [Iwase et al., 1987]. Additionally, not only the extent of β-cell regeneration but also its mechanism seems to be influenced by the timing of STZ injection and/or the strain of the rat: in the Sprague Dawley n2-STZ model the partial replenishment of the β-cell mass has been claimed to reflect mainly replication of existing β-cells [Weir et al., 1986] rather than neogenesis from undifferentiated precursors. n-STZ MODELS AND COMPLICATIONS OF LONGSTANDING NIDDM The n-STZ models are especially attractive in this perspective since the rats can be easily kept for more than 20 months under standard breeding conditions. Their usefulness in studies related to the pathogenesis of nephropathy or hypertension is illustrated thereafter. It is well-known that hypertension is frequently associated with NIDDM in humans. However it is difficult to disclose the mechanism of this combination as well as its long-term effect on the development and progression of nephropathy. It has been shown that neonatal STZ treatment carried out in spontaneously hypertensive rats (SHR) offers appropriate model for studying the interractions between hyperglycemia and hypertension, as well as their influence upon renal damage [Iwase et al., 1987; Wakisaka et al, 1988; Iwase et al, 1994]. Various degrees of hyperglycemia can be achieved in male adult SHR rats by varying the neonatal STZ dose (from 37.5 to 75 mg/kg, i.p.). Exogenous insulin is not required for long term survival and hypertension develops normally (as in non-diabetic SHR) in male SHR rats receiving STZ on day 2 after birth. Using these models it has been demonstrated that the combination of hypertension and hyperglycemia accelerated not only the progression of the established nephropathy but also the early development of nephropathy itself. The more severe hyperglycemia was also associated with the more

N-STZ DIABETIC RATS

231

severe nephropathy [Wakisaka et al., 1988]. Also in conjunction with the development of nephropathy is the report that kidney basolateral membranes from n2-STZ rats exhibited impaired activity of [Ca2+, Mg2+] ATPase and abnormal phospholipid content [Levy et al., 1988]. Clinical studies in humans have revealed the existence of a NIDDM-induced cardiomyopathy that is independent of atherosclerotic coronary artery disease, hypertension, or vascular disease. This syndrome is characterized by both mechanical and metabolic abnormalities of the heart. Studies into the etiology of this myopathy are questionable since most of them were carried out in models more closely resembling insulindependent diabetes with overt hyperglycemia and continued waste throughout their lifetime. A more adequate animal model has become available with the report that Wistar n2-STZ rats develop cardiomyopathy [Schaffer et al., 1985]. Hearts isolated from 12-mo diabetic rats exhibited reduced rates of contractility and relaxation. Associated with the abnormality in contractility was a redistribution in myosin isozyme content. Defects in myocardial relaxation also occurred concomitantly with impaired handling of calcium [Schaffer et al., 1989]. n-STZ RODENTS: MODELS FOR TESTING NEW HYPOGLYCEMIC DRUGS Selection of animal models for studying new hypoglycemic drugs depends upon the particular features of NIDDM required. In the n-STZ rat models of NIDDM, the various lesions contributing to the hypoglycemia represent many potential sites at which new hypoglycemic agents could be directed. To summarize, these targets include 1) the β-cell lesions (n0-STZ model) such as alterated rate of renewal (replication, neogenesis), decreased insulin biosynthesis, decreased insulin release in response to glucose; 2) the insulin target cell lesions (n2-and n5-STZ models); 3) the hepatic glucose output lesions (n2-and n5-STZ models); 4) the imbalance of the glucose-fatty acid cycle (n5-STZ model). Accordingly, the n-STZ rats which are relatively recent additions to the list of animal models of NIDDM have so far been useful for testing many hypoglycemic drugs such as for example, sulfonylureas [Serradas et al., 1989, Ohnota et al., 1996], nateglinide (A4166) or repaglinide [Kergoat, unpublished data], KAD-1229 [Ohnota et al., 1994], M-16209 [Nakayama et al., 1995], JTT-608 [Ohta et al., 1999], SL840418 [Angel et al., 1996], benfluorex [Portha et al., 1993], metformin [Kergoat, unpublished data], pioglitazone [Kergoat, unpublished data], GLP-1 [Dachicourt et al., 1997], phlorizin [Serradas et al., 1991; Blondel et al., 1990], vanadate [Blondel et al, 1990] or tungstate [Barbera et al, 1997]. Besides, they are also convenient models for testing the impact of islet grafting in NIDDM [Elian et al, 1996; Tormo et al, 1997]. REFERENCES Ahren, B. and Skoglund, G. (1989) Insulin secretion in neonatally streptozotocin-injected mice. Diab. Res., 11, 185–190. Angel, I., Burcelin, R., Prouteau, M., Girard, J. and Langer, S.Z. (1996) Normalization of insulin secretion by a selective α2-adrenoceptor antagonist restores Glut-4 glucose transporter in adipose tissues of typeII diabetic rats. Endocrinology, 137, 2022–2027. Barbera, A., Fernandez-Alvarez, J., Truc, A., Gomis, R. and Guinovart, J.J. (1997) Effects of tungstate in neonatally streptozotocin-induced diabetic rats: mechanism leading to normalization of glycemia. Diabetologia, 40, 143–149. Bevilacqua, S., Barett, E.J., Smith, D., Simonson, D.C., Olson, M., Bratusch-Marrain, P., et al.. (1985) Hepatic and peripheral insulin resistance following streptozotocin-induced insulin deficiency in the dog. Metabolism, 34, 817–825. Blondel, O., Bailbe, D. and Portha, B. (1989) Relation of insulin deficiency to impaired insulin action in NIDDM adult rats given streptozotocin as neonates. Diabetes, 36, 610–17.

232

B.PORTHA ET AL.

Blondel, O., Simon, J., Chevalier, B. and Portha, B. (1990) Impaired insulin action but normal insulin receptor activity in the liver of diabetic rats. Effect of vanadate treatment. Am. J. Physiol, 258, E459–467. Blondel, O., Bailbe, D. and Portha, B. (1990) Insulin resistance in rats with non-insulindependent diabetes induced by neonatal (5 days) streptozotocin: evidence for reversal following phlorizin treatment. Metabolism, 39, 787–793. Bonner-Weir, S., Trent, D.E., Honey, R.N. and Weir, G.C. (1981) Responses of neonatal islets to streptozotocin: limited -cell regeneration and hyperglycemia. Diabetes, 30, 64–69. Bonner-Weir, S. and Orci, L. (1982) New perspectives on the microvasculature of the islets of Langerhans in the rat. Diabetes, 31, 883–889. Cantenys, D., Portha, B., Dutrillaux, M.C., Hollande, E., Roze, C. and Picon, L. (1981) Histogenesis of the endocrine pancreas in newborn rats after destruction by streptozotocin. An immunocytochemical study. Virchows Archiv (Cell. Pathol), 35, 109–122. Cunningham, J., Calles, J., Eisikowitz, Zawalich, W. and Felig, P. (1983) Increased efficiency of weight gain and altered cellularity of brown adipose tissue in rats with impaired glucose tolerance during diet-induced overfeeding. Diabetes, 32, 1023–1027. Dachicourt, N., Serradas, P., Bailbé, D., Kergoat, M., Doaré, L. and Portha, B. (1997) GLP-1(7–36) amide confers glucose sensitivity to previously glucose-incompetent β-cells in diabetic rats. In vivo and in vitro studies. J. Endocrinol., 155, 369–376. DeFronzo, R.A., Hendler, R. and Simonson, D. (1982) Insulin resistance is a prominent feature of insulin-dependent diabetes. Diabetes, 31, 795–801. Dutrillaux, M.C., Portha, B., Roze, C. and Hollande, E. (1982) Ultrastructural study of pancreatic B-cell regeneration in newborn rats after destruction by streptozotocin. Virchows Archiv. (Cell Pathol.), 39, 173–185. Eizirik, D., Sandier, S., Welsh, N. and Hellerström, C. (1988) Preferential reduction of insulin production in mouse pancreatic islet maintained in culture after streptozotocin exposure. Endocrinology, 122, 1242–1249. Elian, N., Bensimon, C., Chapa, O., Bethoux, J.P., Cugnenc, P.H. and Altman, J.J. (1996) Pancreatic transplantation in experimental non-insulin-dependent diabetic rats. Transplantation, 61, 696–700. Fantus, I.G., Chayoth, R., O’Dea, L., Marliss, E., Yale, J.F. and Grose, M. (1987) Insulin binding and glucose transport in adipocytes in neonatal streptozotocin-injected rat model of diabetes mellitus. Diabetes, 36, 654–660. Ferrand, N., Astesano, A., Phan, H.H., Lelong, C. and Rosselin, G. (1995) Dynamics of pancreatic cell growth and differentiation during diabetes reversion in STZ-treated newborn rats. Am. J. Physiol, 269, C1250–1264. Friedman, M.I. (1972) Effects of alloxan diabetes on hypothalamic hyperphagia and obesity. Am. J. Physiol., 222, 174–178. Frayon, S., Pessah, M., Giroix, M.H., Mercan, D., Boissard, C., Malaisse, W.J., et al. (1999) Gαolf identification by RTPCR in purified normal pancreatic B-cells and in islets from rat models of non-insulin-dependent diabetes. Biochem. Biophys. Res. Commun., 254, 269– 272. Giroix, M.H., Portha, B., Kergoat, M., Bailbé, D. and Picon, L. (1983) Glucose insensitivity and amino-acid hypersensitivity of insulin release in rats with non-insulin-dependent diabetes: a study with the perfused pancreas. Diabetes, 32, 445–51. Giroix, M.H., Portha, B., Kergoat, M. and Picon, L. (1984) Glucagon secretion in rats with non-insulin-dependent diabetes: an in vivo and in vitro study. Diabète Métabol., 10, 12– 17. Giroix, M.H., Sener, A., Bailbé, D., Portha, B. and Malaisse, W.J. (1990) Impairment of the mitochondrial oxidative response to D-glucose in pancreatic islets from adult rats injected with streptozotocin during the neonatal period. Diabetologia, 33, 654–60. Giroix, M.H., Rasschaert, J., Bailbé, D., Leclercq-Meyer, V., Sener, A., Portha, B., et al. (1991) Impairment of glycerol phosphate shuttle in islets from rats with diabetes induced by neonatal streptozotocin. Diabetes, 40, 227–232. Giroix, M.H., Rasschaert, J., Sener, A., Leclercq-Meyer, V., Bailbé, D., Portha, B., et al. (1992) Study of hexose transport, glycerol phosphate shuttle and Krebs cycle in islets of adult rats injected with streptozotocin during the neonatal period. Mol. Cell. Endocrinol., 83, 95–104. Giroix, M.-H., Baetens, D., Rasschaert, J., Leclercq-Meyer, V., Sener, A., Portha, B., et al. (1992) Enzymic and metabolic anomalies in islets of diabetic rats: relationship to B-cells mass. Endocrinology, 130, 2634–2640.

N-STZ DIABETIC RATS

233

Giroix, M.H., Sener, A., Portha, B. and Malaisse, W.J. (1993) Preferential alteration of oxidative relative to total glycolysis in pancreatic islets of two rat models of inherited or acquired type 2 (non-insulin-dependent) diabetes mellitus. Diabetologia, 36, 305–309. Grill, V. and Rundfeldt, M. (1986) Abnormalities of insulin responses after ambient and previous exposure to glucose in streptozotocin-diabetic and dexamethasone-treated rats: role of hyperglycemia and increased B-cell demands. Diabetes, 35, 44–51. Grill, V., Welsberg, C. and Östenson, C.G. (1987) β-cell sensitivity in a rat model of non-insulin-dependent diabetes. Evidence for a rapidly reversible effect of previous hyperglycemia. J. Clin. Invest., 80, 664–669. Grill, V. and Östenson, C.G. (1988) The influence of a diabetic state on insulin secretion: studies in animal models of non-insulin dependent diabetes. In Pathogenesis of non-insulin-dependent diabetes mellitus, edited by V. Grill and S. Efendic, pp. 93–106. New York: Raven Press. Gutzeit, A., Renold, A.E., Cerasi, E. and Shafrir, E. (1979) Effect of diet-induced obesity on glucose and insulin tolerance of a rodent with a low insulin response (Acomys cahirinus). Diabetes, 28, 777–784. Halban, P.A., Bonner-Weir, S. and Weir, G.C. (1983) Elevated proinsulin biosynthesis in vitro from a rat model of noninsulin-dependent diabetes mellitus. Diabetes, 32, 277–283. Hasegawa, G., Mori, H., Sawada, M., Takagi, S., Shigeta, H., Kitagawa, Y., et al. (1989) Overt diabetes induced by overeating in neonatally STZ-treated impaired glucose tolerant mice: long-term follow up study. Endocrinol. Japon., 36, 471–479. Inoue, K., Hisatomi, A., Umeda, E and Nawata, H. (1992) Relative hypersecretion of amylin to insulin from rat pancreas after neonatal STZ treatment. Diabetes, 41, 723–727. Inoue, K., Cetkovic-Cvrlje, M., Eizirik, D. and Grill, V. (1994) Irreversible loss of normal beta-cell regulation by glucose in neonatally streptozotocin diabetic rats. Diabetologia, 37, 351–357. Iwase, M., Kikuchi, M., Nunoi, K., Wakisaka, M., Maki, Y, Sadoshima, S., et al. (1987) Diabetes induced by neonatal streptozotocin treatment in spontaneously hypertensive rats. Metabolism, 36, 654–657. Iwase, M., Nunoi, K., Himeno, H., Yoshinari, M., Kikuchi, M., Maki, Y, et al. (1994) Susceptibility to neonatal streptozotocin-induced diabetes in spontaneously hypertensive rats. Pancreas, 9, 344–348. Iwase, M., Wakisaka, M., Yoshinari, M., Sato, Y, Yoshizumi, H., Nunoi, K., et al. (1996) Effect of gonadectomy on the development of diabetes mellitus, hypertension, and albuminuria in the rat model. Metabolism, 45, 155–161. Junod, A., Lambert A., Stauffacher, W. and Renold, A.E. (1969) Diabetogenic action of streptozotocin: relationship of dose to metabolic response. J. Clin. Invest., 48, 2129–2139. Kawai, K., Suzuki, S., Watanabe, Y., Murayama, Y. and Yamashita, K. (1991) Neonatal streptozotocin monosodiumglutamate injection in rats: a model of obese NIDDM. In Lessons from animal diabetes. The Tokyo Workshop, edited by Y. Goto and Y. Kanazawa, pp. 45–49. London: Smith-Gordon. Kergoat, M. and Portha, B. (1985) In vivo hepatic and peripheral insulin sensitivity in rats with non-insulin-dependent diabetes induced by streptozotocin. Diabetes, 34, 574–579. Kergoat, M., Bailbé, D. and Portha, B. (1987) Insulin treatment improves glucose-induced insulin release in rats with NIDDM induced by streptozotocin. Diabetes, 36, 971–77. Kergoat, M., Bailbé, D. and Portha, B. (1987) Effect of high sucrose-diet on the insulin secretion and insulin action. A study in the normal rat. Diabetologia, 30, 252–258. Kergoat, M., Bailbé, D. and Portha, B. (1987) Effect of high sucrose diet on insulin secretion and insulin action. A study in rats with non-insulin dependent diabetes induced by streptozotocin. Diabetologia, 30, 666–673. Kergoat, M., Simon, J. and Portha, B. (1988) Insulin receptor binding and tyrosine kinase activity in the liver of rats with non-insulin-dependent diabetes. Biochem. Biophys. Res. Commun., 152, 1015–1022. Kergoat, M., Guerre-Millo, M., Lavau, M. and Portha, B. (1991) Increased insulin action in rats with mild insulin deficiency induced by neonatal streptozotocin. Am. J. Physiol., 260, E561–567. Khan, A., Chandramouli, V., Östenson, C.G., Löw, H., Landau, B.R. and Efendic, S. (1990) Glucose cycling in islets from normal and diabetic rats. Diabetes, 39, 456–459.

234

B.PORTHA ET AL.

Kurose, T., Tsuda, K., Ishida, H., Tsuji, K., Okamoto, Y., Tsuura, Y., et al (1992) Glucagon, insulin and somatostatin secretion in response to sympathetic neural activation in streptozotocin-induced diabetic rats. A study with the isolated perfused rat pancreas in vitro. Diabetologia, 35, 1035–1041. Leahy, J.L., Bonner-Weir, S. and Weir, G.C. (1984) Abnormal glucose regulation of insulin secretion in models of reduced B-cell mass. Diabetes, 33, 667–73, 1984. Leahy, J.L. and Weir, G.C. (1985) Unresponsiveness to glucose in a streptozotocin model of diabetes; inappropriate insulin and glucagon responses to a reduction of glucose concentration. Diabetes, 34, 653–659. Leahy, J.L., Bonner-Weir, S. and Weir, G.C. (1985) Abnormal insulin secretion in a streptozotocin model of diabetes. Effects of insulin treatment. Diabetes, 34, 660–666. LeDoux, S.P., Woodley, S.E., Patton, N.J. and Wilson, G.L. (1986) Mechanisms of nitrosourea- induced β-cell damage. Alterations in DNA. Diabetes, 35, 866–872. Leiter, E.H., Coleman, D.L., Eisenstein, A.B. and Strack, I. (1981) Dietary control of pathogenesis in C57 BL/KsJ db/ db diabetic mice. Metabolism, 30, 554–562. Levy, J., Gavin, J.R., Fausto, A., Gingerich, R.L. and Avioli, L. (1984) Impaired insulin action in rats with non-insulindependent diabetes. Diabetes, 33, 901–906. Levy, J., Suzuki, Y., Avioli, L., Grunberger, G. and Gavin, J. (1988) Plasma membrane phospholipid content in noninsulin-dependent streptozotocin-diabetic rats—effect of insulin. Diabetologia, 31, 315–321. Malaisse, W.J. (1991) Regulation of insulin secretion by nutrients. In Nutrient Regulation of insulin secretion, edited by P.R.Flatt, pp. 83–100. London: Portland Press. Mann, J.I. (1980) Diet and diabetes. Diabetologia, 18, 89–95 McEvoy, R. and Madson, K.L. (1980) Pancreatic insulin, glucagon-and somatostatine positive islet cell populations during perinatal development of the rat. Morphometric quantitation. Biol. Neonate, 37, 248–254. Mende, S., Mühle, W. and Peters, W. (1996) Influence of postnatal overnutrition and pregnancy on non-insulindependent diabetes induced in Wistar rats by neonatal streptozotocin. Norm. Metab. Res., 28, 81–85. Morin, L., Giroix, M.H., Gangnerau, M.-N., Bailbé, D. and Portha, B. (1997) Impaired phosphoinositide metabolism in glucose-incompetent islets of neonatally streptozotocin-diabetic rats. Am. J. Physiol, 272, E737–745. Movassat, J., Saulnier, C. and Portha, B. (1997) Insulin administration enhances growth of the β-cell mass in streptozotocin-treated newborn rat. Diabetes, 46, 1445–1452. Nakayama, K., Murakami, N., Ohta, M., Kato, K., Notsu, T., Mizota, M., et al. (1995) Effects of M16209 on insulin secretion in isolated, perfused pancreases of normal and diabetic rats. Eur. J. Pharmacol., 276, 85–91. Ohta, T., Furukawa, N., Yonemori, F. and Wakitani, K. (1999) JTT-608 controls blood glucose by enhancement of glucose-stimulated insulin secretion in normal and diabetic mellitus rats. Eur. J. Pharmacol., 367, 91–99. Ohnota, H., Koizumi, M., Momose, Y and Sato, F. (1994) Normalization of impaired glucose tolerance by the shortacting hypoglycemic agent calcium(2S)-2-benzyl-3-(cis-hexahydro2-isoindolinylcarbonyl)propionate dihydrate (KAD-1229) in non-insulin-dependent diabetes mellitus rats. Can. J. Pharmacol., 73, 1–6. Ohnota, H., Kitamura, T., Kinikawa, M., Hamano, S., Shibata, N., Miyata, N., et al. (1996) A rapid and short-acting hypoglycemic agent KAD-1229 improves post-prandial hyperglycemia and diabetic complications in streptozotocin-induced non-insulin-dependent diabetes mellitus rats . Jpn. J. Pharmacol., 71, 315–323. Okabayashi, Y., Otsuki, M., Ohki, A., Tani, S. and Baba, S. (1989) Increased B-cell secretory responsiveness to ceruletide and TPA in streptozotocin-induced mildly diabetic rats. Diabetes, 38, 1042–1047. Okamoto, H. (1981) Regulation of proinsulin synthesis in pancreatic islets and a new aspect to insulin-dependent diabetes. Mol. Cell. Biochem., 37, 43–61. Östenson, C.G. and Grill, V. (1985) Glucose exerts opposite effects on muscarinic receptor binding to A and B cells of the endocrine pancreas. Endocrinology, 116, 1741–1745. Östenson, C.G. and Grill, V. (1987) Evidence that hyperglycemia increases muscarinic binding in pancreatic islets of the rat. Endocrinology, 121, 1705–1710. Östenson, C, Grill, V. and Roos, M. (1989) Studies on sex dependency of β-cell susceptibility to streptozotocin in a rat model of type II diabetes mellitus. Exp. Clin. Endocrinol., 93, 241–247.

N-STZ DIABETIC RATS

235

Östenson, C.G., Cattaneo, A.G., Doxey, J.C. and Efendic, S. (1989) α-adrenoceptors and insulin release from pancreatic islets of normal and diabetic rats. Am. J. Physiol., 257, E439-E443. Östenson, C.G., Efendic, S. and Grill, V. (1990) Abnormal regulation by glucose and somatostatin secretion in the perfused pancreas of NIDDM rats. Pancreas, 5, 347–353. Portha, B., Levacher, C., Picon, L. and Rosselin, G. (1974) Diabetogenic effect of streptozotocin during the perinatal period in the rat. Diabetes, 23, 889–895. Portha, B., Picon, L. and Rosselin, G. (1979) Chemical diabetes in the adult rat as the spontaneous evolution of neonatal diabetes. Diabetologia, 17, 371–377. Portha, B. and Picon, L. (1982) Insulin treatment improves the spontaneous remission of neonatal streptozotocin diabetes in the rat. Diabetes, 31, 165–169. Portha, B., Giroix, M.H. and Picon, L. (1982) Effect of diet on glucose tolerance and insulin response in chemically diabetic rats. Metabolism, 21, 1194–1199. Portha, B., Goursot, R., Giroix, M.H., Nicolaïdis, S. and Picon, L. (1983) Experimental hypothalamic or genetic obesity in the non-insulin-dependent diabetic rat. Diabetologia, 25, 51–55. Portha, B., Chamras, H., Broer, Y., Picon, L. and Rosselin, G. (1983) Decreased glucagon-stimulated cyclic AMP production by isolated liver cells of rats with type II diabetes. Mol. Cell. Endocrinol., 32, 13–26. Portha, B. (1985) Decreased glucose-induced insulin release and biosynthesis by islets of rats with non-insulindependent diabetes: effects of tissue culture. Endocrinology, 117, 1735– 41. Portha, B. and Kergoat, M. (1985) Dynamics of glucose-induced insulin release during the spontaneous remission of streptozotocin diabetes induced in the newborn rats. Diabetes, 34, 574–579. Portha, B., Giroix, M.H., Serradas, P., Welsh, N., Hellerström, C., Sener, A. and Malaisse, W.J. (1988) Insulin production and glucose metabolism in isolated pancreatic islets of rats with NIDDM. Diabetes, 37, 1226–33. Portha, B., Blondel, O., Serradas, P., Mc Evoy, R., Giroix, M.H., Kergoat, M., et al. (1989) The rat models of noninsulin dependent diabetes induced by neonatal streptozotocin. Diab. Métab., 15, 61–75. Portha, B., Serradas, P., Blondel, O., Giroix, M.-H. and Bailbé, D. (1990) Relation between hyperglycemia and impairment of insulin secretion and action. Information from the n-STZ rat models. In Frontiers in diabetes research. Lessons from animal diabetes III, edited by E.Shafrir, VII. 1, pp. 334–341. 7 Portha, B., Serradas, P., Bailbé, D., Blondel, O. and Picarel, F. (1993) Effect of Benfluorex on insulin secretion and insulin action in streptozotocin diabetic rats. Diabetes Metabolism Review, 9, 57–63. Portha, B., Kergoat, M., Blondel, O., Bailbé, D., Escriva, E, Pascual-Leone, A.M., et al (1995) Pathogenesis of impaired insulin action in rat models of insulin deficiency. In Frontiers in Diabetes Research; Lessons from Animal Diabetes, edited by E.Shafrir, pp. 83–91. UK: Smith-Gordon Co. Reaven, G.M., Sageman, W.S. and Swenson, R.S. (1977) Development of insulin resistance in normal dogs following alloxan-induced insulin deficiency. Diabetologia, 13, 459–462. Reaven, G.M. (1980) How high the carbohydrate? Diabetologia, 19, 409–413. Rerup, C. (1980) Drugs producing diabetes through damage of the insulin secreting cells. Pharmacol. Rev., 22, 485–518. Schaffer, S., Tan, B. and Wilson G. (1985) Development of a cardiomyopathy in a model of non-insulin-dependent diabetes. Am. J. Physiol., 248, H179–H185. Shaffer, S.W., Seyed-Mozaffari, M., Cutcliff, C.R. and Wilson, G.L. (1986) Postreceptor myocardial metabolic defect in a rat model of non-insulin-dependent diabetes mellitus. Diabetes, 35, 593–597. Schaffer, S., Mozaffari, S., Artman, M. and Wilson G. (1989) Basis for myocardial mechanical defects associated with non-insulin-dependent diabetes. Am. J. Physiol., 256, E25— E30. Serradas, P., Bailbé, D. and Portha, B. (1989) Chronic gliclazide treatment improves the in vitro glucose-induced insulin release in rats with non-insulin-dependent diabetes induced by neonatal streptozotocin. Diabetologia, 32, 577–584. Serradas, P., Bailbé, D., Blondel, O. and Portha, B. (1991) Abnormal B-cell function in rats with non-insulin dependent diabetes induced by neonatal streptozotocin: effect of in vivo insulin, phlorizin or vanadate treatments. Pancreas, 6, 54–62.

236

B.PORTHA ET AL.

Serradas, P., Blondel, O., Bailbé, D. and Portha, B. (1993) Benfluorex normalizes basal hyperglycemia and reverses hepatic resistance in streptozotocin-diabetic rats. Diabetes, 43, 564–570. Thorens, B., Weir, G.C., Leahy, J.L., Lodish, H.F. and Bonner-Weir, S. (1990) Reduced expression of the liver/beta cell glucose transporter isoform in glucose-insensitive pancreatic beta cells of diabetic rats. Proc. Natl. Acad. Sci. USA, 87, 6492–96. Tormo, M.A., Leon-Quinto, T., Saulnier, C., Bailbé, D., Serradas, P. and Portha, B. (1997) Insulin secretion and glucose tolerance after islet transplantation in rats with non-insulin-dependent diabetes induced by neonatal streptozotocin. Cell Transplantation, 6, 23–32. Triadou, N., Portha, B., Picon, L. and Rosselin, G. (1982) Experimental chemical diabetes and pregnancy in the rat. Evolution of glucose tolerance and insulin response. Diabetes, 31, 75–79. Trent, D.F. Fletcher, D.J., May, J.M., Bonner-Weir, S. and Weir, G.C. (1984) Abnormal islet and adipocyte function in young B-cell deficient rats with near normoglycemia. Diabetes, 33, 170–175. Tsuura, Y., Ishida, H., Okamoto, Y., Tsuji, K., Kurose, T., Horie, M., et al. (1992) Impaired glucose sensitivity of ATPsensitive K+ channels in pancreatic B-cells in streptozotocin-induced NIDDM rats. Diabetes, 41, 861–865. Wakisaka, M., Nunoi, K., Iwase, M., Kikuchi, M., Maki, Y., Yamamoto, K., et al (1988) Early development of nephropathy in a new model of spontaneously hypertensive rat with non-insulin-dependent diabetes mellitus. Diabetologia, 31, 291–296. Wang, R.N. Bouwens, L. and Klöppel, G. (1994) Beta-cell proliferation in normal and streptozotocin-treated newborn rats: site, dynamics and capacity. Diabetologia, 37, 1088– 1096. Weir, G.C., Clore, E.E., Zmachinsky, C.J. and Bonner-Weir, S. (1981) Islet secretion in a new experimental model for non-insulin-dependent diabetes. Diabetes, 30, 590–595. Weir, G.C., Leahy, J.L. and Bonner-Weir, S. (1986) Experimental reduction of B-cell mass: implications for the pathogenesis of diabetes. Diabetes Metab. Rev., 2, 125–61. Welsh, N. and Hellerström, C. (1990) In vitro restoration of insulin production in islets from adult rats treated neonatally with streptozotocin. Endocrinology, 126, 1842–1848. Welsh, N., Pääbo, and Welsh, M. (1991) Decreased mitochondrial gene expression in isolated islets of rats injected neonatally with streptozotocin. Diabetologia, 34, 626–631. York, D.A. and Bray, G.A. (1972) Dependence of hypothalamic obesity on insulin, the pituitary and the adrenal gland. Endocrinology, 90, 885–894. Young, T.K. and Liu, A.C. (1965) Hyperphagia, insulin and obesity. Chinese J. Physiol., 19, 247–253.

14. GALACTOSEMIC ANIMAL MODELS W.GERALD ROBISON, JR. National Eye Institute, National Institutes of Health, Bethesda, MD 20892–2735

INTRODUCTION Galactosemic animal models have become important tools for understanding complications of diabetes at the cellular level where some of the earliest changes occur, especially as applied to diabetic ocular complications (Engerman and Kern, 1984; 1995a; Kador et al., 1988; 1990; 1994; 1995; Robison et al., 1983; 1989; 1990a; 1995a; 1995b; Takahashi et al., 1992). It seems incredible that a normal animal fed a diet containing 30% to 64% galactose would develop ocular complications that mimic closely those of diabetes. Yet this is what the evidence indicates, probably because the primary underlying mechanisms causing galactosemic and diabetic complications are the same or very similar (Robison et al., 1995a). Even more surprising, the diabetic-like complications occur sooner and are more severe in galactosemic than in diabetic animals. Although other possibilities will be discussed, the most probable explanation comes from the fact that galactosemia, like hyperglycemia, results in increased flux through aldose reductase, the first enzyme of the polyol pathway, resulting in the accumulation of polyol in all tissues that do not require insulin for hexose uptake (Figure 1). Eventually, all tissues that accumulate polyol (galactitol or sorbitol, respectively) develop ocular complications characteristic of diabetes (Dvornik, 1987; Robison, 1995a). The mimicking of diabetic retinal microangiopathies is extraordinary (Robison et al., 1995a). The spectrum of microangiopathies characteristic of human diabetic retinopathy is very specific, not being completely matched by any other ocular condition in humans (Table 1). Yet galactosemic animals develop, in the same sequence, the entire spectrum of lesions otherwise unique to diabetic retinopathy. The fact that increased tissue polyol accumulation is common to both galactosemic rat and diabetic human retinas implicates flux through aldose reductase. Confirmation of the key role of aldose reductase has come from several studies on diabetic rats (Kojima et al., 1985a; 1985b; Chakrabarti and Sima, 1987; 1989), on galactosemic rats

238

W.G.ROBISON, JR.

Figure 1 Polyol Pathway for Glucose and Galactose—drawn to indicate the greater accumulation of polyol (galactitol) caused by the three- to fourfold greater affinity of aldose reductase for galactose than for glucose and the lack of galactitol metabolism. Reproduced from Robison and Laver (1993) with permission of Smith-Gordon and Co. Ltd.

(Robison et al., 1983; 1989; 1990a; 1990b; 1995b; 1996; Frank et al., 1997), and on galactosemic dogs (Akagi and Kador, 1990; Kador et al., 1994). In these studies various structurally distinct, specific inhibitors of aldose reductase ameliorated or prevented diabetic-like retinopathy and other ocular complications. The demonstration that aldose reductase has a fourfold greater affinity for galactose than for glucose (Sato and Kador, 1990) provides an explanation for the acceleration of diabetic-like ocular complications observed in galactosemic animals. ADVANTAGES OF THE GALACTOSEMIC RAT Galactose-fed animals provide several advantages over other models (Engerman et al., 1982) of diabetic ocular complications. Using the galactosemic rat as an example, the favorable characteristics include the following: 1. accelerates diabetic-like cataract formation—taking only 2 to 3 weeks for developing a mature sugar cataract (Datiles et al., 1982) compared with 2 to 3 months in diabetic rats (Dvornik, 1987); 2. isolates the effects of elevated plasma hexose (galactosemia), because galactose ingestion does not alter plasma glucose or insulin levels (Engerman and Kern, 1995a; Robison et al., 1996); 3. narrows to essentially two the probable candidates for a primary triggering mechanism—increased flux through aldose reductase or increased non-enzymatic glycosylation (glycation)—since the only metabolic perturbations observed result from these two changes (Robison et al., 1995a; 1996); 4. compensates for the short life span of the rat by accelerating diabetic-like retinal damage, thus developing more advanced microangiopathies (Robison et al., 1995a) than possible with diabetic rats (von Sallman and Grimes, 1972; Sima et al., 1983; 1985; Robison et al, 1991; Engerman and Kern, 1995a);

GALACTOSEMIC ANIMAL MODELS

239

Table 1 Diabetic retinopathy compared with other disordersa

a. Condensed from Robison et al, 1995a and reproduced here with permission of the authors and Pergamon Press. For complete details, see the original article. b. Symbols:— = absent or rare; =present but not similar to diabetic retinopathy; =characteristic of diabetic retinopathy c. Retinal Capillary Basement Membrane Thickening: Archer et al, 1991; Ashton, 1983; Ashton, 1974; Bloodworth, 1967; Bloodworth, 1963; Bloodworth, 1962; Green et al, 1980; Sugi, 1966 d. Pericyte Loss: Archer et al., 1991; Ashton, 1971; Cogan and Kuwabara, 1967a; Garner, 1970; Green et al, 1980; Speiser et al, 1968; Yanoff, 1966 e. Dilations: Archer, 1976; Ashton, 1963; Benson et al, 1988; Cogan et al, 1961; Kohner and Henkind, 1970; Palmer et al, 1994; Weinberg and Seddon, 1994; Wise, 1957 g f. Hard Exudates: Brown and Benson, 1984; Benson et al., 1988; Chaudhuri et at., 1981; Garner, 1994; Hayreh, 1970; Keith etal., 1939; Maguire and Schachat, 1994; Weinberg and Seddon, 1994; Yanko et al., 1975 g. Acellularity: Ashton, 1971; Ashton, 1963; Ashton, 1959; Ashton, 1950; Benson et al, 1988; Bresnick, 1994; Bresnick et al, 1976; Brown and Benson, 1984; Brown et al., 1985; Cogan and Kuwabara, 1967b; Chaudhuri et al., 1981; Cogan et al., 196I; de Venecia et al., 1976; Garner, 1994; Garner, 1970; Goldberg, 1976; Hayreh, 1970; Kohner and Henkind, 1970; Maguire and Schachat, 1994; Shimizu et al, 1981; Sugi, 1966; Toussaint, 1968; Weinberg and Seddon, 1994; Wise, 1957 h. Microaneurysms: Ashton, 1971; Ashton, 1963; Ashton, 1950; Ashton, 1949; Ballantyne, 1945; Benson et al, 1988; Brown, 1994a; 1994b; Chaudhuri et al, 1981; Cogan, 1974; Cogan and Kuwabara, 1967b; de Venecia et al, 1976; Duke et al, 1968; Garner, 1994; Goldberg, 1976; Hayreh, 1970; Keith et al, 1939; Kuwabara et al, 1961; Maguire and Schachat, 1994; Michaelson, 1980; Sugi, 1966; Toussaint, 1968; Weinberg and Seddon, 1994; Wise, 1957 i. IRMA: Cogan and Kuwabara, 1967b; de Venecia et al, 1976; Weinberg and Seddon, 1994; Wise, 1957 j. Cotton-wool Spots: Ashton and Harry, 1963; Benson et al, 1988; Brown et al, 1985; Chaudhuri et al, 1981; Garner, 1994; Goldberg, 1976; Hayreh, 1970; Keith et al, 1939; Kohner and Dollery, 1969; Kohner et al, 1969; Maguire and Schachat, 1994; Michaelson, 1980; Mansour et al, 1990; Weinberg and Seddon, 1994 k. Neovascularization: Archer, 1976; 1977; Benson et al, 1988; Brown, 1994a; 1994b; Chaudhuri et al, 1981; Garner, 1994; Hayreh, 1970; Maguire and Schachat, 1994; Patz, 1984; Weinberg and Seddon, 1994; Wise, 1957 1. Hemorrhage: Benson et al., 1988; Chaudhuri et al., 1981; Hayreh, 1970; Maguire and Schachat, 1994 m. Edema: Benson et al, 1988; Chaudhuri et al, 1981; Garner, 1994; Hayreh, 1970; Keith et al, 1939; Maguire and Schachat, 1994

5. mimics faithfully two important characteristics of human diabetic retinopathy —accumulation of retinal polyol and development of microangiopathies that are similar in both spectrum of types (Table 1) and topographical patterns of development (Robison et al., 1995a); 6. provides evidence that increased aldose reductase activity is linked to the microangiopathies of diabetic retinopathy, because specific inhibitors of aldose reductase block polyol accumulation and

240

W.G.ROBISON, JR.

development of microangiopathies without decreasing glycation levels (Robison et al., 1996; Frank et al., 1997); 7. permits return to a normal metabolic state (normalized tissue polyol and percent glycated hemoglobin), without treatment or operation, by merely removing the galactose from the diet (Robison et al., 1998); 8. facilitates determination of how late therapeutic intervention can occur and still have a beneficial effect, using removal of galactose as a standard for the best possible elimination of the cause (Robison et al., 1998); and 9. provides a convenient and relatively economic model—even though the galactose diet is expensive, the costs and maintenance of rats is much less than for larger species. For retinal studies, the galactose-fed dog provides the additional advantage of permitting partial fundus exams since, unlike the rat, the lens opacities induced by galactose are incomplete (Sato et al., 1991), thus permitting some views of the fundus. In addition, dogs can be made aphakic at an early age without much change in the progression of diabetic retinopathy (Kador, personal communication, 1999). Theoretically, galactose-fed mice should provide a good model for diabetic retinopathy. Since mice have very little aldose reductase in their lens, cataracts are not induced by either making them diabetic or galactosemic. Therefore, the development of lesions could be followed by fundoscopy as in diabetic humans and aphakic galactosemic dogs. A few microaneurysms have been reported (Kern and Engerman, 1996). However little other work has yet been carried out with the galactosemic mouse model. DIETS AND CARE FOR GALACTOSE-FED RATS Either male or female Sprague-Dawley rats can be used successfully, and probably most other strains of rats will respond similarly. Since most experiments are relatively long term, the rats must be purchased from a clean facility and be maintained in laminar flow racks with appropriate filters in a clean animal care facility. Even so, a 20% to 30% attrition from natural causes can be expected in both control and galactose-fed rats in experiments with durations of 24 months or more. Spontaneous tumor development is more frequent and the death rate is higher in rats on the control diet, perhaps because the galactose-fed rats have less body fat. Galactose-fed rats, whether untreated or treated with an aldose reductase inhibitor, exhibit decreased weight gain. The decreases in weight gain have ranged from 24% to 55% in different experiments (Robison et al., 1995b; 1996; 1997), possibly depending on the initiating time of the diet, the animal source, or the sex represented in the different studies, but the reason for the differential weight gains awaits further study. Apparently the lower weights are not caused by the 50% dilution of other nutrients in the diet by the galactose, since control diets having 50% starch or non-nutrient fiber (Robison et al., 1995b) represent a similar dilution yet render normal rat weights. To avoid retinal light damage, cage levels of illumination not exceeding 25 foot-candles should be used with light cycles of 12-hours-on/12-hours-off or 14-hours-on/ 10-hours-off. The percentage of galactose in the diet can range from 30% to 64% without compromising essential nutrients, but 50% is the level most commonly used. The most appropriate control diet is provided by substituting starch for the galactose, since the amount of feed rats consume is based on caloric content. Fresh batches of chow should be mixed regularly and the unused diet should be discarded at intervals determined by the shelf-life of the ingredients and storage temperature. To avoid the possibility of dental malocclusion the animals should be checked regularly and the teeth trimmed as needed. The diet and tap water are administered ad libitum. Cataract development can be assessed by slit lamp biomicroscopy after anesthesia using an intramuscular injection with 5 mg Rompun and 50 mg Ketamine/kg body weight followed by 1 to 2 drops in each eye of 1.0% Mydriacyl and 2.5% Neo-Synephrine to dilate the pupils. To

GALACTOSEMIC ANIMAL MODELS

241

minimize the time between food intake (nocturnal feeding) and tissue collection, animals should be killed in the morning. To equalize the effects of the unavoidable variation of the length of time between feed intake and tissue collection times on tissue carbohydrate levels among the treatment groups, an alternating killing schedule should be employed. For instance, only one or two animals from each group should be killed until all groups are represented and then this cycle can be repeated until all the animals in each group are killed. COSTS OF RAT DIET IN A TYPICAL PROTOCOL BY DURATION Using an average food consumption of 500 g/rat/month, the current rat food/ galactose costs, and the quantity of aldose reductase inhibitor needed based on the potency of sorbinil (the first potent inhibitor developed and industry standard), Table 2 shows the total diet costs for the recommended number of rats per killing time for testing three aldose reductase inhibitors or other compounds simultaneously. An adult rat (400 g) eats approximately 10 to 15 g diet per day (300– 450 g/month). However, if wasted food is included in the calculations, a rat actually uses 25 g/day (750 g/month). The cost of the galactose is high, but can be offset somewhat with bulk purchases (Table 2). The costs indicated do not include compound costs, rat purchases, or animal care charges. TISSUE CARBOHYDRATE MEASUREMENTS Plasma glucose can be measured with a clinical chemical analyzer (Abbott, Chicago, IL) by quantitating the reduction of NAD (nicotinamide-adenine dinucleotide) in a reaction that couples hexokinase and glucose-6 phosphate dehydrogenase activities (Bergmeyer et al., 1974). The percent glycated hemoglobin can be determined using GlycoTest II affinity columns (Pierce, Inc., Rockford, IL). The retinal hexose and polyol levels can be determined by pooling two whole retinas from fresh eyes. The galactose and galactitol levels in retina can be measured as their aldonitrile, alditol, and cyclitol acetate derivatives, respectively, by capillary gas liquid chromatography with mass spectrometric detection (Guerrant and Moss, 1984). For these assays, Table 2 Diet costs and compound requirements for retinopathy experimenta. Rats Per Group Duration

CON

3 months 10 8 months 10 16 months 10 24 months 20 Total 50 Experimental groups CON GAL

= control diet, 50% starch = 50% galactose diet

GAL

ARI1 ARI2 ARI3 Total Total Costsb Drug (g)c

10 10 10 20 50 Cost per rat per yearb ($60.00/rat/yr)

10 10 10 20 50

($240.00/rat/yr)

10 10 10 20 50

10 10 10 20 50

50 50 50 100 250

$ 2,550 $ 6,800 $13,600 $40,800 $63,750

7.5 20 40 120 187.5 g/drug

242

W.G.ROBISON, JR.

Rats Per Group Duration

CON

GAL

ARI−1

= 50% galactose diet with compound #1 = 50% galactose diet with compound #2 = 50% galactose diet with compound #3

($240.00/rat/yr)

ARI-2

ARI-3

ARI1 ARI2 ARI3 Total Total Costsb Drug (g)c

($240.00/rat/yr)

($240.00/rat/yr)

a No animal purchase or animal care costs are included. b Diet costs are based on a food consumption of 500 g/rat/ month, when the control diet costs $10.00/ kg and the 50% galactose diet costs $40.00/kg The costs are approximately those currently expected with purchases of 50 kg quantities of diet from Bio-Serve (Frenchtown, NJ). Discounts are available for larger purchases. No compound costs are included. c Total amounts needed for each aldose reductase inhibitor based on 0.05% in diet and a potency equivalent to sorbinil, the first potent aldose reductase inhibitor and industry standard.

retinas are homogenized in distilled water. Following tissue disruption, deuterium-labeled sorbitol (Isotec Inc., Miamisburg, OH) is added as an internal standard. The samples are deproteinized with 5% trichloroacetic acid, extracted with ether, lyophilized, derivatized with hydroxylamine hydrochloride, 4(dimethylamino) pyridine and acetic anhydride (Guerrant and Moss, 1984), and injected into a gas chromatograph equipped with a SPB-1 fused silica capillary column (Supelco Inc., Bellefonte, PA). Concentrations are calculated on the basis of an 8-point standard curve constructed by the addition of varying amounts of galactose, galactitol, and myo-inositol to samples of rat retina or erythrocytes from normal animals. Processing and analysis should be identical to those used for the unknown samples. ELASTASE RETINAL DIGEST PROCEDURES Intact retinal vasculatures can be isolated and mounted flat on slides for examination of retinal microangiopathies in the entire capillary plexus. This is accomplished by a series of incubations, agitations, and manipulations in mild enzyme concentrations to effect digestion and removal of all non-vascular retinal components. The procedure is called retinal digestion and the product is referred to as a retinal digest. Although digests were performed for many years with a crude trypsin extract or the extract in combination with pepsin or collagenase (reviewed by Robison et al., 1995a), it was found (Laver et al., 1993) that digestion with purified elastase (the most active component of the crude extract) provided a high quality preparation more consistently. The following is a more detailed and updated version of the original publication on the procedures for elastase digestion of fixed retinas to isolate intact retinal vascular beds. For more background and theory, see the original work. Enucleation of the Rat Eye:

GALACTOSEMIC ANIMAL MODELS

243

Figure 2 Rat Eye Diagram—indicating the position of the nictitans which can be used as a “handle” during enucleation and as a marker of the nasal side once the eye is removed.

1. Before enucleation, assure that the rat is well anesthetized by gently touching the eyelids to test for the lack of a lid closure response. 2. The eyelids may be removed at this point, to allow easy access, but this is not necessary. 3. Use the attached diagram (Figure 2) and a dissecting microscope, if necessary, to locate the nictitans in the nasal region of the eye. The nictitans can be used as a “handle” during enucleation and, because of its position, can also be used as a marker of the nasal region. 4. Grasp the nictitans with toothed forceps (e.g., Roboz RS #5103). 5. While maintaining a good grasp with the toothed forceps, gently pull the eye towards you, and use curved scissors with blunt ends (e.g., Roboz RS #5673) to cut the conjunctiva and the underlying muscles, Harderian gland, etc. from all sides of the eye. 6. While still pulling the eye gently, reach behind it to cut the optic nerve well behind the globe. The optic nerve should never be cut too short or myelin will be forced into the posterior regions of the retina and make digestion difficult or impossible. Leave 1.5 mm or more of optic nerve attached to the eye. 7. As an option if storage is desirable, proceed as follows: Rinse the eye briefly in phosphate buffered saline (PBS), blot to remove excess liquid, and immerse in optimal cutting temperature compound

244

W.G.ROBISON, JR.

(O.C.T., Tissue-Tek #4583) in a numbered plastic cryomold (Tissue-Tek II #4566). Freeze immediately on dry ice, wrap with aluminum foil (also numbered), and ship and/or store on dry ice or at—70°C until needed. When needed, thaw the eye(s) of interest in PBS at room temperature and rinse off the OCT compound. Removal and Preparation of Retina: 1. Gently slit the eye at the limbus (Figure 2) and immerse in a 10% formalin fixative buffered to pH 7.2 with a 50 mM solution of Na-K phosphate or in a 4.0% paraformaldehyde fixative to be mixed as follows using deionized water in all cases where water is indicated: Ultimate Composition: 1300 mM paraformaldehyde (4.0%) 50 mM Na-K phosphate buffer (pH 7.2) 233.7 mM sucrose (MW 342.3) (8.0%) Preparation of 8% paraformaldehyde: (a) Add 16 g paraformaldehyde to 200 ml H2O; heat to about 80°C in hood (b) Remove from heat, but continue to stir; add several drops of NaOH (IN) to solution until it clears (a few grains may remain) (c) Allow solution to come to room temperature Buffer Stock Solutions (750 mM) Solution A Solution B

102 g/liter KH2PO4 201 g/liter Na2HPO4 ●7H2O

Mix IX amount Mix 3X amount

Preparation of Buffer (200 mM): (a) 26.7 ml of Soln. A to 73.3 ml H2O (b) 80.1 ml of Soln. B to 219–9 ml H2O (c) Adjust to pH 7.2 with same solutions

Preparation of Fixative: 50.0 ml 8.0% paraformaldehyde 25.0 ml 200 mM Na-K phosphate buffer (pH 7.2) 8.0 g sucrose 25.0 ml H2O 2. Fix at room temperature for 4 hours or several days. 3. Cut around the limbus to remove the cornea and lens. 4. Keeping the retina intact, carefully dissect it free of the retinal pigment epithelium (RPE) and choroid. A fine artist brush works best.

GALACTOSEMIC ANIMAL MODELS

245

5. Cut away the sciera until you can flip the retina over to expose the optic nerve well—a spiral cut from the equator to the posterior pole works well. 6. Remove the optic nerve, leaving as small a hole as possible in the posterior pole. 7. Transfer the retina to deionized water to rinse out the fixative for 4 to 24 hours at room temperature. 8. In order to minimize damage, retinas should always be transferred between solutions using the large opening of a Pasteur pipette that has had the narrow end cut off and has been inverted to use the large end—never use forceps for transferring. Digestion and Mounting 1. Measure the elastase as follows: A Obtain elastase from porcine pancreas (CalBiochem #324689) B Calculate amount of elastase necessary (see Table 3): mg elastase=(ml buffer)(Units/ml conc.)(mg/Units) [e.g., if you have 239 U/mg elastase and you want a 20 ml solution at 40 units/ml, then calculate (20 ml)(40 Units/ml)(mg/239 Units) =3.35 mg elastase] Table 3 Recommended digestion times and enzyme concentrations for the initial digestion. Initial Digestion Tissue

Rinse

Conc. U/mla

Time

Amount

TRIS

Mouse Rat (control) Rat (galactosem ic)

overnight 5 hours

20 20 U/mla

6 min 15 min

10 ml 10 ml

overnight overnight

4–7 hours

40 U/ml

10ml

overnight

Dog Human

overnight overnight

10 U/ml 40 U/ml

10–30 min, depending on duration of galactosem ia 40 min 40 min

20 ml 20 ml

3 nights overnight

a The retinas of young control mice and some rats are so delicate that it is best to use 20 U/ml (less than 10 is usually too little) and digest for longer times. Also, rinsing in water for 2 to 3 nights following digestion can render a good result and less damage because less brushing is needed.

2. Prepare Digestion Buffer (100 mM Na-K phosphate buffer w/ 150 mM NaCl and 5mM EDTA, pH=6. 5): A Buffer Stock Solutions (750 mM) 102 g/liter KH2PO4 201 g/liter Na2HPO●7H2O B Prepare 100 mM Na-K phosphate buffer, pH 6.5 Solution A Solution B

Mix 1X amount Mix 3X amount

246

W.G.ROBISON, JR.

(a) 40.05 ml of Soln. A to 219–9 ml H2O (deionized) (b) 120.15 ml of Soln. B to 659.7 ml H2O (deionized) (c) Adjust to pH 6.5 with same solutions C Add 8.7 g NaCl per 1000 ml of buffer D Add 1.68 g EDTA per 1000 ml of buffer E Readjust pH with IN HC1 if necessary NB: All solutions (Digestion Buffer, Na-K Phosphate Buffer, Solutions A and B) can be stored in refrigerator for an extended time. 3. Add the elastase to the digestion buffer, vortex the solution, and place in a shaking water bath at 37°C for 10 minutes to warm up. 4. Place the retina in the digestion solution for an initial digestion following the general recommendations of Table 3. The ideal times will vary depending on the rat age and duration of galactosemia. Make sure shaking is gentle— not so vigorous that it damages the retina. 5. Place retina in TRIS buffer overnight at room temperature. The TRIS Buffer should be made up fresh each week as follows: 4.42 g TRIMZA HC1 8.72 g TRIMZA Base 1000 ml H2O Adjust to pH 8.5 6. Place the retina in deionized water and examine through a dissecting (operating) microscope with darkfield illumination. 7. Use fine brushes to gently agitate and remove loose tissue while observing carefully. Check around the edges of the retina to see if the capillaries are beginning to separate from the inner limiting membrane (ILM). Eventually, the ILM and the attached vitreous must be brushed completely free of the intact vasculature. After gentle attempts to free all loose tissue and the ILM without damaging the vessels, additional digestion for a few minutes should be performed followed by further gentle attempts to remove nonvascular tissues. Then the preparation should be rinsed again in TRIS buffer overnight at room temperature. 8. Repeat steps 5–7 one or more times, until ILM and all debris are removed. If more than three overnight rinses are required, the last ones should be done at 4°C, and perhaps similar retinas should be digested for longer times on the first day. 9. Flatten the capillary bed: Make a radial cut from the outer rim to the optic nerve region. Make a circumferential cut to remove all remaining portions of the optic nerve with as little retina as possible. Cut away any large vessels and tissue at the optic nerve region which would prevent the coverslip from lying flat. If necessary, make additional, smaller cuts (notches), around the border to completely flatten. 10. Fill a finger bowl or other container, which is approximately 2" deep and 4" in diameter, to the brim with deionized water or DPBS (–Ca, –Mg) and “sweep” free of dust by drawing a lint-free tissue across the surface. Place the vessel preparation on the water surface and flatten out the capillary bed. 11. Mount on slide: Place a silanized microscope slide under the retinal vasculature and slowly bring the slide up and out of the DPBS with the retina flattened out on top. Alternatively, the slide can be placed on a pedestal under the water surface, and the vasculature lowered onto it by removing the DPBS from the dish with a large syringe until the DPBS level is below slide level. For best results, ensure that the bowl and pedestal are level. It is important to make sure all areas of the retina stick to the silanized

GALACTOSEMIC ANIMAL MODELS

247

slide evenly to avoid folding and distortion of the natural vessel pattern. Silanized microscope slides can be purchased as catalog number 51308 from Oncor (Gaithersburg, MD) or as catalog number 1010– 1001 from Digene (Beltsville, MD). They generally perform better than gelatin-coated slides. 12. Place the slide on toothpicks or applicator sticks in a petri dish. Using small pieces of filter paper cut to triangular shapes, remove the excess DPBS from around the retina, until only a 1.0 mm-thick margin remains surrounding the retina. Once finished, cover to protect from dust and leave at room temperature until completely dry. For best results, ensure that the petri dish is level. 13. Stain the dried slides by the periodic-acid-Schiff procedure and counterstain with hematoxylin procedures. Best results are obtained when rat retinas are left in well-filtered Schiff reagent for 60 minutes. MEASUREMENTS OF CAPILLARY BASEMENT MEMBRANE THICKNESS Thickening of capillary basement membranes greater than that which occurs with aging alone has been considered to be an ultrastructural hallmark of diabetic microangiopathies (Williamson and Kilo, 1977). Although the cause and consequences of the thickening are not fully understood, it serves as a marker of diabetic complications. If measurements of capillary basement membrane thickness are done carefully they can provide a quantitative basis for estimates of duration and disease progression. Since the measurements require electron microscopy and, therefore, examination of a very small region, great care must be taken to minimize sampling error. Retinal studies permit sampling from similar, accurately defined, locations in different retinas. As shown in several studies (Fischer and Gärtner, 1983; Nagata et al., 1986), it is very important to distinguish whether the measurements of retinal capillary basement membrane (RCBM) thickness are made in the ganglion cell layer or the outer plexiform layer. Since RCBM thickness is greater and more variable in the ganglion cell layer, it is recommended that measurements for assessing the progression of diabetes-related retinal complications be limited to capillaries of the outer plexiform layer. It is also important that the tissue samples be taken from the same retinal quadrant, that more than one sample be taken from each retina, that electron micrographs be taken of all reasonably well transected capillaries found within the outer plexiform layer, that only those micrographs with capillaries transected in a plane that is close to perpendicular to the long axis be chosen for morphometric measurement, that the person choosing the micrographs have an understanding of the quality needed, and that the selection be done under fully masked conditions. An obliquely transected capillary would give a measurement higher than the actual RCBM thickness. Using prints from quality electron micrographs of perpendicular sections enlarged to final magnifications of 12,000 to 25,000 is recommended. Although several different morphometric methods have been utilized for determining basement membrane thickness over the years (Williamson and Kilo, 1977), use of some type of computer-assisted morphometry generally gives the most reproducible results (Robison et al., 1983; 1986; McEwen et al, 1987). Whether the micrographs are digitized for computer analysis by tracing over a digitizing tablet or are captured by video, they need to be traced by a person or persons having expertise in capillary structure including knowledge of the various compartments and RCBM fenestrations seen in transections, but having no knowledge of the experimental groups. For video capture, especially, the tracing is done best on micrographs printed for low contrast and low density. As viewed in transection, the RCBM can be separated into different regions based on the cell types with which it interfaces: 1) endothelial cell and glial cell; 2) endothelial cell and pericyte; 3) pericyte and glial cell; 4) pericyte and pericyte. Some investigators have limited their measurements to one or two of these regions, often to the endothelial cell/ glial cell region. Since that region occupies only 25% to 30% of the surface of retinal capillaries, which have greater pericyte coverage than do capillaries of other tissues, such limiting of sample region could be problematic in retinal

248

W.G.ROBISON, JR.

studies. Some compensation can be achieved by modifications like those used by Frank et al. (1983). However, measurement of all the regions of RCBM and computer-assisted calculation of the average thickness for each capillary are recommended (Robison et al., 1983; 1986; McEwen et al, 1987). These procedures provide the largest sample, minimize the effect of focal thickening, and give the most conservative measurement of any deviations from normal. A Sony 3CCD color video camera (model DXC-930, Sony Corporation, Japan) with 640×480×24-bit resolution and the C-Imaging 1280 Image Analysis System (Compix, Inc. Imaging Systems, Cranberry Township, PA) can be used successfully for digitizing and analysis of micrographs traced so that each capillary compartment is color coded. MORPHOMETRY OF RETINAL VASCULAR WHOLE MOUNTS Isolation of intact retinal vasculatures from rat retinas with very little distortion of the original patterns of vessels is now possible by using the elastase digestion procedures described in detail above. Consistent preparation of clean and undistorted Table 4 Definitions of measurements of retinal vessel whole mounts performed by computer-assisted image analysisa. Measurement

Definition

Retinal area measured

Area (mm2) of vascular bed from which 10 fields of retinal capillaries are digitized (total area to be ca. 5.3 mm2) The total area (µm2) per animal occupied by the capillaries themselves within the Retinal area measured The total length of capillaries (mm) in the Retinal area measured Average width (µm) of capillaries determined as the Capillary area divided by the Capillary length The percent of the Retinal area measured occupied by Capillary area Dilated channels Percent of Capillary area occupied by dilated channels (>20 µm wide) which appeared to be focally enlarged capillaries Number of microaneurysms in the Retinal area measured

Capillary area Capillary length Capillary width Capillary density

Microaneurysms

a Modified from Robison et al. (1995b) and reproduced with permission of the Association for Research in Vision and Ophthalmology.

vessel mounts permits semi-automated analysis of several vessel parameters according to the definitions provided in Table 4. All vessels in a predetermined region of the central and mid-periphery of each retina are systematically digitized and analyzed (Robison et al., 1995b; 1996). It is recommended that the measured region consist of an arc-shaped row often contiguous fields centered 1 mm from the margin of the cut used to remove the optic nerve. In this way, the same retinal region can be utilized in all rats. The total area measured in the retina of each rat should generally be at least 5.0 mm2. Digitization and analysis of the retinal vasculatures can be done using a Dage-MTI 81 series black and white video camera with resolution of 1600 TV lines (Dage-MTI, Inc., Michigan City, IN) and the CImaging 1280 Image Analysis System with 1024×1024×8 bit resolution (Compix, Inc. Imaging Systems, Cranberry Township, PA). Digitized images of the retinal vasculature can be measured for changes in various parameters: capillary length, capillary width, capillary density, percent dilated channels, and numbers of microaneurysms using the definitions in Table 4, based on the proportions of pixels involved in the digitized images. For example, the total length of all capillaries in the Retinal Area Measured is determined by a program which converts

GALACTOSEMIC ANIMAL MODELS

249

the digitized images of all the capillaries to their central axes of only one or two pixels in width (“skeletonization”) and then determines the total number of such pixels that are aligned in tandem in a given field. The pixel size for the magnification used is calibrated in usable units of measure (e.g. µm) using a stage micrometer. Having this information, the computer quickly determines the total length of capillaries in a digitized field and likewise adds the capillary lengths from all fields measured, thus giving data from the total Retinal Area Measured. Mean capillary width is determined by the capillary area (based on the number of pixels occupied by all the capillaries) before “skeletonization” divided by the total length of all capillaries in the Retinal Area Measured. By automated computer size selection, only vessels with widths < 15 µm are used for length and width measurements of capillaries, and only vessels with widths > 20 µm are used for the measurements of dilated channels. Using this computer-assisted criterion for vessel size selection, the amount of minor capillary dilation (mainly diffuse dilation) can be differentiated from measurements of dilated channels, which are used to assess the development of major capillary enlargements (mainly focal, IRMA type lesions). MECHANISM(S) UNDERLYING DIABETIC COMPLICATIONS The Diabetes Control and Complications Trial (DCCT) (1993) has established the long-suspected (Davis, 1988) primary role of hyperglycemia in diabetic complications. Similarly, studies on galactose-fed animals have demonstrated a primary role for galactosemia in causing diabetic-like retinal complications (Robison et al., 1989; 1990a; 1995a; 1996). What might be the common mechanism that can be triggered by chronically elevated plasma levels of either of the corresponding hexoses, glucose or galactose? The following similarities among the ocular tissues affected provide some elucidation: 1) insulin is not required for glucose or galactose uptake in any of the ocular tissues that develop complications; 2) all these tissues have high concentrations of aldose reductase, an enzyme which reduces either of these aldohexoses to its polyol form; 3) all exhibit marked accumulations of polyol, either sorbitol or galactitol, with hyperglycemia or galactosemia, respectively; and 4) all show that blocking of polyol accumulation is accompanied by amelioration or prevention of the complications when the animals are treated from disease onset with an inhibitor of aldose reductase. The cumulative evidence suggests that a substrate-driven increased flux through aldose reductase is the primary event that triggers what has been considered to be a cascade of events leading to cell and tissue complications in both diabetes and galactosemia (Robison et al., 1995a). Although more than one metabolic perturbation results from flux through aldose reductase, only one of these, the accumulation of polyol, appears to be necessary in the lens to cause the cell toxicity that results in cataracts (Chiou et al., 1980; Kinoshita and Nishimura, 1988). The definitive proof of polyol accumulation as the main causative factor in cataracts comes from mice made transgenic for human aldose reductase inhibitor and having a lens promoter (Lee et al., 1995). Normally, mouse lenses have very low levels of aldose reductase and do not develop cataracts even when made galactosemic. The investigators were able to induce cataracts in the transgenic mice by making them galactosemic or diabetic. The rate of cataract development corresponded directly to the amount of polyol accumulation, whether sorbitol or galactitol. Unlike sorbitol, galactitol is not metabolized (Figure 1). Therefore, galactosemic animal models not only permit the isolation of hexose as the primary cause of diabetic ocular complications, but eliminate the possibility that metabolic alterations caused by the second enzyme of the polyol pathway, sorbitol dehydrogenase [such as its product (fructose)] are needed to induce diabetic-like tissue damage.

250

W.G.ROBISON, JR.

DIABETIC RETINOPATHY—PREVENTION STUDIES Several studies of galactosemic dogs using different aldose reductase inhibitors alone or in combination, including sorbinil (Pfizer, Groton, CT) and M79175 (Eisai, Tokyo, Japan), have shown that both cataract formation and diabetic retinopathy are ameliorated or significantly delayed in a dose-related manner by blocking polyol formation without altering the levels of glycated hemoglobin (Kador, 1990; Kador et al, 1988; 1990; 1994; Takahashi et al, 1992;1993). Although some negative findings regarding the efficacy of aldose reductase inhibitors have been reported (Engerman and Kern, 1989; 1991; 1993; Kern and Engerman, 199D, these were based on limited experimental results and have been refuted by several lines of evidence (Robison, 1994; 1995; Robison et al, 1995a; 1995b; 1996). Many findings from inhibitor-treated, galactose-fed rats demonstrate conclusively that inhibitor treatment not only prevents cataracts, but also prevents capillary basement membrane thickening and the entire spectrum of diabetic-like microangiopathies through the mild proliferative stage (Figure 3). In four such studies totaling 136 rats (30 of which were galactosemic for >24 months), galactose-fed rats had high blood levels of galactose (ca. 200 mg/dL) and polyol (ca. 35 mg/ dL) at 4, 6, 8, 16, 18, and 24 month endpoints, but were relatively healthy and had normal life spans. The retinas of untreated rats which were galactosemic Table 5 24-month rats: retinal microangiopathies in relation to carbohydrate levels in plasma, erythrocytes, and retinaa.

Retinal microangiopathies: Capillary length (mm) Capillary width (µm) Capillary density (%) Dilated channels (%) Microaneurysms (number) Carbohydrate levels: Plasma glucose (mM) Glycated hemoglobin (%) Erythrocyte galactose (nmol/mg hemoglobin) Erythrocyte galactitol (nmol/mg hemoglobin) Retinal galactose (nmol/mg wet weight) Retinal galactitol (nmol/mg wet weight) Retinal myo-inositol (nmol/ mg wet weight)

Control

Galactose

GAL+ARIb

243.1±3.6 8.2±0.2 38.0±1.2 1.1±0.2 0.0

306.5±8.2C 10.1±0.3C 58.8±3.0C 10.9±2.4C 5.0±1.6C

253.4±5.5d 8.7±10.2d 41.9±1.4d 4.1±11.4d 1.0±0.5d

5.99±0.39 2.4±0.25 NDe

5.46±0.41 10±0.81C 17.8±6.73C

5.7±0.82 9.5±0.66C 17.0±5.67C

NDe

2.7±1.3C

0.25±0.15d

NDe

6.2±2.44C

4.7±1.69C

NDe

5.1±0.4C

0.26±0.03

1.1±0.11

1.8±0.11

1.4±0.24

a Modified from Robison et al. (1996) and reproduced with permission of the Assocviztion for Research in Vision and Ophthalmology. Values are means standard error, using 6 to 8 separate rats for each measurement at the 24month time point. b GAL+ARI=50% galactose-fed rats treated with the aldose reductase inhibitor WAY-121, 509. c Significantly decreased (p<0.01) compared with control rats. d Significantly decreased (p<0.05) compared with galactose-fed untreated rats. e ND=Non-detectable (below the limit of quanification).

GALACTOSEMIC ANIMAL MODELS

251

Figure 3 Prevention of Diabetic-like Retinal Microangiopathies—showing normal retinal capillary meshworks (m) with intact pericytes (p) and endothelial cells (e) in a rat fed 50% galactose for 24 months but treated with an aldose reductase inhibitor (A), compared with the dilated, hypercellular meshworks (dm), acellular (ac), dilated (dc), and occluded (oc) capillaries, as well as microaneurysms (ma), dilated and tortuous veins (V), and focal losses of pericytes in an untreated galactose-fed rat (B). X300. The calibration bar represents 50 µm. Reproduced from Robison and Laver (1993) with permission of Smith-Gordon and Co. Ltd.

252

W.G.ROBISON, JR.

for 24 months or more exhibited the full range of diabetic-like retinal microangiopathies through the mild proliferative stage. All the lesions were essentially prevented with AL-3152, sorbinil, tolrestat, or WAY-121,509, four structurally distinct aldose reductase inhibitors. The inhibitors were evenly mixed with fresh diet in quantities sufficient to maintain a daily ingestion of 14, 65, 50, and 10 to 25 mg/kg body weight, respectively (Robison and Laver, 1993; Robison et al, 1983; 1986; 1988; 1989; 1990a; 1995b; 1996). Aldose reductase inhibitor treated galactosemic rats had lowered blood polyol levels (<3.0 mg/dL), but did not differ significantly from untreated galactosemic rats in blood glucose, galactose, or glycohemoglobin levels. The treated rats also remained free of cataracts throughout a period of 28 months, indicating that the continual oral administration of an aldose reductase inhibitor was sufficient to prevent damage in at least one tissue besides the retina. In the most recent study (Robison et al, 1996), retinal aldose reductase activity was inhibited by 95% with no decrease in retinal glucose, galactose, or glycated hemoglobin levels (Table 5), clearly indicating a causative role for increased aldose reductase activity. Studies from two independent laboratories have confirmed the development of diabetic-like retinopathy through the severe nonproliferative stage in galactosemic rats with either 30% or 50% dietary galactose. One of those (Kern and Engerman, 1995), however, did not demonstrate prevention with aldose reductase inhibitors, probably due primarily to the utilization of a lower inhibitor dose, which decreased retinal polyol by only 62%. Other possibilities for the discrepancies on prevention have been discussed in detail elsewhere (Robison, 1994; 1995; Robison et al., 1995a; 1995b; 1996). The other independent study (Frank et al, 1997) confirmed both the validity of the galactosemic rat model for diabetic retinopathy and the efficacy of aldose preproliferative retinopathy. The investigators demonstrated that both aminoguanidine and an aldose reductase inhibitor prevented galactose-induced increases in retinal levels of vascular endothelial growth factor, but only the aldose reductase inhibitor prevented the galactose-induced retinopathy. These clear findings on the efficacy of aldose reductase inhibitors in preventing retinopathy in galactosefed rats treated from the onset provide strong support for the positive effects of aldose reductase inhibitors reported in galactose-fed dogs and diabetic rats (Kojima et al., 1985a; 1985b; Chakrabarti and Sima, 1987; 1989). Together, these independent studies on the rat and dog models suggest that diabetic retinopathy in humans may be ameliorated by treatment with a sufficiently potent aldose reductase inhibitor, if treatment begins at the time of disease onset (Tomlinson et al., 1992). CLINICAL TRIALS Insofar as plasma glucose can be maintained at normal levels, the complications of diabetes should be preventable. Insulin treatment, when designed to provide the tightest possible blood glucose control by multiple injections or continuous subcutaneous insulin infusion pumps (DCCT Research Group, 1993; Reichard et al., 1993; Santiago, 1992; Zinman, 1989), has significantly delayed the progression of several diabetic complications, including diabetic retinopathy. However, the methods used to obtain tight glucose control would be difficult to implement in the general population of diabetic patients without significant risks. Therefore, a new approach to preventing diabetic complications is needed. Aldose reductase inhibitors used in addition to conventional insulin treatment have the potential of blocking the toxic effects of the supernormal levels of glucose which the best insulin therapy is unable to avert. Although a three-year, multicenter clinical trial to test the aldose reductase inhibitor sorbinil as a complementary treatment of diabetic retinopathy showed no clear beneficial effect (Sorbinil Retinopathy Trial Research Group, 1990), Pfeifer et al. (1997) suggest that new clinical trials with aldose reductase

GALACTOSEMIC ANIMAL MODELS

253

inhibitors should be initiated. They propose that future trials should be designed with increased statistical power, more quality control, and longer duration so that beneficial effects would be more likely to be detected even though a reversal were not effected. DIABETIC RETINOPATHY—EARLY AND LATE INTERVENTION STUDIES For ethical reasons, the intervention therapies applied in clinical trials are usually delayed until after retinal damage has progressed to the point that it can be seen by fundus exam. Intervention studies in animal models and their extrapolation to the human condition are of inestimable value for determining how late therapy can be implemented and still have an effect. Now that diabetic-like retinopathy can be induced within a relatively short time in the new, galactose-fed animal models, intervention studies which were unfeasible previously can be undertaken. Thus, two studies (Robison et al., 1997; 1998) which combined prevention and intervention were carried out using relatively new aldose reductase inhibitors (AL-3152 and WAY-121,509) on galactosemic rats to test the following possibilities: 1) preventing diabetic-like retinopathy by constant inhibitor treatment from initiation of 50%-galactose feeding; and 2) delaying or halting diabetic-like retinopathy in spite of postponing intervention by addition of inhibitor or withdrawal of galactose until 4, 6, or 8 months after galactosemia was induced. Unfortunately, the findings from both studies showed that even the earliest intervention used did not halt the progression of microangiopathy development and a significant delay was not seen until after 16 to 20 months of intervention. At 4 months, when the earliest intervention was begun (Robison et al., 1998), the untreated galactose-fed rats exhibited a 23%, statistically significant (p <.01), increase in capillary basement membrane thickness and pericyte degeneration. The RCBM thickness continued to increase in both the inhibitor treated and galactose withdrawal groups, even though they showed 95% and 100% decreases in retinal galactitol, respectively. The RCBM thickness had become another 45% thicker in the galactose withdrawal group at the 24-month endpoint. Since not even withdrawing galactose, and thus decreasing retinal polyol to undetectable levels, halted the progression, apparently biochemical perturbations occur early in galactosemia that are not readily reversible. Because only rats which had elevated retinal polyol levels for at least 4 months also exhibited retinopathy, it would appear that this diabetic-like complication, like diabetic cataracts, is probably initiated by polyol accumulation and not by excessive tissue glycation. Although the levels of glycated hemoglobin increased with galactose feeding, as in the prevention studies (Table 5), they did not correlate with either cataracts or retinopathy. The intervention findings in galactose-fed rats are consistent with intervention studies in diabetic animals (Engerman and Kern, 1987; 1995b; McCaleb et al., 1991), which also indicate that reversal does not occur and halting of damage is limited. The effect of intervention therapy was studied in diabetic dogs receiving poor glycemic control for 2½ years followed by good glycemic control for another 21/2 years (Engerman and Kern, 1987). No microaneurysms developed in the 2½ years of poor glycemic control. However, in spite of intervention with good glycemic control for the next 21/2 years, many microaneurysms formed and the overall vessel damage was greater than that found in diabetic dogs receiving good glycemic control for 5 years. Intervention in humans by laser photocoagulation (Aeillo et al., 1985; Early Treatment Diabetic Retinopathy Study Research Group (ETDRS), 1991a; 1991b; Aeillo, 1994) or pituitary ablation (Aeillo et al, 1985; Speakman et al., 1966) decreases hemorrhages and proliferative changes but does not reverse all the lesions of diabetic retinopathy. Pancreas transplantation in humans must be relatively early in order to show a clear effect (Petersen et al., 1990).

254

W.G.ROBISON, JR.

The cumulative evidence on diabetic complications indicates that probably only a delay or slow halting of ongoing degenerations is all that can be effected once diabetes-related tissue damage has progressed significantly. Perhaps there is a damage threshold beyond which reversal is impossible. In sugar cataracts, intervention must take place within 5 days to permit reversibility (Akagi et al., 1988; Hu et al, 1983; Simard-Duquesne and Dvornik, 1973). Perhaps swelling of even a few cells to the point of membrane damage and/or rupture may define the irreversible point for the whole lens since its cells are so interdependent. In diabetic retinopathy, degeneration of intramural pericytes, which have a high concentration of aldose reductase (Akagi et al, 1983; Nishimura et al, 1988; Hohman et al, 1989) and normally provide structural support and autoregulation in the microcirculation, may be the turning point. Once lost, pericytes would be replaced very slowly if at all (Engerman et al., 1967). However, unlike lens cells, degeneration of a few retinal vasculature cells and their sequelae could remain very focal. Pericyte loss and microaneurysm formation are correlated topographically, both occurring predominantly in the central retina (Takahashi et al., 1993). So, while prevention therapy could completely prevent loss of pericytes and its sequelae throughout the retina, intervention therapy would only be expected to prevent loss of pericytes in regions not yet affected. Assuming that pericyte loss initiates most of the subsequent vessel lesions, halting of pericyte loss by intervention would not halt subsequent lesions in those capillaries which had already lost several pericytes. The disease progression would be expected to continue in such capillaries until the effects of pericyte degeneration were exhausted. Evidence of beneficial effects of intervention therapy may be masked for a long time, until such existing tissue damage completely reveals its cascade of consequences. Upon contemplating the extent of retinal vessel damage revealed by the newly developed histological methods of his day, Ashton (1950) stated: It is believed that it has not previously been realised how surprisingly numerous micro-aneurysms are and the picture is a depressing one for one wonders how it can be possible to reverse such a gross and widespread process by the administration of drugs or the control of diet. At best we can only hope to prevent the development of such lesions or, once the condition is established, to attempt to control the haemorrhages. In summary, cataracts and diabetic-like retinopathy are both prevented in galactose-fed rats if treatment with aldose reductase inhibitors is started simultaneous with the galactose insult. But, intervention for retinopathy after 4 months (Robison et al., 1998) or for cataracts after 5 days (Akagi et al., 1988; Hu et al., 1983; Simard-Duquesne and Dvornik, 1973) by removal of galactose or addition of an aldose reductase inhibitor provides no reversal of retinopathy or cataract, but does cause a marginally significant delay of retinopathy, if the intervention is continued long enough. RELATIVE ROLES OF GLYCATION AND OXIDATION Several investigations have been launched to discover other factors contributing to the origin and/or causing the irreversible nature of diabetic retinopathy. Glycation and oxidation are two of the more probable candidates for having an influence. The evidence cited herein indicates that aldose reductase inhibitors prevent diabetic cataracts (Chiou et al., 1980) and diabetic-like retinal microangiopathies (Robison et al., 1996) without altering the galactose-induced elevation of protein glycation levels. Before glycation can be considered to be a major contributor, there needs to be an explanation of the observations summarized in previous reviews (Robison and Laver, 1993; Robison et al., 1995a) as follows:

GALACTOSEMIC ANIMAL MODELS

255

Figure 4 Non-enzymatic Glycosylation (Glycation) of Glucose Compared with Galactose —drawn to indicate the three- to fourfold greater glycation of cell proteins and other components by galactosemic in contrast to diabetic conditions.

1. although blood glucose levels can be very high (ca. 500 mg/dl) in the diabetic rat, the blood galactose levels remain relatively low (ca. 200 mg/dl) in the galactose-fed rat (Robison et al., 1989; 1996), yet sugar-induced cataracts and retinopathy take much longer to develop in the diabetic than in the galactose-fed rat (Datiles et al, 1982; Robison et al, 1989; 1990a; 1995a); 2. congenitally hyperglycemic mice with high glycation levels do not develop a cataract, even when fed galactose, apparently because their lenses exhibit only one-tenth the aldose reductase activity found in rat lenses (Varma and Kinoshita, 1974); 3. the degu rodent which is essentially normo-glycemic (150 mg/dl blood glucose) but has high lens aldose reductase activity develops a cataract after ingesting normal laboratory rat chow and develops a cataract much sooner than any other rodent (within 10 to 12 days) if made diabetic (465 mg/dl blood glucose) by streptozotocin injection (Varma et al., 1977); 4. glycated hemoglobin levels remain unchanged in galactose-fed dogs and rats treated with aldose reductase inhibitors (Kador, 1990; Kador et al., 1990; 1994; Engerman and Kern, 1991; Kern and Engerman, 1991; Robison and Laver, 1993; Robison et al., 1996), yet cataract development and retinal vascular abnormalities are ameliorated (Kador et al., 1994; Robison et al., 1996); 5. inhibition of aldose reductase prevents the galactose-induced cataracts in rats without decreasing the elevated glycosylation levels of the lens proteins (Chiou et al., 1980); 6. inhibition of aldose reductase prevents hexose-induced retinal capillary basement thickening (Frank et al, 1983; Robison et al, 1983; 1986; 1998; Chakrabarti and Sima, 1989) and advanced retinal microangiopathies in rats without altering the blood insulin, glucose, galactose, (Robison et al, 1989; 1990a; Robison et al, 1996) or glycohemoglobin levels (Robison et al, 1996); 7. although pentosidine starts accumulating soon after galactose feeding (Nagaraj et al, 1994), it remains to be demonstrated that advanced glycation end products could accumulate fast enough to cause the rapid development of some complications of hyperglycemia such as the sugar cataracts (Robison et al., 1990b; 1995a), which correlate closely with the incidence of diabetic retinopathy; and 8. transgenic mice which express aldose reductase activity in the lens develop cataracts at rates directly correlated with the amounts of polyol which accumulate in diabetes and galactosemia (Lee et al, 1995), leaving little doubt that increased flux through aldose reductase alone triggers the cascade of events that leads to sugar cataract formation.

256

W.G.ROBISON, JR.

Nevertheless, glycation levels (Bunn and Higgins, 1981; Urbanowski et al, 1982) as well as polyol accumulation (Sato and Kador, 1990) are increased markedly (three- to four-fold) with galactosemia (Figure 4); and there are indications (Baynes, 1996; Bucala et al, 1995) that oxidative stress in combination with glycation might have detrimental consequences through the cumulative formation of highly oxidized advanced glycation end products (AGES). Therefore, microangiopathies in yet another animal model, the vitamin E deficient rat, have been compared with those of galactose-induced diabetic-like retinopathy to assess the relative role of oxidative stress in the development of diabetic retinopathy. The retina experiences unusually high oxygen flux (Yancey and Linsenmeirer, 1989), and the outer segment membranes of its photoreceptor cells are very susceptible to lipid peroxidation (Farnsworth and Dratz, 1976; De La Paz and Anderson, 1992), owing to their extraordinarily high content of polyunsaturated fatty acids (De La Paz and Anderson, 1992; Fleisler and Anderson, 1983). The photoreceptor outer segments are normally rich in α-tocopherol, a potent endogenous antioxidant (Farnsworth and Dratz, 1976; Dilley and McConnell, 1970; Stephens et al., 1988). As might be predicted, dietary deficiency in αtocopherol causes damage to photoreceptor rod outer segment (ROS) membranes and marked accumulations of oxidized membrane remnants, including lipofuscin (aging pigment) granules in the retinal pigment epithelium (Farnsworth and Dratz, 1976; Hayes, 1974; Katz et al., 1978; Robison et al., 1979; 1982). Eventually, most photoreceptor cells degenerate, the cones being the last to be lost. The oxidative stress induced by deficiency in α-tocopherol might be expected to damage other retinal tissues as well. Normal retinal capillaries have relatively high proportions of polyunsaturated fatty acids (Lecomte et al., 1996), suggesting they are susceptible to autoxidation. However, relatively little information is available concerning the effects of α-tocopherol deficiency on the retinal vasculature. Investigators have proposed that oxidative stress induced by hyperglycemia may be involved in the development of diabetic complications, including retinal microangiopathy (Kowluru et al., 1997; Ansari et al., 1998). Supporting evidence includes diabetic tissue measurements of free radicals (Giugliano et al., 1996), increases in oxidation products (Domininiguez et al., 1998; Traverse et al., 1998), decreases in antioxidant defense mechanisms (Altomare et al., 1997; Kowluru et al., 1994; 1997), and normalizations with antioxidant treatments (Ansari et al., 1998). If oxidative stress does, indeed, play a significant role in diabetic retinopathy, one would predict that impaired antioxidant defenses in the retinas of α-tocopherol deficient rats would induce vascular lesions similar to those seen in diabetes. A study was designed to determine whether microangiopathies might be induced in retinal capillaries by α-tocopherol deficiency and whether these would be similar enough to those characteristic of diabetic retinopathy to implicate oxidative stress as one of the factors leading to diabetic microvascular disease (Robison et al., 2000). Rats were fed a basal, chemically defined diet either with (adequate group) or without (deficient group) α-tocopherol. After 6 and 8 months, some rats were killed, the eyes removed, and the retinas were digested by elastase to evaluate lipofuscin-specific autofluorescence (LFSAF) and cell integrity within isolated retinal vessels. Other eyes were prepared for localization of LFSAF in cryostatsectioned retinas. After 8 and 14 months, tissue samples from the central retina of one eye per rat were examined by electron microscopy for RCBM thickening and other ultrastructural changes (Figure 5). At 6 and 8 months the deficient rats exhibited extensive disruption of rod outer segment (ROS) membranes, marked loss of photoreceptor cells, and pronounced increases in the numbers of granules with LFSAF in retinal vessels as well as in the retinal pigment epithelium. At 14 months, the same changes were more severe and there was a 13.6% increase in RCBM

GALACTOSEMIC ANIMAL MODELS

257

Table 6 Retinal capillary basement membrane (RCBM) thickness. Treatments! Months

n

Thickness (nm)a

% Increase by age

% Increase by treatment

Adequate α-tocopherol/8 m Deficient α-tocopherol/8 m Adequate α-tocopherol/14 m Deficient α-tocopherol/14 m STZ diabetic/8 m STZ diabetic/14 m

6 8 8 8 3 5

83.2±1.4 89.0±2.1 86.9±2.9 98.7±2.6b 142.5±16.1C 149.1±3.8d

— — 4.5 11.0 — —

— 7.0 — 13.6 45.1 53.1

a Mean ± standard error of the mean in nanometers b Significantly increased (p<0.05) compared with age-matched adequate control c Hotta et al., 1988 d Sima et al., 1988

thickness (98.7±2.6 nm vs. 86.9±2.9 nm). There is a spectrum of more than 10 lesions (Table 1) that defines the histopathological alterations characteristic of diabetic retinopathy (Robison et al., 1995a), and only one of these (RCBM thickening) was induced by α-tocopherol deficiency, and this, to a much more moderate degree. Although there was no observable qualitative difference between the RCBM thickening induced by α-tocopherol deficiency and that characteristic of diabetic retinopathy, there was a marked quantitative difference. The increases in the RCBM thicknesses that have been reported in diabetic retinopathy are much greater than both those observed in the α-tocopherol-deficient rats of this study and those characteristic of aging. Within 3 to 5 months of streptozotocin-induced diabetes in rats there is a 13– 4% increase in RCBM thickness compared with controls (Hotta et al., 1988). Similar rats with durations of streptozotocin-induced diabetes of 8 to 14 months exhibited 34% and 53.1% increases in RCBM thickness, respectively (Hotta et al., 1988; Fischer and Gärtner, 1983; McCaleb et al., 1991; Sima et al., 1988). This is almost four times as great as the RCBM thickening that occurred as a result of α-tocopherol deficiency over the same time period (Table 6). The mechanism underlying the moderate RCBM thickening seen in response to the oxidative stress of α-tocopherol deficiency is not known. RCBM thickening occurs normally with aging, amounting to approximately 13% when 11-month-old rats are compared with 5-month-old rats (Nagata et al, 1986). Deficiency of α-tocopherol has been thought to accelerate the aging process via acceleration of cumulative oxidative damage (Robison et al., 1979; 1982). All evidence taken together suggests that the RCBM thickening and accumulations of lipofuscin granules in capillaries are probably more related to a modification in the aging process than to diabetic retinopathy. Both lipofuscin accumulation and a moderate RCBM thickening associate α-tocopherol deficiency with aging while only RCBM thickening links α-tocopherol deficiency to diabetic retinopathy, and this link is weak owing to the marked differences in thickening. Data from the present study suggest that, while oxidative stress may contribute to RCBM thickening in diabetic retinopathy, other unknown mechanisms play far more significant roles. In all aspects other than RCBM thickening, the effects of α-tocopherol deficiency on retinal capillaries differ greatly from those of diabetic retinopathy. One of the earliest alterations in retinal capillary cells in diabetes is the selective degeneration of pericytes, while in α-tocopherol deficiency the endothelial cells appear to be affected first, as evidenced by their greater accumulation of lipofuscin. Tortuous, varicose vessels, dilated capillaries, acellular capillaries, microaneurysms, intraretinal microvascular abnormalities (IRMA), pre-retinal membranes, and vitreal neovascularization are typical vessel lesions of diabetic retinopathy, all of which are limited to the vitreous and inner retina. The photoreceptor cells and retinal

258

W.G.ROBISON, JR.

Figure 5 Retinal Oxidative Damage Induced by α-tocopherol Deficiency—demonstrated by degeneration of the rod photoreceptor outer segment (ROS) membranes, by marked accumulation of lipofuscin pigment granules (1p) in the retinal pigment epithelium (RPE), purportedly representing highly peroxidized membrane remnants (A), and by similar lipofuscin granules, mainly in the endothelial cells (en), but also in some pericytes (p) of a representative retinal capillary with slightly thickened basement membranes (bm) and a patent lumen (L) of a rat deficient in α-tocopherol for 14 months (B). Note, in A, the juxtaposition of the Müller’s cell microvilli (mmv) to the microvilli of the RPE (arrowheads) owing to the loss of the ROS layer and many photoreceptor cells. The magnification for A is X12,500 and for B is X22,000. The calibration bars represent 1.0 (µm.

GALACTOSEMIC ANIMAL MODELS

259

pigment epithelium show relatively few changes even after all the vitreal and inner retinal changes have developed. In contrast, α-tocopherol deficiency causes extensive damage to the photoreceptor cells and retinal pigment epithelium, while the vitreous and inner retina remain unchanged except for a non-diabeticlike accumulation of lipofuscin in many endothelial cells and a few pericytes, accompanied by a moderate increase in RCBM thickness. The accumulation of lipofuscin in the capillary cells demonstrates that αtocopherol deficiency subjects these cells to substantial oxidative stress. The fact that similar accumulation does not occur in these cells in diabetic or galactose-fed animals fed adequate α-tocopherol suggests that neither hyperglycemia nor galactosemia alone subject the retinal capillaries to equivalent oxidative stress. In summary, the lesions of diabetic retinopathy are distinct from those induced by the oxidative stress of α-tocopherol deficiency, suggesting that oxidative stress is not an initiating factor in the development of diabetic retinopathy. Oxidative damage probably contributes to age-related changes in the retinal capillaries and may exacerbate diabetic retinopathy, but oxidation alone does not play a significant role in the retinal vascular changes that accompany diabetic retinopathy through the severe non-proliferative stage. However, a possible role in the proliferative stages should not be overlooked. NEOVASCULARIZATION/VASCULOGENESIS: POTENTIAL ROLE OF OXYGEN Oxygen has long been suspected of having a dual involvement in the development and survival of the retinal vasculature. The evidence suggests that, on the one hand, low oxygen tension induces new vessel formation except when otherwise inhibited and, on the other hand, high oxygen tension induces vessel degeneration except when inhibited. The mechanisms have not been fully established. However, it has been proposed that retinal ischemia induces new vessel formation through the modulation of various growth factors, such as vascular endothelial growth factor (Cines et al., 1998) and that high oxygen levels are involved in the degeneration of vessels through oxidative stress and leukocyte damage to endothelial cells (Harris et al, 1994; Hatchell and Sinclair, 1994; Hatchell et al., 1994). There appears to be a link between oxygen and the proliferative stages of diabetic retinopathy, in spite of the differences cited between oxidative stress and the origin of other microangiopathies of diabetic retinopathy. Hyperglycemic- and galactosemic-induced neovascularization is associated with areas of acellular capillaries showing nonperfusion. Although increased aldose reductase activity has been established as the mechanism of selective pericyte loss in diabetic retinopathy, most evidence points to an involvement of oxygen and activated leukocytes in the degeneration of endothelial cells. The respiratory burst of superoxide (O2−) by polymorphonuclear leukocytes in the diabetic cat model is 33% higher than in non-diabetic controls (Freedman and Hatchell, 1992), and leukocytes have a greater propensity for binding to endothelial cells under diabetic conditions (Kim et al., 1994). A large burst of oxygen-derived free radicals and a decrease in endogenous antioxidants occur (Willy et al., 1995) concomitantly with endothelial cell swelling, luminal membrane blebbing, and capillary dysfunction when there is reperfusion of capillaries following ischemia (Ward and Scoote, 1997; Bron et al., 1995). Antioxidant protection has been reported (Ward and Scoote, 1997) to reduce the adhesion of leukocytes to the endothelial cell lining. Since an important role has been ascribed to increased leukocyte-endothelial cell adhesion in the development of acellular capillaries and neovascularization in diabetic retinopathy (Hatchell and Sinclair, 1994; Harris et al., 1994), it is possible that α-tocopherol may ameliorate adhesion related vessel damage in diabetic retinopathy.

260

W.G.ROBISON, JR.

SIGNIFICANCE OF THE GALACTOSEMIC RAT MODEL The galactose-fed rat model provides the best evidence to date that aldose reductase activity is firmly linked to all the retinal lesions of diabetic retinopathy. The fact that galactose-induced retinopathy mimics so well the otherwise unique spectrum of microangiopathies that develop in human diabetic retinopathy (Table 1) indicates similar mechanisms must be involved in galactosemia and diabetic retinopathy. The cumulative findings indicate that the reduction of excess hexose by aldose reductase is not only the triggering event, but that it causes a cascade of cellular and tissue responses which ultimately results in all the microangiopathies of diabetic retinopathy through the mild proliferative stage. This model permits the dissociation of the effects of flux through aldose reductase from those of elevated blood glucose, elevated glycation levels, decreased insulin production and/or recognition, as well as other metabolic and genetic factors which might be involved in the development of diabetic complications. Only aldose reductase-related complications and associated aspects should be manifest in a galactose-fed rat. The model is based on the fact that aldose reductase has a higher affinity for galactose than for glucose. This results in increased flux through aldose reductase and causes accelerated development of aldose reductase related diabetic complications, a real advantage in studies of complications which take a long time to develop, such as retinopathy. Although the evidence strongly suggests that the activity of aldose reductase is the primary causative factor, galactose feeding not only increases aldose reductase activity, but also increases glycation levels. Prevention of the retinopathy with aldose reductase inhibitors eliminates the possibility that glycation could be a major factor, since the inhibitors are specific and do not alter the plasma or tissue levels of hexose or glycation in the galactose-fed rat (Chiou et al., 1980; Robison et al., 1996). A complete prevention of cataracts and essential prevention of diabetic-like retinopathy has been documented in rats fed a 50% galactose diet plus one of four different aldose reductase inhibitors (AL-3152, sorbinil, tolrestat, or WAY-121, 509) for 22 to 28 months (Frank et al., 1983; 1997; Robison et al., 1989; 1990a; 1990b; 1995b; 1996; 1997; 1998). One of the concerns that has been expressed regarding the galactose-fed rat as a model is why retinopathy is uncommon in galactosemic patients. It should be noted that essentially all patients with congenital galactosemia have cataracts, as would be predicted from studies of galactose-fed rats. However, the clinical manifestation of retinopathy in these patients is uncommon, probably because galactosemia is strictly controlled and/or the patients usually die before notable retinal changes would be expected. Other concerns have been addressed previously (Robison et al., 1995a). The discovery that galactose feeding induces development of diabetic-like retinal microangiopathies through the mild proliferative stage has been confirmed independently by two other laboratories (Frank et al., 1983; 1997; Kern and Engerman, 1995). The finding that aldose reductase inhibitors can prevent the diabetic-like lesions has been confirmed independently in one other laboratory for galactose-fed rats (Frank et al., 1983; 1997) and in several laboratories for diabetic rats (Kojima et al., 1985a; 1985b; Chakrabarti and Sima, 1987; 1989; McCaleb et al., 199D-Novel aldose reductase inhibitors with increased potency and lower toxicity (Jacot and Sredy, 1999; Pfeifer et al., 1997;) as well as other therapeutic approaches (Inoguchi et al., 1992; Ishii et al. 1996; Jacot and Sredy, 1999) for the prevention and amelioration of diabetic ocular complications are being developed currently. As suggested by Porte and Schwartz (1996), the development of such new potential therapies should include the gathering of adequate data on prevention of diabetic complications in reliable animal models before initiating extensive clinical trials. The galactose-fed rat provides one of the most reliable and economical animal models for testing the efficacy of new approaches. Diabetic retinopathy is the major cause of blindness in adults (20 to 74 years old) in the industrialized countries. Whereas systemic diabetes mellitus results from lowered availability and/or cellular recognition

GALACTOSEMIC ANIMAL MODELS

261

of insulin, the complications of diabetes, such as diabetic retinopathy, are caused by the chronic hyperglycemia itself. Although conventional insulin therapy lowers blood glucose levels enough to preserve life, it does not permit complete euglycemia nor prevent the long-term complications of chronic supernormal levels of blood glucose. Recent extensive clinical trials demonstrate that even intensive insulin treatment only delays diabetic complications, and it causes a two- to threefold increase in severe hypoglycemia. This report presents experimental animal evidence for the efficacy of a novel approach to preventing diabetic complications, which would be used in addition to conventional insulin therapy, not to control blood glucose, but, instead, to decrease the toxic effects of hyperglycemia on cells. The findings identify a treatment which should be beneficial in humans if used to compliment conventional insulin therapy. Aldose reductase inhibitors would protect against the toxic effects of the relatively low yet supernormal levels of glucose that can not be eliminated with insulin therapy. However, treatment should begin early, long before clinical signs of the complications; drug pharmacokinetics should be considered thoroughly; and treatment should be continued on a long-term basis in order to prevent complications which take a long time to develop. ACKNOWLEDGMENTS The author wishes to thank Dr. Jin H.Kinoshita for his constant encouragement; Dr. Jorge L.Jacot, Dr. Martin L.Katz, Dr. Nora M.Laver, Dr. Ying Li, Dr. Noemi Lois, and Mr. Joel P.Glover for intellectual and research contributions to many of the studies cited; Ms. Jennifer C.Cook, Mr. Brian L.Feldman, Ms. Margaret A. Kelley, Ms. Carolyn Lemke, Ms. Erin Packer, and Ms. Arezoo Zomorrodi for their dedicated assistance as NIH summer interns; Ms. Evita G.Bynum for help with the manuscript and figures; and Mrs. Anne B.Groome and Mr. Joseph J.Hackett for the expert electron microscopy and preparation of the micrographs for publication. REFERENCES Aiello, L.M. (1994) Diagnosis, management and treatment of nonproliferative diabetic retinopathy and macular edema. In Principles and Practice of Ophthalmology, Clinical Practice, edited by D.M.Albert and F.A.Jakobiec, pp. 747–760. Philadelphia: W.B.Saunders Company. Aiello, L.M., Rand, L.I., Sebestyen, J.G., Weiss, J.N., Bradbury, M.J., Wafai, M.Z., et al. (1985) The eyes and diabetes. In joslin’s Diabetes Mellitus, 12th edn., edited by C.R. Kahn and G.C. Weir, pp. 600–634. Philadelphia: Lea & Febiger. Akagi, Y. and Kador, P.P. (1990) Effect of aldose reductase inhibitors on the progression of retinopathy in galactose-fed dogs. Exp. Eye Res., 50, 635–639. Akagi, Y., Kador, P.F, Kuwabara, T. and Kinoshita, J.H. (1983) Aldose reductase localization in human retinal mural cells. Invest. Ophthalmol. Vis. Sci., 24, 1516–1519. Akagi, Y., Tasaka, H., Terubayashi, H., Kador, P.F. and Kinoshita, J.H. (1988) Aldose reductase localization in rat sugar cataract. In Polyol Pathway and Its Role in Diabetic Complications, edited by N.Sakamoto, J.H.Kinoshita, P.F.Kador and N.Hotta, pp. 170–181. Netherlands: Elsevier Science Publishers BV (Biomedical Division). Altomare, E., Grattagliano, I., Vendemaile, G., Micelli-Ferrari, T., Signorile, A. and Cardia, L. (1997) Oxidative protein damage in human diabetic eye: Evidence of a retinal participation. European J. Clinical Invest., 27, 141–147. Ansari, N.H., Zhang, W., Fulep, E. and Mansour A. (1998) Prevention of pericyte loss by trolox in diabetic rat retina. J. Toxicol. & Environ. Health, 54, 467–475. Archer, D.B. (1976) Neovascularization of the retina. Trans. Ophthal. Soc. UK, 96, 471–493. Archer, D.B. (1977) Intraretinal, preretinal and subretinal new vessels. Trans. Ophthalmol. Soc. UK, 97, 449–456.

262

W.G.ROBISON, JR.

Archer, D.B., Amoaku, W.M.K. and Gardiner, T.A. (1991) Radiation retinopathy—Clinical, histopathological, ultrastructural and experimental correlations. Eye, 5, 239–251. Ashton, N. (1949) Vascular changes in diabetes with particular reference to the retinal vessels. Preliminary report. Brit. J. Ophthalmol., 33, 407–420. Ashton, N. (1950) Injection of the retinal vascular system in the enucleated eye in diabetic retinopathy . Brit. J. Ophthalmol., 34, 38–41. Ashton, N. (1959) Diabetic retinopathy. A new approach. The Lancet, 2, 625–630. Ashton, N. (1963) Studies of the retinal capillaries in relation to diabetic and other retinopathies. Brit. J. Ophthalmol., 47, 521–538. Ashton, N. (1971) Diabetic retinopathy. Some current concepts. In Proc. Int. Symp. Fluorescein Angiography, pp. 334–345. Basel: Karger. Ashton, N. (1974) Vascular basement membrane changes in diabetic retinopathy. Brit. J. Ophthalmol., 58, 344–366. Ashton, N. (1983) Pathogenesis of diabetic retinopathy. In Diabetic Retinopathy, edited by H.L. Little, R.L.Jack, A.Patz and P.H.Forsham, pp. 85–106. New York: Thieme-Stratton Inc. Ashton, N. and Harry, J. (1963) The pathology of cotton wool spots and cytoid bodies in hypertensive retinopathy and other diseases. Trans. Ophthalmol. Soc. UK, 83, 91–114. Ballantyne, A.J. (1945) Retinal changes associated with diabetes and with hypertension. Archiv. Ophthalmol., 33, 97–105. Baynes, J.W. (1996) The role of oxidation in the Maillard reaction in vivo. In The Maillard Reaction: Consequences for the Chemical and Life Sciences, edited by R. Ikan, pp. 55–72. New York: Wiley. Benson, W.E., Brown, G.C. and Tasman, W. (1988) Diabetes and Its Ocular Complications. Philadelphia: W.B.Saunders Company. Bergmeyer, H.U., Bernt, E., Schmidt, F. and Stork, H. (1974) Determination with hexokinase and glucose-6-phosphate dehydrogenase. In Methods of Enzymatic Analysis, 2nd edn., pp. 1196–1201. New York: Academic Press. Bloodworth, J.M.B. Jr. (1962) Diabetic retinopathy. Diabetes, 2, 1–22. Bloodworth, J.M.B. Jr. (1963) Diabetic microangiopathy. Diabetes, 12, 99–114. Bloodworth, J.M.B. Jr. (1967) Fine structure of retina in human and canine diabetes mellitus. In Vascular Complications of Diabetes Mellitus, with Special Emphasis on Microangiopathy of the Eye, edited by S.J.Kimura and W.M.Caygill, pp. 73–98. St. Louis: CV Mosby Company. Bresnick, G.H. (1994) Nonproliferative diabetic retinopathy. In Retina, Volume 2: Medical Retina, edited by S.J.Ryan, A.P.Schachat and R.B.Murphy, pp. 1277–1318. St. Louis: C.V.Mosby Company. Bresnick, G.H., Engerman, R., Davis, M.D., de Venecia, G. and Myers, F.L. (1976) Patterns of ischemia in diabetic retinopathy. Trans. Amer. Acad. Ophthalmol. Otolaryngol., 81, 694– 709. Bron, A.M., Maupoil, V., Garcher, C., Guyonnet, G., Chelqi, E.H. and Rochette, L. (1995) Modification of vitamin E during ischemia-reperfusion in rat retina. Invest. Ophthalmol. Vis. Set., 36, 1084–1087. Brown, G.C. (1994a) Ocular ischemic syndrome. In Retina, Volume 2. Medical Retina, edited by S.J.Ryan, A.P.Schachat and R.B.Murphy, pp. 1515–1527. St. Louis: The C.V. Mosby Company. Brown, G.C. (1994b) Retinal arterial obstructive disease. In Retina, Volume 2: Medical Retina, edited by S.J.Ryan, A.P.Schachat and R.B.Murphy, pp. 1361–1377. St. Louis: The C.V. Mosby Company. Brown, G.C. and Benson, W.E. (1984) The fundus manifestations of malignant hypertension. Trans. PA Acad. Ophthalmol. Otolaryngol., 37, 19–25. Brown, G.C., Brown, M.M., Hiller, T., Fischer, D., Benson, W.E. and Magargal, L.E. (1985) Cotton-wool spots. Retina, 5, 206–214. Bucala, R., Cerami, A. and Vlassara, H. (1995) Advanced glycosylation end products in diabetic complications. Biochemical basis and prospects for therapeutic intervention. Diabetes Reviews, 3, 258–268. Bunn, H.F. and Higgins, P.J. (1981) Reaction of monosaccharides with proteins: Possible evolutionary significance. Science, 213, 222–224. Chakrabarti, S. and Sima, A.A.F. (1987) Pathogenetic heterogeneity in retinal capillary basement membrane thickening in the diabetic BB-rat. Diabetologia, 30, 966–968.

GALACTOSEMIC ANIMAL MODELS

263

Chakrabarti, S. and Sima, A.A.F. (1989) Effect of aldose reductase inhibition and insulin treatment on retinal capillary basement membrane thickening in BB rats. Diabetes, 38, 1181–1186. Chaudhuri, P.R., Austin, D.J. and Rosenthal, A.R. (1981) Treatment of radiation retinopathy. Brit. J. Ophthalmol., 65, 623–625. Chiou, S.-H., Chylack, L.T. Jr., Bunn, H.F. and Kinoshita, J.H. (1980) Role of nonenzymatic glycosylation in experimental cataract formation. Biochem. Biophys. Res. Com., 95, 894– 901. Cines, D.B., Pollack, E.S., Buck, C.A., Loscalzo, J., Zimmerman, G.A., McEver, R.P., et al. (1998) Endothelial cells in physiology and in the pathophysiology of vascular disorders. Blood, 91, 3527–3561. Cogan, D.G. (1974) Diabetes. Ophthalmic manifestations of systemic vascular disease, Vol. III. In Major Problems in Internal Medicine, edited by L.H.Smith, Jr., pp. 81–180. Philadelphia: W.B.Saunders Co. Cogan, D.G. and Kuwabara, T. (1967a) The mural cell in perspective. Archiv. Ophthalmol., 78, 133–139. Cogan, D.G. and Kuwabara, T. (1967b) Ocular microangiopathy in diabetes. In Vascular Complications of Diabetes Mellitus, with Special Emphasis on Microangiopathy of the Eye, edited by S.J.Kimura and W.M.Caygill, pp. 53–72. St. Louis: C.V.Mosby Co. Cogan, D.G., Toussaint, D. and Kuwabara, T. (1961) Retinal vascular patterns. IV. Diabetic retinopathy. Archiv. Ophthalmol., 66, 366–378. Datiles, M., Fukui, H., Kuwabara, T. and Kinoshita, J.H. (1982) Galactose cataract prevention with sorbinil, an aldose reductase inhibitor: A light microscopic study. Invest. Ophthalmol. Vis. Sci., 22, 174–179. Davis, M.D. (1988) Diabetic retinopathy: A clinical overview. Diabetes/Metabolism Reviews, 4, 291–322. de la Paz, M.A. and Anderson, R.E. (1992) Lipid peroxidation in rod outer segments. Role of hydroxyl radical and lipid hydroperoxides. Invest. Ophthalmol. Vis. Sci., 33, 2091–2096. de Venecia, G., Davis, M. and Engerman, R. (1976) Clinocopathologic correlations in diabetic retinopathy. I. Histology and fluorescein angiography of microaneurysms. Archiv. Ophthalmol., 94, 1766–1773. Diabetes Control and Complications Trial Research Group (The) (1993) 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., 329, 977–986. Dilley, R.A. and McConnell, D.G. (1970) Alpha-tocopherol in the retinal outer segment of bovine eyes. J. Membr. Biol., 2, 317–323. Dominguez, C., Ruiz, E., Gussinye, M. and Carrascosa, A. (1998) Oxidative stress at onset and in early stages of type 1 diabetes in children and adolescents. Diabetes Care, 21, 1736– 1742. Duke, J.R., Wilkinson, C.P. and Sigelman, S. (1968) Retinal microaneurysms in leukaemia. Brit. J. Ophthalmol., 52, 368–374. Dvornik, D. (1987) Aldose Reductase Inhibition: An Approach to the Prevention of Diabetic Complications, Biomedical Information Corporation. New York: McGrawHill. Early Treatment Diabetic Retinopathy Study Research Group (1991a) Grading diabetic retinopathy from stereoscopic color fundus photographs—An extension of the modified Airlie House classification, ETDRS Report Number 10. Ophthalmology, 98, 786–806. Early Treatment Diabetic Retinopathy Study Research Group (1991b) Fundus photographic risk factors for progression of diabetic retinopathy, ETDRS Report Number 12. Ophthalmology, 98, 823–833. Engerman, R., Finkelstein, D., Aguirre, G., Diddie, K.R., Fox, R.R., Frank, R.N., et al (1982) Ocular complications. Diabetes, 31(Suppl. 1), 82–88. Engerman, R.L. and Kern, T.S. (1984) Experimental galactosemia produces diabetic-like retinopathy. Diabetes, 33, 97–100. Engerman, R.L. and Kern, T.S. (1987) Progression of incipient diabetic retinopathy during good glycemic control. Diabetes, 36, 808–812. Engerman, R.L. and Kern, T.S. (1989) Determinants of experimental diabetic retinopathy. In Diabetes 1988, edited by R.Larkins, P.Zimmer and D.Chisholm, pp. 221–224. Netherlands: Elsevier Science Publishers B.V. (Biomedical Division).

264

W.G.ROBISON, JR.

Engerman, R.L. and Kern, T.S. (1991) Aldose reductase and retinopathy—current controversies. In Diabetes 1991, Proceedings of the 14th International Diabetes Federation Congress, Washington, DC, 23–28 June 1991, edited by H.Rifkin, J.A.Colwell and S.I.Taylor, pp. 462–465. Amsterdam: Elsevier Science Publishers B.V. Engerman, R.L. and Kern, T.S. (1993) Aldose reductase inhibition fails to prevent retinopathy in diabetic and galactosemic dogs. Diabetes, 42, 820–825. Engerman, R.L. and Kern, T.S. (1995a) Retinopathy in animal models of diabetes. Diabetes/ Metabolism, 11, 109–120. Engerman, R.L. and Kern, T.S. (1995b) Retinopathy in galactosemic dogs continues to progress after cessation of galactosemia. Archiv. OphthalmoL, 113, 355–358. Engerman, R.L., Pfaffenbach, D. and Davis, M.D. (1967) Cell turnover of capillaries. Lab. Invest, 17, 738–743. Farnsworth, C.C. and Dratz, E.A. (1976) Oxidative damage of retinal rod outer segment membranes and the role of vitamin E. Biochim. Biophys. Acta, 433, 556–570. Fischer, F. and Gärtner, J. (1983) Morphometric analysis of basal laminae in rats with long-term streptozotocin diabetes 1. II. Retinal capillaries. Exp. Eye Res., 37, 55–64. Fliesler, S.J. and Anderson, R.E. (1983) Chemistry and metabolism of lipids in the vertebrate retina. Prog Lipid Res., 22, 79–131. Frank, R.N., Amin, R., Kennedy, A. and Hohman, T.C. (1997) An aldose reductase inhibitor and aminoguanidine prevent vascular endothelial growth factor expression in rats with long-term galactosemia. Archiv. OphthalmoL, 115, 1036–1047. Frank, R.N., Keirn, R.J., Kennedy, A. and Frank, K.W. (1983) Galactose-induced retinal capillary basement membrane thickening: Prevention by sorbinil. Invest. OphthalmoL Vis. Sci., 24, 1519–1524. Freedman, S.F. and Hatchell, D.L. (1992) Enhanced superoxide radical production by stimulated polymorphonuclear leukocytes in a cat model of diabetes. Exp. Eye Res., 55, 767– 773. Garner, A. (1970) Pathology of diabetic retinopathy. Brit. Med. Bull., 26, 137–142. Garner, A. (1994) Vascular diseases. In Pathobiology of Ocular Disease. A Dynamic Approach, 2nd edn. Part B., edited by A.Garner and G.K.Klintworth, pp. 1625–1710. New York: Marcel Dekker Inc. Giugliano, D., Ceriello, A. and Paolisso, G. (1996) Oxidative stress and diabetic vascular complications. Diabetes Care, 19, 257–267. Goldberg, M.F. (1976) Retinal vaso-occlusion in sickling hemoglobinopathies. Birth Defects: Original Article Series, 12, 475–515. Green, W.R., Quigley, H.A., de la Cruz, Z. and Cohen, B. (1980) Parafoveal retinal telangiectasis. Light and electron microscopy studies. Trans. OphthalmoL Soc. UK, 100, 162–170. Guerrant, G.O. and Moss, C.W. (1984) Determination of monosaccharides as aldononitrile, 0-methyloxime, alditol, and cyclitol acetate derivatives by gas chromatography. Anal. Chem., 56, 633–638. Harris, A.G., Skalak, T.C. and Hatchell, D.L. (1994) Leukocyte-capillary plugging and network resistance are increased in skeletal muscle of rats with streptozotocin-induced hyperglycemia. Int. J. Microcirc., 14, 159–166. Hatchell, D.L. and Sinclair, S.H. (1994) Role of leukocytes in diabetic retinopathy. In Physiology and Pathophysiology of LeuLocyte Adhesion, edited by N.Granger and G.W.Schmid-Schoenbein, pp. 289–293. London: Oxford Univ. Press. Hatchell, D.L., Wilson, C.A. and Saloupis, P. (1994) Neutrophils plug capillaries in acute experimental retinal ischemia. Microvascular Research, 47, 344–354. Hayes, K.C. (1974) Retinal degeneration in monkeys induced by deficiencies of vitamin E or A. Invest. Ophthalmol., 13, 499–510. Hayreh, S.S. (1970) Post-radiation retinopathy. A fluroescence fundus angiographic study. Brit. J. Ophthalmol. , 54, 705–714. Hohman, T.C., Nishimura, C. and Robison, W.G. Jr. (1989) Aldose reductase and polyol in cultured pericytes of human retinal capillaries. Exp. Eye Res., 48, 55–60. Hotta, N., Kakuta, H., Fukasawa, H., Koh, N., Matsumae, H., Kimura, M., et al. (1988) Prevention of diabetic retinopathy by an aldose reductase inhibitor in fructose-fed streptozotocin-diabetic rats. In Polyol Pathway and Its

GALACTOSEMIC ANIMAL MODELS

265

Role in Diabetic Complications, edited by N.Sakamoto, J.H.Kinoshita, P.P.Kador and N.Hotta, pp. 311–318. Netherlands: Elsevier Science Publishers BV (Biomedical Division). Hu, T.-S., Datiles, M. and Kinoshita, J.H. (1983) Reversal of galactose cataract with sorbinil in rats. Invest. Ophthalmol. Vis. Sci., 24, 640–644. Inoguchi, T., Battan, R., Handler, E., Sportsman, J.R., Heath, W. and King, G.L. (1992) Preferential elevation of protein kinase C isoform βII and diacylglycerol levels in the aorta and heart of diabetic rats: Differential reversibility to glycemic control by islet cell transplantation. Proc. Natl. Acad. Sci., 89, 11059–11063. Ishii, H., Jirousek, M.R., Koya, D., Takagi, C., Xia, P., Clermont, A., et al. (1996) Amelioration of vascular dysfunctions in diabetic rats by an oral PKC β inhibitor. Science, 272, 728– 731. Jacot, J.L. and Sredy, J. (1999) Emerging therapeutics for diabetic retinopathy: potential therapies for the new millennium. Emerging Therapeutic Targets, 3, 1–29. Kador, P.P. (1990) Preventive medicine: The contribution of pharmaceuticals to better health. Prevention of ocular diabetic complications by aldose reductase inhibitors. Swiss Pharma., 12, 33–38. Kador, P.F., Akagi, Y., Takahashi, Y, Ikebe, H., Wyman, M. and Kinoshita, J.H. (1990) Prevention of retinal vessel changes associated with diabetic retinopathy in galactose-fed dogs by aldose reductase inhibitors. Archiv. Ophthalmol., 108, 1301–1309. Kador, P.P., Akagi, Y., Terubayashi, H., Wyman, M. and Kinoshita, J.H. (1988) Prevention of pericyte ghost formation in retinal capillaries of galactose-fed dogs by aldose reductase inhibitors. Archiv. Ophthalmol., 106, 1099–1102. Kador, P.F., Takahashi, Y., Sato, S. and Wyman, M. (1994) Amelioration of diabetes-like retinal changes in galactosefed dogs. Preventive Med., 23, 717–721. Kador, P.F., Takahashi, Y., Wyman, M. and Ferris, F. III (1995) Diabetes like proliferative retinal changes in galactosefed dogs. Archiv. Ophthalmol., 113, 352–354. Katz, M.L., Stone, W.L. and Dratz, E.A. (1978) Fluorescent pigment accumulation in retinal pigment epithelium of antioxidant-deficient rats. Invest. Ophthalmol. Vis. Sci., 17, 1049– 1058. Keith, N., Wagener, H.P. and Barker, N.W. (1939) Some different types of essential hypertension: Their course and prognosis. Am. J. Med. Sci., 197, 332–343. Kern, T.S. and Engerman, R.L. (1991) Retinal polyol and myo-inositol in galactosemic dogs given an aldose-reductase inhibitor. Invest. Ophthalmol. Vis. Sci., 32, 3175–3177. Kern, T.S. and Engerman, R.L. (1995) Galactose-induced retinal microangiopathy in rats. Invest. Ophthalmol. Vis. Sci., 36, 490–496. Kern, T.S. and Engerman, R.L. (1996) A mouse model of diabetic retinopathy. Archiv. Ophthalmol., 114, 986–990. Kim, J.A., Berliner, J.A., Natarajan, R.D. and Nadler, J.L. (1994) Evidence that glucose increases monocyte binding to human aortic endothelial cells. Diabetes, 43, 1103–1107. Kinoshita, J.H. and Nishimura, C. (1988) The involvement of aldose reductase in diabetic complications. Diabetes/ Metabolism Reviews, 4, 323–337. Kohner, E.M. and Dollery, C.T. (1969) The natural history of diabetic retinopathy. In Symposium on the Treatment of Diabetic Retinopathy, U.S. Public Health Service Publication No. 1890, edited by M.F.Goldberg and S.L. Fine, pp. 65–79. Washington, D.C., U.S.: Government Printing Office. Kohner, E.M., Dollery, C.T. and Bulpitt, C.J. (1969) Cotton-wool spots in diabetic retinopathy. Diabetes, 18, 691–704. Kohner, E.M. and Henkind, P. (1970) Correlation of fluorescein angiogram and retinal digest in diabetic retinopathy. Amer. J Ophthalmol., 69, 403–414. Kojima, K., Matsubara, H., Harada, T., Mizuno, K., Suzuki, M., Hotta, N., Kakuta, W., et al. (198 5 a) Effects of aldose reductase inhibitor on retinal microangiopathy in streptozotocin-diabetic rats. Jpn. J. Ophthalmol., 29, 99–109. Kojima, K., Matsubara, H., Mizuno, K., Harada, T., Suzuki, M., Hotta, N., et al. (1985b) Experimentally induced diabetic retinal microangiopathy and polyol pathway. II. Effect of aldose reductase inhibitors on microangiopathy. Nippon Ganka Gakkai Zasshi, 89, 247– 256. Kowluru, R., Kern, T.S. and Engerman, R.L. (1994) Abnormalities of retinal metabolism in diabetes or galactosemia II. Comparison of γ-glutamyl transpeptidase in retina and cerebral cortex, and effects of antioxidant therapy. Current Eye Res., 13, 891–896.

266

W.G.ROBISON, JR.

Kowluru, R.A., Kern, T.S. and Engerman, R.L. (1997) Abnormalities of retinal metabolism in diabetes or experimental galactosemia. IV. Antioxidant defense system. Free Radical Biology & Medicine, 22, 587–592. Kuwabara, T., Carroll, J.M. and Cogan, D.G. (1961) Retinal vascular patterns. III. Age, hypertension, absolute glaucoma, injury. Archiv. Ophthalmol., 65, 708–716. Laver, N.M., Robison, W.G. Jr. and Pfeffer, B.A. (1993) Novel procedures for isolating intact retinal vascular beds from diabetic humans and animal models. Invest. Ophthalmol. Vis. Sci., 34, 2097–2104. Lecomte, M., Paget, C., Ruggiero, D., Wiernsperger, N. and Lagarde, M. (1996) Docosahexaenoic acid is a major n-3 polyunsaturated fatty acid in bovine retinal microvessels. J. Neurochem., 66, 2160–2167. Lee, A.Y.W., Chung, S.K. and Chung, S.S.M. (1995) Demonstration that polyol accumulation is responsible for diabetic cataract by the use of transgenic mice expressing the aldose reductase gene in the lens. Proc. Natl. Acad. Sci., 92, 2780–2784. Maguire, A.M. and Schachat, A.P. (1994) Radiation retinopathy. In Retina, Volume 2: Medical Retina, edited by S.J.Ryan, A.P.Schachat and R.B.Murphy, pp. 1509–1514. St. Louis: The C.V.Mosby Company. Mansour, A.M., Rodenko, G. and Dutt, R. (1990) Half-life of cotton-wool spots in the acquired immunodeficiency syndrome. Int. J. STD & AIDS, 1, 132–133. McCaleb, M.L., McKean, M.L., Hohman, T.C., Laver, N. and Robison, W.G. Jr. (1991) Intervention with the aldose reductase inhibitor, tolrestat, in renal and retinal lesions of streptozotocin-diabetic rats. Diabetologia, 34, 695–701. McEwen, T.A.J., Chakrabarti, S. and Sima, A.A.F. (1987) A rapid reproducible method for determination of basement membrane thickness in biological structures. Comput. Biol. Med., 17, 193–197. Michaelson, I.C. (1980) Textbook of the Fundus of the Eye, 3rd edn. London: Churchill Livingstone. Nagaraj, R.H., Prabhakaram, M., Ortwerth, B.J. and Monnier, V.M. (1994) Suppression of pentosidine formation in galactosemic rat lens by an inhibitor of aldose reductase. Diabetes, 43, 580–586. Nagata, M., Katz, M.L. and Robison, W.G. Jr. (1986) Age-related thickening of retinal capillary basement membranes. Invest. Ophthalmol. Vis. Sci., 27, 437–440. Nishimura, C., Graham, C., Hohman, T.C., Nagata, M., Robison, W.G. Jr. and Carper, D. (1988) Characterization of mRNA and genes for aldose reductase in rat. Biochem. Biophys. Res. Com., 153, 1051–1059. Palmer, E.A., Patz, A., Phelps, D.L. and Spencer, R. (1994) Retinopathy of prematurity. In Retina, Volume Two. Medical Retina, edited by S.J.Ryan, A.P.Schachat and R.B.Murphy, pp. 1473–1498. St. Louis: The C.V.Mosby Company. Patz, A. (1984) Retinal neovascularisation: Early contributions of Professor Michaelson and recent observations. Brit. J. Ophthalmol., 68, 42–46. Petersen, M.R., Vine, A.K. and The University of Michigan Pancreas Transplant Evaluation Committee (1990) Progression of diabetic retinopathy after pancreas transplantation. Ophthalmology, 97, 496–502. Pfeifer, M.A., Schumer, M.P. and Gelber, D.A. (1997) Aldose reductase inhibitors: The end of an era or the need for different trial designs? Diabetes, 46, S82-S89. Porte, D. Jr. and Schwartz, M.W. (1996) Diabetes complications: Why is glucose potentially toxic? Science, 272, 699–700. Reichard, P., Nilsson, B.-Y. and Rosenqvist U. (1993) The effect of long-term intensified insulin treatment on the development of microvascular complications of diabetes mellitus. N. Engl. J. Med., 329, 304–309. Robison, W.G. Jr. (1994) Aldose reductase inhibition and retinopathy. Diabetes, 43, 337–338. Robison, W.G. Jr. (1995) Diabetic retinopathy: Galactose-fed rat model. Invest. Ophthalmol. Vis. Sci., 36, 1742–1743. Robison, W.G. Jr., Jacot, J.L., Glover, J.P., Basso, M.D. and Hohman, T.C. (1998) Diabetic-like retinopathy: Early and late intervention therapies in galactose-fed rats. Invest. Ophthalmol. Vis. Sci., 39, 1933–1941. Robison, W.G. Jr., Jacot, J.L., Katz, M.L. and Glover, J.P. (2000) Retinal vascular changes induced by the oxidative stress of α-tocopherol deficiency contrasted with diabetic microangiopathy. J. Ocular Pharmacognd Therapeutics, 16, 109–120. Robison, W.G. Jr., Kador, P.P., Akagi, Y., Kinoshita, J.H., Gonzalez, R. and Dvornik, D. (1986) Prevention of basement membrane thickening in retinal capillaries by a novel inhibitor of aldose reductase, tolrestat. Diabetes, 35, 295–299.

GALACTOSEMIC ANIMAL MODELS

267

Robison, W.G. Jr., Kador, P.F. and Kinoshita, J.H. (1983) Retinal capillaries: Basement membrane thickening by galactosemia prevented with aldose reductase inhibitor. Science, 221, 1177–1179. Robison, W.G. Jr., Kuwabara, T. and Bieri, J.G. (1979) Vitamin E deficiency and the retina: photoreceptor and pigment epithelial changes. Invest. Ophthalmol. Vis. Sci., 18, 683–690. Robison, W.G. Jr., Kuwabara, T. and Bieri, J.G. (1982) The roles of vitamin E and unsaturated fatty acids in the visual process. Retina, 2, 263–281. Robison, W.G. Jr. and Laver, N.M. (1993) Ocular lesions in animal models of human diabetes. In Lessons from Animal Diabetes IV, Biennial Review Volume: 1992, edited by E.Shafrir, pp. 145–163. London: Smith-Gordon and Company Limited. Robison, W.G. Jr., Laver, N.M., Jacot, J.L., Chandler, M.L., York, B.M. and Glover, J.P (1997) Efficacy of treatment after measurable diabetic like retinopathy in galactose-fed rats. Invest. Ophthalmol. Vis. Sci., 38, 1066–1073. Robison, W.G. Jr., Laver, N.M., Jacot, J.L. and Glover, J.P. (1995b) Sorbinil prevention of diabetic-like retinopathy in the galactose-fed rat model. Invest. Ophthalmol. Vis. Sci., 36, 2368–2380. Robison, W.G. Jr., Laver, N.M., Jacot, J.L., Glover, J.P., Basso, M.D., Blouin, P., et al (1996) Diabetic-like retinopathy ameliorated with the aldose reductase inhibitor WAY-121,509. Invest. Ophthalmol. Vis. Sci., 37, 1149–1156. Robison, W.G. Jr., Laver, N.M. and Lou, M.F. (1995a) The role of aldose reductase in diabetic retinopathy: Prevention and intervention studies. In Progress in Retinal and Eye Research, edited by N.N.Osborne and G.J.Chader, vol. 14, pp. 593–640. Oxford: Pergamon. Robison, W.G. Jr., McCaleb, M.L., Feld, L.G., Michaelis, O.E. IV, Laver, N. and Mercandetti, M. (1991) Degenerated intramural pericytes (‘ghost cells’) in the retinal capillaries of diabetic rats. Current Eye Res., 10, 339–350. Robison, W.G. Jr., Nagata, M. and Kinoshita, J.H. (1988) Aldose reductase and retinal capillary basement membrane thickening. Exp. Eye Res., 46, 343–348. Robison, W.G. Jr., Nagata, M., Laver, N., Hohman, T.C. and Kinoshita, J.H. (1989) Diabetic-like retinopathy in rats prevented with an aldose reductase inhibitor. Invest. Ophthalmol. Vis. Sci., 30, 2285–2292. Robison, W.G. Jr., Tillis, T.N., Laver, N. and Kinoshita, J.H. (1990a) Diabetes-related histopathologies of the rat retina prevented with an aldose reductase inhibitor. Exp. Eye Res., 50, 355–366. Robison, W.G. Jr., Holder, N. and Kinoshita, J.H. (1990b) The role of lens epithelium in sugar cataract formation. Exp. Eye Res., 50, 641–646. Santiago, J.V. (1992) Intensive management of insulin-dependent diabetes: Risks, benefits, and unanswered questions. J. Clin. Endocrin. Metab., 75, 977–982. Sato, S. and Kador, P.P. (1990) NADPH-dependent reductases of the dog lens. Exp. Eye Res., 50, 629–634. Sato, S., Takahashi, Y., Wyman, M. and Kador, P.F. (1991) Progression of sugar cataract in the dog. Invest. Ophthalmol. Vis. Sci., 32, 1925–1931. Shimizu, K., Kobayashi, Y. and Muraoka, K. (1981) Midperipheral fundus involvement in diabetic retinopathy. Ophthalmology, 88, 601–612. Sima, A.A.F., Chakrabarti, S., Garcia-Salinas, R. and Basu, P.K. (1985) The BB-rat—an authentic model of human diabetic retinopathy. Current Eye Res., 4, 1087–1092. Sima, A.A.F., Garcia-Salinas, R. and Basu, P.K. (1983) The BB Wistar rat: An experimental model for the study of diabetic retinopathy. Metabolism, 32, 136–140. Sima, A.A.F., Nathaniel, V., Bril, V., Lattimer, S.A. and Greene, D.A. (1988) The effect of sorbinil treatment on diabetic neuropathy. In Polyol Pathway and Its Role in Diabetic Complications , edited by N.Sakamoto, J.H.Kinoshita, P.P.Kador and N.Hotta, pp. 416– 421. Netherlands: Elsevier Science Publishers BV (Biomedical Division). Simard-Duquesne, N. and Dvornik, D. (1973) Galactitol accumulation and irreversible lens opacities in galactosemic rats. Invest. Ophthalmol. Vis. Sci., 12, 82–83. Sorbinil Retinopathy Trial Research Group (1990) A randomized trial of sorbinil, an aldose reductase inhibitor, in diabetic retinopathy. Archiv. Ophthalmol., 108, 1234–1244. Speakman, J.S., Mortimer, C.B., Briant, T.D.R., Ezrin, C., Lougheed, W.M. and Clarke, W.T.W. (1966) Pituitary ablation for diabetic retinopathy. Canad. Med. Ass. J., 94, 627– 635.

268

W.G.ROBISON, JR.

Speiser, P., Gittelsohn, A.M. and Patz, A. (1968) Studies on diabetic retinopathy, III. Influence of diabetes on intramural pericytes. Archiv. Ophthalmol., 80, 332–337. Stephens, R.J., Negi, D.S., Short, S.M., van Kuijk, F.J., Dratz, E.A. and Thomas, D.W. (1988) Vitamin E distribution in ocular tissues following long-term dietary depletion and supplementation as determined by micro-dissection and gas chromatography-mass spectrometry. Exp. Eye Res., 47, 237–245. Sugi, K. (1966) Studies on the pathological changes in the retinal vessels of human eyes, using the trypsin digestion method. Japanese J. Ophthalmol., 10, 252–266. Takahashi, Y., Wyman, M., Ferris, F. III and Kador, P.P. (1992) Diabetic-like preproliferative retinal changes in galactose-fed dogs. Archiv. Ophthalmol., 110, 1295–1302. Takahashi, Y., Augustin, W., Wyman, M. and Kador, P.F. (1993) Quantitative analysis of retinal vessel changes in galactose-fed dogs. J. Ocular Pharmacol., 9, 257–269. Tomlinson, D.R., Willars, G.B. and Carrington, A.L. (1992) Aldose reductase inhibitors and diabetic complications. Pharmac. Ther., 54, 151–194. Toussaint, D. (1968) Contribution à l’étude anatomique et clinique de la retinopathie diabetique chez I’homme et chez l’animal. Pathologia Europaea, 3, 11–167. Traverse, N., Menini, S., Cosso, L., Odetti, P., Albano, E., Pronzato, M.A., et al. (1998) Immunological evidence for increased oxidative stress in diabetic rats. Diabetologia, 41, 265–270. Urbanowski, J.C., Cohenford, M.A. and Dain, J.A. (1982) Nonenzymatic galactosylation of human serum albumin. In vivo preparation. J. Biological Chem., 257, 111–115. Varma, S.D. and Kinoshita, J.H. (1974) The absence of cataracts in mice with congenital hyperglycemia. Exp. Eye Res., 19, 577–582. Varma, S.D., Mizuno, A. and Kinoshita, J.H. (1977) Diabetic cataracts and flavonoids. Science, 195, 205–206. von Sallmann, L. and Grimes, P. (1972) Histopathologic changes in the eyes and kidneys of rats with long term streptozotocin diabetes. In Contemporary Ophthalmology Honoring Sir Stewart Duke-Elder, edited by J.G. Bellows, pp. 502–511. Baltimore: Williams and Wilkins Company. Ward, B.J. and Scoote, M. (1997) Antioxidants attenuate postischemic endothelial cell swelling and luminal membrane blebbing in cardiac capillaries. Microvascular Res., 53, 179– 186. Weinberg, D.V. and Seddon, J.M. (1994) Venous occlusive diseases of the retina. In Principles and Practice of Ophthalmology, Clinical Practice, edited by D.M.Albert and F.A.Jakobiec, Vol. 2, pp. 735–746. Philadelphia: W.B.Saunders Company. Williamson, J.R. and Kilo, C. (1977) Current status of capillary basement-membrane disease in diabetes mellitus. Diabetes, 26, 65–73. Willy, C., Thiery, J., Menger, M., Messmer, K., Arfors, K.E. and Lehr, H.-A. (1995) Impact of vitamin E supplement in standard laboratory animal diet on microvascular manifestation of ischemia/reperfiision injury. Free Radical Biology & Medicine, 19, 919–926. Wise, G.N. (1957) Retinal neovascularization. In Transactions of the American Ophthalmological Society, pp. 729–826. New York: Columbia University Press. Yancey, C.M. and Linsenmeier, R.A. (1989) Oxygen distribution and consumption in the cat retina at increased intraocular pressure. Invest. Ophthalmol. Vis. Sci., 30, 600–611. Yanko, L., Rosenmann, E. and Cohen, A.M. (1975) A correlative study on the occurrence of retinopathy and nephropathy in sucrose-fed diabetic rats. Acta Diabet., 12, 52–65. Yanoff, M. (1966) Diabetic retinopathy. N. Engl. J. Med., 274, 1344–1349. Zinman, B. (1989) The physiologic replacement of insulin. N. Engl. J. Med., 321, 363–370.

15. THE RHESUS MONKEY (MACACA MULATTA): A UNIQUE AND VALUABLE MODEL FOR THE STUDY OF SPONTANEOUS DIABETES MELLITUS AND ASSOCIATED CONDITIONS NONI L.BODKIN Obesity and Diabetes Research Center, School of Medicine, Dept. of Physiology, 10 South Pine St. MSTF 6–00, University of Maryland, Baltimore, Baltimore, MD 21201 USA. Telephone: (410) 706–3904; Fax: (410) 706–7540; Email: [email protected]

ABSTRACT For almost 3 decades, the rhesus monkey (Macaca mulatta) has been a valuable model for the study of Type 2 diabetes mellitus, the insulin resistance syndrome, and associated metabolic abnormalities. Advantages of the rhesus monkey include the spontaneous development of obesity and Type 2 diabetes mellitus, the ability to carry out prospective, longitudinal studies within a compressed time frame (compared to humans), and the remarkable similarity of the pathophysiological defects to that of human Type 2 diabetes. Technological advances in the research field, including the development of the euglycemic hyperinsulinemic clamp and the ability to carry out multiple tissue biopsies under basal and insulin-stimulated conditions have led to a greater understanding of the defect(s) in the insulin action pathway. Attention to specialized staff training, specialized primate health monitoring and experimental methods, as well as an awareness of environmental risks, are essential to the successful care and maintenance of the diabetic primate research laboratory. The rhesus monkey will continue to provide valuable insight into the study of obesity, Type 2 diabetes mellitus, and associated disorders, and the application of such knowledge to the prevention and treatment of human diabetes mellitus. The Rhesus Monkey in Diabetes Research The nonhuman primate rhesus monkey (Macaca mulatto) has long been of interest as an animal model of metabolic disorders. Almost 3 decades ago, Hamilton and associates (Hamilton et al., 1978) presented

270

N.L.BODKIN

findings showing evidence of diabetes mellitus in the rhesus monkey. Other hallmark studies documenting diabetes mellitus in individual rhesus monkeys include Kirk and colleagues (Kirk et al., 1972), DiGiacomo and associates (DiGiacomo et al., 1971), and Uno and co-workers (Uno et al., 1985). A number of comprehensive reviews have been published regarding diabetes mellitus in nonhuman primates (Howard, 1983; Howard and Yasuda, 1990; Hansen, 1992; Hansen, 1996), and Hansen and colleagues (Hansen et al., 1995); a summary listing of the primate species that have been identified to develop Type 2 diabetes mellitus is shown in Table 1. However, with the exception of the rhesus monkey, and the previous work of Howard and colleagues in the Macaca nigra primate (Howard, 1988), the reports have been limited. The Obesity and Diabetes Research Center (ODRC) primate colony, of the University of Maryland, is unique in the large number of diabetic and prediabetic Table 1 Primate species with diabetes*. Cebus apella (capuchin) Cercopithecus aethiops (African green monkey, vervet, or grivet) Cercopithecus cephus (moustached green guenon) Cercopithecus diana (Diana monkeys) Cercopithecus mitis (Blue, Sykes, Silver, Golden, or Samango monkey) Cercopithecus mona (Mona monkey) Colobus polykomos (King Colobus monkey) Galago crassicaudatus (African bushbaby) Macaca cyclopis (Taiwan or Formosan rock macaque) Macaca fasicularis (cynomolgus or crab-eating macaque) Macaca mulatta (rhesus monkey) Macaca nemestrina (pig-tailed macaque) Macaca radiata (bonnet macaque) Mandrillus leucophaeus (mandrill baboon) Papio hamadryas (sacred baboon) Saguinus fuscicollis (tamarin) Saguinus oedipus (tamarin) Saimiri sciureus (squirrel monkeys) Pan troglodytes (chimpanzee) * reproduced with permission, Hansen, B.C., in press, 2000.

rhesus monkeys which are under longitudinal study. In a review published in 1988 (Bodkin et al., 1988), 7 monkeys having fasting plasma glucose>140 mg/dl were diagnosed as diabetic; according to the earlier criteria of the American Diabetes Association (National Diabetes Data Group, 1979). At the time of this earlier review we had identified and were maintaining an additional 9 diabetic rhesus monkeys; in studies since that time, we have identified 55 diabetic rhesus monkeys. Currently, the ODRC includes a colony of >65 rhesus monkeys, of which approximately 1/3 of the monkeys have spontaneous adult-onset obesity-associated Type 2 diabetes mellitus. In addition, >50 monkeys under longitudinal study are in the process of developing Type 2 diabetes (Hansen et al., 1986; Bodkin et al., 1989; Hansen et al., 1990). This identification has become possible through careful study of the natural history of the disease in rhesus monkeys.

RHESUS MONKEY (MAC AC A MULATTA)

271

The present facilities were designed and customized to provide for integrated care and study of this unique and important animal resource, and have laboratory support and personnel areas directly associated with the husbandry areas. This proximity promotes continual monitoring of the animals, and provides for optimal laboratory handling of specimens, and of the extensive daily record keeping. Overview of Discoveries Related to Obesity and Diabetes Acquisition of obese monkeys and early characterization of Type 2 diabetes The colony began in 1969, under the direction of Dr. Barbara Hansen, with a few young adult lean rhesus monkeys, and in its first phase, increased adiposity was experimentally induced by forced overfeeding. In 1980 with the help of Dr. Eliot Stellar, then Provost of the University of Pennsylvania, a large group of obese middle-aged monkeys was obtained from the primate colony originally studied by Dr. Charles Hamilton (Hamilton et al., 1978), who had passed away. These unrelated laboratory maintained monkeys became the initial group of longitudinally studied primates. This animal resource has grown and developed over the last 30 years, with gradual increase in the colony size to more than 65 monkeys. The characterization of the spontaneous development of obesity and physiological and morphological characteristics of adiposity in the rhesus monkey were important findings (Jen et al., 1985; Hansen et al., 1988), as well as the association of various metabolic disorders, including hyperinsulinemia, hyperleptinemia (Bodkin et al., 1996), insulin resistance, and dyslipidemia, including increased plasma VLDL triglyceride and decreased HDL cholesterol concentrations (Hannah et al., 1991; Bodkin et al., 1993). Early studies in the ODRC primates also provided evidence of increased B-cell tropin concentrations in obese monkeys (Morton et al., 1992) and decreased plasma insulin-like growth factor-I (IGF-I) concentrations in aging monkeys (Bodkin et al., 1991). Identification of natural history of Type 2 diabetes mellitus Between the ages of 10 and 25 years, many of those monkeys which have become obese have been shown to subsequently develop classical noninsulin-dependent Type 2 diabetes mellitus (Hansen et al., 1985; Hansen et al., 1986). The discovery of Type 2 diabetes in several rhesus monkeys in the ODRC colony led to the determination of the natural history of the disease in nonhuman primates (Hansen et al., 1986), which to date has been shown to be virtually identical to both the in vivo and in vitro pathophysiological changes found in human Type 2 diabetes. In many cases, the metabolic findings published in the rhesus preceded those same findings in both rodents and humans, establishing the prediabetic and diabetic rhesus monkey as an exciting and valuable model with which to provide insight into human diabetes. Initially, serial observations in young, lean monkeys were obtained to establish the reference values under well-controlled conditions, as well as measurements in older obese, but metabolically normal monkeys. All monkeys were systematically followed longitudinally (for up to 12 years) with measurements taken at least annually. Results of these studies showed that Type 2 diabetes was a progressive disorder with distinct serial changes in body weight, body fat, and plasma insulin (Hansen et al., 1986) and acute insulin response to glucose and glucose tolerance (Hansen et al., 1985). The metabolic changes were found to be transitional and progressive rather than clustered, and Hansen and co-workers proposed the term “phases” to identify the metabolic breakpoints in the diabetic progression. Briefly, Phase 1 comprised normal lean young adult monkeys; Phase 2: aging, nonobese normal monkeys and obese normal monkeys; Phases 3

272

N.L.BODKIN

through 7: monkeys in a period of progression with obesity and hyperinsulinemia, and Phases 8 and 9: overt clinical Type 2 diabetes (Hansen et al., 1986; Hansen et al., 1988). These initial primate studies were instrumental in establishing that the earliest detectable abnormality in the development of Type 2 diabetes in primates was not an absolute deficiency of insulin, but rather an

RHESUS MONKEY (MAC AC A MULATTA)

273

Figure 1 The figure shows a diagrammatic representation of the natural history and progressive development of Type 2 diabetes mellitus in the spontaneously obese rhesus monkey. The critical metabolic variables include changes in body weight, fasting plasma glucose, glucose tolerance, fasting plasma insulin, acute insulin response to intravenous glucose, and insulin sensitivity (M), the latter as determined by peripheral glucose uptake during an euglycemic hyperinsulinemic clamp technique. ●--------● represents the approximate effect of aging when those monkeys who do not become obese and/or diabetic are not included in the data. (Reprinted with permission of publisher.)

enhanced basal insulin secretion and insulin response to glucose, a finding that has now been well-supported in other diabetes models, including other nonhuman primate species, rodents, and humans. The late change in fasting plasma glucose, and hence its unsuitability for use as an early diagnostic criterion, was also an important finding. A scheme of the phase concept and critical metabolic variables is shown in Figure 1.

274

N.L.BODKIN

Insulin resistance and the use of the euglycemic clamp An important step in the study of the natural history of diabetes in the rhesus monkey followed from the development of the euglycemic-hyperinsulinemic clamp (DeFronzo et al., 1979) for the assessment of insulin sensitivity in vivo in humans. This method was then adapted for use in rhesus monkeys (Bodkin et al., 1989) and is still considered to be the gold standard in diabetes research, for the measurement of in vivo insulin-stimulated glucose uptake. In the rhesus, it is carried out under light ketamine hydrochloride sedation (10–15 mg/kg) or under fentanyl citrate anesthesia (1–2 µg/kg.h). Briefly, a priming dose of regular insulin is administered followed by continuous insulin infusion (400 mU×m2 body surface area×min −1 for 120 min) with a variable 20% glucose infusion in order to maintain plasma glucose at a target steadystate glucose level of 85 mg/dl. The insulin dose was selected to assure maximum effect, as previously determined in dose response curves (Bodkin et al., 1989). The glucose disposal rate (M rate) is calculated during the steady state plasma glucose level and is adjusted for fat-free mass, as determined by tritiated water (Bodkin et al, 1993). In the primates, the euglycemic-hyperinsulinemic clamp technique allowed the estimation of insulin sensitivity, which usually develops concurrently with defective B-cell function, and considered by many to be the primary defect in Type 2 diabetes. In the decade following 1985, the study of insulin sensitivity became the focus of many diabetes research groups, who actively sought to identify the major defect(s) in the insulin action pathway. Many groups have contributed to the advances made in this period and to the present-day understanding of the pathology of insulin resistance. Tissue biopsies, obtained under basal and insulin-stimulated conditions during the euglycemic hyperinsulinemic clamp, were key to these studies. The optimal size of the rhesus monkey allowing concurrent tissue biopsies of key organs involved in the development of Type 2 diabetes (muscle, liver, subcutaneous and omental adipose tissue) during the euglycemic hyperinsulinemic clamp has provided insight into the possible defect(s) in the insulin action pathway in both monkeys and humans. Findings in the Type 2 diabetic rhesus have included defective insulin action on skeletal muscle glycogen synthase (Ortmeyer et al, 1993) and cAMP-dependent protein kinase activity (Ortmeyer, 1997) and adipose tissue glycogen synthase (Ortmeyer et al., 1993), although there is normal insulin action on liver glycogen synthase (Ortmeyer et al., 1997; Ortmeyer et al., 1998). In addition, findings of a decrease in urinary excretion of D-chiroinositol (a component of a putative mediator or modulator of insulin action) from diabetic monkeys and humans (Kennington et al., 1990; Ortmeyer et al, 1993), enhanced insulin action on muscle glycogen synthase and phosphorylase activity when D-chiroinositol was administered to monkeys (Ortmeyer et al., 1995), and enhanced glucose and insulin lowering in insulin-resistant and diabetic monkeys with D-chiroinositol administration (Ortmeyer et al., 1993; Ortmeyer et al., 1995) have also been carried out. Studies of pharmacological agents Rhesus monkeys have served in this and other laboratories as excellent models for the study of mechanisms of action and efficacy of many pharmaceutical agents. Over the past ten years in the ODRC colony we have reported studies involving lipid lowering agents (Hannah et al., 1995), anti-hypertensives (Bodkin et al., 1995), insulin sensitizers (Ortmeyer et al., 1999), and glucose lowering agents (Bodkin et al., 1991). These studies in the primates offer many advantages in the complex steps of pharmaceutical compound development, including the identification of appropriate target groups (e.g., prediabetic versus diabetic subjects), appropriate dose response data, and bioavailability and pharmacokinetic data obtained under wellcontrolled conditions.

RHESUS MONKEY (MAC AC A MULATTA)

275

METHODS FOR OBTAINING OPTIMAL RESULTS Well-controlled Environmental Conditions As part of the ODRC standard operating procedures, temperature and humidity are consistently maintained in each of the animal areas at 72°F or 22°C and 30– 70% relative humidity as noted in the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Room temperature is verified at least once daily as part of routine husbandry procedures by the husbandry technicians. Because the ODRC staff is in each room feeding, watering, carrying out husbandry and/ or monitoring the diabetic monkeys a majority of the day, any deviations from normal can also be directly and readily noted. Ad libitum Food Intake The food intake and body weights of each animal are monitored continuously. Food intake measurement is carried out and recorded a minimum of twice daily on each monkey. Fresh food (Monkey Diet #5038, Purina-Mills, Inc., St. Louis, MO) is provided to the monkey and recorded each morning and each afternoon according to the monkey’s assigned allotment. A careful system of recording and replacing biscuits is used such that intake records are adjusted for any discarded biscuits. Thus, our feeding and recording system assures that the food intake records of each monkey are accurate and dependable. Daily food intake evaluation is particularly important to the management of the diabetic monkeys due to the need to have ample fresh food available throughout the day and night in order to minimize the possibility of a hypoglycemic reaction. Any alterations from normal feeding behavior are immediately investigated. A monkey with a decreased food intake from normal is provided with a fruit supplement and carefully evaluated to determine the cause of the change in appetite. All monkeys receive a daily multivitamin. Body Weight Monitoring Body weights of all monkeys are determined a minimum of once weekly and usually twice weekly. These weights are recorded on the individual food intake sheets so that fluctuations or variability can be noted immediately. We have found that a decreased food intake and decreased body weight, particularly in the older-aged and/ or diabetic monkeys, is almost always indicative of an impending disorder or illness, and thus, this regimen allows for early detection and follow-up of such problems. Activity While there is some variation in normal activity levels across monkeys, most adult middle-aged rhesus monkeys are relatively sedentary. Thus, if left to themselves, adult monkeys do not choose to exercise in any large measure. The genetic abnormalities which underlie the expression of obesity and diabetes in humans are likely to be present in a substantial number of rhesus monkeys, although the phenotypic identification of such genetic contributors is still in the future. The phenotypic expression may be enhanced by ad libitum fed conditions. The development of obesity and diabetes in free-ranging monkeys on the island of Cayo Santiago has been also described (Schwartz et al., 1993). The Cayo Santiago monkeys live in an environment protected from predators, and with ad libitum provisioning of food by humans. Their activity is unrestricted.

276

N.L.BODKIN

Health Monitoring The monkeys of the ODRC colony are generally in excellent health, especially given the advanced ages of many of the prediabetic and diabetic monkeys. A practical overview of basic protocols and clinical monitoring has been previously published. Presently, the oldest monkey in the ODRC colony is a female, aged over 37 years with Type 2 diabetes, receiving insulin therapy. As in humans, older monkeys may develop occasional medical problems. We have found such medical problems in the monkeys to include occasional rashes and mild skin breakdown in the very obese monkeys, particularly in the abdominal and inguinal areas, infrequent but occasional diabetic-related skin ulcers of the heel or knee and tartar-buildup or dental caries. In our experience, these medical conditions have always been quickly resolved by appropriate treatment, such as a short period of skin care with topical antibiotics, appropriate antibiotics and dressings in the care of the diabetics, and dental scaling and/or tooth extraction as required. Baseline chemistry and hematology data are available on each monkey in the colony, and these data are invaluable in evaluating changes in white cell and differential counts, BUN and creatinine and/ or liver function tests. For the monkey’s health and well being, specific important parameters such as food intake, body weight, blood pressure, and hematocrit are carefully checked and recorded prior to each experiment. Hematocrit is particularly important as it reflects the overall general health of the monkey and potential to tolerate blood sampling during an experimental procedure. The general health of the animals is monitored, including routine tuberculosis testing and physical examinations and all veterinary, medical, surgical and/or emergency care, including but not limited to radiological examinations and cardiopulmonary evaluations are performed as needed. Special Care Considerations Special care is required for many of the animals in this colony due to aging, obesity or diabetes-associated complications. The longitudinal and acute metabolic status of each monkey is well-known at any given point in time; therefore, all prediabetic (fasting plasma glucose from 90 to 125 mg/dl) monkeys are identified prospectively and carefully monitored in regard to stability of body weight, food and water intake, degree of glucosuria and degree of hyperglycemia. Further, diabetic monkeys (fasting plasma glucose > 126 mg/dl; as identified by the 1997 revised recommendations of the American Diabetes Association, (Diabetes, 1997)) are carefully maintained on appropriate insulin therapy so as to assure adequate and appropriate food intake, mild glycosuria (to minimize any possibility of a hypoglycemic reaction), and stable body weight. Blood and urine glucose levels are monitored and insulin therapy is carefully adjusted for each individual diabetic monkey. All diabetic monkeys (fasting plasma glucose >126 mg/dl) are carefully maintained so as to assure adequate and appropriate food intake and a stable body weight. The insulin doses are adjusted based on twice-daily urine glucose and ketone measurements and on blood glucose determinations. For over 11 years, we have had very good success with glucose and ketone monitoring, probably due to the combination of carefully monitored food intake and activity levels of the diabetic monkeys, the regularity and structure of the insulin treatment/monitoring regimen, and close vigilance. NPH Humulin insulin is administered in the morning, and Ultralente insulin in the afternoon. Humulin insulin is used since monkey insulin has been shown to be structurally identical to human insulin (Naithani et al., 1984) and therefore the development of insulin antibodies in the diabetic monkeys is minimized. Only the free insulin assay is used to determine insulin concentrations of plasma samples from the diabetic monkeys.

RHESUS MONKEY (MAC AC A MULATTA)

277

The daily insulin dose of the diabetic monkeys ranges from 20–250 units/day. The insulin (U-100) is given subcutaneously in a standard U-100 syringe with a 28-gauge micro-fine needle. All of the monkeys are accustomed to their daily dose and tolerate it well, and most will readily present their arm or leg when shown the syringe. A small food reward is then given. Primate Environmental Enrichment The objectives of the ODRC primate enrichment program are to utilize unique enrichment resources and techniques to provide a consistently nonthreatening and interactive environment for the primates. Toys and other objects are available for manipulation. As all people who work closely with nonhuman primates know, environmental enrichment is a very complex subject, and yet, an integral component of primate husbandry and care. The primate environment can be the richest place in the world and yet it can also be the most lethal. In fact, “free roaming” monkeys, such as those on the island of Cayo Santiago, are in a very rich environment, and yet rarely live past the age of 15 years, unless they are placed in an individually protected environment. Another environmentally enriched setting is the corral-maintained colony. However, it is clear from reports of other colonies, that corral maintenance produces significant problems of aggression, in fact, lethal problems for some primates. The dominance hierarchy means that some animals will be injured or be deprived, while others will exist at the opposite end of the spectrum. Unfortunately, corral maintenance also makes difficult the early identification of a sick monkey or of impending illness. In fact, most diabetic animals identified in the corral situation are identified only after major weight loss occurs, and at a point where their disease has become life threatening. This contrasts with the very early detection of and treatment of laboratorymaintained diabetic primates to maintain their health. Within the care program of the Obesity, Diabetes and Aging Animal Resource there are several key contributors to environmental enrichment. Social grouping and facilitation of visual and auditory interactions between monkeys and between monkeys and caretakers is carefully considered. This colony consists of normal, pre-diabetic, diabetic, obese and aged monkeys. Many of these animals have special requirements such as daily insulin treatment, specific feeding protocols and close monitoring of behavior and feeding. The research of the ODRC requires that each primate have individual access to ad libitum food intake which is documented daily; therefore, all macaques in this research program are required to be individually housed. Although these macaques are individually housed, each primate is able to see and hear other monkeys in the room. Two rooms are equipped with mirrors to facilitate the viewing of other monkeys in the room. Generally, cages are arranged so that the primates can readily see one another. Very importantly, human contact with the monkeys occurs throughout the day in all areas, and due to the close monitoring of the diabetic and older monkeys, is likely to be much more frequent than in standard primate facilities. Specialized Primate Methods Sedation and anesthesia The monkeys are lightly sedated (ketamine hydrochloride, 10–15 mg/kg) prior to handling, thereby eliminating any distress that might occur with direct handling. Ketamine has been shown to be safe and useful as a tranquilizing agent in large species and is the usual drug of choice for handling primates. It has a wide therapeutic index and is administered i.m. (Green, 1982). None of our research procedures involves

278

N.L.BODKIN

pain; however, for the safety of the staff and the monkeys, the monkeys are maintained under light ketamine sedation during the experiments. In addition, our own observations have confirmed the reports of Kemnitz and Kraemer (1982) and of Brady and Koritnik (1985) that ketamine has no or minimal effect on glucose and insulin levels. Blood sampling techniques Blood sampling is carried out by using either an Abbocath T-20 or T-22 gauge×1¼" radiopaque teflon catheter inserted into either the saphenous or cephalic vein. The catheter is connected by a stopcock to a syringe filled with 0.9% NaCl or 0.9% NaCl and heparin (1000 U/ml) at a dilution of 7.5 units heparin/ml saline. Basal fasting samples for substrates and hormones include four 3-ml samples which are drawn in 3-min intervals and inverted gently for thorough mixing. These 4 samples are centrifuged, plasma aliquoted, pooled, stored at — 20°C and assayed for insulin, C-peptide, glucagon, and other hormones and substrates. Glucose levels are assayed the day of experiment, using a Beckman glucose analyzer II (Beckman Instruments, Fullerton, CA). Extra plasma (1–2 ml) is placed in long-term storage tubes in 0.5 ml aliquots and stored at—80°C. Other experiments requiring plasma, such as the i.v. glucose tolerance test, oral glucose tolerance test and the euglycemic clamp, are sampled in a similar manner. In addition, under appropriate anesthesia and with sterile surgical technique, basal and insulin-stimulated skeletal muscle (vastus lateralis), subcutaneous and omental adipose tissue, and liver were obtained for the determination of activities of multiple enzymes (glycogen synthase and phosphorylase, protein kinase, and protein phosphatase), and substrate concentrations (glucose-6-phosphate and glycogen). Muscle was frozen in-situ in liquid nitrogen with freeze-clamps and stored in liquid nitrogen until processing. Adipose tissue and liver were frozen ex-situ with freeze-clamps and stored in liquid nitrogen until processing. The activities of the enzymes and concentrations of the substrates are then compared, basal vs. insulinstimulated values lending important insight into the pathology involved in the development of type 2 diabetes mellitus. SUPPLIERS AND SOURCES FOR THIS MODEL The Regional Primate Center of the University of Washington maintains a regular publication, The Primate Clearinghouse, which announces the availability of both nonhuman primates and of tissue or other nonhuman primate specimens. Other sources include commercial breeding companies, which also may be obtained from the Primate Clearinghouse data base. The cost of a young (5 to 8 years of age) male rhesus monkey generally ranges from $2,500 to $3,500. These costs can only be estimated initially, that is, for a young research-naïve primate, because obesity and diabetes develop over 2 to 10 years, and the costs are cumulative beyond the initial purchase price. During these developmental years, the longitudinal studies that span the monkey’s lifetime are extremely expensive, requiring a specialized primate facility with a trained and knowledgable husbandry and research staff.

RHESUS MONKEY (MAC AC A MULATTA)

279

MOST SUITABLE AREAS OF RESEARCH WHICH PARTICULARLY APPLY TO THIS MODEL Insulin resistance syndrome and Type 2 diabetes The spontaneously obese rhesus monkey is an excellent model for studying the metabolic components of the insulin resistance syndrome, including its prospective development, sequence and contribution to the eventual development of Type 2 diabetes. The ability to carry out well-controlled primate studies with subjects of known metabolic status is an advantage not present in human studies. A summary of the characteristics of insulin resistance, prediabetes, and Type 2 diabetes in the rhesus monkeys, compared with humans, is shown in Table 2. Table 2 (reprinted with permission of publisher). Insulin resistance (early prediabetes)

Monkeys

Humans

Central obesity Normoglycemia Fasting hyperinsulinemia (greater in monkeys than in humans) Increased insulin response to glucose Insulin molecule structure identical Slightly reduced glucose tolerance Reduced hepatic clearance of insulin Declining whole-body insulin-mediated glucose uptake Reduced insulin activation of muscle glycogen synthase Reduced basal and insulin-stim. total muscle glycogen synthase Reduced insulin activation of adipose tissue glycogen synthase Higher insulin-stim. muscle glucose-6-phosphate content Beginning hypertriglyceridemia (sometimes) Hypertension (sometimes) Impaired glucose tolerance (late prediabetes) Postprandial glucosuria Slight hyperglycemia Impaired glucose tolerance Decreasing insulin response to glucose Reduced whole-body insulin-stimulated glucose uptake Reduced insulin activation of muscle glycogen synthase Reduced insulin activation of adipose tissue glycogen synthase Reduced basal and insulin-stim. total muscle glycogen synthase Higher insulin-stim. muscle glucose-6-phosphate content Hypertriglyceridemia (sometimes) Hypertension (sometimes) Non-insulin-dependent diabetes mellitus “Middle-age” onset most common (in years)

Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes

Yes Yes Yes Yes Yes Yes Yes Yes Yes ? ? ? Yes Yes

Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes

Yes Yes Yes Yes Yes Yes ? ? ? Yes Yes

12–25

30–60

280

N.L.BODKIN

Insulin resistance (early prediabetes)

Monkeys

Humans

Polyphagia Polydipsia Glucosuria Hyperglycemia Hyperfructosemia Normo- or hyperinsulinemia (insulin levels higher in monkeys than in humans) Impaired glucose tolerance Decreased insulin response to glucose Low whole-body insulin-mediated glucose uptake Lower basal and insulin-stim. muscle total glycogen synthase Higher insulin-stim. muscle glucose-6-phosphate content Reduced insulin action on: muscle glycogen synthase adipose tissue glycogen synthase adipocyte glucose oxidation adipocyte lipid synthesis Increased hepatic glucose production Hypertriglyceridemia (common) ReducedHDL-cholesterol (common) Hypertension (common) Microalbuminuria (common) Amyloid deposition in pancreatic islets (significant) Complications (frequent) neuropathy nephropathy retinopathy

Yes Yes Yes Yes Yes Yes

Yes Yes Yes Yes Yes Yes

Yes Yes Yes Yes Yes

Yes Yes Yes ? ?

Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes

Yes ? Yes Yes Yes Yes Yes Yes Yes Yes

Yes Yes Yes

Yes Yes Yes

In addition, our previous studies have documented the compressed time-frame (3–6 years in monkeys, in contrast to 10–25 years in humans) during which the insulin resistance syndrome may develop, and/or the progression of a subject from normal metabolic status to prediabetes (obesity, hyperinsulinemia, and insulin resistance), and finally, to overt Type 2 diabetes. At the present time in the ODRC, ongoing studies of the mechanisms underlying these disorders are underway, with complex in vivo and in vitro methods not possible in human subjects, such as serial measurements of insulin resistance with the euglycemic hyperinsulinemic clamp, and concurrent tissue biopsies. Studies of islet transplant and function The insulin-requiring Type 2 diabetic rhesus monkey is an excellent model for studies of islet transplantation and islet function. Such studies would have exciting potential to provide a more normal insulin replacement therapy for either the insulin-requiring Type 1 or Type 2 diabetic subject. The size of

RHESUS MONKEY (MAC AC A MULATTA)

281

the rhesus monkey, the potential to successfully undergo intra-abdominal surgical procedures, the identical structure of rhesus monkey insulin and human insulin (Naithani et al., 1984), and the similarity of the pancreatic islet cell morphology both in normal monkeys and in the insulin-requiring Type 2 diabetic monkey (Clark et al., 1990) are distinct advantages over many other species. Complications associated with diabetes The naturally occurring diabetes in the rhesus monkey is associated with all of the complications of Type 2 diabetes known in man, including hypertension (Bodkin et al, 1995), neuropathy (Cornblath et al., 1989), nephropathy (Kopp et al, 1990), retinopathy and cataracts (Robison et al., 1991; Laver et al., 1993) and infiltration of the pancreatic islets by amyloid (Clark et al., 1990; Clark et al., 1991; deKoning et al., 1993; Clark et al, 1995; deKoning et al, 1995; deKoning et al, 1999). Thus, the rhesus monkey offers exciting opportunities to provide insight into these often-occurring conditions in the Type 2 diabetic human. Table 2 provides a comparison of the rhesus monkey and humans summarizing several characteristics associated with insulin resistance, prediabetes, and Type 2 diabetes. TIPS ON POSSIBLE PITFALLS Importance of Longitudinal Characterization A number of measurements are regularly and periodically carried out to assess longitudinally some of the pathophysiological processes that develop spontaneously and naturally in these monkeys as they undergo the aging process and as they progress from normal, lean, young animals to obese animals with or without various physiological disturbances. Usually these disturbances develop at varying rates in different monkeys and include hypertension, hyperlipidemia, hyperinsulinemia, hyperglycemia, impaired glucose tolerance, insulin resistance, peripheral neuropathy, impaired renal clearance, and cardiovascular disease—in short, the entire syndrome which becomes manifest in aged and diabetic humans. The longitudinal studies depend specifically on regular assessment of each animal and determination of the rate at which that animal is or is not progressing in the development of pathophysiological disturbances. This allows one to estimate where a given monkey is in regard to the progression to overt diabetes. Environmental Risks The procedures, plasma samples, and tissue samples in the rhesus monkey involve situations or materials which may be hazardous to personnel. Therefore, special precautions must be used. Research with nonhuman primates carries the risk of B-virus infection, which is known to be fatal to most humans who contract it. Most cases reported to date have involved the handling of primates without anesthesia and/or without proper protective apparel. Because of the prevalence of B-virus in macaques, all personnel must be well-trained regarding the risk of B-virus and how to avoid potential exposure. At the ODRC in the event an employee undergoes a potential exposure (bite, scratch, or needle-stick), a strict B-virus protocol is carried out under the direction of the University Employee Health Services and Veterinary Resources. Briefly, the protocol includes immediate and thorough cleansing of the wound, and reporting of the incident for prompt medical attention.

282

N.L.BODKIN

Baseline serology is drawn on the exposed person for examination as well as baseline serology and virology (buccal and conjunctival samples) on the monkey. The employee is instructed regarding evidence of infection and closely observed over the next 2 weeks (symptoms include blisters at the site of the wound, itching, pain, or numbness, any neurological illness, or other evidence of the B-virus syndrome). The employee has a follow-up serology drawn 14–28 days after the incident. It should be noted that only about 2–3% of macaques shed B-virus at any one time, and in spite of the large number of primates and primate handlers involved in research, less than 40 cases of B-virus have been reported since 1920. To reduce the potential for hazardous exposure, it is recommended that needles not be recapped. Staff training concerning protective laboratory apparel, bloodborne pathogens, and the handling of radioactive substances and wastes is required. In assays and protocols involving radioactive materials, appropriate shielding devices and/or dosimeter badges are used as needed and all Environmental Health and Safety guidelines are followed. ACKNOWLEDGMENTS The author gratefully acknowledges the scientific expertise of Dr. Barbara Hansen and Dr. Heidi Ortmeyer, in the preparation of the manuscript, and without whom these longitudinal studies of obesity and diabetes in the rhesus monkey would not have been possible. In addition, many faithful and skilled technical staff have contributed to the challenging husbandry, research, and managerial responsibilities associated with the primate colony of the Obesity and Diabetes Research Center, University of Maryland. These include Theresa Alexander, Joe Haney, Wallace Evans, Maryne Glowacki, Karen Brocklehurst, Dosu Doherty, Susan Fluck, Dennis Harman, Charla Sweeley, Terry Russell, and many others. Finally, we gratefully acknowledge the skilled veterinary expertise provided by the staff of the University of Maryland Veterinary Resources, including Dr. Kyle Stump, Dr. Joanne Smith, Dr. S.Srinivas, Dr. Mary Martin, and K.Wesley. REFERENCES Bodkin, N.L. (1996) The rhesus monkey: providing insight into obesity and diabetes. Lab. Animal, 25, 33–36. Bodkin, N.L., Hannah, J.S., Ortmeyer, H.K. and Hansen, B.C. (1993) Central obesity in rhesus monkeys: association with hyperinsulinemia, insulin resistance, and hypertriglyceridemia? Int. J. Obes., 17, 53–61. Bodkin, N.L. and Hansen, B.C. (1988) Discrete yet serial indications of the development of Type 2 diabetes in the rhesus monkey: Marked similarity to human NIDDM. In Monographs in Primatology: Nonhuman Primate Studies in Diabetes, Carbohydrate Intolerance, and Obesity, edited by C.F.Howard, pp. 7–27. New York: Alan R.Liss, Inc. Bodkin, N.L. and Hansen, B.C. (1995) Antihypertensive effects of captopril without adverse effects on glucose tolerance in hyperinsulinemic rhesus monkeys . J. Med. Primatol., 24, 1–6. Bodkin, N.L., Metzger, B.L. and Hansen, B.C. (1989) Hepatic glucose production and insulin sensitivity preceding diabetes in monkeys. Am. J. Physiol., 256, E676-E681. Bodkin, N.L., Nicolson, M., Ortmeyer, H.K. and Hansen, B.C. (1996) Hyperleptinemia: Relationship to adiposity and insulin resistance in the spontaneously obese rhesus monkey. Horm. Metab. Res., 28, 674–678. Bodkin, N.L., Sportsman, R., DiMarchi, R.D. and Hansen, B.C. (1991) Insulin-like growth factor-I in non-insulindependent diabetic monkeys: Basal plasma concentrations and metabolic effects of exogenously administered biosynthetic hormone. Metabolism, 40, 1131–1137.

RHESUS MONKEY (MAC AC A MULATTA)

283

Brady, A. and Koritnik, D. (1985) The effects of ketamine anesthesia on glucose clearance in African green monkeys. J. Med. Primatology, 14, 99–107. Clark, A., deKoning, E., Hansen, B.C., Bodkin, N. and Morris, J.F. (1990) Islet amyloid in glucose intolerant and spontaneous diabetic ‘Macaca mulatta’ monkeys. In Lessons from Animal Diabetes, edited by E.Shafrir pp. 502–506. London: Smith-Gordon. Clark, A., deKoning, E.J.P., Hattersley, A.T., Hansen, B.C., Yajnik, C.S. and Poulton, J. (1995) Pancreatic pathology in non-insulin dependent diabetes (NIDDM). Diabet. Res. Clin. Prac., 28, S39–S47. Clark, A., Morris, J.F., Scott, LA., McLay, A., Foulis, A.K., Bodkin, N.L., et al (1991) Intracellular formation of amyloid fibrils in β-cells of human insulinoma and pre-diabetic monkey islets. In Amyloid and Amyloidosis, edited by J.B.Natvis, pp. 453–456. Amsterdam: Kluwer. Cornblath, D.R., Hillman, M.A., Striffler, J.S., Herman, C.N. and Hansen, B.C. (1989) Peripheral neuropathy in diabetic monkeys. Diabetes, 38, 1365–1370. DeFronzo, R.A., Tobin, J.D. and Andres, R. (1979) Glucose clamp technique: A method for quantifying insulin secretion and resistance. Am. J. Physiol, 237, E214-E223. deKoning, E., Charge, S., Morris, J., Hansen, B., Bodkin, N. and Clark, A. (1995) Macrophages in pancreatic islet amyloidosis.In Amyloid and Amyloidosis, edited by J.B. Natvis, pp. 405– 407. Amsterdam: Kluwer. deKoning, E.J., Brand, J.J.V.d, Mott, V.L., Charge, S.B., Hansen, B.C., Bodkin, N.L., et al (in press 1999) Macrophages and pancreatic islet amyloidosis. Amyloid. deKoning, E.J.R, Bodkin, N.L., Hansen, B.C. and Clark, A. (1993) Diabetes mellitus in Macaca mulatta monkeys is characterized by severe islet amyloidosis and reduction in 6-cell population. Diabetologia, 36, 378–384. Di Giacomo, R.F., Myers, R.E. and Baez, L.R. (1971) Diabetes mellitus in a rhesus monkey (Macaca mulatta): a case report and literature review. Lab. Anim. Sci., 21, 572–574. Diabetes M.E.Cot.Da.Co. (1997) Report of the expert committee on the diagnosis and classification of diabetes mellitus. Diabetes Care, 20, 1183–1187. Green, C. (1982) Pharmacology of drugs acting on the central nervous system. In Animal Anaesthesia, edited by C. Green, pp. 29–43. London: Laboratory Animals Ltd. Hamilton, C.L. and Ciaccia, P. (1978) The course of development of glucose intolerance in the monkey (Macaca mulatta). J. Med. Primatol., 7, 165–173. Hannah, J.S., Bodkin, N.L., Paidi, M.S., Anh-Le, N., Howard, B.V. and Hansen, B.C. (1995) Effects of acipimox on the metabolism of free fatty acids and VLDL triglyceride. Acta Diabetologia, 32, 279–293. Hannah, J.S., Verdery, R.B., Bodkin, N.L., Hansen, B.C., Le, N.-A. and Howard, B.V. (1991) Changes in lipoprotein concentrations during the development of noninsulin dependent diabetes mellitus in obese rhesus monkeys (Macaca mulatta). J. Clin. Endocrinol. Metab., 72, 1067–1072. Hansen, B.C. (1992) Obesity and diabetes in monkeys. In Obesity, edited by P.Bjorntorp and B.N.Brodoff, pp. 256–265. New York: J.B. Lippincott, Co. Hansen, B.C. (1996) Primate animal models of non-insulin dependent diabetes mellitus. In Diabetes mellitus: A fundamental and clinical text, edited by D.LeRoith, S.I.Taylor and J.M. Olesfky, pp. 595–603. Philadelphia: Lippincott-Raven Publishers. Hansen, B.C. and Bodkin, N.L. (1985) Beta cell hyperresponsiveness to glucose precedes both fasting hyperinsulinemia and reduced glucose tolerance. (Abstract) Diabetes, 34(Suppl 1), 8A. Hansen, B.C. and Bodkin, N.L. (1986) Heterogeneity of insulin responses: phases in the continuum leading to noninsulin-dependent diabetes mellitus. Diabetologia, 29, 713– 719. Hansen, B.C. and Bodkin, N.L. (1990) β-cell hyperresponsiveness: earliest event in development of diabetes in monkeys. Am. J. Physiol., 259, R612-R617. Hansen, B.C., Bodkin, N.L., Schwartz, J. and Jen, K.-L.C. (1988) Beta cell responses, insulin resistance and the natural history of non insulin dependent diabetes in obese rhesus monkeys. In Frontiers in Diabetes Research. Lessons from Animal Diabetes, edited by E.Shafrir and A.E.Renold, pp. 279–287. London: John Libbey. Hansen, B.C., Jen, K.-L. and Schwartz, J. (1988) Changes in insulin responses and binding in adipocytes from monkeys with obesity progressing to diabetes. Int. J. Obes., 12, 391– 401 .

284

N.L.BODKIN

Hansen, B.C., Ortmeyer, H.K. and Bodkin, N.L. (1995) Obesity, insulin-resistance and noninsulin-dependent diabetes in aging monkeys: Implications for NIDDM in humans. in Lessons from Animal Diabetes, edited by E.Shafrir and A.E.Renold, pp. 93–105. London: Smith-Gordon. Howard, C. and Yasuda, M, (1990) Diabetes mellitus in nonhuman primates: Recent reserach advances and current husbandry practices. J. Med. Primatol., 19, 609–625. Howard, C.F. (1988) Spontaneous diabetes in Macaca nigra: Relevance to human beings with NIDDM, Adv. Expe. Med. Biol., 246, 33–39. Howard, J.C.F. (1983) Diabetes and carbohydrate impairment in nonhuman primates. In Nonhuman Primate Models for Human Diseases, edited by E.W.R.Dukelow, pp. 1–36. Florida: CRC Press. Jen, K.-L.C., Hansen, B.C. and Metzger, B.L. (1985) Adiposity, anthropometric measures, and plasma insulin levels of rhesus monkeys. Int. J. Obes., 9, 213–224. Johnson, D.O. (1985) History. In Nonhuman primates in biomedical research—biology and management, edited by B.Bennett, C.Abee and R.Henrickson, pp. 1–14. New York: Academic Press. Kemnitz, J. and Kraemer, G. (1982) Assessment of glucoregulation in rhesus monkeys sedated with ketamine. Am. J. Med. Primatology, 3, 201–210. Kennington, A.S., Hill, C.R., Craig, J., Bogardus, C., Raz, I., Ortmeyer, H.K., et al. (1990) Low urinary chiro-inositol excretion in non-insulin-dependent diabetes mellitus. N. Engl. J. Med., 323, 373–378. Kirk, J.H., Casey, H.W. and Harwell, J.F. (1972) Diabetes mellitus in two rhesus monkeys. Lab. Anim. Set., 22, 245–248. Kopp, J.B., Marinos, N.J., Bodkin, N.L., Hansen, B.C. and Klotman, P.E. (1990) Increased laminin in nodular glomerulosclerosis affecting rhesus monkeys with non-insulin dependent diabetes mellitus (NIDDM) (Abstract). J. Am. Soc. Nephrol., 1, 551 A. Laver, N.M., W.G.Robison, J. and Hansen, B.C. (1993) Demonstration of retinal histopathologic conditions in spontaneously diabetic monkeys. (Abstract) Am. J. Clin. Path., 99, 349. Morton, J.L., Davenport, M. and Beloff-Chain, A. (1992) Correlation between plasma B-cell tropin concentrations and body weight in obese rhesus monkeys. Am. J. Physiol., 262, E963-E967. Naithani, V.K., Steffens, G.J. and Tager, H.S. (1984) Isolation and amino-acid sequence determination of monkey insulin and proinsulin. Hoppe-Seyler’s Z Physiol. Chem., 365, 571– 575. National Diabetes Data Group (1979) National Diabetes Data Group: Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. Diabetes, 28, 139–157. Ortmeyer, H. (1997) Insulin decreases skeletal muscle cAMP-dependent protein kinase (PKA) activity in normal monkeys and increases PKA activity in insulin-resistant monkeys. J. Basic Clin. Physiol. Pharmacol., 8, 223–235. Ortmeyer, H. and Bodkin, N. (1998) Lack of defect in insulin action on hepatic glycogen synthase and phosphorylase in insulin-resistant monkeys. Am. J. Physiol., 274, G1005– G1010. Ortmeyer, H.K., Bodkin, N.L. and Hansen, B.C. (1993) Adipose tissue glycogen synthase activation by in vivo insulin in spontaneously insulin-resistant and Type 2 (non-insulin-dependent) diabetic rhesus monkeys. Diabetologia, 36, 200–206. Ortmeyer, H.K., Bodkin, N.L. and Hansen, B.C. (1993) Insulin-mediated glycogen synthase activity in muscle of spontaneously insulin-resistant and diabetic rhesus monkeys. Am. J. Physiol., 266, R552—R558. Ortmeyer, H.K., Bodkin, N.L. and Hansen, B.C. (1997) Insulin regulates liver glycogen synthase and glycogen phosphorylase activity reciprocally in rhesus monkeys. Am. J. Physiol., 272, E133-E138. Ortmeyer, H.K., Bodkin, N.L., Hansen, B.C. and Larner, J. (1995) In vivo D-chiroinositol activates skeletal muscle glycogen synthase and inactivates glycogen phosphorylase in rhesus monkeys. J. Nutr. Biochem., 6, 499–503. Ortmeyer, H.K., Bodkin, N.L., Lilley, K., Larner, J. and Hansen, B.C. (1993) Chiroinositol deficiency and insulin resistance. I. Urinary excretion rate of chiroinositol is directly associated with insulin resistance in spontaneously diabetic rhesus monkeys. Endocrinol., 132, 640–645. Ortmeyer, H.K., Bodkin, N.L., Yoshioka, S., Horikoshi, H. and Hansen, B.C. (1999) A thiazolidinedione improves insulin action on skeletal muscle glycogen synthase in insulin resistant rhesus monkeys. Diabetes, 48, A231.

RHESUS MONKEY (MAC AC A MULATTA)

285

Ortmeyer, H.K., Huang, L.C., Zhang, L., Hansen, B.C. and Larner, J. (1993) Chiroinositol deficiency and insulin resistance. II. Acute effects of D-chiroinositol administration in streptozotocin-diabetic rats, normal rats given a glucose load and spontaneously insulin-resistant rhesus monkeys. Endocrinology, 132, 646–651. Ortmeyer, H.K., Greene, H.L. and Larner, J. (1995) Effects of D-chiroinositol added to a meal on plasma glucose and insulin in hyperinsulinemic rhesus monkeys. Obes. Res., 3, 605S– 608S. Robison, W. and Laver, N. (1991) Diabetic retinal microangiopathies in humans and animals models: Similar histological progressions. Inv. Ophthal. Vis. Set., 32, 915. Schwartz, S.M., Kemnitz, J.W. and Howard, C.F. (1993) Obesity in free-ranging rhesus macaques. Int. J. Obesity, 17, 1–10. Uno, H., Kemnitz, J., Warner, J. and Brigham, T. (1985) Spontaneous diabetes mellitus in aged captive rhesus monkeys. Lab. Invest., 52, 572–574.

16. PSAMMOMYS OBESUS: PRIMARY INSULIN RESISTANCE LEADING TO NUTRITIONALLY INDUCED TYPE 2 DIABETES EHUD ZIV1 and RONY KALMAN1,2 1Diabetes

Research Unit, Hadassah-Hebrew University Medical Center, Jerusalem, Israel

2Animal

Facility, Hebrew University-Hadassah Medical Center, Jerusalem, Israel

ABSTRACT The gerbil Psammomys obesus (sand rat) is a model of nutritionally induced Type 2 diabetes. The progression of diabetes in Psammomys resembles in many respects the development of insulin resistance and diabetes in certain human populations. Psammomys is well adapted to its environment, and its high metabolic efficiency, a feature of the thrifty metabolism, represents a natural adaptation to life on low energy diet. Psammomys is prone to develop hyperinsulinemia, hyperglycemia and obesity when transferred to high energy (HE) diet and expresses four phenotypic states: State A—Normoinsulinemia and normoglycemia. State B— Hyperinsulinemia but normoglycemia. State C—Hyperinsulinemia and hyperglycemia. State D — Hypoinsulinemia and hyperglycemia as a result of loss of insulin secretion capacity. The animals in the Jerusalem colony were separated into two outbred distinct lines differing phenotypically and genotypically: diabetes prone (DP) and diabetes resistant (DR) animals. Psammomys from the DP line become diabetic in young age when transferred to HE diet: 81 to 90% of animals reach blood glucose levels >200 mg/dl within 7 to 14 days after weaning to HE diet. In animals older than 8 months, the potential to develop diabetes and obesity decreases. These changes are in correlation with the decrease in fertility in Psammomys, both in DP and DR lines. There is no hyperphagia in DP or DR lines when fed on HE, or LE diets. Metabolic efficiency in DP line Psammomys fed on all diets is 6.0–6.6 Kcal per g of weight increase whereas it is 9.0–9.6 Kcal per g of weight increase in the DR line. Primary insulin resistance is a species characteristic of Psammomys. On LE diet they do not become hypoglycemic even if administered high amounts of exogenous insulin. In hyperinsulinemic-euglycemic clamp studies, the hepatic glucose production was only partially reduced by insulin infusion (from 10.0

PSAMMOMYS OBESUS

287

Table 1 Composition, energy and digestibility of Psammomys diets.

±0.6 to 3.8±0.4 mg/ min.kg). Small elevation of total glucose transport in the clamped Psammomys attests to the fact that the peripheral glucose utilization is low enough to be compensated by gluconeogenesis, avoiding lapse into hypoglycemia during exogenous insulin treatment. It is emphasized that the combined responses of the reactions of the two different organs determines the diabetic potential of individual Psammomys. The adaptation of Psammomys to desert conditions and the primary insulin resistance characterized by the low expression of certain cellular proteins: low receptor tyrosine kinase and low PTPase activity, low amount of GLUT4 mRNA and protein, but overexpression of protein kinase C epsilon, believed to promote serine phosphorylation. Primary insulin resistance and lack of gluconeogenesis restraint in Psammomys, representative of both animal and human populations, in the fact of rich nutrient intake lead to overstimulation of insulin secretion followed by (β-cell exhaustion and apoptosis. Psammomys, Environment and Diets The gerbil Psammomys obesus often referred to as sand rat is a model of nutritionally induced Type 2 diabetes. The progression of diabetes in Psammomys resembles in many respects the development of insulin resistance and diabetes in certain human populations, such as Pima Indians, Australian Aborigines, Asian Indians, Mexican Americans and other populations in transition from scarce to affluent nutrition. In its native habitat in North Africa and Eastern Mediterranean Psammomys obesus is a herbivorous desert rodent belonging to the Gerbillinae family, subsisting on a low energy, electrolyte rich diet and weighing up to 180–200 g. Psammomys consumes mainly plants belonging to the Chenopodiacae family. In the Israeli desert region, Psammomys obesus feeds mainly on the succulent salt bush plant (Atripex halimus). These animals are well adapted to their environment with suitable shelter under the salt bush plants and drink very small, if any, quantities of water. The low, but constant, energy supply determines the type of metabolism of the Psammomys. As detailed latter, the high metabolic efficiency of the animals is part of the thrifty metabolism, and it represents a life pattern on low energy diet. The composition of Atriplex halimus and of laboratory diets on which the Psammomys may be maintained are given in Table 1. The digestible portion of this plant is 1.92 Kcal/g of dry leaves. The Psammomys consumes about 62.4 ± 2.0g of fresh leaves (10.3±0.4g dry leaves)/day. In various laboratories the animals were fed on other natural diets, such as beet (Duhault et al., 1994) and spinach (Habito et al.,

288

E.ZIV AND R.KALMAN

1995). In our laboratory the natural diet has been substituted by several dry artificial diets (Table 1). The main diet in use is a low energy (LE) pelleted diet suitable for Psammomys to maintain normoglycemia. The other two diets are high energy diets, “HE” extruded diet, and “KE” pelleted diet. These two diets differ in their physical properties rather than in their composition and as a result, in their diabetogenic potential. The HE diet is more accessible, causing a higher percentage of animals with hyperglycemia. Animals are maintained on these diets from weaning and adapt to them immediately. Usage of artificial diets enables a reproducible breeding and research environment. Maintenance of Psammomys The maintenance of several outbred lines of animals in the same environment requires particular adherence to the breeding procedures to avoid mixture. Detailed records should be kept of all breeding activity and each weaned and retained Psammomys is given an identification number that enables the familiarity with the pedigree of each animal. Since some lines are relatively small (60–70 breeding females) it is necessary to ensure minimal inbreeding while performing assortative matings to select the most suitable animals for breeding. The colony is maintained in a closed barrier and has a conventional health status. Sentinel animals are tested routinely and health reports are available. Psammomys are mated in monogamous pairs for a period of 9–12 months. Weanlings are separated at three weeks of age. Animals are fed a commercial diet according to the genetic line and research requirements. The cages are solid bottom shoebox polypropylene cages equipped with water bottles. The bedding consists of white pine woodchips. Selection of Defined Lines Psammomys obesus are prone to develop hyperglycemia, hyperinsulinemia and obesity when fed the HE or KE diets. As described in several previous reviews from our laboratory (Shafrir and Gutman, 1993; Ziv and Shafrir, 1995; Shafrir and Ziv, 1999; Shafrir et al., 1999) there is a strong genetic basis for the development of diabetes but the essential factor—prolonged consumption of high energy diet—is required for the disease to be expressed. The development of hyperglycemia is fast—7 to 14 days on HE diet. In the original colony the reaction of randomly chosen individual animals to the same HE diet may differ by the time hyperglycemia has developed. A few animals may remain normoglycemic even on HE diet (Kalderon et al., 1986; Kalman et al., 1993). We used the HE diet to examine the diabetic potential of each line, and to identify the diabetic and non-diabetic margins of the population. By using an assortative mating system based on a minimal inbreeding method (Baker, 1979) it is possible to separate the animals in the colony into two distinct lines differing phenotypically and genotypically: diabetes prone (DP) and diabetes resistant (DR) animals (Kalman et al, 1993). Animals to be mated were chosen according to phenotypic parameters (post-prandial blood glucose and plasma insulin levels). We isolated two lines that differ in their diabetic sensitivity, without diluting their total genetic pool. The pattern of diabetes of the Psammomys of the diabetes prone (DP) line and the diabetes resistant (DR) line is demonstrated in Figure 1. Previous studies (Kalderon et al, 1986) have indicated the existence of four different phenotypic states in the Psammomys: State A—Normoinsulinemic normoglycemic Psammomys. State B—Hyperinsulinemic normoglycemic Psammomys. State C—Hyperinsulinemic hyperglycemic Psammomys. State D— Hypoinsulinemic hyperglycemic Psammomys as a result of loss of β-cell function. During the development of diabetes each animal expresses, generally consecutively, all four states. Relation between plasma insulin and blood glucose levels in the four states is demonstrated in Figure 2 (Adler et al., 1988).

PSAMMOMYS OBESUS

289

Figure 1 Evolution of diabetes in Diabetic Prone (DP) and Diabetes Resistant (DR) Psammomys fed on HE diet from weaning. Animals in the DP line become diabetic in 7 to 14 days, and if left on HE diet die after 40 to 70 days. Animals from the DR line may become diabetic (although in a more attenuated manner than animals in the DP line), hyperinsulinemic but normoglycemic or not develop diabetes.

Figure 2 Plasma insulin levels correlated to blood glucose levels in Psammomys Values are given as mean±SE (from Adler et al., 1988)

A characteristic feature of Psammomys is their capacity to revert within 1–7 days from either state B or C to state A when transferred from HE to LE diet. Food withdrawn for a period of 6 h already normalizes their blood glucose and plasma insulin levels. Psammomys that reach state D on HE diet cannot revert to state A by diet restriction but may be returned to state C by treatment with exogenous insulin (Ziv et al., 1996). In Stage D islet overstimulation is evident leading to the loss of the insulin secretion due to apoptosis (Bar-On et al., 1999 and Donath et al., 1999). Psammomys from the DP line become diabetic in very young ages

290

E.ZIV AND R.KALMAN

Figure 3 Diabetic animals, relative size of epididymal fat weight and relative size of kidney weight in Psammomys from the DP line fed on KE diet at different ages. Values are given as mean±SE for 16–10 animals at each point.

when transferred to HE diet. Within 7 days from weaning 81% of the animals reach blood glucose levels >200 mg/dl. Percentage of diabetic animals on HE diet increases to >90% within 14 days from weaning. Age Related Diabetogenicity The potential of Psammomys to become diabetic decreases with age. In Psammomys at ages of 1–12 months maintained on the LE diet from weaning and transferred Table 2 Breeding performances of Psammomys from the DP and DR lines. Characteristic

DP line

DR line

Number of breeding pairs Non-fertile pairs Time interval between first two litters (weeks) Age of female in first birth (weeks) Breeding span (weeks) Number and percentage of females that gave birth at least once

394 96(22%) 7.0±0.26 22.7±0.69 37.4±1.02 308(78%)

132 54(41%) 6.1±0.39 32.2±1.9 20.5±1.74 78(59%)

P

NS <0.001 <0.001

PSAMMOMYS OBESUS

Characteristic

DP line

DR line

P

Number and percentage of females that gave birth at least three times Number and percentage of females that gave birth at least seven times Number of litters per female Number of newborns per litter Number of newborns in first litter Number of weaned per litter Number of weaned in first litter Total number of newborns per female Total number of weaned per female Number of newborns per female per week Number of weaned per female per week

230(58%) 74(19%) 3.3±0.12 2.8±0.05 1.9±0.07 1.9±0.05 1.6±0.09 11.5±0.42 8.4±0.4 0.35±0.01 0.28±0.01

26(20%) 2(1%) 1.3±0.19 2.7±0.11 1.8±0.13 2.2±0.12 1.6±0.08 5.4±0.56 5.3±0.7 0.36±0.03 0.28±0.01

<0.001 NS NS NS NS <0.001 <0.001 NS NS

291

Values are given as Mean±SEM NS—non significant

at different ages to KE diet (Ziv et al., 1999), the sensitivity to the development of diabetes increases from weaning to the peak at about 5 months of age and decreases thereafter (Figure 3). At 5 months of age, obesity factor measured as the proportion of epididymal fat weight to total body weight, is highest, while the proportion of other organs to total body weight remains unchanged. In animals older than 7–8 months, the potential to develop diabetes and obesity decreases. These changes are in correlation with the decrease in fertility in Psammomys from both DP and DR lines as demonstrated in Table 2. Metabolic and Reproductive Efficiency The essence of thrifty metabolism is high metabolic efficiency that enables the existence in an environment characterized by constant supply of low energy diets. The artificial laboratory condition of an ad-libitum accessible HE diet that creates a continuous input of energy, leads to hyperinsulinemia and hyperglycemia. The different sensitivity to the development of diabetes between the DP and the DR lines can be caused by one or more factors: differences in food intake, in hepatic and peripheral resistance, in pancreatic activity, or in metabolic efficiency. To measure the metabolic efficiency we followed animals during the period of their most rapid growth after weaning (2.5–3.0 g/day in males and 2.4–2.9 g/day in females) and calculated metabolic efficiency as the relation between digestible energy intake to weight increment (Kalman et al., 1993). There is no hyperphagia in both DP and DR lines when fed on HE, KE or LE diets, hyperglycemia is related to energy availability in the diet. Quantity of feces was significantly higher in animals fed on LE diet compared with animals fed on HE or KE diet. As indicated in Figure 4, metabolic efficiency in DP line Psammomys fed on all diets was 6.0– 6.6 Kcal/g of weight increase while in the DR line metabolic efficiency was 9.0–9–6 Kcal/g of weight increase. Other studies were performed on a branch of the Israeli colony in Australia, (Collier et al., 1997a, 1997b; Barnett et al., 1994a; Barnett et al., 1994b; Barnett, 1995). The insulin resistance developing in Psammomys was suggested to result from an excessive energy intake secondary to the hyperinsulinemia which might have induced hyperphagia. However, the weight gain between groups C and A did not exceed 15% (Barnett et al., 1995). Differences between DP and DR lines are not limited to their dietary induced diabetes but also to other zootechnical and reproductive characteristics (Kalman et al., 1996). Table 2 compares various reproductive characteristics in both lines. Reproductive efficiency in Psammomys is lower than in outbred rat strains

292

E.ZIV AND R.KALMAN

Figure 4 Metabolic efficiency in DP and DR Psammomys fed on various diets. Values are given as mean ± SE for 6–27 animals at each point.

Figure 5 Body weights of male and female Psammomys from weaning, fed on LE diet. Values are given as mean ± SE for 110 males and 109 females *—indicates significant difference between male and female weights.

(Baker, 1979). Average number of weaned per female per week is 0.28 versus 1.0–1.5 in outbred rats (Weihe, 1987). Reproductive efficiency is higher in the DP line than in DR line females, due to the difference in the proportion of nonreproductive females (22% in the DP line vs. 41% in the DR line) and due to the difference in the average number of births per female (3.3 in the DP line vs. 1.3 in the DR line). These two parameters create a difference in the total number of newborn per female during its reproductive life (11.5 newborn per female in the DP line vs. 5.4 in the DR line). No difference was observed in the average number of newborn per birth (2.8 vs. 2.7 respectively).

PSAMMOMYS OBESUS

293

Figure 6 Relative change in blood glucose levels (expressed in %) following IP injection of 2U of insulin in fed and fasted states. Values are given as mean ± SE for 8 animals.

Figure 5 demonstrates growth curves of male and female Psammomys from the DP line when fed on LE diet. Period of most rapid growth is up to 65 days (2.5– 3.0 g/day in males and 2.4–2.9 g/day in females). Growth continues up to 180 days but at a slower rate (1.4 and 0.8 g/day respectively). Average weight of males is 264±5 g and females 223.8±7 g. The relative weight of most organs remains unchanged (Kalman et al, 1996). Adrenal glands are the only organ which relative weight is significantly different between males and females in all age groups (p < 0. 001), and is always higher compared with albino rats (Kalman et al., 1996). Primary insulin resistance Insulin resistance in Psammomys is an inherent innate species characteristic even in the normoglycemic— normoinsulinemic state. In experiments where plasma insulin was elevated by i.p. administration of exogenous bovine insulin to normoglycemic normoinsulinemic animals (state A) (Figure 6), only mild hypoglycemia was observed compared with albino rats that received similar dosage of insulin (Ziv et al., 1996). External bovine insulin is effective in Psammomys as demonstrated in its hypoglycemie and hypotriglyceridemic effect in insulin-deficient Psammomys in state D. Exogenous insulin in form of s.c. implants that release 2U/24h for 10 days was implanted in state D Psammomys that became metabolically similar to the insulin resistant state C, characterized by endogenous hyperinsulinemia (Ziv et al., 1996). Figure 7 shows that despite the strong hypoglycemic effect in fasted state D Psammomys, and it’s capacity to lower triglycerides, the superimposed exogenous insulin is not capable to lower blood glucose levels in the non fasting—HE diet fed—state C Psammomys. Experiments with prediabetic, weanling, state A Psammomys given exogenous insulin and fed either LE or HE diet for 14 to 28 days, were aimed in order to learn if the cause of diabetes in Psammomys is lack of biologically active insulin, and if insulin resistance is primary or secondary to obesity. Figure 8 demonstrates the effect of diet and exogenous insulin on growth, plasma insulin and blood glucose levels.

294

E.ZIV AND R.KALMAN

Figure 7 Blood glucose and plasma triglyceride levels in state D Psammomys following administration of external insulin. Values are given as mean ± SE for 8 animals.

Both groups of Psammomys, with and without exogenous insulin, fed on HE diet, grew faster and became more obese than Psammomys fed on LE diet. No significant difference was observed between the two groups fed on the same diet (Ziv et al., 1996). Addition of exogenous insulin produces high plasma insulin levels in both LE and HE fed Psammomys. High levels of exogenous insulin do not succeed in preventing hyperglycemia in HE fed Psammomys even though administered to weaning, non-obese, animals. Feeding

PSAMMOMYS OBESUS

295

Figure 8 Effect of diet and exogenous insulin on growth, plasma insulin and blood glucose levels of Psammomys in a 14 day period. Values are means ± SE for 6 animals at each point.

Psammomys on HE diet causes by itself, as expec, a dramatic increase in blood glucose levels. Psammomys fed on LE diet keep normoglycemia in the fed state and do not lapse into hypoglycemia, even if high doses of exogenous insulin are administered to them. These results clearly indicate that hyperglycemia in HE fed Psammomys is not due to lack of insulin, and that the primary cause for the development of diabetes in Psammomys is a primary, inherited insulin resistance expressed mainly in the fed state. Our premise of liver and muscle primary insulin resistance as a characteristic exemplified by Psammomys was confirmed by the hyperinsulinemic-euglycemic clamp studies (Ziv et al., 1996). Quantitative data obtained on hepatic glucose production (HGP) and total glucose transport (TGT), indicate that insulin infusion did suppress the HGP and increase the TGT, again demonstrating the effectiveness of exogenous insulin. However, the HGP was only partially reduced (from 10.0±0.6to 3.8±0.4 mg/min.kg), whereas in albino rats under the same conditions, the HGP was completely abolished (from 11.0±0.5 to 0.7±0.3 mg/ min.kg). Lack of complete suppression of HGP in Psammomys was evident at a higher and longer lasting level of hyperinsulinemia than in rats. Also the limited elevation of TGT in the hyperinsulinemic clamped Psammomys (TGT Psammomys=16.0±1.9 vs. 39.0±1.9 mg/min.kg in albino rats) attests to the fact that the peripheral glucose utilization is low enough to be compensated by gluconeogenesis, avoiding lapse into hypoglycemia during exogenous insulin treatment. While clamped albino rats have very similar reactions among all individual animals, clamped Psammomys show individual phenotypic differences. Some with reduced peripheral response to insulin, and others with reduced hepatic response. Figure 9 shows the

296

E.ZIV AND R.KALMAN

Figure 9 HGP and TGT values in Psammomys in course of an euglycemic—hyperinsulinemic clamp, r=0.624 for 15 animals.

correlation between HGP and TGT levels in each clamped individual. Thus, it should be emphasized that the combination of the responses of these two different organs to insulin, determines the diabetic potential of each individual Psammomys. Overexpression of Certain Intracellular Proteins—the Possible Mechanism of Insulin Resistance The binding of insulin to the liver and muscle insulin receptors was very low even in stage A, compared with the albino rat. Insulin binding curve indicated that the muscle and liver membranal insulin receptor density in Psammomys is about one fifth compared with albino rats (Kanety et al., 1994). The tyrosine kinase activity was assessed in Psammomys both by receptor autophosphorylation and by phosphorylation of synthetic tyrosine kinase substrate, poly (glu:tyr) 4:1 (Kanety et al., 1994). Basal phosphorylation of the isolated insulin receptor tyrosine kinase was comparable in the stages B and C of Psammomys and similar to that in the albino rat. The extent of the tyrosine kinase activation by insulin was markedly lower at stages B and C, compared with stage A both in liver and in muscle in association with hyperinsulinemia. The deterioration of the insulin receptor function in Psammomys is reversible upon return to LE diet for 7–14 days or by diet restriction (Kanety et al., 1994). The recovery of tyrosine kinase activity was complete when the animals at stage C returned to stage A. The reduced tyrosine kinase-catalyzed tyrosine phosphorylation may be assumed to be caused by enhanced dephosphorylation of the receptor β-subunit and IRS-1, carried out by the action of phosphotyrosine phosphatase (PTPase). Psammomys in stage A exhibited low LAR (leukocyte antigen receptor)-PTPase activity in liver and muscle, in parallel with the low density of insulin receptors (Ziv et al., 1997). However, Psammomys tissues in stage C did not show an increase in cytosolic or membranal LAR-PTPase activity, compared with stage A (Balta et al., 1997), suggesting that LAR-PTPase is unlikely to be responsible for their additional reduced receptor tyrosine phosphorylation in the hyperinsulinemic Psammomys. Also, there is a low amount of GLUT-4 mRNA and GLUT-4 protein in the

PSAMMOMYS OBESUS

297

muscle of the Psammomys even in the basal normoglycemic-normoinsulinemic stage, compared with the albino rat (Ziv et al., 1997). It was recently demonstrated (Ikeda et al., 1999; Shafrir et al., 1999) that several protein kinase C (PKC) isoforms (alpha, epsilon and zeta) which include members from all three subclasses of PKC, are overexpressed in the skeletal muscle of diabetic animals. This is most prominent for the epsilon isotype of PKC (82.2% increase). Increased expression of PKCε (593% increase) could already be detected in A state DP animals when compared with A state DR animals. In addition, plasma membrane associated fractions of PKCε were increased in skeletal muscle of both diabetic and prediabetic animals, suggesting increased activation of this PKC isotype in the DR line. Overexpression and/or increased activity of PKCε may contribute to the development of diabetes in these animals, possibly through inhibition of insulin receptor tyrosine kinase activity mediated by serine/threonine phosphorylation of the insulin receptor or the IRS-1. In addition, Overexpression and/or chronic activation of PKCε may contribute to the development of insulin resistance in the Psammomys, also by enhancing the degradation of insulin receptors (Ikeda et al., 1999). These results indicate that the particular composition of intracellular proteins active in the insulin signalling pathway, determines the adaptation of’Psammomys to the desert conditions. This adaptation is recognized as “primary insulin resistance” in accordance with the thrifty gene hypothesis (Neel, 1963). Potential use of Psanimomys Potential uses of Psammomys include research on nutritionally induced insulin resistance, contribution of different nutrients (such as carbohydrate, fat and protein), β cell function and nature of overstimulation whether glyco- or lipotoxic or caused by other mechanisms, insulin receptors and 1RS activation and deactivation, GLUT transporter localization, mechanism of the insulin signalling system, conditions favoring obesity coincident with insulin resistance, lipid metabolism, physiology of desert animals. Among the complications which can be studied in Psammomys obesus are cataract formation (Gutman et al., 1975; Zahnd and Adler, 1984), hepatic thymic and uterine neoplasms in old Psammomys (Adler et al., 1976; Rosenmann et al., 1982; Czernobilsky et al., 1982), diabetic angiolopathy (Caffier et al., 1977; Marquie et al., 1991), skeletal bone deterioration (Moskowitz et al, 1990; Amir et al, 1991; Silberberg et al., 1986; Silberberg et al., 1979), diabetic neuropathy (Wuarin-Bierman et al., 1987) and hyperlipidemia (Chajek-Shaul et al., 1988; Gutman et al., 1991). Pharmacologic effects on hyperlipidemia, obesity and atherosclerosis were also investigated (Tzur et al., 1988; Marquie et al., 1996; Marquie et al., 1997). In the physiologic experiments, particularly in those aimed at inducing hyperglycemia and obesity, the DP animals in Stage A (on LE diet) should serve as controls. Also the DR animals maintained on HE diet may serve as controls of DP animals subjected to HE diet. Availability and Shipping At the time of writing, Psammomys are available from the original breeding colony through one of the following addresses: Rony Kalman DVM, Animal Facility, Hebrew University-Hadassah Medical School, Jerusalem 91120, Israel, Fax: (972–2)-6424654, e-mail: [email protected]. and/or Harlan Laboratories, Ein Kerem, Jerusalem 91120, POB 12085, Israel, Tel: (972–2)-6439398, Fax: (972– 2)-6439403 Cost of Psammomys ranges from 60 to 80 USD excluding freight costs.

298

E.ZIV AND R.KALMAN

CONCLUSIONS The thrifty metabolism enables Psammomys an efficient and controlled usage of its natural energy sources, which in the desert are represented by the presence of LE diet and only occasional short-term availability of HE diet. The need of Psammomys for such metabolism is highest during its growing and breeding periods. Such beneficial characteristics become a disadvantage when Psammomys is fed on HE diet for long period. They cause a permanent hyperglycemia and a secondary damage to other organs especially β cells. Normoglycemic animals from the DR line represent animals that are not well adapted to desert conditions, their low breeding efficiency is one of the signs demonstrating this property. It is probable that the adaptation of both animals and humans to plentiful nutrition either in laboratory conditions or as a result of affluence of the Western lifestyle have attenuated or selected out the “inborn” insulin resistance characteristics of populations living in nutritional scarcity. Presence of primary insulin resistance and absence of gluconeogenesis restraint in such populations, together with rich nutrient intake would lead to enhancement of insulin secretion, excessive lipogenesis followed by hyperglycemia. This path of events with eventual obesity with potentiation of insulin resistance is well exemplified in Psammomys. ACKNOWLEDGEMENTS We wish to thank Prof. Eleazar Shafrir for his advice during the preparation of this Chapter. REFERENCES Adler, J.H.Roderig, H. and Ungar, H. (1976) Neoplastic liver nodules of unknown cause in a colony of sand rats (Psammomys obesus). Isr. J. Med. Set., 12, 1212–15. Adler, J.H., Lazarovici, G. and Marton, M., et al. (1988) Patterns of hyperglycemia, hyperinsulinemia and pancreatic insufficiency in sand rats (Psammomys obesus). In Lessons from Animal Diabetes II, edited by E.Shafrir and A.E.Renold, pp. 384–388. London: John Libbey. Amir, G., Adler, J.H. and Menczel, J. (1991) Histomorphometric analysis of weight bearing bones of diabetic and nondiabetic sand rats (Psammomys obesus). Diab. Res., 17, 135–137. Baker, D.E.J. (1979) Reproduction and breeding. In The laboratory rat, edited by H.J.Baker, J.R.Lindsey and S.H.Weisbroth, pp. 154–166. Orlando FL: Academic Press. Balta, Y., Ziv, E., Kalman, R., Sack, J., et al. (1997) Psammomys obesus, a model of nutritionally induced diabetes mellitus is grossly deficient of both receptor-like protein—tyrosine phosphatase LAR and insulin receptor. Isr. J. Med. Sci., 33(Suppl. 1), S—10. Barnett, M., Collier, G.R., Collier, F.M., Zimmet, P. and O’Dea, K. (1994) A cross-sectional and short-term longitudinal characterization of NIDDM in Psammomys obesus. Diabetologia, 37, 671–676. Barnett, M.M., Collier, G.R., Zimmet, P. and O’Dea, K. (1994) The effect of restricting energy intake on diabetes in Psammomys obesus. Int. J.Obes. Relat. Metab. Disord., 18, 789–94. Barnett, M., Collier, G.R., Zimmet, P. and O’Dea, K. (1995) Energy intake with respect to the development of diabetes mellitus in Psammomys obesus. Diab. Nutr. Metab., 8, 1–6. Bar-On, H., Ben-Sasson, R., Ziv, E., Arar, N., Shafrir, E. (1999) Irreversibility of nutritionally induced NIDDM in Psammomys obesus is related to β-cell apoptosis. Pancreas, 18, 259— 265. Caffier, P., Fuchs, U., Wohlrab, F., et al. (1977) The problem of diabetic angiolopathy in sand rats (Psammomys obesus). Endokrinologie, 70, 269–88. Chajek-Shaul, T., Ziv, E., Friedman, G., et al. (1988) Regulation of lipoprotein lipase activity in the sand rat: Effect of nutritional stage and cAMP modulation. Metabolism, 37, 1152– 58.

PSAMMOMYS OBESUS

299

Collier, G.R., De Silva, A., Sanigorski, A., Walder, K., Yamamoto, A. and Zimmet, P. (1997) Development of obesity and insulin resistance in the Israeli sand rat (Psammomys obesus). Does leptin play a role? Ann. NY Acad. ScL, 827, 50–63. Collier, G.R., Collier, P.M., Sanigorski, A., Walder, K., Cameron Smith, D. and Sinclair, A.J. (1997) Non-insulin dependent diabetes mellitus in Psammomys obesus is independent of changes in tissue fatty acid composition. Lipids, 32, 317–22. Czernobilsky, B., Ungar, H. and Adler, J.H. (1982) Spontaneous uterine neoplasms in the fat sand rat (Psammomys obesus). Lab. Anim., 16, 285–289. Donath, M.Y., Gross, D.J., Cerasi, E. and Kaiser, N. (1999) Hyperglycemia-induced β-cell apoptosis in pancreatic islets of Psammomys obesus during development of diabetes. Diabetes, 48, 738–44. Duhault, J., Boulanger, M., Espinal, J., Marquie, G., et al. (1994) Islet amyloid polypeptide in Psammomys obesus: lack of correlation between insulin resistance and plasma IAPP levels. Cell. Mol. Biol., 40(4), 535–540. Gutman, A., Andreus, A. and Adler, J.H. (1975) Hyperinsulinemia, insulin resistance and cataract formation in sand rats. Isr. J. Med. Set., 11, 714–22. Gutman, A., Kalderon, B., Levy, E. and Shafrir, E. (1991) Pattern of very low density lipoprotein disposal in the sand rat (Psammomys obesus). In Lessons from Animal Diabetes, edited by E.Shafrir, 3, pp. 699–71. London: SmithGordon. Habito, R.C., Barnett, M., Yamamoto, A., Cameron-Smith, D., et al. (1995) Basal glucose turnover in Psammomys obesus. Acta Diabetol., 32, 187–192. Ikeda, Y., Ziv, E., Hansen, L.L., Busch, K.A., et al. (1999) Cellular mechanism of nutritionally induced insulin resistance in Psammomys obesus: Overexpression of Protein Kinase C epsilon in skeletal muscle precedes the onset of hyperinsulinemia and hyperglycemia. Diabetes, 48, A76. Kalderon, B., Gutman, A., Levy, E., Shafrir, E. and Adler, J.H. (1986) Characterization of stages in the development of obesity—diabetes syndrome in the sand rat (Psammomys obesus). Diabetes, 35, 717–724. Kalman, R., Adler, J., Lazarovici, G., et al. (1993) The efficiency of sand rat metabolism is responsible for development of obesity and diabetes. J. of Basic & Clin. Physiol. & Pharmacol., 4, 57–68. Kalman, R., Lazarovici, G., Bar-On, H. and Ziv, E. (1996) Psammomys obesus (sand rat)— morphological, physiological and biochemical characteristics of a model for type II diabetes. Contemp. Topics Lab. Anim. Set., 35 (5), 67–70. Kanety, H., Moshe, S., Shafrir, E., Lunenfeld, B. and Karasik, A. (1994) Hyperinsulinemia induces a reversible impairment in insulin receptor function leading to diabetes in the sand rat model of non-insulin-dependent diabetes mellitus. Proc. Natl. Acad. S ci. USA, 91, 1853–57. Marquie, G., Hadjiiski, P., Arnaud, O. and Duhault, J. (1991) Development of macroangiopathy in sand rats (Psammomys obesus), an animal model of non-insulin-dependent diabetes mellitus: effect of gliclazide. Am. J. Med., 90 (Suppl 6A), 655–61. Marquie, G., Menouar, T., Pieraggi, M.T., Dousset, N. and Bennani, N. (1996) Prevention of preatheromatous lesions M in sand rats by treatment with a nutritional supplement. Arzneimittelforschung, 46, 610–614. Marquie, G., Duhault, J., Espinal, J., Petkov, P., Jablenska, R., Khallayoun, S., et al. (1997) S1 5 261, a novel agent for the treatment of insulin resistance. Studies on Psammomys obesus. Effect on pancreatic islets of insulin resistant animals. Cell. Mol. Biol., 43, 243–251. Moskowitz, R.W., Ziv, I., Denko, C.W., Boja, B., Jones, P.K. and Adler, J.H. (1990) Spondylosis in sand rats: A model of intervertebral disc degeneration and hyperostosis. J. Orthopaedic Res., 8, 401–411. Neel, J.V. (1963) Diabetes mellitus: a “thrifty” genotype rendered detrimental by “progress”? Am. J. Hum. Genet., 14, 353–362. Rosenmann, E., Adler, J.H. and Ungar, H. (1982) Spontaneous thymic tumours in the fat sand rat (Psammomys obesus). J. Comp. Pathol., 92, 349–356. Shafrir, E. and Gutman, A. (1993) Psammomys obesus of the Jerusalem colony. A model for nutritionally induced, noninsulin-dependent diabetes. J. Basic. Clin. Physiol. Pharmacol., 4, 83–99.

300

E.ZIV AND R.KALMAN

Shafrir, E., Ziv, E. and Mosthaf, L. (1999) Nutritionally induced insulin resistance and receptor defect leading to beta cell failure in animal models—human implications. New York Acad Sci, 892, 223–246. Shafrir, E. and Ziv, E. (1999) Cellular mechanism of nutritionally induced insulin resistance: the desert gerbil Psammomys obesus and other animals in which insulin resistance leads to detrimental outcome. J. Basic Clin. Physiol. Pharmacol., 9, 347–385. Silberberg, R., Aufdermaur, M. and Adler, J.H. (1979) Degeneration of the intervertebral disks and spondylosis in aging sand rats. Arch. Pathol. Lab. Med., 103, 231–35. Silberberg, R. and Adler, J.H. (1983) Comparison of truncal and caudal lesions in the vertebral column of the sand rat (Psammomys obesus). Isr. J. Med. Sci., 19, 1064–1071. Silberberg, R., Adler, J.H. and Meier-Ruge, W. (1986) Effects of hyperinsulinism and of diabetes on proteoglycans of the intervertebral disc in weanling sand rats. Exp. Cell. Biol., 54, 121–127. Tzur, R., Rose-Kahn, G., Adler, J.H. and Bar-Tana, J. (1988) Hypolipidemic, antiobesity and hypoglycemichypoinsulinemic effects of β, βmethyl-substituted hexadecanedioic acid in sand rats. Diabetes, 37, 1618–24. Wuarin-Bierman, K., Zahnd, G.R., Kaufmann, F., Burcklen, L. and Adler, J. (1987) Hyperalgesia in spontaneous and experimental animal models of diabetic neuropathy. Diabetologia, 30, 653–57. Weihe, W.H. (1987) The laboratory rat. In The UP AW handbook on the care & management of laboratory animals, edited by T.Poole, pp. 309–330. England: Longman Scientific & Tech. Zahnd, G.R. and Adler, J.H. (1984) Sand rat as a model of diabetic cataract. In Lessons from Animal Diabetes, edited by E.Shafrir and A.E.Renold, 1, 500–2. London: J. Libbey. Ziv, E. and Shafrir, E. (1995) Psammomys obesus: nutritionally induced NIDDM-like syndrome on a “thrifty gene” background. In Lessons from Animal Diabetes, edited by E.Shafrir, 5, 285–300. London: Smith-Gordon. Ziv, E., Kalman, R., Hershkop, K., Barash, V., Shafrir, E. and Bar-On, H. (1996) Insulin resistance in the NIDDM model Psammomys obesus in the normoglycemic-normoinsulinemic state. Diabetologia, 39, 1265–1275. Ziv, E., Nachliel, L, Bar-On, H. and Shafrir, E. (1997) Nutritionally induced insulin resistance in Psammomys obesus, its consequences and its possible prevention. Exp. Clin. Endocr. Diabetes, 105, 12–13. Ziv, E., Shafrir, E., Kalman, R., Galer, S. and Bar-On, H. (1999) Changing pattern of prevalence of insulin resistance in Psammomys obesus, a model of nutritionally induced type 2 diabetes. Metabolism, 48, 1549–1554.

17. THE C57BL/6J MOUSE AS A MODEL OF DIETINDUCED TYPE 2 DIABETES AND OBESITY ANN E.PETRO and RICHARD S.SURWIT Duke University Medical Center, Durham, NC 27710

ABSTRACT Over the years, extensive work has been done to characterize the B6 mouse as an animal model to study the genetics of diet-induced diabetes and obesity. We initially observed that, in comparison to several other strains, B6 mice are susceptible to diabetes/obesity when maintained on a high-fat diet. This has led to a developing literature on the interaction of dietary factors and genetic pre-disposition in the development of diabetes and obesity. Research by our group as well as by others has implicated abnormalities in autonomic nervous system, beta cell, and adipocyte function in the etiology of this condition. Most recently, genetic mapping studies have identified differences in the expression of a novel uncoupling protein, UCP2, in adipocytes of obesity-prone B6 mice. Because of the importance of environmental variables in the development of obesity in these animals, carefully controlled studies that pay close attention to issues of handling, housing and diet promise to yield important information in the understanding of diabetes and obesity. Type 2 diabetes is a syndrome characterized by hyperglycemia, hyperinsulinemia, insulin resistance, and obesity. The effects of chronic hyperglycemia include many serious conditions such as atherosclerosis, nephropathy, neuropathy, retinopathy, and stroke. The increasing prevalence of type 2 diabetes has been attributed to the increasing obesity of the general population and consumption of a high-fat diet has been demonstrated to elicit the syndrome. In addition to promoting obesity, high-fat diets have been associated with risk for developing type 2 diabetes in epidemiological studies and, notably, this association has been found independent of obesity. Since obesity has been shown repeatedly (Morris et al., 1989; Marshall et al., 1991; Marshall et al., 1994; Feskens et al, 1995) to be one of the strongest risk markers for the development

302

A.E.PETRO AND R.S.SURWIT

of diabetes, high-fat diets may not only contribute directly to diabetes risk, but may also do so indirectly through their potential to induce obesity. An increase in dietary fat content has also been shown to produce diabetes and obesity in various strains of mice by West et al. (1992) and of rats by Schemmel et al. (1970). The C57BL/6J (B6) mouse has been wellcharacterized as an animal model to study the influence of diet on genetic background, specifically the genetics of diet-induced hyperinsulinemia, hyperglycemia, and obesity. The B6 mouse develops this syndrome only in response to a high-fat diet (Surwit et al., 1988). On a low-fat diet, this animal remains normal. In contrast, other strains such as the A/J mouse (Rebuffe-Scrive et al., 1993; Surwit et al., 1995; and Surwit et al, 1988) or the C57BL/KsJ (KsJ) (Surwit et al., 1994) are relatively resistant to these effects when fed a high fat diet. The A/J mouse has been used as a control strain throughout most of the research on the B6 mouse. The development of diabetes and obesity in the B6 mouse closely parallels the progression of common forms of the human disease. In humans, the onset of diabetes and obesity occurs gradually and often in the presence of a high-fat diet. In addition, Rebuffe-Scrive et al. (1993) and Surwit et al. (1995) reported that diet-induced diabetes and obesity in the B6 mouse is characterized by selective deposition of fat in the mesentery, an observation (Marshall et al., 1991 and Feskens et al,, 1995) consistent with the finding that abdominal obesity is an independent risk factor for diabetes in humans. Finally, Mills et al. (1993) reported that diet-induced diabetes and obesity in the B6 mouse is accompanied by the development of hypertension. Others have noted that high fat feeding can produce compromised immune function in B6 mice (Crevel et al., 1992). Coleman et al. (1978) noted that unlike the ob/ob and db/db mutant models of obesity, B6 mice develop obesity without hyperphagia. Any evidence of increased food consumption that occurs in comparison to control groups is in proportion to the increased size of the animals. The development of obesity results from increased feed efficiency (weight gained/kcalorie consumed), as demonstrated by Surwit et al. (1995); Brownlow et al. (1996); and Parekh et al. (1996). Yen et al. (1972) observed that while the ob/ob and db/db models are hypoactive, the B6 mouse becomes obese in spite of increased activity levels. Brownlow et al. (1996) reported that obese B6 mice are equally as active as their lean counter parts and nearly three times as active as A/J mice. Thus, the B6 mouse is an example of obesity developing as a result of the interaction of nutritional content of the diet and genetic variables. Data on the relationship between hyperphagia, physical activity, and obesity in humans are conflicting. Reports from several research groups (Braitman et al., 1985; Dreon et al., 1988; Romieu et al., 1988; Miller et al., 1990) show that, contrary to popular belief, obese subjects do not necessarily consume more calories than lean subjects. Recent studies (Hunter et al., 1996 and Hunter et al., 1997) found a relationship between percentage body fat and intraabdominal fat and activity in men and women. Other studies (Maffeis et al., 1998, Waxman et al., 1980, and Gazzaniga et al., 1993) failed to find a relationship between activity and weight. Maffeis et al. (1998), Waxman et al. (1980), and Gazzaniga et al. (1993) noted that diet composition was one main risk factor for the development of obesity and Maffeis et al. (1998) determined that genetic background was another such factor. The development of the B6 mouse as a research tool in which to gain understanding of mechanisms involved in the interactions between diet and genetic background will be invaluable for future research. This animal will also be useful to target and test therapeutic interventions in pre-clinical trials. Within a month of eating a high-fat diet, diabetes is evident in the B6 mouse. It has been reported (Rebuffe-Scrive et al., 1993; Surwit et al., 1995; Surwit et al., 1988) that B6 mice became hyperglycemic and hyperinsulinemic when compared to either B6 mice on a low-fat diet or A/J mice on either low- or highfat diets (Figures 1a and 1b). As shown in Figure 2, these studies also found that after eating a high fat diet for eight weeks B6 mice were significantly heavier when compared to control groups and, as illustrated in Figure 3 the glucose tolerance in B6 mice was impaired. The diabetes/obesity syndrome worsens with time.

C57BL/6J MOUSE AS A MODEL

303

Data at 16 weeks, published by Surwit et al. (1995), showed B6 mice fed a high-fat diet had developed adipocyte hyperplasia, and hypertrophy (Figures 4a and 4b), resulting in animals with a fat mass increased by 93% as reported by Black et al. (1998). In addition, Mills et al. (1993) noted that at 16-weeks high-fat fed B6 mice were hypertensive. Even though fully manifested after 16 weeks of a high-fat diet, the diabetes/obesity syndrome is completely reversible at this stage in these mice. Parekh et al. (1996) showed that treatment with a low-fat diet fed ad libitum alone was sufficient. Furthermore, it has been shown that dietary fat is responsible for the development diabetes/obesity syndrome in B6 mice used as the model. Surwit et al. (1995) demonstrated that low-fat diets that are high in sucrose do not provoke the syndrome regardless of whether the sucrose is provided in the chow or water. In fact, Black et al. (1998) reported that body composition studies show an increased percentage of body protein in B6 mice fed low-fat highsucrose diets. There are a number of other animal models available in which the diabetes/obesity syndrome is studied. However, in many of these models the syndrome occurs spontaneously or is the result of a single gene mutation. This limits their utility as a research tool in which to study the effect of diet on genetic background in this polygenic syndrome. A number of defects in mechanisms that affect glucose and fat metabolism have been identified in these animals. Surwit et al. (1988) demonstrated that obese B6 mice show exaggerated glycemic responses to stress and catecholamines and postulated that abnormal autonomic function is a key aspect of the syndrome (see Surwit and Kuhn 1993; Kuhn et al., 1995). This theory was supported by Ahren et al. (1997) who noted that while glucose stimulated insulin release was blunted in these animals, they showed increased insulin secretion in response to carbochol. Glucose-stimulated insulin release from pancreatic islets is compromised in B6 mice. This impairment is worsened when mice are raised on a high-fat diet, as reported by Lee et al. (1995) and Wencel et al. (1995). Like the ob/ob mouse, obese B6 mice show a defect in β adrenergic function in adipocytes. β1 and 3 receptor mRNA expression is diminished when B6 mice are fed high-fat diets. Furthermore, Collins et al. (1997) reported that adenylyl cyclase activity in response to β adrenergic stimulation is decreased in both white and brown adipose tissue when B6 mice fed a high-fat diet are compared to A/J mice fed a high-fat diet or to B6 mice fed a low-fat diet. As shown in Figure 5, fatfed B6 mice also show a tendency toward decreased glucose transport into white adipose tissue. Plasma leptin levels in response to high-fat feeding are also abnormal in B6 mice. Within the first two weeks of eating a high-fat diet, plasma leptin in A/J mice is significantly increased. This response is absent in the B6 mouse. According to Surwit et al. (1997), plasma leptin does not increase in B6 mice until after obesity is established (Figure 6). While the underlying genetic factors that predispose B6 mice to these abnormalities remain to be identified, Surwit et al. (1991) demonstrated that hyperglycemia and insulin resistance in B6 mice were likely controlled by different genes. Seldin et al. (1994) mapped hyperinsulinemia to a site on distal mouse chromosome 7, while hyperglycemia mapped to a different region on the same chromosome. Fleury et al. (1997) mapped both hyperinsulinemia and hyperleptinemia to the region containing UCP2 and UCP3. Surwit et al. (1998) showed that expression of UCP2, but not of UCP3, is upregulated in WAT in response to dietary fat in A/J and KsJ mice, but not in B6 mice. The differential expression in response to dietary fat in these strains suggests that UCP2 may have a role in the development of diet-induced diabetes and obesity, thus identifying a genetic marker that is influenced by diet and important in energy metabolism. SOURCES AND CARE OF RESEARCH SUBJECTS There are a number of reliable sources that provide laboratory animals. Animals are ordered by strain, age, weight and/or sex. Within strain and sex, the animals should be matched by weight or age. The supplier

304

A.E.PETRO AND R.S.SURWIT

Figure 1 The effect of high- (58% kcal fat) and low-fat (11% kcal fat) diets on plasma glucose in A/J and B6 mouse strains is shown in panel (a) and the effect of plasma insulin is shown in panel (b). B6 mice are hyperglycemic and hyperinsulinemic after 1 month on a high-fat diet. Data are the mean ± sem of 10 animals per group.

C57BL/6J MOUSE AS A MODEL

305

Figure 2 The effect of the high- and low-fat diets on body weight in A/J and B6 mice. Data are the mean±sem. n=10.

must be able to provide a written description of the facility, the animal husbandry procedures, and the genetic quality control program that is in place. A health report detailing the bacteria and parasites found in the animals must be supplied with each shipment. A single supplier should be identified and used. If there are compelling reasons to change suppliers, a small cohort of animals from each supplier should be bought and tested simultaneously prior to starting experiments. For example, animals used for studies of the genetics diet-induced type 2 diabetes and obesity would be followed to assure that the syndrome developed in the same time frame and to the same extent as seen in previous work. Plasma glucose and insulin and body weight are the criteria that would be assessed. Minimum cage space requirements for laboratory animals are described in the “Guide for the Care and Use of Laboratory Animals” (NIH 86–23). The number of mice permitted per cage is based on the weight of the animals relative to the square area of the cage. For most research purposes, mice are usually housed in groups of three to five. However, there are some experimental designs that require Table 1 The effect of housing density on the development of the diabetic/obesity phenotype in A/J and B6 mice.

Weight (g) Glucose (mg/dl) Insulin (uU/ml)

A/J Group of 5

A/J Individual

B6 Group of 5

B6 Individual

33.8±1.2 152±4 34.7±4.4

33.2±0.8 161±5 91.6±11

45.6±1.0 216±8 107±16

41.5±0.9 192±7 137.7±19

individual housing, i.e., pair feeding or animals that have been cannulated. Fewer animals per cage increases the operating costs of the facility. The number of mice per cage can affect study results. Some strains are more sensitive to housing density than others. A study comparing the effect of group (5 mice/cage) vs. individual housing on B6 and A/J mice showed B6 mice housed individually had significantly lower body

306

A.E.PETRO AND R.S.SURWIT

Figure 3 The effect of diet on glucose tolerance in B6 mice. A 0.50 glucose/g/kg challenge was given by i.p. injection 30 minutes before sample collection. Data are the mean ± sem of 5 mice per group.

weight and plasma glucose when compared to group housed animals. Plasma insulin was not affected. There was no affect on weight, glucose, or insulin in A/J mice (Table 1). On the other hand, overcrowding also induced hyperglycemia. Research with ob/ob mice (Surwit et al., 1993) has shown that increasing the number of animals in a cage to 20 mice/cage increases hyperglycemia, even though the cage size is increased proportionately. Thus, it can be expected that varying housing conditions from those described above will alter the diabetes phenotype. The diets must meet the nutritional needs of the species and the American Institute of Nutrition (1977, 1980) outlined the dietary requirements for rodents. There are a number of firms that can provide purified diets. Choose a manufacturer that can ensure the diets are made from high quality ingredients and that these ingredients are consistent in their composition. The producer should have the capability to prepare and ship the diets promptly. The exact composition of the diet will vary with experimental design. Test compounds can be admixed into the diets at the time of manufacture, should that be desired. Our original observation (Surwit et al., 1988) that a high-fat diet would induce obesity and diabetes utilized a diet high in both sucrose and fat formulated by Bio-Serve, Inc. (Frenchtown, NJ). The fat content of the diet was obtained from lard. The composition of lard varies from batch to batch and in more recent studies the high fat diet was based on coconut oil (Surwit et al., 1995). Research on the comparative contribution of sucrose and saturated fat to the development of hyperglycemia demonstrated that fat was the nutrient primarily responsible for both diabetes and obesity. Surwit et al. (1995) noted that while adding sucrose to fat exacerbated the effects of fat alone, high sucrose in the absence of fat failed to induced either obesity or diabetes. A subsequent study (Opara et al., 1996) revealed that polyunsaturated fat produced a similar diabetic effect. Rodents are normally maintained on a 12-hour light/dark cycle in a room that is temperature controlled (68–7 5 °F). In areas of research where feeding behavior affects the results (even though it may not be the

C57BL/6J MOUSE AS A MODEL

307

Figure 4 The effect of diet and strain on adipocyte hyperplasia (panel a) and hypertrophy (panel b) in the mesenteric and inguinal fat pads. A high-fat diet induces both adipocyte hyperplasia and hypertrophy in the B6 mouse. Data are the mean ± sem of 10 animals per group.

308

A.E.PETRO AND R.S.SURWIT

Figure 5 The effect of high- and low-fat diets on glucose transport into white adipose tissue (as assessed by the accumulation of 14C 2 deoxyglucose) in B6 mice. Data are the mean ± sem of 5 animals per group.

primary focus of the research), it is best to have the light portion of the cycle timed to occur during normal working hours. Rodents are nocturnal and, as such, a large portion of their food intake occurs during the dark cycle. If the dark cycle occurs during working hours, disruptions in feeding may occur when room lights are turned on to allow caretakers to work in the facility. Experience in our laboratory (unpublished) showed that reversing the day/night cycle disrupted feeding behavior to the extent that the diet-induced diabetes/obesity phenotype was compromised and the genetic mapping studies were nearly impossible. In addition, it is important to maintain room temperature within a few degrees. Drastic changes in room temperature will affect the basal metabolic rate of the animals, potentially alter food intake, and lead to inconsistent results. Each animal in a cage needs a unique identifying mark, and toe clips or ear punches are commonly used. Animals should be weighed the day they arrive. Food must also be weighed, especially if the diet is different than the weaning diet. The animals and food should be weighed daily for the first few days after arrival. Animals should not lose weight and should be eating. The most common cause of weight loss and decreased food intake is failure to adapt to automatic watering systems. The animals will rapidly dehydrate and set off a downward spiral from which they may not recover. Water can be provided in dishes along with moistened food for the first few days. Any difficulties the animals experience in adjusting to the new facility and/or diet can be identified early and likely remedied. Even after the animals have acclimated to the facility, they need to be observed on a daily basis. Body weight and food intake measurements can be decreased to once or twice per week unless otherwise dictated by the experimental protocol. Visual inspection of the animals is sufficient to identify any of the common problems that may develop. Occasionally an animal will suffer from overgrown incisors and be unable to eat.

C57BL/6J MOUSE AS A MODEL

Figure 6 The effect of high-fat (panel a) and low-fat (panel b) on plasma leptin concentration in A/J and B6 mice.

309

310

A.E.PETRO AND R.S.SURWIT

The teeth can be carefully trimmed with a pair of sharp scissors. Sluggishness and ruffled fur are indicators of an ill animal. If the problem cannot be identified and corrected in a day or two, the animal needs to be sacrificed. Skin lesions may be indicative of an infection or they may be bite wounds inflicted by a cage mate. Observing the behavior of the animals in the cage will clarify this. If the wounds are due to aggressive behavior try housing the animals in smaller groups. If the lesions appear to be caused by parasitic infestation, a quick and proper diagnosis is essential. The institutional veterinary staff can prescribe proper corrective action. If there are ailments outside of these or recurring problems, a consultation with the veterinary staff is necessary. Animals should be handled on a regular basis before procedures are performed so they are accustomed to being touched. It is easy to incorporate this into the weekly routine without a large investment in time. The animals are handled when they are weighed and can be touched during the daily visual inspection. This will minimize any affect that a novel, stressful experience might have. It has been shown that glycemic response to both major behavioral stress and adrenergic stimulation is increased when mice are obese. As reported by Surwit et al. (1988), this stress-induced hyperglycemia is far greater in the morbidly obese B6 mice than in the moderately obese A/J. Stress effects on glycemia can be extremely subtle. In one experiment with ob/ob mice, Surwit et al. (1985) demonstrated that mice exposed to the sound of a metronome presented simultaneously with unexpected movement would show severe hyperglycemia when the metronome was later presented alone. Other minor changes can be disruptive and have deleterious consequences that may not be foreseen. Some changes affect strains differentially. For example, when B6 and A/J mice were transferred from one laboratory to another for a three-day behavioral testing regime, A/J but not B6 mice lost weight (unpublished) even though there were no changes within the cage or to the diet. Further increases in blood glucose in diabetic/obese B6 mice were seen when the animals were moved to another laboratory immediately prior to sample collection. See also the effect of changing light/dark cycles. Therefore, subtle variations in the environment can greatly influence the phenotypic characteristics of these strains. To avoid or minimize these complications, the room maintenance schedule should follow as consistent a routine as possible on a weekly basis. It is best to avoid collecting samples on days when the workload is heavy in the animal rooms. For example, providing fresh bedding to a cage causes a period of aggressive behavior in some strains of mice as the animals adjust to the new bedding and mark their territory. Some neuroendocrine values may be influenced under these circumstances. When samples are to be collected at regular intervals throughout the experiment, care should be taken to make the collection under the same circumstances. Some variables, i.e., body temperature, are influenced by time of day. If samples are to be collected in the post-absorptive state, the number of hours since the food was removed should always be the same. The order of collection should be randomized across the treatments to minimize the effect of the time lag. If the number of subjects is large and the time lag long, it may be necessary to spread the collection across two half days, being consistent about the time of day the samples will be collected. REFERENCES Ad Hoc Committee on Standards for Nutritional Studies (1980) Second Report. Journal of Nutrition, 110, 1726. Ahren, B., Simonsson, E., Scheurink, A.J., Mulder, H., Myrsen, U. and Sundler, F. (1997) Dissociated insulinotropic sensitivity to glucose and carbachol in high-fat diet-induced insulin resistance in C57BL/6J mice. Metabolism, 46, 97–106. American Institute of Nutrition (1977) Report of the Ad Hoc Committee on Standards for Nutritional Studies. Journal of Nutrition, 107, 1340–1348.

C57BL/6J MOUSE AS A MODEL

311

Black, B.L., Croom, J., Eisen, E.J., Petro, A.E., Edwards, C.L. and Surwit, R.S. (1998) Differential effects of fat and sucrose on body composition in A/J and C57BL/6J mice. Metabolism, 47, 1354–59. Braitman, L.E., Adlin, E.V. and Stanton, J.L. (1985) Obesity and caloric intake: The National Health and Nutrition Examination Survey of 1971–1975 (HANES I). Journal of Chronic Diseases, 38, 727–732. Brownlow, B.S., Petro, A., Feinglos, M.N. and Surwit, R.S. (1996) The role of motor activity in diet-induced obesity in C57BL/6J mice. Physiological Behavior, 60, 37–41. Coleman, D.L. (1978) Obese and diabetes: Two mutant genes causing diabetes-obesity syndromes in mice. Diabetologia, 14, 141–148. Collins, S., Daniel, K.W., Petro, A.E. and Surwit, R.S. (1997) Strain-specific response to beta3-adrenergic receptor agonist treatment of diet-induced obesity in mice. Endocrinology, 138, 405–413. Crevel, R.W.R., Friend, J.V., Goodwin, B.F.J. and Parish, W.E. (1992) High-fat diets and the immune response of C57 B1 mice. British Journal of Nutrition, 67, 17–26. Dreon, D.M., Frey-Hewitt, B., Ellsworth, N., Williams, P.T., Terry, R.B. and Wood, P.D. (1988) Dietary fat: Carbohydrate ratio and obesity in middle-aged men. American Journal of Clinical Nutrition, 47, 995–1000. Feskens, E.J.M., Virtanen, S.V., Rasanen, L., Tuomilehto, J., Stengard, J., Pekkanen, J., et al. (1995) Dietary factors determining diabetes and impaired glucose tolerance: A 20-year follow-up of the Finnish and Dutch cohorts of the seven countries study. Diabetes Care, 18, 1104–1112. Fleury, C., Neverova, M., Collins, S., Raimbault, S., Champigny, O., Levi-Meyrueis, C., et al. (1997) Uncoupling protein-2: A novel gene linked to obesity and hyperinsulinemia. Nature Genetics, 15, 269–272. Gazzaniga, J.M. and Burns, T.L. (1993) Relationship between diet composition and body fatness, with adjustment for resting energy expenditure and physical activity, in preadolescent children. American Journal of Clinical Nutrition, 58, 21–28. Hunter, G.R., Kekes-Szabo, T., Treuth, M.S., Williams, M.J., Goran, M. and Pichon, C. (1996) Intra-abdominal adipose tissue, physical activity and cardiovascular risk in pre- and post-menopausal women. International Journal of Obesity and Related Metabolic Diseases, 20, 860– 865. Hunter, G.R., Kekes-Szabo, T., Snyder, S.W., Nicholson, C., Nyikos, I. and Berland, L. (1997) Fat distribution, physical activity, and cardiovascular risk factors. Medicine and Science in Sports and Exercise, 29, 362–369. Kuhn, C.M., Surwit, R.S. and Feinglos, M.N. (1994) Glipizide stimulates sympathetic outflow in diabetes-prone mice. Life Sciences, 56, 661–666. Lee, S.K., Opara, E.G., Surwit, R.S. and Akwari, O.E. (1995) Defective glucose-stimulated insulin release from perifused islets of C57BL/6J mice. Pancreas, 11, 206–211. Maffeis, C., Talamini, G. and Tato, L. (1998) Influence of diet, physical activity and parents’ obesity on children’s adiposity: A four-year longitudinal study. International Journal of Obesity and Related Metabolic Disorders, 22, 758–764. Marshall, J.A., Hamman, R.F. and Baxter, J. (1991) High-fat, low-carbohydrate diet and the etiology of non-insulindependent diabetes mellitus: The San Luis Valley diabetes study. American Journal of Epidemiology, 134, 590–603. Marshall, J.A., Hoag, S., Shetterly, S. and Hamman, R.F. (1994) Dietary fat predicts conversion from impaired glucose tolerance to NIDDM: The San Luis Valley diabetes study. Diabetes Care, 17, 50–56. Miller, W.C. (1991) Diet composition, energy intake, and nutritional status in relation to obesity in men and women. Medicine and Science in Sports and Exercise, 23, 280–284. Mills, E., Kuhn, C.M., Feinglos, M.N. and Surwit, R. (1993) Hypertension in C57BL/6J mouse model of non-insulin dependent diabetes mellitus. American Journal of Physiology, 264 (Regulatory Integrative Comparative Physiology 33), R73—R78. Morris, R.D., Rimm, D.L., Hartz, A.J., Kalkhoff, R.K. and Rimm, A.A. (1989) Obesity and heredity in the etiology of non-insulin-dependent diabetes mellitus in 32, 662 adult white women. American Journal of Epidemiology, 130, 112–121. National Diabetes Data Group, National Institutes of Health (1995) Diabetes in America, 2nd ed. Washington, D.C.

312

A.E.PETRO AND R.S.SURWIT

Opara, E.G., Petro, A., Tevrizian, A., Feinglos, M.N. and Surwit, R.S. (1996) L-Glutamine supplementation of a highfat diet reduces body weight and attenuates hyperglycemia and hyperinsulinemia in C57BL/6J mice. Journal of Nutrition, 126, 273–279. Parekh, P.I., Petro, A.E., Tiller, J.M., Feinglos, M.N. and Surwit, R.S. (1998) Reversal of diet-induced obesity and diabetes in C57BL/6J mice. Metabolism, 47, 1089–1096. Rebuffe-Scrive, M., Surwit, R., Feinglos, M., Kuhn, C. and Rodin, J. (1993) Regional fat distribution and metabolism in a new mouse model (C57BL/6J) of non-insulin-dependent diabetes mellitus. Metabolism, 42, 1405–1409. Romieu, I., Willett, W.C., Stampfer, M.J., Colditz, G.A., Sampson, L., Rosner, B., et al (1988) Energy intake and other determinants of relative weight. American Journal of Clinical Nutrition, 47, 406–412. Schemmel, R., Michelsen, O. and Gill, J.L. (1970) Dietary obesity in rats: Body weight and body fat accretion in seven strains of rats. Journal of Nutrition, 100, 1041–1048. Surwit, R.S., McCubbin, J.A., Livingston, E.G. and Feinglos, M.N. (1985) Classically conditioned hyperglycemia in the obese mouse. Psychosomatic Medicine, 47, 565–568. Surwit, R.S., Seldin, M.F., Kuhn, C.M., Cochrane, C. and Feinglos, M.N. (1991) Control of expression of insulin resistance and hyperglycemia in diabetic C57BL/6J mice by different genetic factors. Diabetes, 40, 82–87. Surwit, R.S., Seldin, M.F., Kuhn, C.M., Secor, C. and Feinglos, M.N. (1994). Diet-induced obesity and diabetes in C57BL/6J and C57BL/KsJ Mice. Mouse Genome, 92, 523–525. Surwit, R.S., Kuhn, C.M., Cochrane, C., McCubbin, J.A. and Feinglos, M.N. (1988) Diet-induced Type II diabetes in C57BL/6J mice. Diabetes, 37, 1163–1167. Surwit, R.S., Feinglos, M.N., Rodin, J., Sutherland, A., Petro, A.E., Opara, E.C., Kuhn, C.M. and Rebuffé-Scrive, M. (1995) Differential effects of fat and sucrose on the development of obesity and diabetes in C57BL/6J and A/J mice. Metabolism, 44, 645–651. Surwit, R.S. and Kuhn, C.M. (1993) Diet-induced type 2 diabetes and obesity in the C57BL/ 6J mouse: A stress-related model of human disease. In Lessons From Animal Diabetes, edited by E.Shafrir, pp. 219–228. London: SmithGordon. Surwit, R.S. (1993) Of mice and men: Behavioral medicine in the study of type II diabetes. Annals of Behavior Medicine, 15, 227–235. Surwit, R.S., Petro, A.E., Parekh, P. and Collins, S. (1997) Low plasma leptin in response to dietary fat in diabetes- and obesity-prone mice. Diabetes, 46, 1516–1520. Surwit, R.S., Wang, S., Petro, A.E., Sanchis, D., Raimbault, S., Ricquier, D., et al. (1998) Diet-induced changes in uncoupling proteins in obesity-prone and obesity-resistant strains of mice. PNAS, 95, 4061–4065. U.S. Department of Health and Human Services, NIH Publication 86–23, Guide for the Care and Use of Laboratory Animals. Waxman, M. and Stunkard, A.J. (1980) Caloric intake and expenditure of obese boys. Journal of Pediatrics, 96, 187–193. Wencel, H., Smothers, C., Opara, E.C., Kuhn, C.M., Feinglos, M.N. and Surwit, R.S. (1995) Impaired second phase insulin response of diabetes-prone C57BL/6J mouse islets. Physiological Behavior, 57, 1215–1220. West, D.B., Boozer, C.N., Moody, D.L. and Atkinson, R.L. (1992) Dietary obesity in nine inbred mouse strains. American Journal of Physiology, 262, R1025—R1032. Yen, T.T.T. and Acton, J.M. (1972) Locomotor activity of various types of genetically obese mice. Proceedings of the Society for Experimental Biology and Medicine, 140, 647–650.

INDEX

A/J mouse 343 adipocyte hyperplasia 347 high fat diet 346 low fat diet 346 plasma glucose 345 plasma insulin 345 acarbose 138 age related diabetogenicity 331–332 AGES (advanced glycation end products) 293 aging and retinopathy 275 AIDS and retinopathy 275 aldose reductase inhibition 9, 287, 290, 293 α-tocopherol 294, 295 amylin 118 ANS 72 antihypertensive drugs 155 apoptosis of β-cells 116 arterial occlusion and retinopathy 275 autoreactive cells beta cell killing 17 cytokines 17 generation 17–18 identity and specificity 16–17 origin of 15–16 phenotype 16

B6 (C57BL/6J) mouse 64–66, 343–355 adipocyte hyperplasia 347 glucose transport 349 high fat diet 344, 346 housing 351 low fat diet 344, 346 origin 64–65 plasma leptin concentration 350 source and care 349–353 B6-Lepob mouse adipose tissue cellularity 73–74 adipose tissue localization 73 adipose tissue size 73 atherosclerosis resistance 85–86 autonomous nervous system 72 behavioral characteristics 71–72 blood glucose 78 central effects 72 control 88 DHEA effect 8–87 experimental techniques 93–95 food intake 71–72 high ambient temperature requirement 71 immunodeficiency 87–88 insulin resistance 77–78 insulin secretion 76–77 313

314

INDEX

ketone metabolization 85 lipogenesis 75 lipolysis 75 pancreas 77 plasma insulin 75–76 reproductive characteristics 83–84 thermogenesis 74 BB rat 1–41 as a model for human autoimmune diabetes 28–29 cellular autoimmunity 11 cellular ontogeny 12–15 clinical features and pathology 6–10 complications 9 cytokines 17, 22 diet 23 generation of autoreactive cells 15–18 genetic basis of IDDM susceptibility 25–28 humoral immunity 10–11 immunomodulation 22–23 immunosuppression 21 infection 24 maintaining and using 5–6 metabolic abnormalities 14 prevention of spontaneous diabetes 21 regulatory cells 18–19 RTE 14 T cell 14 T cell maturation 13–14 thymocyte development 12–13 BB/Wor rat 2, 3 BBBA/Wor rat 4 BB-DP (diabetes prone)/Ed rat 4 BB-DR (diabetes resistant)/Ed rat 2, 4 induction of IDDM 24–25 prevention of induced diabetes 25 BBDR/Wor rat 4 BBVB/Wor rat 4 β-cell mass 116–117 regeneration 263–264 trophin 311 BKS (C57BLKS/J) mouse 64–66, 343 origin 64–65 plasma glucose 345 plasma insulin 345 BKS-Leprdb mouse adipose tissue 79 atherosclerosis resistance 85–86 autonomous nervous system 72 behavioral characteristics 71–72

blood glucose 80–81 body weight development 78–79 carbohydrate sensitivity 87 central effects 72 control 88–89 DHEA effect 86–87 experimental techniques 93–95 food intake 71–72 gestational diabetes 84–85 high ambient temperature requirement 71 immunodeficiency 87–88 insulin resistance 81 ketone metabolization 85 kidney pathology 85 lipogenesis 79 myocardial disease 83 nephropathy 82 neuropathy 82 plasma and pancreas insulin 81 reproductive characteristics 84 retinopathy 83 thermogenesis 79 vasculopathy 81 blood glucose 335, 336 blood pressure 146–148 C57BL/6J mouse, see B6 mouse C57BL/KSJ mouse, see BKS mouse candesartan cilexetil 167 capillary basement membrane thickness 283–284 cardiomyopathy 264 cardiovascular characteristics 146–148 cardiovascular disease 231–234 cataract formation, 339 ceramide metabolism 116 chromosome 3 55–56 ciglitazone 137 clinical trials 289–290 Coat’s disease and retinopathy 275 congenitally hyperglycemie rat, and cataracts 292 cytokines 17, 22 DCCT (diabetes control and complications trial) 286 degu rodent 292 diabetic angiolopathy 339 complications 9, 135–136, 197, 207, 320 at the cellular level 273 mechanisms underlying 286 genetics of 48–56

INDEX

neuropathy 339 retinopathy 275 early and late intervention 290–292 microangiopathies 288 prevention 287–289 diet 261 DP-BB rat 28 DR-BB/Wor rat 4 dyslipidemia 311 Eales disease and retinopathy 275 elastase retinal digest procedure 278–283 endocrine function 150–152 endocrine pancreas 151 environmental manipulation 23–24 extrapancreatic abnormalities and diabetic conditions 9 FFA (free fatty acids) 115–116, 173 fructose-1, 6-bisphosphatase 171 galactosemia 273 animal models 273–308 dog 287 galactosemic rat model 298–299 advantages 274–276 capillary basement membrane thickness 283–284 cost of diet 277 diet and care 276–277 elastase retinal digest procedure 278–283 retinal vascular whole mounts 284–286 tissue carbohydrate measurements 277–278 gastric mobility 120 genetics of diabetes 48–53 gestational diabetes 84–85 GFR (glomerular filtration rate) 165 GK (Goto-Kakizaki) rat 197–211 colonies 198 diabetic complications 207 genetics 199–200 glucose tolerance 198–199 glucose-induced insulin secretion 202–205 insulin response to nonglucose secretagogues 205 insulin sensitivity 206 islet and β cell function 202–205 micro-circulation 205–206 morphology and hormone content 200–202 structure and composition 200–202 isolation of pancreatic islets 202 macroangiopathy 207

microangiopathy 207 nephropathy 207 neuropathy 207 origin and breeding 197–199 osteopathy 207 pancreatic and islet hormone content 202 plasma insulin levels 198–199 practical considerations 198 retinopathy 207 gluconeogenesis 171 glucose intolerance 197 tolerance 1173 transporters 119 glycation and oxidation 292–297 glycogen synthase 171 HGP (hepatic glucose production) 161, 206, 337 husbandry 92–93 hyperglucagonemia 174 hyperglycemia 165, 172, 329 hyperinsulinemia 172, 311 hyperleptinemia 311 hyperlipidemia 148, 339 hyperphagia 71–72 hypertension 264, 344 hypertension and retinopathy 275 hypoglycemic drug testing 265 hypothalamic-pituitary-adrenal axis 150–151 IDDM (insulin-dependent diabetes mellitus) 1–41 iddml 26–27 iddm2 26 iddm3 27 iddm4 27 iddm5 27 IGF-1 (insulin-like growth factor) 311 inbreeding 197 incretin hormone 118 induced DR-BB rat 28 insulin resistance 77–78, 81, 118, 171, 197, 311 syndrome 318–320 secretion 76–77, 113–114, 171, 197 Intraepithelial lymphocytes 15 IRMA (intraretinal microvascular abnormalities) 297 IRS-1 (insulin-receptor substrate) 206 ischemic syndrome and retinopathy 275 islet transplant 320 JCR: LA-cp rat 227–245

315

316

INDEX

anesthesia 238 animal handling and drug administration 237–238 availability 242 cardiovascular disease 231–234 compared to other obese rat strains 234–235 diurnal cycle 238–239 historical background 227–228 hyperphagia and body weight 229 insulin and glucose metabolism 239–240 insulin resistance and hyperlipidemia 229–231 leptin receptors 229 maintenance of the strain 236–237 metabolic response to drugs 240–241 pathophysiology 229–234, 240–241 uses 241 kidney disease 148–149 KK mouse 129–142 BAT (brown adipose tissue) 135 chromosome mapping 136–137 derivation 129–130 histological changes 135–136 insulin resistance 131–134 strain difference 130–131 TZD (thiazolidinedione) 137, 138 WAT (white adipose tissue) 135 KKAy mouse 130 chromosome mapping 136–137 histological changes 135–136 insulin resistance 133–134 TZD (thiazolidinedione) 137 138 KRV (Kilham rat virus) 6 LCMV (lymphocytic choriomeningitis virus) 6 Lepob mouse, source colonies 89–90 Leprdb mouse breeding schemes 91–92 source colonies 90–91 leptin 63–107, 150 effect of administration 70–71 identification 68–71 receptor 63–107, 114, 229 identification 68–71 LETL (Long-Evans Tokushima Lean) rat 213 LFSAF (lipofuscin-specific autofluorescence) 294 lipid and lipoprotein 117 Macca mulatta 309 macroangiopathy 207 magnesium 122

mapped loci 192–193 metabolic defects in the mouse 67–68 metabolic syndrome X 143–158 metformin 122 MHC gene 53–55 MHC-linked susceptibility genes 54–55 microangiopathy 207, 273 mitochondrial enzymes 119–120 monoxidine 122 morphometry 284–286 mouse husbandry 92–93 mouse mutations, current nomenclature 63–66 moxonidine treatment 155–156 myocardiopathy 9, 83 NAG (N-acetyl-beta-D-glucoanimidase) 165 neovascularization 297–298 nephropathy 9, 58, 82, 120, 135, 160, 164–167, 193, 207, 264 neuropathy 9, 82, 136, 207 Nidd/gkl 199 Nidd/gk2 199 Niddlnsy 191–192 candidate gene 193 Nidd2nsy 192 Nidd3nsy 192 Nidc4nsy 192 NIDDM 112, 164, 171–183, 197, 218–219, 247, 309, 327, 343 complications of 264 Niddml 199 Niddm2 199 nitric oxide 114–115 NK cells 15 NK T cells 15 NMRI mouse, and STZ 24–250 NOD (nonobese diabetic) mouse 43–61 biomedical research using 58–59 characteristics 46 cogenic strains 48–53 environmental effects 47–48 maintenance 44–46 MHC genes 53–55 multifactoral diseases in 48 origin 43–44 NON (nonobese nondiabetic) mouse 43, 56–58 genetic background 58 origin 56–57 phenotypes 57–59

INDEX

renal lesions 58 n-STZ (neonatal streptozotocin) rat 247–271 advantages 250 β-cell function 250–257 β-cell regeneration 263–264 complications of NIDDM 264 diet 261 hypoglycemic drug testing 265 insulin action in 257–261 insulin secretion 250–251 intracellular mechanisms 251–254 obesity 261–262 pregnancy 262 reversible glucose incompetence 255–257 NSY (Nagoya-Shibata-Yasuda) mouse 185–195 candidate gene for Niddlnsy 193 growth curve 188–189 inheritance of diabetes 190–191 insulin action 190 insulin secretion 189–190 mapped loci 192–193 nephropathy 193 Niddlnsy 191–192 Nidd2nsy 192 Nidd3nsy 192 Nidd4nsy 192 origin 186–187 phenotypic characteristics 188–190 practical information 193–194 susceptibility for diabetes 191–193 type 2 diabetes incidence 188 type 2 diabetes pathogenesis 188–190 NZO (New Zealand Obese) mouse 171–183 FFA, 173 glucose tolerance 1173 hepatic insulin resistance 178–180 hyperglucagonemia 174 hyperglycemia 172 hyperinsulinemia 172 immune abnormalities 180–181 insulin action 175–180 insulin secretion 174–175 metabolic characteristics 172–174 muscle and fat insulin resistance 176–178 origins 171–172 obesity 145–146, 171, 261–262 mutations in the mouse 63–107 OB-R over-expression 115

317

OGTT (oral glucose tolerance test) 216 OLETF (Otsuka Long-Evans Tokushima Fatty) rat 213–225 breeding 214 genetics 215 NIDDM model 218–219 OGTT (oral glucose tolerance test) 216 pathophysiology 215–216 renal complications 216–218, 219–223 supply of 222–223 osteopathy 207 oxygen 297–298 pancreatic endothelium 8 pancreatic insulitis 7–8 pancreatic islet morphology 197 pentosidine 293 PGU (peripheral glucose utilization) 161 pioglitazone 137, 138, 162, 167 polyol pathway 274 PPARgamma 137 agonists 121 preadipocyte 137 pregnancy 262 primate species with diabetes 310 Psammomys obesus (sand rat) 327–324 age related diabetogenicity 331–332 availability 339 blood glucose 335, 336 breeding performance 332 diet 328–329 DP (diabetes prone) 327, 329–331 DR (diabetes resistant) 327, 329–331 environment 328–329 HGP (hepatic glucose production) 337 hyperglycemia 329 insulin resistance 334–337 insulin resistance mechanism 338 maintenance 329 metabolic efficiency 332–334 plasma insulin 330, 336 TGT (total glucose transport) 337 uses, 339 radiation retinopathy 275 rat eye diagram 279 enucleation 279–280 RCBM (retinal capillary basement membrane) thickness 296 renal glomeruli 166

318

INDEX

renal tubuli 166 reproductive function 151–152 retinal changes 119 retinal digestion 278 retinal microangiopathies 273 retinal microangiopathies, in rats 289 retinal vascular whole mounts 284–286 retinopathy 9, 83, 207 of prematurity 275 rhesus monkey 309–325 activity 315 ad libitum food intake 314 areas for research 318–320 blood sampling 317–318 body weight monitoring 314–315 diabetes research using 309–310 diabetic complications 320 environmental conditions 314 environmental risks 321 health monitoring 315 insulin resistance 313 insulin resistance syndrome 318–320 islet transplant 320 longitudinal characterization 320–321 natural history 311–312 obesity and diabetes 310–314 pharmacological agents 313–314 pitfalls 320–321 primate environmental enrichment 316–317 sedation and anesthesia 317 special care 315–316 suppliers 318 RT6 rat 19–21 immunomodulation 20–21 nomenclature 20 Secondary complications 81–83 Sherman rat, and STZ 247–249 SHROB rat 143–158 antihypertensive drugs 155 availability 156 blood pressure 146–148 cardiovascular characteristics 146–148 dietary manipulation 153 endocrine function 150–152 endocrine pancreas 151 experimental model 153–156 high salt diet 154 high sucrose diet 154–155

historical background 144–145 hyperlipidemia 148 hypothalamic-pituitary-adrenal axis 150–151 kidney disease 148–149 leptin 150 lifespan and cause of death 153 low calorie diet 153 morphology and anatomy 152 moxonidine treatment 155–156 obesity 145–146 pharmacotheraphy 155–156 phenotypic features 145–153 reproductive function 151–152 vascular disease 149–150 weight cycling 153–153 sickle cell retinopathy 275 skeletal bone deterioration 339 spontaneously diabetic BB rat 1–3 BioBreeding colony 1 genetic heterogeneity 1–3 lymphopenia 12 nomenclature 3 Sprague Dawley rat, and STZ 249 SSPG (steady state plasma glucose) 162 SSPI (steady state plasma insulin) 161 STZ (streptozotocin) 247 sugar-induced cataracts 292 Syndrome X 143 TGT (total glucose transport) 337 thymic epithelial defects 9–10 thyroiditis 8 tissue carbohydrate measurements 277–278 TNF-alpha (tumor necrosis factor-alpha) 163 transgenic mouse, cataracts 286, 293 troglitazone 137 type 1 diabetes see IDDM type 2 diabetes, see NIDDM vanadium 121–122 vascular changes 118–119 vascular disease 149–150 vasculogenesis 297–298 vasculopathy 81 venous occlusion and retinopathy 275 voglibose 138, 165, 167 Wistar fatty rat 159–169 compared to Zucher fatty rat 160

INDEX

establishment and utilization 159–160 GFR (glomerular filtration rate) 165 HGP (hepatic glucose production) 161 hyperglycemia 165 insulin resistance 161–164 NAG (N-acetyl-beta-D-glucoanimidase) 165 nephropathy 164–167 PGU (peripheral glucose utilization) 161 renal glomeruli 166 renal tubuli 166 SSPG (steady state plasma glucose) 162 SSPI (steady state plasma insulin) 161 Wistar Kyoto rat 159 Wistar rat, and STZ 247–249 wound healing and infections 120 ZDF (Zucker diabetic fatty) rat 109–128 abnormalities 17–120 amylin 118 apoptosis of -cells 116 background 110–111 β-cell mass 116–117 ceramide metabolism 116 development 111 FFA (free fatty acids) 115–116 gastric mobility 120 genetics 112 glucose transport 113, 119 incretin hormones 118 insulin levels 110 insulin resistance 118 insulin secretion 113–114 leptin receptor effects 114 lipids and lipoproteins 117 mitochondrial enzymes 119–120 nephropathy 120 neuropathy 120 nitric oxide effects 114–115 obese female 112 obese male 109 over-expression of OB-R 115 pancreatic abnormalities 112–117 phenotypic appearance 111–112 PPAR agonists 121 retinal changes 119 treatment 121–122 vascular changes 118–119 wound healing and infections 120 ZDFB rat 111

ZDT rat, blood glucose level 110

319